Cervical and Vaginal Cancer
BS. Nguyen Hong Anh
KEY POINTS
1 Human papillomavirus (HPV) infection is the causal agent of cervical cancer.
2 Screening programs are effective at decreasing the incidence of cervical cancer.
2446Vaccines help decrease the incidence of cervical cancer.
3 The most common histologic type of cervical cancer is squamous, and the relative
and absolute incidence of adenocarcinoma is increasing. Both histologies are caused
by HPV infection.
4 Cervical cancer is clinically staged, although modern radiographic modalities such as
computed tomography, magnetic resonance imaging, ultrasound, or positron
emission tomography, if available, may be beneficial for individual treatment
planning. The staging system has been updated in 2018.
5 Treatment of cervical cancer is based on stage of disease. Early-stage disease (stages
I to IIA) can be treated with either radical surgery or radiation therapy. Advanced
stage disease (stages IIB to IV) is best treated with chemoradiation, including
brachytherapy.
6 Vaginal cancer is a rare disease with many similarities to cervical cancer. Radiation
therapy is the mainstay of treatment for most patients; however, select patients may
be treated with radical surgery.
7 Minimally invasive radical hysterectomy. Laparoscopic and robotic radical
hysterectomies were being performed with larger degree of frequency in highly
selected patients world wide.
Cervical cancer ranks as the third most common gynecologic neoplasm in the
United States, behind cancer of the uterine corpus and ovary, mainly as a result of
the effectiveness of screening programs. Worldwide, cervical carcinoma
continues to be a significant health care problem. In developing countries,
where health care resources are limited, cervical carcinoma is the second
most frequent cause of cancer death in women. Because cervical cancer is
preventable, it is imperative that gynecologists and other primary health care
providers for women be familiar with vaccination programs, screening
techniques, diagnostic procedures, and risk factors for cervical cancer and
management of preinvasive disease. Vaginal cancer is a rare tumor that shares an
epidemiology and risk factor profile that is similar to cervical cancer.
CERVICAL CANCER
Epidemiology and Risk Factors
Invasive cancer of the cervix is considered a preventable disease because it
has a long preinvasive state, cervical cytology screening programs are
available, and the treatment of preinvasive lesions is effective. In spite of the
preventable nature of this disease, the CDC reported 12,578 new cases of invasive
cervical cancer resulting in 4,115 deaths in the United States in 2014 (1). This
2447data are similar to the 2011 statistics, with only a slight decrease (1,2). Nationally,
the lifetime probability of developing cervical cancer is 1:128. Although
screening programs in the United States are well established, it is estimated that
30% of cervical cancer cases will occur in women who have never had a
Papanicolaou (Pap) test. In developing countries, this percentage approaches 60%
(3). Nevertheless, the worldwide incidence of invasive disease is decreasing, and
cervical cancer is being diagnosed earlier, leading to better survival rates (2,4).
The mean age for cervical cancer in the United States is 47 years, and the
distribution of cases is bimodal, with peaks at 35 to 39 years and 60 to 64 years of
age (2).
There are numerous risk factors for cervical cancer: young age at first
intercourse (younger than 16 years), multiple sexual partners, cigarette
smoking, race, high parity, low socioeconomic status, and chronic immune
suppression. The relationship to oral contraceptive use was debated. Some
investigators proposed that use of oral contraceptives might increase the incidence
of cervical glandular abnormalities; however, this hypothesis was not consistently
supported (5,6). [1] Many of these risk factors are linked to sexual activity and
exposure to sexually transmitted diseases. Infection with the herpes virus was
thought to be the initiating event in cervical cancer; however, infection with
human papillomavirus (HPV) was determined to be the causal agent in the
development of cervical cancer, with herpes virus and Chlamydia trachomatis
likely acting as cofactors. The role of human immunodeficiency virus (HIV) in
cervical cancer is mediated through immune suppression (5). The Centers for
Disease Control and Prevention described cervical cancer as an acquired immune
deficiency syndrome (AIDS)–defining illness in patients infected with HIV (7).
[1] The initiating event in cervical dysplasia and carcinogenesis is infection
with HPV. HPV infection was detected in up to 99% of women with squamous
cervical carcinoma. HPV is the causative agent in both squamous and
adenocarcinoma of the cervix, but the respective tumors may have different
carcinogenic pathways (8). There are more than 100 different types of HPV, more
than 30 of which can affect the lower genital tract. There are 15 high-risk HPV
subtypes; two of the high-risk subtypes, 16 and 18, are found in up to 70% of
cervical carcinomas. The mechanism by which HPV affects cellular growth and
differentiation is through the interaction of viral E6 and E7 proteins with tumor
suppressor genes p53 and Rb, respectively. Inhibition of p53 prevents cell cycle
arrest and cellular apoptosis, which normally occurs when damaged DNA is
present, whereas inhibition of Rb disrupts transcription factor E2F, resulting in
unregulated cellular proliferation (9). Both steps are essential for the malignant
transformation of cervical epithelial cells. Initially, there were two HPV vaccines,
the quadrivalent Gardasil and the bivalent Cervarix, approved by the U.S. Food
2448and Drug Administration (FDA). Both protect against HPV subtypes 16 and 18.
[2] The efficacy of Gardasil was 97% to 100% for preventing cervical
intraepithelial neoplasia (CIN) grades 2 and 3 caused by HPV 16 or 18 in females
who were not previously infected with either HPV 16 or 18 before vaccination;
however, efficacy was only 44% in those who were infected prior to vaccination
(10). Recently, the FDA has approved a third HPV vaccine, Gardasil-9, which
protects against the four subtypes of the quadrivalent vaccine plus an additional
five high-risk subtypes (HPV 31/33/45/52/58). In a head-to-head,
immunoresponse trial, the 9-valent HPV vaccine showed non-inferior anti-HPV
6/11/16/18 antibody response (11). The efficacy of the 9-valent vaccine in
preventing CIN 2,3, VIN 2 or 3, and VAIN 2 or 3 was 97% in the HPV-naïve
patient population (12). Long-term data for superior efficacy in HPV
31/33/45/52/58-related clinical outcomes is still pending. The FDA also extended
the recommended age indication for Gardisil-9 to include men and women ages 9
to 45. Because these vaccines do not protect against all HPV subtypes, vaccinated
women need to continue to receive cervical cancer screening according to
published guidelines.
Evaluation
Vaginal bleeding is the most common symptom occurring in patients with
cancer of the cervix. Most often, this is postcoital bleeding, but it may occur as
irregular or postmenopausal bleeding. Patients with advanced disease may present
with a malodorous vaginal discharge, weight loss, or obstructive uropathy. In
asymptomatic women, cervical cancer is most commonly identified through
evaluation of abnormal cytologic screening tests. The false-negative rate for Pap
tests in the presence of invasive cancer is up to 50%, so a negative Pap test should
never be relied on in a symptomatic patient (13).
Initially, all women suspected of having cervical cancer should have a general
physical examination performed to include evaluation of the supraclavicular,
axillary, and inguinofemoral lymph nodes to exclude the presence of metastatic
disease. On pelvic examination, a speculum is inserted into the vagina, and the
cervix is inspected for suspicious areas (Fig. 38-1). The vaginal fornices should
be closely inspected. With invasive cancer, the cervix is usually firm and
expanded, and these features should be confirmed by digital examination. Rectal
examination is important to help establish cervical consistency and size,
particularly in patients with endocervical carcinomas. Rectal examination is the
only way to determine cervical size if the vaginal fornices have been obliterated
by menopausal changes or by the extension of disease. Parametrial extension of
disease is best determined by the finding of nodularity beyond the cervix on rectal
examination.
2449When obvious tumor growth is present, a cervical biopsy is usually
sufficient for diagnosis. If gross disease is not present, a colposcopic
examination with cervical biopsies and endocervical curettage is warranted.
If the diagnosis cannot be established conclusively with colposcopy and
directed biopsies, which may be the case with adenocarcinoma, cervical
conization may be necessary.
Colposcopic Findings of Invasion
Colposcopic examination is mandatory for patients with suspected early
invasive cancer based on cervical cytology and a grossly normal-appearing
cervix. Colposcopic findings that suggest invasion are (i) abnormal blood
vessels, (ii) irregular surface contour with loss of surface epithelium, and (iii)
color tone change. Colposcopically directed biopsies may permit the
diagnosis of frank invasion and thus avoid the need for diagnostic cone
biopsy, allowing treatment to be administered without delay. If there is debate
about the depth of invasion based on the cervical biopsies, and if the clinical stage
may be upstaged to stage IA2 or IB1, the patient should undergo a conization. In
the presence of a large cervical biopsy specimen showing invasion greater than 3
mm, or two biopsy specimens separated by 7 mm showing invasive cervical
carcinoma, therapy should proceed without delay, and the patient could undergo
radical surgery or radiation therapy.
2450FIGURE 38-1 Gross appearance of cervical cancer on examination.
Abnormal Blood Vessels
Abnormal vessels may be looped, branched, or reticular. Abnormal looped
vessels are the most common colposcopic finding and arise from the
punctated and mosaic vessels present in CIN. As the neoplastic growth process
proceeds and the need for oxygen and nutrition increases, angiogenesis occurs as
a result of tumor and local tissue production of vascular endothelial growth factor
(VEGF), platelet-derived growth factor (PDGF), epidermal growth factor (EGF),
and other cytokines, resulting in the proliferation of blood vessels and
neovascularization. Punctate vessels push out over the surface of the epithelium in
an erratic fashion, producing the looped, corkscrew, or J-shaped pattern of
abnormal vessels characteristic of invasive disease. Abnormal blood vessels arise
from the cervical stroma and are pushed to the surface as the underlying cancer
invades. The normally branching cervical stromal vessels are best observed over
nabothian cysts. In this area, the branches are generally at acute angles, with the
caliber of vessels becoming smaller after branching, much like the arborization of
a tree. The abnormal branching blood vessels seen with cancer tend to form
2451obtuse or right angles, with the caliber sometimes enlarging after branching.
Sharp turns, dilations, and luminal narrowing characterize these vessels. The
surface epithelium may be lost in these areas, leading to irregular surface contour
and friability.
Abnormal reticular vessels represent the terminal capillaries of the
cervical epithelium. Normal capillaries are best seen in postmenopausal women
with atrophic epithelium. When cancer involves this epithelium, the surface is
eroded, and the capillary network is exposed. These vessels are very fine and
short and appear as small comma-shaped vessels without an organized pattern.
They are not specific to invasive cancer; atrophic cervicitis may also have this
appearance.
Irregular Surface Contour
Abnormal surface patterns are observed as tumor growth proceeds. The surface
epithelium ulcerates as the cells lose intercellular cohesiveness secondary to loss
of desmosomes. Irregular contour may occur as a result of papillary
characteristics of the lesion. This finding can be confused with a benign HPV
papillary growth on the cervix. For that reason, biopsies should be
performed on all papillary cervical growths to avoid missing invasive disease.
Color Tone
Color tone may change as a result of increasing vascularity, surface epithelial
necrosis, and in some cases, production of keratin. The color tone is yelloworange rather than the expected pink of intact squamous epithelium or the red of
the endocervical epithelium.
Adenocarcinoma
Adenocarcinoma of the cervix does not have a specific colposcopic
appearance. All of the aforementioned blood vessels may be seen in these
lesions. Because adenocarcinomas tend to develop within the endocervix,
endocervical curettage is required as part of the colposcopic examination, and
traditional screening methods are less reliable (13).
Histologic Appearance of Invasion
Cervical conization is required to assess correctly the depth and the linear
extent of involvement when microinvasion is suspected. Early invasion is
characterized by a protrusion of malignant cells from the stromal–epithelial
junction. This focus consists of cells that appear better differentiated than the
adjacent noninvasive cells and have abundant pink-staining cytoplasm,
hyperchromatic nuclei, and small-to-medium–sized nucleoli (14). These early
invasive lesions form tonguelike processes without measurable volume and
2452are classified as International Federation of Gynecology and Obstetrics
(FIGO) stage IA1. With further progression, more tonguelike processes and
isolated malignant cells appear in the stroma, followed by a proliferation of
fibroblasts (desmoplasia) and a bandlike infiltration of chronic inflammatory cells
(Fig. 38-2). With increasing depth of invasion, lesions occur at multiple sites,
and the growth becomes measurable by depth and linear extent. Lesions that
are less than 3 mm in depth are classified as FIGO stage IA1. Lesions that
are 3 to 5 mm or more in depth and up to 7 mm in linear extent are classified
as FIGO stage IA2 (15). As the depth of stromal invasion increases, so does the
risk of capillary lymphatic space involvement. Dilated capillaries, lymphatic
spaces, and foreign-body multinucleated giant cells containing keratin debris are
often seen in the stroma.
The depth of invasion should be measured with a micrometer from the base of
the epithelium to the deepest point of invasion. Depth of invasion is a significant
predictor for the development of pelvic lymph node metastasis and tumor
recurrence. Although lesions that have invaded 3 mm or less rarely
metastasize, patients in whom lesions invade between 3 and 5 mm have
positive pelvic lymph nodes in 3% to 8% of cases (16). The significance of the
cutoff at 3 mm is not identified completely; it was postulated that small capillary–
lymphatic spaces at this level are incapable of facilitating the transport of
malignant cells. Uneven shrinkage of tissue by fixative often creates space
between the tumor nests and the surrounding fibrous stroma, simulating vascular
lymphatic invasion (Fig. 38-2). Therefore, suspected vascular–lymphatic
involvement with invasion of less than 3 mm should be interpreted with care. A
lack of endothelial lining indicates that the space is a fixation artifact rather than
true vascular invasion.
2453FIGURE 38-2 Microinvasive squamous carcinoma. Multiple irregular tonguelike
processes and isolated nests of malignant cells are seen, some surrounded by clear spaces,
simulating capillary lymphatic invasion. This is an artifact caused by tissue shrinkage. The
depth of stromal invasion is measured from the basement membrane of the overlying
cervical intraepithelial neoplasia (CIN). In this case, it is 1.2 mm.
Staging
The FIGO staging system was updated in 2018 to include imaging and pathologic
findings and is presnted in Table 38-1 and Figure 38-3. The old 2008 FIGO
clinical staging system is shown in Table 38-2. The FIGO staging system is
applicable to all histologic types of cervical cancer. The reported data presented in
this chapter are based on the 2008 staging system.
When there is doubt concerning the stage to which a cancer should be
allocated, the earlier stage should be selected. After a stage is assigned and
treatment is initiated, the stage must not be changed because of subsequent
findings. The upstaging of patients during treatment will produce an erroneous
perception of improvement in the results of treatment of low-stage disease.
Following is a breakdown of the incidence of cervical cancer by the clinical
2454staging system at diagnosis: 38%, stage I; 32%, stage II; 26%, stage III; and 4%,
stage IV (4,16,17).
In the 2018 staging system, horizontal spread is no longer considered in
stage IA, only the depth of invasion is used. Stage IB is now divided into 3
stages, Stages IB1, IB2, and IB3. Stage IB1 denotes lesions that are <2 cm in
greatest dimension, reflecting advances fertility-sparing procedures that are
now recommended for selected patients with these smaller tumors. Last,
Stage III has added a stage with substages to reflect pelvic and paraaortic
lymph node metastasis; this is due to the worsening of outcomes when lymph
node metastasis is present. See Table 38-1 for the revised 2018 FIGO staging
system.
Table 38-1 FIGO Staging of Carcinoma of the Cervix Uteri (2018)
Stage Description
I The carcinoma is strictly confined to the cervix (extension to the uterine
corpus should be disregarded)
IA Invasive carcinoma that can be diagnosed only by microscopy, with
maximum depth of invasion <5 mma
IA1 Measured stromal invasion <3 mm in depth
IA2 Measured stromal invasion ≥3 mm and <5 mm in depth
IB Invasive carcinoma with measured deepest invasion ≥5 mm (greater than
Stage IA), lesion limited to the cervix uterib
IB1 Invasive carcinoma ≥5 mm depth of stromal invasion, and <2 cm in
greatest dimension
IB2 Invasive carcinoma ≥2 cm and <4 cm in greatest dimension
IB3 Invasive carcinoma ≥4 cm in greatest dimension
II The carcinoma invades beyond the uterus, but has not extended onto the
lower third of the vagina or to the pelvic wall
IIA Involvement limited to the upper two-thirds of the vagina without
parametrial involvement
IIA1 Invasive carcinoma <4 cm in greatest dimension
IIA2 Invasive carcinoma ≥4 cm in greatest dimension
2455IIB With parametrial involvement but not up to the pelvic wall
III The carcinoma involves the lower third of the vagina and/or extends to the
pelvic wall and/or causes hydronephrosis or nonfunctioning kidney and/or
involves pelvic and/or para-aortic lymph nodesc
IIIA The carcinoma involves the lower third of the vagina, with no extension to
the pelvic wall
IIIB Extension to the pelvic wall and/or hydronephrosis or nonfunctioning
kidney (unless known to be due to another cause)
IIIC Involvement of pelvic and/or para-aortic lymph nodes, irrespective of
tumor size and extent (with r and p notations)c
IIIC1 Pelvic lymph node metastasis only
IIIC2 Para-aortic lymph node metastasis
IV The carcinoma has extended beyond the true pelvis or has involved
(biopsy proven) the mucosa of the bladder or rectum. (A bullous edema, as
such, does not permit a case to be allotted to Stage IV)
IVA Spread to adjacent pelvic organs
IVB Spread to distant organs
When in doubt, the lower staging should be assigned.
aImaging and pathology can be used, where available, to supplement clinical findings with
respect to tumor size and extent, in all stages.
b
The involvement of vascular/lymphatic spaces does not change the staging. The lateral
extent of the lesion is no longer considered.
cAdding notation of r (imaging) and p (pathology) to indicate the findings that are used to
allocate the case to Stage IIIC. Example: If imaging indicates pelvic lymph node
metastasis, the stage allocation would be Stage IIIC1r, and if confirmed by pathologic
findings, it would be Stage IIIC1p. The type of imaging modality or pathology technique
used should always be documented.
From Bhatla N, Berek JS, Cuello Frede M, et al. FIGO Committee Report. Revised
FIGO staging for carcinoma of the cervix uteri. Int J Gynecol Obstet 2019:1–7.
Additional Staging Modalities
2456Various investigators used lymphangiography, computed tomography (CT),
ultrasonography, magnetic resonance imaging (MRI), and positron emission
tomography (PET) in an attempt to improve the accuracy of clinical staging
(18–28). These modalities suffer from poor sensitivity and high false-negative
rates. Evaluation of the para-aortic lymph nodes with lymphangiography is
associated with a false-positive rate of 20% to 40% and a false-negative rate of
10% to 20% (18–20). Overall, lymphangiography has a sensitivity of 79% and
specificity of 73% (23). CT has poor sensitivity (34%) but excellent specificity
(97%) (24). The accuracy of CT scanning is 80% to 85%; the false-negative rate
is 10% to 15%, and the false-positive rate is 20% to 25% (19–21). Ultrasound has
a high false-negative rate (30%), low sensitivity (19%), but high specificity (99%)
(22). Early data showed that MRI results were comparable to those of CT
scanning, a finding confirmed on meta-analysis (24,25). A systematic review
comparing CT scan with MRI showed that MRI is significantly more
sensitive with equivalent specificity. MRI has excellent sensitivity on T2-
weighted images for the detection of parametrial disease (26).
PET scans are increasingly being utilized alone or in conjunction with CT or
MRI to detect metastatic disease. Studies suggest that PET may be more useful
than other imaging techniques for the detection of extrapelvic disease, with
comparable or better sensitivity and specificity (27). A meta-analysis of 72
studies found PET (75%, 98%) to have superior sensitivity and specificity
compared with CT (58%, 92%) and MRI (56%, 93%) for the detection of lymph
node metastases (28). PET scans may be better predictors of eventual treatment
outcome. Although these studies show promise for the use of PET scans in
detecting metastases, its sensitivity in detecting pelvic nodal disease may be
limited (29). Integrated PET/CT imaging may prove more sensitive than PET
alone, especially for the detection of pelvic adenopathy.
2457FIGURE 38-3 FIGO Cervical Cancer Staging (2018). (Adapted from Benedet J, Odicino
F, Maisonneuve P, Beller U, Creasman W, Heintz A, Ngan H, and Pecorelli S.
Carcinoma of the cervix uteri. International Journal of Gynecology & Obstetrics.
2003;83:41–78.)
When abnormalities are noted on CT, MRI, or PET, radiographic-guided
fine-needle aspirations (FNA) can be performed to confirm metastatic
disease and individualize treatment planning. [4] The FIGO 2018 staging
system incorporates imaging and pathologic measurements in determining
the size and extent of the primary tumor and the presence of pelvic and para-
2458aortic metastasis.
The accuracy of clinical staging is limited, and surgical evaluation,
although not practical or feasible in all patients, can more accurately identify
metastatic disease (30).
Pathology
Squamous Cell Carcinoma
Invasive squamous cell carcinoma is the most common variety of invasive
cancer in the cervix. [3] Histologically, variants of squamous cell carcinoma
include large cell keratinizing, large cell nonkeratinizing, and small cell types
(31). Large cell keratinizing tumors consist of tumor cells forming irregular
infiltrative nests with laminated keratin pearls in the center. Large cell
nonkeratinizing carcinomas reveal individual cell keratinization but do not form
keratin pearls (Fig. 38-4). The category of small-cell carcinoma includes poorly
differentiated squamous cell carcinoma and small-cell anaplastic carcinoma. If
possible, these two tumors should be differentiated. The former contains cells that
have small-to-medium-sized nuclei and more abundant cytoplasm than those of
the latter. The designation of small-cell anaplastic carcinoma should be reserved
for lesions resembling oat cell carcinoma of the lung. Small-cell anaplastic
carcinoma infiltrates diffusely and consists of tumor cells that have scanty
cytoplasm, round to oval small nuclei, coarsely granular chromatin, and high
mitotic activity. The nucleoli are absent or small. Immunohistochemistry or
electron microscopy can differentiate the small-cell neuroendocrine tumors.
Patients with the large cell type of carcinoma, with or without keratinization, have
a better prognosis than those with the small-cell variant. Small-cell anaplastic
carcinomas behave more aggressively than poorly differentiated squamous
carcinomas that contain small cells. Infiltration of parametrial tissue and pelvic
lymph node metastasis affect the prognosis.
Table 38-2 FIGO Clinical Staging of Carcinoma of the Cervix Uteri (2008)
Stage I The carcinoma is strictly confined to the cervix (extension to the corpus
would be disregarded)
IA Invasive carcinoma which can be diagnosed only by microscopy, with
deepest invasion ≤5 mm and largest extension ≤7 mm
IA1 Measured stromal invasion of ≤3.0 mm in depth and extension of ≤7.0 mm
IA2 Measured stromal invasion of >3.0 mm and not >5.0 mm with an extension
2459of not >7.0 mm
IB Clinically visible lesions limited to the cervix uteri or preclinical cancers
greater than stage IAa
IB1 Clinically visible lesion ≤4.0 cm in greatest dimension
IB2 Clinically visible lesion >4.0 cm in greatest dimension
Stage II Cervical carcinoma invades beyond the uterus, but not to the pelvic wall or
to the lower third of the vagina
IIA Without parametrial invasion
IIA1 Clinically visible lesion ≤4.0 cm in greatest dimension
IIA2 Clinically visible lesion >4 cm in greatest dimension
IIB With obvious parametrial invasion
Stage III The tumor extends to the pelvic wall and/or involves lower third of the
vagina and/or causes hydronephrosis or nonfunctioning kidneyb
IIIA Tumor involves lower third of the vagina, with no extension to the pelvic
wall
IIIB Extension to the pelvic wall and/or hydronephrosis or nonfunctioning
kidney
Stage IV The carcinoma has extended beyond the true pelvis or has involved (biopsy
proven) the mucosa of the bladder or rectum. A bullous edema, as such,
does not permit a case to be allotted to stage IV
IVA Spread of the growth to adjacent organs
IVB Spread to distant organs
aAll macroscopically visible lesions, even those with superficial invasion, are allotted to
stage IB carcinomas. Invasion is limited to a measured stroma invasion with a maximal
depth of 5.0 mm and a horizontal extension greater than 7.0 mm. Depth of invasion should
not be greater than 5 mm taken from the base of the epithelium of the original tissue
squamous or glandular. The depth of invasion should always be reported in millimeters,
even those cases with “early minimal stromal invasion” (∼ 1 mm). The involvement of
vascular/lymphatic spaces should not change stage allotment.
b
On rectal examination, there is no cancer-free space between the tumor and the pelvic
2460wall. All cases with hydronephrosis or nonfunctioning kidney are included, unless they are
known to be due to another cause.
FIGO Committee on Gynecologic Oncology. Revised FIGO staging for carcinoma of the
vulva, cervix, and endometrium. Int J Obstet Gynecol 2009;105:103–104.
FIGURE 38-4 Invasive squamous cell carcinoma, large cell nonkeratinizing type. Tumor
cells form irregular nests and have abundant eosinophilic cytoplasm and distinct cell
borders indicative of squamous differentiation.
Other less common variants of squamous carcinoma include verrucous
carcinoma and papillary (transitional) carcinoma. Verrucous carcinomas may
resemble giant condyloma acuminatum, are locally invasive, and rarely
metastasize. Papillary carcinomas histologically resemble transitional cells of the
bladder and may have more typical squamous cell invasion at the base of the
lesion. Papillary carcinomas behave and are treated in a manner similar to
traditional squamous cell cancers, except that late recurrences were noted.
Adenocarcinoma
2461There are an increasing number of cervical adenocarcinomas reported in
women in their 20s and 30s. [3] Although the total number of cases of
adenocarcinoma is relatively stable, this disease is appearing more frequently
in young women, especially as the number of cases of invasive squamous cell
carcinoma decreases. Older reports indicated that 5% of all cervical cancers
were adenocarcinomas, whereas newer reports show a proportion as high as
18.5% to 27% (32–34). Much of this proportional increase is related to a
decreasing incidence of squamous carcinoma secondary to screening programs
(which are less accurate at identifying preinvasion adenocarcinoma), greater
exposure to oral contraceptives, and a greater exposure to HPV (5,6).
Adenocarcinoma in situ (AIS) is believed to be the precursor of invasive
adenocarcinoma, and it is not surprising that the two often coexist (35). In
addition to AIS, intraepithelial or invasive squamous neoplasia occurs in
30% to 50% of cervical adenocarcinomas (36). A squamous intraepithelial
lesion may be observed colposcopically on the ectocervix, and the coexistent
adenocarcinoma often is higher in the cervical canal.
Patients with AIS who are treated with conization should undergo close
clinical follow-up. Endocervical curettage, often used in surveillance, may miss
residual or invasive disease, and false-negative rates as high as 50% were reported
(37). In addition, skip lesions not resected at the time of conization may be
present. For these reasons, hysterectomy should be considered the standard
therapy for patients who have completed their childbearing. In two reports,
patients with negative cone biopsy margins were followed conservatively, with
few requiring repeat surgical procedures (38,39). Because cervical AIS tends to
affect women during their reproductive years, a thorough discussion of risks and
benefits should take place, and treatment should be individualized.
Adenocarcinoma of the cervix is managed in the same manner used for
squamous cell carcinoma. Adenocarcinoma was believed to be associated with a
worse prognosis and outcome when compared with squamous cell carcinoma. A
study of 203 women with adenocarcinoma and 756 women with squamous
carcinoma supported this assertion (33). This study showed 5-year survival rates
of 90% versus 60%, 62% versus 47%, and 36% versus 8% for stages I, II, and III,
respectively. Although some attributed these rates to a relative resistance to
radiation, they are more likely a reflection of the tendency of adenocarcinomas to
grow endophytically and to be undetected until a large volume of tumor is
present. Adenocarcinomas may be detected by cervical sampling, but less reliably
so than squamous carcinomas. A definitive diagnosis often requires cervical
conization. When adjusted for tumor size, it appears that there is no difference in
prognosis between the two histologic subtypes. In a Gynecologic Oncology
Group (GOG) review of patients treated with concurrent cisplatin chemotherapy
2462with radiotherapy, survival was equivalent among squamous, adenocarcinoma,
and adenosquamous carcinoma subgroups (40).
The clinical features of stage I adenocarcinomas are well studied (33,41,42).
These studies identified size of tumor, depth of invasion, grade of tumor, and age
of the patient as significant correlates of lymph node metastasis and survival.
When matched with squamous carcinomas for lesion size, age, and depth of
invasion, the incidence of lymph node metastases and the survival rate appear to
be the same (41,42). Patients with stage I adenocarcinomas can be selected for
treatment according to the same criteria as for those with squamous cancers
(42).
The choice of treatment for bulky stages I and II tumors is controversial. Some
advocated treatment with radiation alone, whereas others support radiation plus
extrafascial hysterectomy (43–45). In 1975, Rutledge et al. reported an 85.2% 5-
year survival rate for all patients with stage I disease treated with radiation alone
and an 83.8% survival rate for those who had radiation plus surgery (44). The
central persistent disease rate was 8.3%, compared with 4% for those who had
radiation plus surgery. In stage II disease, the 5-year survival rate was 41.9% for
radiation alone and 53.7% for radiation plus surgery. A subsequent report
revealed no significant difference in survival among patients treated with
radiation alone or radiation plus extrafascial hysterectomy (46).
Invasive adenocarcinoma may be pure (Fig. 38-5A,B) or mixed with squamous
cell carcinoma. Within the category of pure adenocarcinoma, the tumors are quite
heterogeneous, with a wide range of cell types, growth patterns, and
differentiation (33). About 80% of cervical adenocarcinomas consist
predominantly of the endocervical type cells with mucin production. The
remaining tumors are populated by endometrioid cells, clear cells, intestinal
cells, or a mixture of more than one cell type. By histologic examination
alone, some of these tumors are indistinguishable from those arising
elsewhere in the endom etrium or ovary. Within each cell type, the growth
patterns and nuclear abnormalities vary according to the degree of differentiation.
In well-differentiated tumors, tall columnar cells line the well-formed branching
glands and papillary structures, whereas pleomorphic cells tend to form irregular
nests and solid sheets in poorly differentiated neoplasms. The latter may require
mucicarmine and periodic acid–Schiff (PAS) staining to confirm their glandular
differentiation.
In a retrospective review of GOG trials of chemoradiation for cervix
carcinoma, treatment with radiation alone was associated with worse overall
survival (OS) in adenocarcinoma and adenosquamous carcinoma when compared
with squamous carcinoma. In contrast, treatment with cisplatin-based
chemoradiation was associated with similar progression-free and OS between the
2463adenocarcinoma and squamous groups (40).
24642465FIGURE 38-5 Invasive adenocarcinoma of the cervix, well differentiated. A: Irregular
glands are lined with tall columnar cells with vacuolated mucinous cytoplasm resembling
endocervical cells. B: Nuclear stratification, mild nuclear atypism, and mitotic figures are
evident in higher power.
There are several special variants of adenocarcinoma. Minimal deviation
adenocarcinoma (adenoma malignum) is an extremely well-differentiated form
of adenocarcinoma in which the branching glandular pattern strongly simulates
that of the normal endocervical glands. The lining cells have abundant mucinous
cytoplasm and uniform nuclei (47,48). Because of this, the tumor may not be
recognized as malignant in small biopsy specimens, thereby causing considerable
delay in diagnosis. Special immunohistochemical staining may be required to
establish the diagnosis. Earlier studies reported a dismal outcome for women with
this tumor, but other studies found a favorable prognosis if the disease is detected
early (49). Although rare, similar tumors were reported in association with
endometrioid, clear, and mesonephric cell types (50).
An entity described as villoglandular papillary adenocarcinoma deserves
special attention (51). It primarily affects young women, some of whom are
pregnant or users of oral contraceptives. Histologically, the tumors have smooth,
well-defined borders, are well differentiated, and are either in situ or superficially
invasive. Follow-up information is encouraging. None of these tumors recurred
after cervical conization or hysterectomy, and no metastasis was detected among
women undergoing pelvic lymphadenectomy. This tumor appears to have limited
risk for spread beyond the uterus.
Adenosquamous Carcinoma
Carcinomas with a mixture of malignant glandular and squamous components are
known as adenosquamous carcinomas. Patients with adenosquamous carcinoma
of the cervix were reported to have a poorer prognosis than those with pure
adenocarcinoma or squamous carcinoma but a review of patients treated with
cisplatin-based chemoradiation has shown no difference in survival among the
three groups (52).
In mature adenosquamous carcinomas, the glandular and squamous carcinomas
are readily identified on routine histologic evaluation and do not cause diagnostic
problems. In poorly differentiated or immature adenosquamous carcinomas,
glandular differentiation can be appreciated only with special stains, such as
mucicarmine and PAS. In one study, 30% of squamous cell carcinomas
demonstrated mucin secretion when stained with mucicarmine (50). These
squamous cell carcinomas with mucin secretion have a higher incidence of pelvic
lymph node metastases than do squamous cell carcinomas without mucin
2466secretion, and they are similar to the signet-ring variant of adenosquamous
carcinoma (50,53).
Glassy cell carcinoma is recognized as a poorly differentiated form of
adenosquamous carcinoma (53). Individual cells have abundant eosinophilic,
granular, ground-glass cytoplasm, large round to oval nuclei, and prominent
nucleoli. The stroma is infiltrated by numerous lymphocytes, plasma cells, and
eosinophils. Approximately half of these tumors contain glandular structures or
stain positive for mucin. The poor diagnosis of this tumor is linked to
understaging and resistance to radiotherapy.
Other variants of adenosquamous carcinoma include adenoid basal carcinoma
and adenoid cystic carcinoma. Adenoid basal carcinoma simulates the basal cell
carcinoma of the skin (54). Nests of basaloid cells extend from the surface
epithelium deep into the underlying tissue. Cells at the periphery of tumor nests
form a distinct parallel nuclear arrangement, so-called peripheral palisading. An
“adenoid” pattern occasionally develops, with “hollowed-out” nests of cells.
Mitoses are rare, and the tumor often extends deep into the cervical stroma.
Adenoid cystic carcinoma of the cervix behaves much like such lesions
elsewhere in the body. The tumors tend to invade into the adjacent tissues and
metastasize late, often 8 to 10 years after the primary tumor was removed. Like
other adenoid cystic tumors, they may metastasize directly to the lung. The
pattern simulates that of the adenoid basal tumor, but there is a cystic component,
and the glands of the cervix are involved (54). Mitoses may be seen but are not
numerous.
Sarcoma
The most important sarcoma of the cervix is embryonal rhabdomyosarcoma,
which occurs in children and young adults. The tumor has grapelike polypoid
nodules, known as botryoid sarcoma, and the diagnosis depends on the
recognition of rhabdomyoblasts. Leiomyosarcomas and mixed mesodermal
tumors involving the cervix may be primary but are more likely to be secondary
to uterine tumors. Cervical adenosarcoma is described as a low-grade tumor
with a good prognosis (55).
Malignant Melanoma
On rare occasions, melanosis is seen in the cervix. Malignant melanoma may
arise de novo in this area. Histopathologically, it simulates melanoma elsewhere,
and the prognosis depends on the depth of invasion into the cervical stroma.
Neuroendocrine Carcinoma
The classification of neuroendocrine cervical carcinoma includes four histologic
2467subtypes: (i) small cell, (ii) large cell, (iii) classical carcinoid, and (iv) atypical
carcinoid (56). Neuroendocrine tumors of the cervix are rare, and treatment
regimens are based on small case series of patients.
Small-cell (neuroendocrine type) carcinoma of the cervix is aggressive in
nature and is similar to cancer arising from the bronchus (57). The hallmark
of neuroendocrine tumors is their aggressive malignant behavior with the
propensity to metastasize. At the time of diagnosis, it is usually disseminated,
with bone, brain, liver, and bone marrow being the most common sites of
metastases. In one study of 11 patients with disease apparently confined to the
cervix, a high rate of lymph node metastasis was noted (58). Pathologically, the
diagnosis is aided by the finding of neuroendocrine granules on electron
microscopy and by immunoperoxidase studies that are positive for a variety of
neuroendocrine proteins such as calcitonin, insulin, glucagon, somatostatin,
gastrin, and adrenocorticotropic hormone (ACTH). In addition to the traditional
staging for cancer of the cervix, these patients should undergo bone, liver, and
brain scanning, bone marrow aspiration, and biopsy to evaluate the possibility of
metastatic disease. Therapy consists of surgery, chemotherapy, and radiation.
Because patients with early-stage disease have distant metastases, multimodal
therapy is recommended. The main active chemotherapeutic agent is etoposide.
Local therapy alone gives almost no chance of cure of small-cell
carcinoma. Regimens of combination chemotherapy improved the median
survival rates in small-cell bronchogenic carcinoma, and these regimens are
used for treatment of small-cell carcinoma of the cervix. Combination
chemotherapy may consist of vincristine, doxorubicin, and cyclophosphamide
(VAC) or VP-16 (etoposide) and cisplatin (EP) (59). Patients must be monitored
carefully because they are at high risk for developing recurrent metastatic disease
(60).
Patterns of Spread
Cancer of the cervix spreads by (i) direct invasion into the cervical stroma,
corpus, vagina, and parametrium; (ii) lymphatic metastasis; (iii) blood-borne
metastasis; and (iv) intraperitoneal implantation. The incidence of pelvic and
para-aortic nodal metastasis is shown in Table 38-3.
The cervix is commonly involved in cancer of the endometrium and
vagina. The latter is rare, and most lesions that involve the cervix and vagina
are designated cervical primaries. Consequently, the clinical classification is
that of cervical neoplasia extending to the vagina, rather than vice versa.
Endometrial cancer may extend into the cervix by three modes: direct extension
from the endometrium, submucosal involvement by lymph vascular extension,
and multifocal disease. The latter is most unusual, but occasionally a focus of
2468adenocarcinoma may be seen in the cervix, separate from the endometrium. This
lesion should not be diagnosed as metastasis but rather as multifocal disease.
Malignancies involving the peritoneal cavity (e.g., ovarian cancer) may be found
in the cul-de-sac and extend directly into the vagina and cervix. Carcinomas of
the urinary bladder and colon occasionally extend into the cervix. Cervical
involvement by lymphoma, leukemia, and carcinoma of the breast, stomach, and
kidney is usually part of the systemic pattern of spread for these malignancies.
Isolated metastasis to the cervix in such cases may be the first sign of a primary
tumor elsewhere in the body.
Table 38-3 Incidence of Pelvic and Para-Aortic Lymph Node Metastasis by Stage
Based on FIGO Clinical Staging (2008)
Stage No. of
Patients
Positive Pelvic Nodes
(%)
Positive Para-Aortic Nodes
(%)
IA1 (≤3 mm) 179a 0.5 0
IA2 (>3–5
mm)
84a 4.8 <1
IB 1,926b 15.9 2.2
IIA 110c 24.5 11
IIB 324c 31.4 19
III 125c 44.8 30
IVA 23c 55 40
aReferences: (77,111,118,121,122,170).
b
References: (17,77,79,94,95,98–102,189).
cReferences: (17,18,95,98,99,103,150).
Treatment Options
The treatment of cervical cancer is similar to the treatment of any other type of
malignancy in that both the primary lesion and potential sites of spread should be
evaluated and treated. The therapeutic modalities for achieving this goal include
primary treatment with surgery, radiotherapy, chemotherapy, or [5]
2469chemoradiation. While radiation therapy can be used in all stages of disease,
surgery is limited to patients with stages I to IIA disease. The 5-year survival
rate for stage I cancer of the cervix is approximately 85% with either
radiation therapy or radical hysterectomy. A study using the National Cancer
Institute’s Surveillance Epidemiology and End Results data by an intent-to-treat
analysis showed that patients in the surgery arm had an improved survival when
compared with patients in the radiation arm (61). Optimal therapy consists of
radiation, or surgery alone, to limit the increased morbidity that occurs when the
two treatment modalities are combined. Recent improvements in the treatment of
cervical carcinoma include adjuvant chemoradiation in patients discovered to
have high-risk cervical carcinoma after radical hysterectomy and in patients with
locally advanced cervical carcinoma.
Surgery
There are advantages to the use of surgery instead of radiotherapy,
particularly in younger women for whom conservation of the ovaries is
important. Chronic bladder and bowel problems that require medical or surgical
intervention occur in up to 8% of patients undergoing radiation therapy (62). Such
problems are difficult to treat because they result from fibrosis and decreased
vascularity. This is in contrast to surgical injuries, which usually can be repaired
without long-term complications. Sexual dysfunction is less likely to occur after
surgical therapy than radiation, because of vaginal shortening, fibrosis, and
atrophy of the epithelium associated with radiation. Surgical therapy shortens the
vagina, but gradual lengthening can be brought about by sexual activity. The
epithelium does not become atrophic because it responds either to endogenous
estrogen or to exogenous estrogens if the patient is postmenopausal.
Radical hysterectomy is reserved for women who are in good physical
condition. Advanced chronologic age should not be a deterrent. With
improvements in anesthesia, elderly patients withstand radical surgery almost as
well as their younger counterparts (63). It is prudent not to operate on lesions
that are larger than 4 cm in diameter because these patients will require
postoperative radiation therapy. When selected in this manner, the urinary
fistula rate is less than 2%, and the operative mortality rate is less than 1%
(64,65). A summary of the management of cervical cancer is presented in Table
38-4.
If radiation therapy is needed, transposing the ovaries out of the planned
radiation field may preserve ovarian function. Although transposition
provides some protection, studies suggest that normal ovarian function is
preserved in fewer than 50% of patients (66,67). Metastasis to the ovaries
occurs in 0.9% of cases of early-stage cervical cancer, so preservation of the
2470ovaries, particularly with adenocarcinoma, may confer a small recurrence risk
(68).
Cone Biopsy of the Cervix
Cone biopsy of the cervix serves a diagnostic and therapeutic role in cervical
cancer. The procedure is indicated to confirm the diagnosis of cancer, and to
definitively treat stage IA1 disease when preservation of fertility is desired. For
effective treatment, there must be no evidence of lymph–vascular space invasion,
and both endocervical margins and curettage findings must be negative for cancer
or dysplasia. Because stage IA1 cancers have less than a 1% risk of lymph node
metastasis, lymphadenectomy is not necessary. If the endocervical margin or
curettage is positive for dysplasia or malignancy, further treatment is necessary
because these findings are strong predictors of residual disease. For squamous cell
carcinoma, the risk of residual disease is 4% if both the endocervical margin and
curettage are negative for dysplasia or malignancy, 22% if the endocervical
margin alone is positive, and 33% if both are positive (69). In cases of AIS, the
status of the cone margins is particularly important, with residual preinvasive and
invasive disease noted in up to 25% and 3%, respectively, of cases with negative
margins, and up to 80% and 7%, respectively, in cases with positive margins
(70,71).
Table 38-4 Management of Invasive Cancer of the Cervix by Stage
Stage Disease Treatmenta
IA1 <3 mm - LVSI Conization or Extrafascial Hyst
<3 mm + LVSI Modified Rad Trachel or Modified Rad Hyst + pelvic
lymph or SLN
IA2 ≥3 mm <5 mm Modified Rad Trachel or Modified Rad Hyst + pelvic
lymph or SLN
IB1 ≥5 mm <2 cm Mod Rad/Rad Trachel or Mod Rad/Rad Hyst +
pelvic lymph or SLN
IB2 ≥2 cm <4 cm Rad Hyst + pelvic lymph
IB3 ≥4 cm Chemoradiation, pelvic field
IIA1 <4 cm + upper vagina Rad Hyst + pelvic lymph or chemoradiation
IIA2 ≥4 cm + upper vagina Chemoradiation, pelvic field
2471IIB + Parametria not to
pelvic wall
Chemoradiation, pelvic field
IIIA + Lower vagina Chemoradiation, pelvic field
IIIB + Pelvic wall or
hydronephrosis
Chemoradiation, pelvic ± extended field
IIIC1 + Pelvic lymph nodes Chemoradiation, pelvic ± extended field
IIIC2 + Para-aortic lymph
nodes
Chemoradiation, pelvic + extended field + systemic
chemotherapy
IVA + Adjacent pelvic
organs
Chemoradiation, pelvic + extended field or pelvic
exenteration
IVB + Distant organs Systemic chemotherapy ± radiation, pelvic or
modified field
LVSI, lymphvascular space invasion; Mod, modified; Rad, radical; Trachel,
trachelectomy; Hyst, hysterectomy; lymph, lymphadenectomy; SLN, sentinel lymph node
biopsy in selected cases.
aTreatment recommendations must be individualized based on the patient’s status and
specific disease variables.
Simple (Extrafascial) Hysterectomy
Type I hysterectomy is an appropriate therapy for patients with stage IA1
tumors without lymph–vascular space invasion who are not desirous of
future fertility. In such cases, lymphadenectomy is not recommended. If
lymph-vascular space invasion is found, a type II modified radical
hystectormy with pelvic sentinel lymph node biospy or lymphandectomy is
appropriate.
2472FIGURE 38-6 Abdominal radical trachelectomy.
Radical Trachelectomy
Radical trachelectomy is a procedure that is gaining popularity as a surgical
management option for women with stage 1A2 and IB1 disease who desire
uterine preservation and fertility. This procedure may be performed
vaginally, abdominally, laparoscopically, or robotically (Fig. 38-6), and it
usually is accompanied by pelvic lymphadenectomy and cervical cerclage
placement. The risk of positive pelvic lymph nodes with stage IA2 cancer may be
as high as 8%, indicating the need for lymphadenectomy. Lymphadenectomy may
be performed laparoscopically, robotically, or by the open laparotomy technique.
Experience with this therapeutic modality is limited, although early results are
promising, and it is uncertain whether the long-term outcome is similar to that of
traditional therapy. Patients who are ideal candidates for this procedure have
tumors less than 2 cm in diameter and have negative lymph nodes (2018
FIGO Stage IB1). Lymphadenectomy can be performed at the beginning of the
procedure, and depending on those results, the procedure can be continued or
abandoned. A retrospective trial comparing patients who had tumors with these
2473attributes and were treated with either laparoscopic radical hysterectomy or
laparoscopic radical trachelectomy showed similar outcomes and recurrence (72).
There are limited data on subsequent pregnancy outcomes after radical
trachelectomy; however, successful outcomes were reported. A study found that
for women attempting to conceive after radical trachelectomy, the 5-year
cumulative pregnancy rate was 52.8%, with an increased risk of miscarriage (73).
Although radical trachelectomy and lymphadenectomy are performed with
curative intent, it should be remembered that if a recurrence develops, definitive
therapy with surgery or radiation is necessary.
24742475FIGURE 38-7 A: Radical hysterectomy. An intraoperative photograph showing the lateral
dissection during a radical hysterectomy. Note the ureter running beneath the uterine artery
(tissue in the clamp). B: Radical hysterectomy specimen.
Radical Hysterectomy
The radical hysterectomy (Fig. 38-7A,B) performed most often in the United
States is that described by Meigs in 1944 (74). The operation includes pelvic
lymphadenectomy along with removal of most of the uterosacral and
cardinal ligaments and the upper one-third of the vagina. This operation is
referred to as type III radical hysterectomy (75).
The hysterectomy described by Wertheim is less extensive than a radical
hysterectomy and removes the medial half of the cardinal and uterosacral
ligaments (65). This procedure is often referred to as modified radical or type
II hysterectomy. Wertheim’s original operation did not include pelvic
lymphadenectomy but instead included selective removal of enlarged lymph
nodes. The modified radical hysterectomy (type II) differs from the radical
hysterectomy (type III) in the following ways:
1. The uterine artery is transected at the level of the ureter, thus preserving
the ureteral branch to the ureter.
2. The cardinal ligament is not divided near the sidewall but instead is
divided close to its midportion near the ureteral dissection.
3. The anterior vesicouterine ligament is divided, but the posterior
vesicouterine ligament is conserved.
4. A smaller margin of vagina is removed.
Radical hysterectomies can be further classified as extended radical
hysterectomy (type IV and type V). In the type IV operation, the periureteral
tissue, superior vesicle artery, and as much as three-fourths of the vagina are
removed. In the type V operation, portions of the distal ureter and bladder are
resected. This procedure is rarely performed because radiotherapy should be used
when such extensive disease is encountered (75).
The abdomen is opened through a midline incision or a low transverse
incision after the methods of Maylard or Cherney. The low transverse
incision requires division of the rectus muscles and provides excellent
exposure of the lateral pelvis. It allows adequate pelvic lymphadenectomy and
wide resection of the primary tumor. After the abdomen is entered, the peritoneal
cavity is explored to exclude metastatic disease. The stomach is palpated to
ensure that it has been decompressed to facilitate packing of the intestines. The
2476liver is palpated, and the omentum is inspected for metastases. Both kidneys are
palpated to ensure their proper placement and lack of congenital and other
abnormalities. The para-aortic nodes are palpated transperitoneally.
During exploration of the pelvis, the fallopian tubes and ovaries are inspected
for any abnormalities. In premenopausal patients, the ovaries can be
conserved. The peritoneum of the vesicouterine fold and the rectouterine pouch
should be inspected for signs of tumor extension or implantation. The cervix is
palpated between the thumb anteriorly and the fingers posteriorly to determine its
extent, and the cardinal ligaments are palpated for evidence of lateral tumor
extension or nodularity.
Lymphadenectomy
After inspection of the abdomen and pelvis, the pelvic and para-aortic lymph
nodes should be inspected and palpated. Lymph nodes suspicious for gross
disease should be excised and evaluated by frozen section. If metastatic disease
is identified, consideration should be given to abandoning radical surgery in favor
of primary chemoradiation therapy. If the patient has no gross evidence of
metastatic disease, the pelvic lymphadenectomy is begun.
Pelvic Lymphadenectomy
The pelvic lymphadenectomy is begun by opening the round ligaments at the
pelvic sidewall and developing the paravesical and pararectal spaces. The ureter is
elevated on the medial flap by a Deaver retractor to expose the common iliac
artery. The common iliac and external iliac nodes are dissected, with care taken to
avoid injuring the genitofemoral nerve, which lies laterally on the psoas muscle.
At the bifurcation of the common iliac artery, the external iliac node chain is
divided into lateral and medial portions.
The lateral chain is stripped free from the artery to the circumflex iliac vein
distally. A hemoclip is placed across the distal portion of the lymph node chain to
reduce the incidence of lymphocyst formation. The medial chain is dissected. The
obturator lymph nodes are dissected; for this procedure, the lymph nodes are
grasped just under the external iliac vein, and traction is applied medially. In most
patients, the obturator artery and vein are dorsal to the obturator nerve; however,
10% have an aberrant vein arising from the external iliac vein. The node chain is
separated from the nerve and vessels and clipped caudally. Dissection continues
cephalad to the hypogastric artery. The cephalad portion of the obturator space
should be entered lateral to the external iliac artery and medial to the psoas
muscle, where the remainder of the obturator node tissue can be dissected as far
cephalad as the common iliac artery. Drainage of the pelvic and para-aortic lymph
node beds is not performed because of the increase in complications in patients in
2477whom drains were used (76).
Patients who have bulky cervical tumors or grossly positive pelvic nodes,
or for whom frozen section evaluation will be performed, should undergo
para-aortic lymph node evaluation to determine the full extent of disease and
to guide adjuvant therapy.
Para-Aortic Lymph Node Evaluation
The bowel is packed to expose the peritoneum overlying the bifurcation of the
aorta. The peritoneum is incised medial to the ureter and over the right common
iliac artery. A retractor is placed retroperitoneally to expose the aorta and the vena
cava. Any enlarged para-aortic lymph nodes are removed, hemoclips are applied
for hemostasis, and specimens are sent for analysis by frozen section. If the
lymph nodes are positive for metastatic cancer, an option is to discontinue the
operation and treat the patient with radiation therapy (74). If the lymph nodes are
negative for disease, the left side of the aorta is palpated through the peritoneal
incision with a finger passed under the inferior mesenteric artery. The lymph
nodes on this side of the aorta are more lateral and nearly behind the aorta and the
common iliac artery. If the left para-aortic lymph nodes appear healthy and the
cervical tumor is small with no suspicious pelvic lymph nodes, these additional
lymph nodes are not submitted for frozen-section analysis. If they are removed,
they may be dissected through the incision made for the right para-aortic nodes,
or they may be dissected after reflection of the sigmoid colon medially.
Development of Pelvic Spaces
The pelvic spaces are developed by sharp and blunt dissection (Fig. 38-8).
The paravesical space is bordered by the following structures:
1. The obliterated umbilical artery running along the bladder medially
2. The obturator internus muscle along the pelvic sidewall laterally
3. The cardinal ligament posteriorly
4. The pubic symphysis anteriorly
2478FIGURE 38-8 The pelvic ligaments and spaces. (From Hacker NF, Vermorken JB.
Cervical cancer. In: Berek JS, Hacker NF. Berek & Hacker’s Gynecologic Oncology. 6th
ed. Philadelphia, PA: Wolters Kluwer; 2015:345.)
The attachments of the vagina to the tendinous arch form the floor of the
paravesical space.
The pararectal space is bordered by the following structures:
1. Rectum medially
2. Cardinal ligament anteriorly
3. Hypogastric artery laterally
4. Sacrum posteriorly
The coccygeus (levator ani) muscle forms the floor of the pararectal space.
The development of these spaces before pelvic lymphadenectomy will aid in
identification and dissection of the pelvic lymph nodes and dissection of the
ureter as it passes into the vesicouterine ligament tunnel.
Dissection of the Bladder
2479The dissection of the bladder from the anterior part of the cervix and vagina
is a critical step. Occasionally, tumor extension into the base of the bladder
(which cannot be detected with cystoscopy) precludes adequate mobilization
of the bladder flap, leading to the abandonment of the operation. Therefore,
this portion of the operation should be undertaken early in the procedure.
The bladder should be mobilized off of the upper third of the vagina to remove
the tumor safely and with adequate margins.
Dissection of the Uterine Artery
The superior vesicle artery is dissected away from the cardinal ligament at a point
near the uterine artery. The uterine artery, which usually arises from the
superior vesicle artery, is thus isolated and divided, preserving the superior
vesicle arteries. The uterine vessels are brought over the ureter by application of
gentle traction. Occasionally, the uterine vein passes under the ureter.
Dissection of the Ureter
The ureter is dissected free from the medial peritoneal flap at the level of the
uterosacral ligament. As the ureter passes near the uterine artery, there is a
consistent arterial branch from the uterine artery to the ureter. This branch is
sacrificed in the standard radical (type III) hysterectomy but preserved in the
modified radical (type II) hysterectomy. Dissection of the ureter from the
vesicouterine ligament (ureteral tunnel) may now be accomplished. If the patient
has a deep pelvis, ligation of the uterosacral and cardinal ligaments may be
undertaken first to bring the ureteral tunnel dissection closer to the operator. The
roof of the ureteral tunnel is the anterior vesicouterine ligament. It should be
ligated and divided to expose the posterior ligament. The posterior ligament is
divided in the radical (type III) hysterectomy but conserved in the modified
radical (type II) hysterectomy.
Posterior Dissection
The peritoneum across the cul-de-sac is incised, exposing the uterosacral
ligaments. The rectum is rolled free from the uterosacral ligaments, which
are divided midway to the sacrum in a radical (type III) hysterectomy and
near the rectum in the modified radical (type II) operation. This allows the
operator to isolate and separate the cardinal ligament from the rectum. A
surgical clamp is placed on the cardinal ligament at the lateral pelvic sidewall in a
radical hysterectomy and at the level of the ureteral bed in the modified radical
procedure. A clamp is placed on the specimen side to maintain traction and to
ensure that the full cardinal ligament is excised with the specimen. A right-angled
clamp is placed caudad to this clamp across the paravaginal tissues. A second
paravaginal clamp is usually needed to reach the vagina.
2480The vagina is entered anteriorly, and a suitable margin of proximal vagina
is removed with the specimen. More vaginal epithelium can be excised if
necessary, depending on the previous colposcopic findings. The vaginal edge may
be sutured in a hemostatic fashion and left open with a drain from the pelvic space
or closed with a suction drain placed percutaneously. The ureteral fistula and
pelvic lymphocyst rates from these two techniques are similar.
Complications of Radical Hysterectomy
Acute Complications
The acute complications of radical hysterectomy include (77):
Blood loss (average, 0.8 L)
Ureterovaginal fistula (1% to 2%)
Vesicovaginal fistula (1%)
Pulmonary embolus (1% to 2%)
Small bowel obstruction (1%)
Febrile morbidity (25% to 50%)
Febrile morbidity is most often caused by pulmonary infection (10%) and is
seen frequently with pelvic cellulitis (7%) and urinary tract infection (6%).
Wound infection, pelvic abscess, and phlebitis all occur in fewer than 5% of
patients (78).
Subacute Complications
The subacute effects of radical hysterectomy are postoperative bladder
dysfunction and lymphocyst formation. For the first few days after radical
hysterectomy, bladder volume is decreased, and filling pressure is increased. The
sensitivity to filling is diminished, and the patient is unable to initiate voiding.
The cause of this dysfunction is unclear. It is important to maintain adequate
bladder drainage during this time to prevent overdistention. Bladder drainage is
usually accomplished with a suprapubic catheter. It is more comfortable for the
patient and allows the physician to perform cystometrography and determine
residual urine volume without the need for frequent catheterization. In addition,
the patient is able to accomplish voiding trials at home by clamping the catheter,
voiding, and releasing to check the residual urine level. Cystometrography may
be performed 3 to 4 weeks after surgery. For the catheter to be discontinued, the
patient must be able to sense the fullness of the bladder, initiate voiding, and void
with a residual urine level of less than 75 to 100 mL. Otherwise, voiding trials
should continue at home until these criteria can be fulfilled.
Lymphocyst formation occurs in fewer than 5% of patients, and the cause
2481is uncertain (78). Adequate drainage of the pelvis after radical hysterectomy may
be an important step in prevention. However, routine placement of retroperitoneal
drains did not reduce this morbidity (75). Ureteral obstruction, partial venous
obstruction, and thrombosis may occur from lymphocyst formation. Simple
aspiration of the lymphocyst is generally not curative, but percutaneous catheters
with chronic drainage may allow healing. If this treatment is unsuccessful,
operative intervention with excision of a portion of the lymphocyst wall and
placement of either large bowel or omentum into the lymphocyst should be
performed.
Chronic Complications
The most common chronic effect of radical hysterectomy is bladder
hypotonia or, in extreme instances, atony. This condition occurs in about 3% of
patients, regardless of the method of bladder drainage used (79,80). It may be a
result of bladder denervation and not simply a problem associated with bladder
overdistention (81). Voiding every 4 to 6 hours, increasing intra-abdominal
pressure with Credé’s maneuver, and intermittent self-catheterization may be used
to manage bladder hypotonia.
Ureteral strictures are uncommon in the absence of postoperative
radiation therapy, recurrent cancer, or lymphocyst formation (81). If the
stricture is associated with lymphocyst formation, treatment of the lymphocyst
usually alleviates the problem. Strictures that occur after radiation therapy should
be managed with ureteral stenting. If a ureteral stricture is noted in the absence of
radiotherapy or lymphocyst formation, recurrent carcinoma is the most common
cause. A CT scan of the area of obstruction should be obtained and cytologic
assessment by FNA should be performed if there is a target lesion to exclude
carcinoma. If the results of these tests are negative, a ureteral stent may be placed
to relieve the stricture. Close observation for recurrent carcinoma is necessary,
and the diagnosis of recurrence may ultimately require laparotomy.
Nerve-Sparing Radical Hysterectomy
Nerve-sparing radical hysterectomies were described in an attempt to
diminish the bladder dysfunction, sexual dysfunction, and colorectal motility
disorders commonly encountered after traditional radical hysterectomy.
Multiple techniques were described involving the identification of the pelvic
autonomic nerves at the sacral promontory followed by various surgical methods
of nerve preservation as the nerves transit the cardinal ligaments. These
techniques are promising and in small series did reduce postoperative bladder
dysfunction (82,83).
Minimally Invasive Radical Hysterectomy
2482[7] Laparoscopic and Robotic Radical Hysterectomies were being performed
with large degree of frequency in highly selected patients world wide. The use
of laparoscopy in cervical cancer patients is appealing because it may lead to less
blood loss, improved cosmetic results, shorter duration of hospitalization, and
faster recovery. In one large series of 200 women with stages IA1 to IIB cervical
cancer treated with laparoscopic lymphadenectomy followed by radical vaginal
hysterectomy, the authors found a 5-year survival rate comparable to patients
treated with a similar abdominal approach and a comparable rate of intraoperative
complications (84).
One study reports comparable body mass index, operative times, parametrial
margin, and number of lymph nodes collected when compared with open cases.
Robotic cases had significantly shorter hospital stays and blood loss, while having
significantly larger incidence of postoperative bladder dysfunction (85–88).
Minimally Invasive Versus Open Radical Hysterectomy
Although there are benefits to minimally invasive procedures, a non-inferiority
randomized control trial showed that minimally invasive radical hysterectomies
were associated with a lower rate disease-free survival at 86.0% compared to
open radical hysterectomy at 96.5% at 4.5 years (89). This study also showed
worse overall survival at the 3-year rate of 91.2% versus 97.1% in the minimally
invasive versus open radical hysterectomy respectively. This study reported that
minimally invasive radical hysterectomy is inferior to open radical hysterectomy
in respect to disease-free survival and overall survival (90). While the explanation
for this finding is unclear, it was suggested that spread of malignant cells by the
uterine manipulator or CO2 gas insufflation are potential causes.
Another study utilizing the National Cancer data base also showed that patient
with IA2 and IB1 undergoing a minimally invasive radical hysterectomy had a
higher 4 year mortality rate compared to those who underwent and open radical
hysterectomy (91).
Sentinel Lymph Node Evaluation
Sentinel lymph node detection has become an integral part of the management
strategy for breast cancer and melanoma and is being investigated as a diagnostic
tool in multiple human malignancies, including carcinoma of the cervix. The
sentinel node is a specific lymph node (or nodes) that is the first to receive
drainage from a malignancy and is a primary site of nodal metastasis. In theory,
the presence or absence of metastatic disease in the sentinel node should reflect
the status of the nodal basin as a whole. Thus, a negative sentinel lymph node
would allow omission of lymphadenectomy of the involved nodal basin. Sentinel
lymph nodes are detected through perilesional injection of radiolabeled
2483technetium-99 or blue dye followed by intraoperative identification of the sentinel
lymph nodes utilizing handheld gamma probes or visual identification of bluestained nodes. These techniques are primarily applicable in patients with earlystage disease and clinically negative lymph nodes, in whom lymph node status
may influence the extent of the procedure or the use of adjuvant treatment.
National Comprehensive Cancer Network guidelines on the use of sentinel
lymph node mapping in cervical cancer recommend the technique be considered
in tumors less than 2 cm (90). The procedure is performed by direct cervical
injection with dye or radiocolloid technetium-99 into the cervix, usually at
the 3 and 9 o’clock positions of the cervix. The sentinel lymph nodes are
identified at by direct visualization at time of laparotomy or laparoscopy of
colored dye, a fluorescent camera if indocyanine green (ICG) was used, or a
gamma probe if 99Tc was used. The nodes are commonly located medial to
the external iliac, ventral to the hypogastric, or in the superior aspect of the
obturator space (90).
Sentinel nodes can be detected in 80% to 100% of cervical cancer patients, and
these rates were confirmed by both laparotomy and laparoscopy. A combination
of dye and radiolabeled techniques appears to be superior for the detection of
sentinel lymph nodes over either technique used alone. Test sensitivity of 65% to
87% can be expected with a 90% to 97% negative predictive value. The
likelihood of detecting sentinel nodes may depend on the tumor volume, the time
from injection to retrieval of the sentinel nodes, and the volume of dye or
radiolabeled tracer injected. Sentinel node detection rates do not appear to be
influenced by prior cold knife cone biopsy. Although increasingly used, the role
of sentinel node detection in early-stage cervix cancer continues to be defined
(92).
Postoperative Management
Prognostic Variables for Early-Stage Cervical Cancer (IA2–IIA)
The survival of patients with early-stage cervical cancer after radical
hysterectomy and pelvic lymphadenectomy depends on the presence or absence
of several intermediate and high-risk pathologic factors (79,93–107).
Intermediate risk factors for recurrent disease are:
1. Large tumor size
2. Cervical stromal invasion to the middle or deep one-third
3. Lymph–vascular space invasion
High risk factors for recurrent disease are:
24841. Positive or close margins
2. Positive lymph nodes
3. Microscopic parametrial involvement
Patients treated with radical hysterectomy who have intermediate or high
risk factors have a 30% and 40% risk, respectively, of recurrence within 3
years (108–110).
Lesion Size
Lesion size is an independent predictor of survival. Patients with lesions
smaller than 2 cm have a survival rate of approximately 90%, and patients
with lesions larger than 2 cm have a 60% survival rate (97). When the
primary tumor is larger than 4 cm, the survival rate drops to 40% (95,105).
An analysis of a GOG prospective study of 645 patients showed a 94.6% 3-year
disease-free survival rate for patients with occult lesions, 85.5% for those with
tumors smaller than 3 cm, and 68.4% for patients with tumors larger than 3 cm
(106).
Depth of Invasion
Patients in whom depth of invasion is less than 1 cm have a 5-year survival
rate of approximately 90%, but the survival rate falls to 63% to 78% if the
depth of invasion is more than 1 cm (79,106,110–113).
Parametrial Spread
Patients with spread to the parametrium have a 5-year survival rate of 69%,
compared with 95% when the parametrium is negative. When the
parametrium is involved and pelvic lymph nodes are positive, the 5-year survival
rate falls to 39% to 42% (98,114).
Lymph–Vascular Space Involvement
The significance of finding lymph–vascular space involvement is somewhat
controversial. Several reports show a 50% to 70% 5-year survival rate when
lymph–vascular space invasion is present and a 90% 5-year survival rate
when invasion is absent (79,97,101,115,116). Others found no significant
difference in survival if the study is controlled for other risk factors
(106,107,117–120). Lymph–vascular space involvement may be a predictor of
lymph node metastasis and not an independent predictor of survival.
Lymph Nodes
The variable that is most independently predictive of survival is the status of
the lymph nodes. Patients with negative nodes have an 85% to 90% 5-year
2485survival rate, whereas the survival rate for those with positive nodes ranges
from 20% to 74%, depending on the number of nodes involved and the
location and size of the metastases (102–104,107,110,113,119–121).
Data on lymph node status is summarized as follows:
1. When the common iliac lymph nodes are positive, the 5-year survival rate is
about 25%, compared with about 65% when only the pelvic lymph nodes are
involved (114,122,123).
2. Bilateral positive pelvic lymph nodes portend a less favorable prognosis (22%
to 40% survival rate) than unilateral positive pelvic nodes (59% to 70%)
(122,123).
3. The presence of more than three positive pelvic lymph nodes is accompanied
by a 68% recurrence rate, compared with 30% to 50% when three or fewer
lymph nodes are positive (102,120).
4. Patients in whom tumor emboli are the only findings in the pelvic lymph node
have an 82.5% 5-year survival rate, whereas the survival rate is 62.1% and
54% with microscopic invasion and macroscopic disease, respectively (90).
Given the high risk of recurrent disease in surgically treated patients with
early-stage cervical cancer who exhibit intermediate- or high-risk pathologic
factors, adjuvant radiation or chemoradiation therapy should be considered.
Primary Radiation Therapy
[5] Radiotherapy can be used to treat all stages of cervical cancer, with cure
rates of about 70% for stage I, 60% for stage II, 45% for stage III, and 18%
for stage IV (4). A comparison of surgery and radiation for treatment of lowstage disease is shown in Table 38-5. Primary radiation treatment plans consist of
a combination of external beam radiation therapy to treat the regional lymph
nodes and to decrease the tumor volume, and brachytherapy delivered by
intracavitary applicators and/or interstitial implants to provide a treatment boost
to the central tumor. Intracavitary therapy alone may be used in patients with
early disease when the incidence of lymph node metastasis is negligible.
The treatment sequence can depend on tumor volume. Patients commonly
receive 5 weeks of pelvic external beam radiation upfront, with brachytherapy
introduced at week 4. In some cases, stage IB lesions smaller than 2 cm may be
treated first with an intracavitary source to treat the primary lesion, followed by
external therapy to treat the pelvic lymph nodes. Larger lesions require external
radiotherapy first to shrink the tumor and to reduce the anatomic distortion caused
by the cancer. Tumor shrinkage averaging >1 cm per week can be achieved with
adequate external beam radiation prior to starting brachytherapy (124,125). Such
2486a treatment strategy enables the radiation oncologist to physically implant
brachytherapy applicators and achieve dosimetric radiation coverage of the tumor
target volume.
Table 38-5 Comparison of Surgery Versus Radiation for Stage IB/IIA Cancer of the
Cervix
Surgery Radiation
Survival 85% 85%
Serious
complications
Urologic fistulas 1–2% Intestinal and urinary strictures
and fistulas 1.4–5.3%
Vagina Initially shortened, but may
lengthen with regular intercourse
Fibrosis and possible stenosis,
particularly in postmenopausal
patients
Ovaries Can be conserved Destroyed
Chronic
effects
Bladder atony in 3% Radiation fibrosis of bowel and
bladder in 6–8%
Applicability Best candidates are younger than 65
years of age, <200 lb, and in good
health
All patients are potential
candidates
Surgical
mortality
1% 1% (from pulmonary embolism
during intracavitary therapy)
Cervix cancer brachytherapy has classically been delivered with low dose rate
(LDR) technique involving inpatient admission and x-ray film–based radiation
dose prescription to “point A.” Over decades, multiple advances in radiation
oncology have resulted in improved brachytherapy techniques (e.g., image-guided
adaptive brachytherapy [IGABT]) that have translated to superior survival, better
tumor control, less side effects, and improved safety. The RetroEMBRACE study
included 731 locally advanced cervix cancer patients from 12 centers who
received external beam radiation ± chemotherapy plus 3D IGABT from 1998 to
2008. Patients achieved excellent local control (91%), pelvic control (87%), OS
(74%), and cancer-specific survival (79%) with limited severe morbidity (126).
One chief modern innovation in cervix cancer brachytherapy is the shift from
classically prescribing dose to a generic “point A” localized via 2D x-ray films, to
volumetrically prescribing dose shaped to the patient’s individual tumor geometry
as defined on 3D MRI or CT imaging. The 2D approach delivers similar
2487brachytherapy treatments to all patients regardless of tumor geometry or patient
anatomy, which can result in underdosing of larger tumors or overdosing of
smaller tumors plus over-irradiation of the adjacent bladder, rectum, and sigmoid.
In contrast, the modern 3D approach permits individualization and shaping of the
radiation prescription dose to fit patient’s tumor and normal organ anatomy. The
French multicenter STIC trial compared classical 2D versus modern 3D
approaches in 705 cervix cancer patients, prospectively demonstrating that 3D
brachytherapy resulted in improved local control and half the toxicity seen with
2D brachytherapy (127).
In classical 2D cervix brachytherapy, usual doses delivered are 7,000 to
8,000 cGy to point A (defined as 2 cm superior to the external cervical os and
2 cm lateral to the internal uterine canal) and 6,000 cGy to point B (defined
as 3 cm lateral to point A), limiting the bladder and rectal dosage to less than
6,000 cGy. To achieve this level, it is necessary to adequately pack the bladder
and bowel away from the intracavitary source. Localization films and careful
calculation of dosimetry are mandatory to optimize the dose of radiation and to
reduce the incidence of bowel and bladder complications. Local control depends
on delivering an adequate dose to the tumor from the intracavitary source.
In modern 3D IGABT, usual cumulative radiation doses delivered are 8,500 to
9,500 cGy to the high-risk CTV D90 per GEC-ESTRO recommendations and
8,000 to 9,000 cGy per American Brachytherapy Society recommendations (128).
This is often achieved with a radiation treatment course of 25 to 28 days once
daily external beam fractions, plus 4 to 5 brachytherapy fractions. The cumulative
dose to control a cervix tumor is related to the tumor’s volume, with higher doses
required to control larger tumors (129). Image-guided computerized planning is
performed on CT or MRI images acquired of the patient’s pelvis with
brachytherapy applicators in place. An updated CT or MRI scan is performed
with each brachytherapy applicator insertion, so that the radiation plan can be
adapted to match shrinking geometry of the tumor and sculpted away from the
daily position of the bladder, rectum, and sigmoid. Cumulative dose refers to the
combined dose from external beam radiation plus all brachytherapy fractions,
converted to a 2 Gy per fraction equivalent. The high-risk CTV (clinical target
volume) consists of the entire cervix plus any clinically/radiographically
identifiable tumor at time of brachytherapy (130). “D90” refers to 90% of the HRCTV receiving the highest radiation doses. Typically, dose constraints to organs
at risk are specified to the D2cc, or the 2 cc of the organ receiving the highest
radiation dose. GEC-ESTRO recommendations include keeping the cumulative
D2cc dose to bladder <90 Gy EQD2, rectum <75 Gy EQD2, sigmoid <75 Gy
EQD2 (131). The current EMBRACE II protocol (www.embracestudy.dk) for
cervix cancer suggests more stringent D2cc constraints with the goal of further
2488reducing toxicity.
Another chief innovation in modern cervix brachytherapy has been the use of
newer radiation isotopes and loading techniques. Classic 2D brachytherapy is
often performed with LDR isotopes such as cesium-137. LDR isotopes like Cs-
137 are placed manually by hand, require inpatient admissions for extended
overnight irradiation over 1 to 3 days with prolonged patient immobilization,
DVT prophylaxis, and the need for careful radiation shielding/precautions taken
by potentially exposed medical personnel. LDR isotopes are only capable of
emitting therapeutic radiation at a rate of less than 0.4 Gy per hour, whereas highdose rate (HDR) isotopes can treat at a speed of greater than 12 Gy per hour.
Modern 3D brachytherapy is most often performed with the HDR isotope
iridium-192 (Ir-192). Ir-192 is capable of delivering the prescribed brachytherapy
radiation dose in just a few minutes, typically delivered in the
ambulatory/outpatient setting with a robotic remote afterloader system that
minimizes radiation exposure to medical personnel.
Pulsed-dose rate (PDR) brachytherapy is a relatively uncommon technique
mostly seen at specialized academic centers. PDR uses an Ir-192 HDR source
with a pulsed treatment regimen designed to mimic the radiobiologic effects of
LDR treatment. Like LDR, PDR brachytherapy requires patient immobilization
and admission (132).
Although brachytherapy was traditionally prescribed using a low-dose
rate technique, high-dose rate techniques are becoming more popular.
Proponents of high-dose rate techniques argue that the exposure of radiation to
medical personnel is less, ambulatory therapy is possible, and total treatment time
is less. Advocates of low-dose rate techniques cite literature suggesting that
complication rates are higher with high-dose rate therapy. Several published trials
show that there may be slight stage-related differences in survival between
patients treated with low- and high-dose rate regimens, but the techniques have
comparable survival and complication rates (133–135). A Cochrane review of
published studies comparing LDR versus HDR brachytherapy for locally
advanced cervix cancer showed no significant differences in OS, disease-specific
survival, recurrence-free survival, local control, recurrence, metastases, or
treatment-related complications. However, these conclusions were based on a
small (4) number of trials that met inclusion criteria (136).
Newer combined intracavitary/interstitial brachytherapy applicators
permit successful treatment of bulky and/or irregularly shaped tumor
targets, and can often obviate the need to treat with the more morbid
techniques of Syed and MUPIT template needle implants. An important
innovation in the treatment of cervix cancer is the development of newer
brachytherapy applicators that allow interstitial needles to be incorporated into the
2489standard geometry of classical intracavitary tandem and ovoids, or tandem and
ring, applicators (137). The incorporation of interstitial needles permits more
effective dosing of tumor targets beyond the normal reach of standard
intracavitary brachytherapy applicators while keeping adjacent organs at risk
within safe radiation limits (138). Patients on the retroEMBRACE study with
locally advanced cervix cancer receiving combined intracavitary and interstitial
brachytherapy treatment (versus intracavitary only) had an improved therapeutic
ratio, with a 10% increase in local control seen in those with larger tumors (cervix
+ tumor volume > 30 cm3) (139).
Clinical staging is imprecise and fails to accurately predict disease
extension to the para-aortic nodes in 7% of patients with stage IB, 18% with
stage IIB, and 28% with stage III disease (140). Such patients will have
“geographic” treatment failures if standard pelvic radiotherapy ports are used. As
a result, treatment plans for these patients are individualized based on CT scans,
PET scans, and biopsies of the para-aortic lymph nodes for consideration of
extended-field radiotherapy. The routine use of extended-field radiation for
prophylactic para-aortic radiation without documentation of distant metastasis to
the para-aortic nodes was evaluated and is not practiced because of the increased
enteric morbidity associated with this treatment modality. If available, PET/CT
scans are encouraged as they can detect potential pelvic, para-aortic, or
supraclavicular nodal involvement, which in turn relates to clinical stage,
potential recurrence, and survival outcomes (141).
Intensity-Modulated Radiation Therapy
Intensity-modulated radiation therapy (IMRT) continues to be a significant
therapeutic development. IMRT is an external beam radiation technique that
improves upon older 3D conformal radiation therapy (3D-CRT) external beam
techniques. While 3D-CRT is capable of projecting block or polygonal-shaped
radiation into target tumors, IMRT is capable of projecting curved, even concave,
conformal radiation that is more precisely shaped to the target tumor space. IMRT
permits more accurate irradiation of tumor targets and significantly enhances
sparing of nearby normal organs. In the treatment of cervix cancer, use of IMRT
could reduce radiation dose to multiple organs at risk including bone marrow,
rectum, bladder, small bowel, and femurs. Technically, IMRT is achieved with
volumetric image-based (CT or MRI) radiation treatment planning in which the
radiation oncologist manually draws the tumor/target volume(s) on a
computerized dosimetry system, followed by a computer algorithm that generates
a set of treatment external beam angles and patterns that conform to the
physician-delineated target(s) while meeting prescription goals and normal organ
radiation constraints. IMRT has been used in postoperative and intact cervix
2490cancer patients, and is an advanced technique under active investigation within
radiation oncology.
Prior to the advent of IMRT, women receiving pelvic radiation typically
received 3D-CRT “four field box” plans that delivered full external beam
prescription dose (45 to 50 Gy) to their rectum, bladder, femurs, small bowel, and
bone marrow. IMRT allows the radiation oncologist to essentially sculpt dose
away from these organs at risk to reduce radiation side effects. For postoperative
cervix (and endometrial) cancer patients, the NRG-RTOG 1203 TIME-C trial
demonstrated that 278 women randomized between IMRT and 3D-CRT had, with
IMRT, better patient-reported bowel and bladder function (EPIC, FACT-Cx,
PRO-CTCAE) and less use of antidiarrheal medications. Longer follow-up is
needed to evaluate long-term toxicity (142).
A consensus guideline for IMRT use in intact cervix cancer was published in
2011, with IMRT targeting the gross tumor, entire cervix, uterus, parametria,
ovaries, and vaginal tissues (143). The amount of vaginal length covered can vary
from half to full depending on the extent of vaginal disease. For involvement of
uterosacral ligaments, IMRT coverage of the mesorectum is advised. The
INTERTECC-2 study was an international multicenter phase II clinical trial of 83
stages IB to IVA intact cervix cancer patients who received weekly cisplatin
concurrent with daily IMRT treatments followed by intracavitary brachytherapy.
Patients who received IMRT had reduced acute hematologic and GI toxicity
compared with 3D-CRT treatment (144). However, the cervix and uterus are
known to be quite mobile, capable of up to 3.5 cm shifts (even
retroversion/anteversion) between daily radiation fractions as evidenced by daily
cone beam CT scans acquired on the treating linear accelerator (145). IMRT plans
for intact cervix cancer need to be carefully designed to account for the possibility
of significant interfraction cervix movement caused by cervix/uterus mobility and
variations in daily bladder and rectal filling. In the case of node-positive cervix
cancer patients, IMRT with simultaneous integrated boost (SIB or “dose
painting”) allows simultaneous delivery of higher external beam radiation dose
(e.g., 55 Gy in 2.2 Gy per fraction) to FDG+ nodes beyond the reach of
brachytherapy, while maintaining standard fraction dosing (45 Gy in 1.8 Gy per
fraction) to central disease that will receive the remainder of curative radiation via
brachytherapy boosts (146,147). Cervix cancer patients requiring irradiation of
para-aortic nodes can potentially benefit from IMRT, which is associated with
low rates of GI toxicities (6.5% acute and late) and no duodenal-specific toxicities
even in the setting of concurrent radiosensitizing chemotherapy (148).
Despite the technical sophistication of IMRT, in the field of radiation
oncology, IMRT is considered complementary to brachytherapy and not a
replacement for brachytherapy. A SEER propensity score matching analysis of
24917,359 patients with stages IB2 to IVA cervix cancer indicated that inclusion of
brachytherapy as part of definitive treatment was independently associated with
significantly higher cause-specific survival (CSS; 64.3% vs. 51.5%, P <0.001)
and OS (58.2% vs. 46.2%, P <0.001) (149). The study identified a significant
decrease in American brachytherapy utilization from 83% in 1988 to a nadir of
43% in 2003 (and some recovery to 58% in 2009) quite possibly driven in the
early 2000s by attempted substitution of brachytherapy by IMRT. This suggests a
strong need to improve modern brachytherapy training in the field of radiation
oncology and promote awareness of brachytherapy’s importance to collaborating
gynecologic oncologists.
Adjuvant Radiation
In an effort to improve survival rates, postoperative radiotherapy was
recommended for patients with high and intermediate risk factors such as
metastasis to pelvic lymph nodes, invasion of paracervical tissue, deep
cervical invasion, or positive surgical margins (79,94,97,98,114,119,120,150).
Although most authors agree that postoperative radiotherapy is necessary in the
presence of positive surgical margins, the use of radiation in patients with other
high-risk factors is controversial. Increasing evidence supports the use of adjuvant
radiation. Particularly controversial, and best studied, is the use of radiation in the
presence of positive pelvic lymph nodes. The rationale for treatment is the
knowledge that pelvic lymphadenectomy does not remove all of the nodal and
lymphatic tissue and subsequent radiotherapy can eradicate microscopic disease.
The hesitancy to recommend postoperative radiotherapy derives from the
significant rate of postradiation bowel and urinary tract complications (151). Most
of the available data are retrospective. However, a randomized study by the GOG
comparing radiation with no further treatment for patients at high risk for
recurrence with negative pelvic nodes revealed a 30% serious complication rate,
16% reoperation rate, and a 2% mortality rate as a result of treatment-related
complications (152).
Based on retrospective studies, it appears that postoperative radiation
therapy for positive pelvic nodes can decrease pelvic recurrence but does not
improve 5-year actuarial survival rates. One multi-institutional study showed
no difference in survival in patients with three or fewer positive pelvic nodes
(59% vs. 60%) (120). However, there seemed to be a benefit when radiotherapy
was given to those with more than three positive nodes.
In a study of 60 pairs of irradiated and nonirradiated women matched for age,
lesion size, number, and location of positive nodes after radical hysterectomy, no
significant difference was found in projected 5-year survival rates (72% for
surgery alone, 64% for surgery plus radiation) (153). The proportion of
2492recurrences confined to the pelvis was 67% in patients treated with surgery only
and 27% in patients treated with postoperative radiation (P = 0.03). In a Cox
regression analysis of 320 women who underwent radical hysterectomy, for the
72 who received postoperative radiation, there was a significant decrease in pelvic
recurrence but no survival benefit (103). A multi-institutional retrospective study
was performed on 185 women with positive pelvic nodes after radical
hysterectomy, including 103 who received postoperative radiotherapy (105).
Multivariate analysis disclosed that radiotherapy was not an independent predictor
of survival, whereas age, lesion diameter, and number of positive nodes did
influence survival. These authors concluded that additional treatment is needed to
improve survival rates. Because survival is limited by distant recurrence, the
addition of chemotherapy to postoperative radiotherapy was proposed. A 75%
disease-free survival rate was reported at 3 years in 40 high-risk patients given
cisplatin, vinblastine, and bleomycin after radical hysterectomy, and a 46%
disease-free survival rate was found in 79 comparable patients who refused
treatment (154). Only 4 (11.8%) of 34 patients with positive pelvic nodes had
recurrences, whereas disease recurred in 8 (33%) of 24 untreated patients with
positive nodes. An 82% rate of disease-free survival was reported at 2 years
among 32 patients who were treated postoperatively with radiation therapy plus
cisplatin and bleomycin (155).
The location of lymph node metastases apparently is relevant to
postirradiation recurrence rates. When common iliac lymph nodes are
involved, the survival rate drops to 20%. As the number of positive pelvic nodes
increases, the percentage of positive common iliac and low para-aortic nodes
increases (i.e., 0.6% when pelvic lymph nodes are negative, 6.3% with one
positive pelvic node, 21.4% with two or three positive nodes, and 73.3% with
four or more positive nodes). This information was used to recommend extendedfield radiotherapy to patients with positive pelvic lymph nodes in an attempt to
treat undetected extrapelvic nodal disease (115). A 3-year disease-free survival
rate of 85% occurred in patients with positive pelvic nodes, and a survival rate of
51% occurred in patients with positive common iliac nodes; these rates are better
than the survival rates of 50% and 23%, respectively, for historical control groups
receiving radiotherapy to the pelvis alone.
The GOG reported the results of a randomized controlled trial on patients
with cervical cancer treated by radical hysterectomy and found to have at
least two of the following risk factors: capillary lymphatic space invasion,
more than one-third stromal invasion, and large tumor burden (109). A total
of 277 patients were entered into the study, with 140 patients randomized to no
further therapy and 137 patients randomized to adjuvant pelvic radiotherapy.
Patients with these risk factors who were treated postoperatively with
2493radiation therapy had a statistically significant (47%) decrease in recurrent
disease. After extensive follow-up, there is no statistically significant difference
in mortality rates (156). The morbidity with combination therapy was acceptable,
with a low rate of enteric and urinary complications. A second GOG study of
patients with high-risk cervical cancer randomized patients to concurrent
chemoradiation therapy or radiation therapy alone (108).
Concurrent Chemoradiation
Radiation therapy fails to achieve tumor control in 20% to 65% of patients with
advanced cervical cancer. Chemotherapy, despite its relative lack of success in
treating patients with cervical cancer, was evaluated as neoadjuvant treatment in
combination with surgery. Concomitant use of chemotherapy and radiation
was studied extensively by the GOG and results of five randomized studies
were reported. The concept of chemoradiation encompasses the benefits of
systemic chemotherapy with the benefits of regional radiation therapy. The use of
chemotherapy to sensitize cells to radiation therapy improved local–regional
control. These results changed the way cervical cancer is treated in many medical
centers.
An intergroup trial involving the GOG, the Southwestern Oncology
Group, and the Radiation Therapy Oncology Group evaluated postoperative
chemoradiation therapy in patients with stage IA2, IB, or IIA cervical cancer
who had positive pelvic lymph nodes, positive parametrial extension, or
positive vaginal margins at the completion of radical hysterectomy (155). A
total of 243 patients were assessed in this trial, with 127 receiving chemoradiation
(cisplatin, 5-fluorouracil [5-FU], radiation therapy) and 116 receiving radiation.
The results of this trial showed a statistically significant improvement in
progression-free survival (PFS) and overall survical (OS) at 43 months for
the patients receiving concurrent chemoradiation. The 4-year survival rates for
the patients receiving chemoradiation versus radiation alone were 81% and 71%,
respectively. The toxicity levels in the two groups were acceptable, with a higher
rate of hematologic toxicity in the concurrent chemoradiation arm. This study
showed that in patients with these high-risk factors after radical
hysterectomy for stages IA2, IB, and IIA disease, chemoradiation is the
postoperative treatment of choice.
Concurrent chemoradiation was evaluated in patients with advanced cervical
carcinoma. GOG protocol 85 was a prospective study that enrolled patients with
stages IIb to IVA cervical cancer and compared concurrent chemoradiation (157).
There were 177 patients treated with cisplatin, 5-FU, and radiation. These
patients were compared with 191 patients treated with hydroxyurea and radiation.
The median follow-up of patients who were alive at the time of the analysis was
24948.7 years. Patients who received concurrent chemoradiation and were treated with
cisplatin and 5-FU had a statistically significant improvement in progression-free
interval and OS (157). Hematologic toxicity levels in the two groups were similar.
This study showed that cisplatin-based concurrent chemoradiation was a
superior treatment when compared with hydroxyurea and concurrent
radiation.
GOG Protocol 120 was initiated to evaluate patients with negative para-aortic
nodes and cervical carcinoma stages IIB to IVA treated with concurrent
chemoradiation. The treatment arms in this study consisted of radiation plus
weekly cisplatin; or cisplatin, 5-FU, and hydroxyurea; or hydroxyurea. There
were 176 patients in the weekly cisplatin arm; 173 patients in the cisplatin, 5-FU,
and hydroxyurea arm; and 177 patients in the hydroxyurea arm (158). The two
treatment arms with cisplatin-based chemotherapy and radiation showed an
improvement in progression-free interval and OS at a median follow-up of 35
months. The relative risks for progression of disease or death were 0.55 and 0.57,
respectively, for patients treated with cisplatin-based chemotherapy and radiation,
compared with the patients treated with hydroxyurea and radiation (158). This
study confirmed the findings of GOG Protocol 85 and reaffirmed the finding
that cisplatin-based concurrent chemoradiation is the treatment of choice for
patients with advanced-stage cervical cancer.
A third GOG trial evaluated patients with stages IB to IVA cervical cancer. Of
the patients enrolled in this study, 70% had stage IB or IIA disease (159). A total
of 403 patients were enrolled and evaluated. The 5-year survival rates were 73%
in patients treated with chemoradiation and 58% in patients treated with radiation
therapy alone. The cumulative rates of disease-free survival at 5 years were 67%
in patients treated with concurrent chemoradiation and 40% in patients treated
with radiation therapy alone. Survival and progression-free intervals for patients
receiving concurrent chemoradiation were significantly improved (159). The
results of this study suggested that chemoradiation is the treatment of choice
for stages IIB to IVA disease and that those patients with stages IB2 and IIA
disease may benefit from chemoradiation.
A GOG study of chemoradiation comparing concurrent cisplatin and
radiation with radiation alone in patients with bulky IB cervical cancer
included adjuvant hysterectomy after completion of the radiation (160).
There were 183 patients assigned to the concurrent chemotherapy and radiation
arm and 186 patients treated with radiation alone. The median duration of followup was 36 months, with disease recurrence detected in 37% of the patients treated
with radiation alone, compared with 21% who were treated with concurrent
chemoradiation (160). The 3-year survival rates were 83% in the group who
received concurrent chemoradiation and 74% in the group who received radiation
2495alone (160). The study included adjuvant hysterectomy after completion of
radiation treatment. Because the results did not show an improvement in survival
by using adjuvant hysterectomy, the authors concluded that adjuvant
hysterectomy would not be part of their recommendations. This study supports
the results of previous studies and shows that patients with bulky stages IB
and IIA cervical cancer treated with concurrent chemoradiation have
survival rates superior to those treated with radiation alone. These two
studies indicate that patients with bulky stages IB and IIA disease should
have primary treatment consisting of chemoradiation, with the
chemotherapy agent being weekly cisplatin.
Based on 2,042 patients enrolled on prior GOG trials, modern nomograms
have been developed to predict 2-year PFS, 5-year OS, and pelvic recurrence
risk for women with locally advanced cervix cancer treated with definitive
chemoradiation. Prognostic factors include histology, race/ethnicity,
performance status, tumor size, FIGO stage, grade, presence of pelvic and
para-aortic nodes, and use of cisplatin-based chemotherapy (161).
Surgical Staging Before Radiation
Surgical staging procedures designed to discover positive lymph nodes may
be forgone with use of PET/CT imaging studies. The use of transperitoneal
exploration was associated with a 16% to 33% mortality rate from radiotherapyinduced bowel complications and a 5-year survival rate of only 9% to 12%
(162,163). To avoid these complications, extraperitoneal dissection of the
para-aortic nodes is recommended, and the radiation dose should be reduced
to 5,000 cGy or less (164,165). When this approach is used, postradiotherapy
bowel complications occur in fewer than 5% of patients, and the 5-year
survival rate is 15% to 26% in patients with positive para-aortic nodes
(22,166,167). Survival appears to be related to the amount of disease in the paraaortic nodes and to the size of the primary tumor. In patients whose metastases to
the para-aortic lymph nodes are microscopic and whose central tumor has not
extended to the pelvic sidewall, the 5-year survival rate improves to 20% to 50%
(168,169). Surgical staging techniques have improved to include laparoscopic
assessment of the para-aortic and pelvic lymph nodes. Studies demonstrated
benefit from surgical staging with improved survival and changes in treatment
plans in 40% of patients (165,166). When PET/CT is compared with
surgicopathologic staging of para-aortic lymph nodes, some patients with
histologically positive para-aortic lymph nodes are missed with surgicopathologic
staging (170).
Management of Grossly Positive Para-Aortic Lymph Nodes
2496The management of patients with macroscopic or grossly positive para-aortic
lymph nodes discovered at the time of surgery or by imaging studies is
controversial. It is likely that grossly positive nodes may be beyond the
ability of radiation therapy alone to sterilize. Therefore, to improve survival,
additional therapy is required. In a representative study of the multiple reports in
the literature, lymph node metastases were noted in 133 of 266 patients. Pelvic
and para-aortic nodes were positive in 44 patients and positive para-aortic nodes
were noted in only 2 patients. Five- and 10-year survival rates were similar for
patients with macroscopically positive resectable nodes and microscopically
positive nodes. Patients with unresectable nodal disease had a worse survival rate
than those with resectable disease. All patients underwent extraperitoneal lymph
node resections and subsequent radiotherapy. There was a 10% incidence of
severe morbidity related to radiation use. Consistent with other reports in the
literature, this study showed that extraperitoneal debulking lymphadenectomy
confers a survival advantage similar to that enjoyed by patients with
micrometastatic disease without additional morbidity (30). The topic
nevertheless remains in debate because chemoradiation may permit the treatment
of larger volume disease without the need for surgical debulking.
Prophylactic Para-Aortic Radiation Therapy
Prophylactic extended-field radiation therapy is an alternative to surgical
staging of the para-aortic lymph node chain in women with advanced
cervical cancer judged to be at high risk but without radiologic or clinical
evidence of para-aortic lymph node involvement. This treatment strategy was
evaluated in 441 patients with stages I to III disease (171). High rates of
gastrointestinal toxicity were noted in the treatment group. There was no
difference in disease-free survival or OS between the control and treated groups,
although treated patients had fewer para-aortic failures. A lack of difference in
survival rates in this study may be related to high local and regional failure rates,
suggesting that ideal patients for prophylactic radiotherapy would be those in
whom there is a high likelihood of achieving pelvic control. A survival benefit
was noted in a study by the Radiation Therapy Oncology Group, in which 367
patients with stages IB to IIB disease were randomized to pelvic radiotherapy
versus pelvic and extended field radiotherapy (172). The extended field treatment
arm suffered more grade 4 and 5 toxicity, confirming previous studies.
Complicating the issue is another study from the Radiation Therapy Oncology
Group revealing that in locally advanced cervical cancer, pelvic radiation therapy
with concurrent cisplatin chemotherapy was superior to extended-field radiation
therapy (159). There are no large prospective trials that compare pelvic
chemoradiation with extended-field chemoradiation. Given the potential toxicity,
2497it seems prudent to administer extended-field radiation only to patients with
suspected para-aortic node involvement.
Supraclavicular Lymph Node Biopsy
Although not standard practice, the performance of a supraclavicular lymph node
biopsy was advocated in patients with positive para-aortic lymph nodes before the
initiation of extended-field irradiation and in patients with a central recurrence
before exploration for possible exenteration. The incidence of metastatic disease
in the supraclavicular lymph nodes in patients with positive para-aortic lymph
nodes is 5% to 30% (173). Node enlargement and increased metabolic activity
can be assessed with chest PET/CT scanning. Cytologic assessment by FNA can
obviate the need for an excisional biopsy and should be performed if any enlarged
nodes are present. If the scalene lymph nodes are positive, palliative
chemotherapy should be considered.
Complications of Radiation Therapy
Perforation of the uterus may occur (∼ 9%) with the insertion of the uterine
tandem. This is particularly a problem for elderly patients and those who
had a previous diagnostic conization procedure. When perforation is
recognized, an LDR tandem should be removed, and the patient should be
observed for bleeding or signs of peritonitis. Survival may be decreased in
patients who have uterine perforation, possibly because these patients have more
extensive uterine disease (174). Fever may occur after insertion of the uterine
tandem and ovoids. Fever most often results from infection of the necrotic tumor
and occurs 2 to 6 hours after insertion of the intracavitary system. If uterine
perforation was excluded by ultrasonography, intravenous broad-spectrum
antibiotic coverage, usually with a cephalosporin, should be administered. If the
fever does not decrease promptly or if the temperature is higher than 38.5°C, an
aminoglycoside and a bacteroides species–specific antibiotic should be
administered. If fever persists or if the patient shows signs of septic shock or
peritonitis, any LDR intracavitary system must be removed. Antibiotics are
continued until the patient recovers, and any LDR intracavitary application is
delayed for 1 to 2 weeks. In case of HDR brachytherapy, antibiotics can be given
and the tandem should be repositioned into proper place so that the HDR
radiation treatment (typically only several minutes duration) can be delivered
without delay followed by removal of the applicators (175). The use of
intraoperative real-time transabdominal ultrasound guidance during placement of
the brachytherapy tandem can significantly reduce the chance of perforation and
help ensure an accurate brachytherapy implant.
2498Acute Morbidity
The acute effects of radiotherapy are caused by ionizing radiation on the
epithelium of the intestine and bladder and occur after administration of
2,000 to 3,000 cGy. Symptoms include diarrhea, abdominal cramps, nausea,
frequent urination, and occasional bleeding from the bladder or bowel mucosa.
Bowel symptoms can be treated with a low-gluten, low-lactose, and low-protein
diet. Antidiarrheal and antispasmodic agents may help. Bladder symptoms may be
treated with antispasmodic medication. Severe symptoms may require a week of
rest from radiotherapy.
Chronic Morbidity
The chronic effects of radiotherapy result from radiation-induced vasculitis
and fibrosis and are more serious than the acute effects. These complications
occur several months to years after radiotherapy is completed. The bowel
and bladder fistula rate after pelvic radiation therapy for cervical cancer is
1.4% to 5.3%, respectively (62,64). Other serious toxicity (e.g., bowel bleeding,
stricture, stenosis, or obstruction) occurs in 6.4% to 8.1% of patients (62,64).
Radiation Proctopathy
Rectal bleeding after pelvic radiotherapy (radiation proctopathy) commonly
self-resolves over time. Patients with continued bleeding can be offered at least
flexible sigmoidoscopy to rule out causes unrelated to radiation therapy. For
radiation-induced bleeding, first-line sucralfate enemas can be used. For
refractory bleeding, interventions such as argon plasma coagulation, laser therapy
and formalin can be considered (176).
Rectovaginal Fistula
Rectovaginal fistulas or rectal strictures occur in fewer than 2% of patients.
The successful closure of fistulas with bulbocavernosus flaps or sigmoid colon
transposition was reported (177,178). Occasionally, resection with anastomosis is
feasible. Diversion resulting in colostomy may be the optimal therapy in patients
who have poor vascular supply to the pelvis and a history of an anastomotic leak
or breakdown from prior repairs.
Small Bowel Complications
Patients with previous abdominal surgery are more likely to have pelvic
adhesions and thus sustain more radiotherapy complications in the small
bowel. The terminal ileum may be particularly susceptible to chronic
damage because of its relatively fixed position at the cecum. Patients with
small bowel complications have a long history of crampy abdominal pain,
intestinal rushes, and distention characteristic of partial small bowel obstruction.
2499Often, low-grade fever and anemia accompany the symptoms. Patients who have
no evidence of disease should be treated aggressively with total parenteral
nutrition, nasogastric suction, and early surgical intervention after the
anemia resolves and good nutritional status is attained. The type of procedure
performed depends on individual circumstances (179). Small bowel fistulas that
occur after radiotherapy rarely close spontaneously while total parenteral nutrition
is maintained. Recurrent cancer should be excluded; aggressive fluid replacement,
nasogastric suction, and wound care should be instituted. Fistulography and a
barium enema should be performed to exclude a combined large and small bowel
fistula. The fistula-containing loop of bowel may be either resected or isolated
and left in situ. In the latter case, the fistula will act as its own mucous fistula.
Urinary Tract
Chronic urinary tract complications occur in 1% to 5% of patients and
depend on the dose of radiation to the base of the bladder. Vesicovaginal
fistulas are the most common complication and usually require
supravesicular urinary diversion. Occasionally, a small fistula can be repaired
with either a bulbocavernosus flap or an omental pedicle. Ureteral strictures are
usually a sign of recurrent cancer, and a cytologic sample should be obtained at
the site of the obstruction using FNA guided by a CT scan. If the findings are
negative, the patient should undergo exploratory surgery to evaluate the presence
of recurrent disease. If radiation fibrosis is the cause, ureterolysis may be possible
or indwelling ureteral stents may be passed through the open urinary bladder to
relieve obstruction.
Chemotherapy
Neoadjuvant Chemotherapy
Randomized trials were initiated by the GOG and other large centers to determine
the efficacy of neoadjuvant chemotherapy. In the era of effective
chemoradiation therapy, there is no evidence that neoadjuvant
chemotherapy offers superior results or a survival advantage over standard
therapy.
Chemotherapy for Advanced Disease
Chemotherapy was studied in advanced cervical cancer with mixed results (180–
183). Doublet therapy compared four cisplatin-containing doublets (gemcitabine,
paclitaxel, topotecan, vinorelbine), and there were no major differences in OS,
although the cisplatin and paclitaxel doublet trended toward the best results
(GOG 204) (184). Studies showed comparable survival and lower toxicity
with carboplatin and paclitaxel compared with cisplatin and paclitaxel (185).
2500A survival benefit was demonstrated when bevacizumab, an anti-VEGF-A
monoclonal antibody, was added to the combination of platinum-based
chemotherapy in patients with metastatic, persistent, or recurrent cervical
carcinoma (GOG 240). However, the addition of bevacizumab to the
combination increased the risk of fistula formation (186).
While multiagent regimens offer an improved response rate and several months
longer OS, they are more toxic than single-agent regimens. For women who
cannot tolerate combination chemotherapy, single-agent chemotherapy with
carboplatin is an appropriate treatment.
Treatment of Cervical Cancer by Stage
[5] Stage IA
Until 1985, no FIGO recommendation existed concerning the size of lesion or the
depth of invasion that should be considered microinvasive (stage IA). This led to
considerable confusion and controversy in the literature. Over the years, as many
as 18 different definitions were used to describe microinvasion. In 1974, the
Society of Gynecologic Oncologists recommended a definition that is accepted by
FIGO: A microinvasive lesion is one in which neoplastic epithelium invades
the stroma to a depth of less than 3 mm beneath the basement membrane
and in which lymphatic or blood vascular involvement is not demonstrated.
The purpose of defining microinvasion is to identify a group of patients who are
not at risk for lymph node metastases or recurrence and who therefore may be
treated with less than radical therapy.
Diagnosis must be determined on the basis of a cone biopsy of the cervix.
The treatment decision rests with the gynecologist and should be based on a
review of the conization specimen with the pathologist. It is important that
the pathologic condition be described in terms of (i) depth of invasion, (ii)
presence or absence of lymph–vascular space invasion, and (iii) margin
status. These variables are used to determine the degree of radicality of the
operation and whether the regional nodes should be treated (171).
Stage IA1 <3 mm Invasion
Lesions with invasion less than 3 mm have less than 1% incidence of pelvic
node metastases. Within this group, it appears that the patients most at risk for
nodal metastases or central pelvic recurrence are those with definitive evidence of
tumor emboli in lymph vascular spaces (77,187). Therefore, patients with less
than 3 mm invasion and no lymph–vascular space invasion may be treated with
extrafascial hysterectomy without lymphadenectomy. Therapeutic conization
appears to be adequate therapy for these patients if preservation of childbearing
capability is desired. Surgical margins and postconization endocervical curettage
2501must be free of disease. If there is lymph–vascular space invasion, a type I
(extrafascial) or II (modified radical) hysterectomy with pelvic lymphadenectomy
should be considered.
Treatment of microinvasive cervical adenocarcinoma is complicated by a
lack of agreement on approaches. Reports show that patients with stage IA1
cervical adenocarcinoma may be treated in a fashion similar to patients with this
stage and a squamous lesion (111–113). Some experts disagree with this
interpretation because of the difficulty in establishing a pathologic diagnosis of
microinvasion from a frankly invasive adenocarcinoma. Patients diagnosed with
microinvasive cervical adenocarcinoma should have expert pathologic
assessment before considering treatment with extrafascial hysterectomy or
conization alone.
Stage IA2 ≥3 mm to <5 mm Invasion
Lesions with invasion of greater than or equal to 3 to 5 mm have a 3% to 8%
incidence of pelvic node metastases; thus, pelvic lymphadenectomy or sentinel
lymph node mapping is necessary for these lesions (187,188). The primary tumor
may be treated with a modified radical hysterectomy (type II) or a radical
trachelectomy if preservation of fertility is desired. If intermediate- or high-risk
pathologic factors are identified in the surgical specimen, adjuvant radiation or
chemoradiation therapy is recommended.
[5] Stages IB1, IB2, and IIA1 Invasive Cancer
Stage IB lesions are subdivided into stage IB1, which denotes lesions that
have 5 mm or more stromal invasion and are smaller than 2 cm in maximum
diameter, and stage IB2, for lesions that are 2 cm or larger and less than 4
cm, and IB3, which denotes lesions that are 4 cm or larger. Patient’s with
lesions 2 cm or less may be managed with either a radical trachelectomy or
type III radical hysterectomy with pelvic lymphadenectomy. Radical
trachelectomy should be restricted to candidates who desire future fertility
with low-risk disease and a tumor size less than 2 cm. The para-aortic lymph
node chain should be evaluated if pelvic nodal disease is encountered.
Adjuvant radiation therapy is recommended if intermediate risk factors are
identified postoperatively. Adjuvant chemoradiation is indicated if high-risk
features are found.
Alternatively, primary chemoradiation therapy with curative intent is
appropriate. A comparison of radical hysterectomy with radiation resulted
in similar survival rates for the two treatment modalities. Several studies
comparing patients treated by either radical hysterectomy or radiation therapy
showed similar survival rates and outcomes for both groups (93,189). However,
2502patients treated with type III radical hysterectomy who subsequently received
postoperative radiation had a higher rate of intestinal and urinary morbidity
compared with patients treated with either modality alone. Therefore, some
clinicians advocate using radiation and avoiding surgery in these patients because
many will require adjuvant postoperative radiation.
[5] Bulky Stages IB3 and IIA2 Invasive Cancer
Patients with bulky IB3 and IIA2 disease may be treated with either primary
chemoradiation or radical surgery. Because many of these patients will have
intermediate- or high-risk factors postoperatively, strong consideration
should be given to primary chemoradiation. If surgical therapy is desired, a
type III radical hysterectomy with pelvic and para-aortic lymphadenectomy,
followed by adjuvant chemoradiation if intermediate- or high-risk factors
are present, is appropriate therapy. This option has the benefits of complete
surgical staging and ovarian preservation, if desired. Disadvantages of primary
surgery include increased morbidity if multimodality therapy is utilized (189).
Stages IIB to IIIB Invasive Cancer
Therapy for patients with stage IIB or greater cervical cancer traditionally
was radiation therapy. Primary pelvic radiotherapy fails to control disease
progression in 30% to 82% of patients with advanced cervical carcinoma (4).
Two-thirds of these failures occur in the pelvis (190). A variety of agents were
used in an attempt to increase the effectiveness of radiation therapy in patients
with large primary tumors. Because chemoradiation was superior to radiation
therapy alone, chemoradiation is the preferred treatment strategy for these
patients, with cisplatin the chemotherapy agent of choice. Nodal involvement,
particularly the para-aortic lymph nodes, is the most important factor related to
survival.
Stage IIIC1 and IIIC2
The updated staging system adds this new category to reflect metastatic disease to
the pelvic lymph nodes (Stage IIIC1) and/or the para-aortic lymph nodes (Stage
IIIC2). If the pelvic lymph nodes only are involved, chemoradiation to the pelvis
is the recommended treatment, while if the para-aortic lymph nodes are involved,
extended-field radiation with chemosensitization can be used and this can be
followed by systemic chemotherapy (191–195).
[5] Stages IVA and IVB Cancer
Primary exenteration may be considered for patients with direct extension to the
rectum or bladder, but it is rarely performed. For patients with extension to the
bladder, the survival rate with radiation therapy is as high as 30%, with a urinary
2503fistula rate of only 3.8% (196). The presence of tumor in the bladder may prohibit
cure with radiation therapy alone; thus, consideration must be given to removal of
the bladder on completion of external beam radiation treatment. This is
particularly true if the disease persists at that time and the geometry is not
conducive to brachytherapy. Rectal extension is less commonly observed but may
require diversion of the fecal stream before chemoradiation to avoid septic
episodes from fecal contamination. In certain clinical situations, such as with
patients who have stage IVA disease and present with vesicovaginal or
rectovaginal fistula, urinary or rectal diversion may be performed, followed by
chemoradiation.
Patients with stage IVB cervical carcinoma are candidates for
chemotherapy and palliative pelvic radiation therapy. Control of symptoms
with the least morbidity is of primary concern in this patient population.
Patient Evaluation and Follow-Up After Therapy
Patients who receive radiotherapy should be monitored closely to assess
treatment response. Tumors may be expected to regress for up to 3 months
after radiotherapy. During the pelvic examination, progressive shrinkage of the
cervix and possible stenosis of the cervical os and surrounding upper vagina are
expected and should be noted. During rectovaginal examination, careful palpation
of the uterosacral and cardinal ligaments for nodularity is important. Cytologic
assessment by FNA of suspicious areas should be performed to allow early
diagnosis of persistent disease. In addition to the pelvic examination, the
supraclavicular and inguinal lymph nodes should be carefully examined, and
cervical or vaginal assessment should be performed every 3 to 6 months for 2
years and then every 6 to 12 months for the next 3 to 5 years. Endocervical
curettage may be performed in patients with large central tumors (90).
Patients with isolated recurrences may benefit from surgical resection.
Metastasis to the lung is reported in 1.5% of cases of patients with advanced
disease. Solitary nodules are present in 25% of cases with metastasis. Resection
of a solitary nodule in the absence of other persistent disease may yield some
long-term survivors (197). After radical hysterectomy, about 80% of recurrences
are detected within 2 years (198). The larger the primary lesion, the shorter the
median time is to recurrence (199). Although CT, PET, or IVP is not a part of
routine postradiotherapy surveillance, it should be performed if a pelvic mass is
detected or if other symptoms warrant evaluation. The finding of ureteral
obstruction after radiotherapy in the absence of a palpable mass may indicate
unresectable pelvic sidewall disease, but this finding should be confirmed, usually
by FNA cytologic assessment (200).
2504Special Considerations
Cervical Cancer During Pregnancy
The incidence of invasive cervical cancer associated with pregnancy is 1.2 in
10,000 (201). A Pap test should be performed on all pregnant patients at the initial
prenatal visit, and any grossly suspicious lesions should be biopsied. Diagnosis is
often delayed during pregnancy because bleeding is attributed to pregnancyrelated complications. If the result of the Pap test is positive for malignant
cells, and invasive cancer cannot be diagnosed using colposcopy and biopsy,
a diagnostic conization procedure may be necessary. Conization in the first
trimester of pregnancy is associated with hemorrhagic and infectious
complications, and an abortion rate as high as 33% (202,203). Because
conization subjects the mother and fetus to complications, it should not be
performed before the second trimester and only in patients with colposcopy
findings consistent with cancer, biopsy-proven microinvasive cervical cancer,
or strong cytologic evidence of invasive cancer. Inadequate colposcopic
examination may be encountered during pregnancy in patients who had prior
ablative therapy. Close follow-up throughout pregnancy may allow the cervix to
evert and develop an ectropion, allowing satisfactory colposcopy in the second or
third trimester. Patients with obvious cervical carcinoma may undergo cervical
biopsy and clinical staging similar to that of nonpregnant patients.
After conization, there appears to be no harm in delaying definitive
treatment until fetal maturity is achieved in patients with stage IA cervical
cancer (202,204,205). Patients with less than 3 mm of invasion and no lymphatic
or vascular space involvement may be followed to term. Historically, these
patients were allowed to deliver vaginally, and a hysterectomy was performed 6
weeks postpartum if further childbearing was not desired. However, in a
multivariate analysis of 56 women with cervical cancer diagnosed during
pregnancy and 27 women with cervical cancer diagnosed within 6 months of
delivery, vaginal delivery was the most significant predictor of recurrence. In
addition, most recurrences after vaginal delivery involved distant sites. The ideal
delivery method for these patients is not known definitively; however, strong
consideration should be given to performing a cesarean birth in women with
cervical cancer of any stage (206). If vaginal delivery is chosen, close inspection
of the episiotomy site is required during follow-up because of rare reports of
metastatic cervical cancer at these locations (207).
Patients with 3 to 5 mm of invasion and those with lymph–vascular space
invasion may be followed to term or delivered early after establishment of
fetal pulmonary maturity (202,205). They may have cesarean delivery,
immediately followed by modified radical hysterectomy and pelvic
2505lymphadenectomy. Patients with more than 5 mm invasion should be treated
as having frankly invasive carcinoma of the cervix. Treatment depends on the
gestational age of the pregnancy and the wishes of the patient. Modern neonatal
care affords a 75% survival rate for infants delivered at 28 weeks of gestation and
90% for those delivered at 32 weeks of gestation. Fetal pulmonary maturity can
be determined by amniocentesis, and prompt treatment can be instituted when
pulmonary maturity is documented. Although timing is controversial, it is
probably unwise to delay therapy for longer than 4 weeks (204,205). The
recommended treatment is classic cesarean delivery followed by radical
hysterectomy with pelvic lymphadenectomy. There should be a thorough
discussion of the risks and options with both parents before any treatment is
undertaken.
Patients with stages II to IV cervical cancer should be treated with
radiotherapy. If the fetus is viable, it is delivered by classic cesarean birth,
and therapy is begun postoperatively. If the pregnancy is in the first trimester,
external radiation therapy can be started with the expectation that spontaneous
abortion will occur before the delivery of 4,000 cGy. In the second trimester, a
delay of therapy may be entertained to improve the chances of fetal survival. If
the patient wishes to delay therapy, it is important to ensure fetal pulmonary
maturity before delivery is undertaken. Neoadjuvant chemotherapy has been
administered to women during pregnancy with cervical cancer after 13 weeks
gestation, without clear short-term harm to the fetus, although longer clinical
follow-up is necessary (208).
The clinical stage is the most important prognostic factor for cervical
cancer during pregnancy. OS for these patients is slightly better because an
increased proportion of these patients have stage I disease. For patients with
advanced disease, there is evidence that pregnancy impairs the prognosis
(202,205). The diagnosis of cancer in the postpartum period is associated with a
more advanced clinical stage and a corresponding decrease in survival (206).
Cancer of the Cervical Stump
Cancer of the cervical stump was more common many decades ago when
supracervical hysterectomy was popular; because this operation is being
performed more frequently, this situation may become increasingly familiar.
Early-stage disease is treated surgically, with very little change in technique from
that used when the uterus is intact (209). Radical parametrectomy with upper
vaginectomy and pelvic lymphadenectomy is the standard procedure.
Advanced-stage disease may present a therapeutic problem for the radiation
oncologist if the length of the cervical canal is less than 2 cm. This length is
necessary to allow satisfactory placement of the uterine tandem. If the uterine
2506tandem cannot be placed, radiation therapy can be completed with vaginal ovoids
or external beam techniques such as IMRT. Radiation oncology departments
skilled in combined interstitial/intracavitary brachytherapy offer the ability to
achieve curative brachytherapy coverage of cervical stump carcinomas.
Intravaginal ovoids abutting the stump can be loaded with interstitial needles
arrayed through the entire diameter and thickness of the stump to achieve
comprehensive tumor coverage.
Pelvic Mass
The origin of a pelvic mass must be clarified before treatment is initiated. A
CT urogram can exclude a pelvic kidney, and a barium enema helps to identify
diverticular disease or carcinoma of the colon. An abdominal x-ray film may
show calcifications typically associated with benign ovarian teratomas or uterine
leiomyomas. Pelvic ultrasonography differentiates between solid and cystic
masses and indicates uterine or adnexal origin. Solid masses of uterine origin are
most often leiomyomas and do not need further investigation.
Pyometra and Hematometra
An enlarged fluid-filled uterine cavity may be a pyometra or a hematometra.
The hematometra can be drained by dilation of the cervical canal and will
not interfere with treatment. The pyometra should be drained, and the
patient should be given antibiotics to cover Bacteroides species, anaerobic
Staphylococcus and Streptococcus species, and aerobic coliform bacterial
infection. Placement of a large mushroom catheter through the cervix was
advocated, but the catheter may become obstructed, leading to further occlusion
of the drainage. Repeated dilation of the cervix with aspiration of pus every 2 to 3
days is more effective.
If the disease is stage I, a radical hysterectomy and pelvic
lymphadenectomy may be performed. However, a pyometra is usually found
in patients with advanced disease, and thus radiotherapy is required.
External-beam therapy can begin after the pyometra is healed. Patients often have
a significant amount of pus in the uterus or a tubo-ovarian abscess without signs
of infection; therefore, a normal temperature and a normal white blood cell count
do not necessarily exclude infection. Repeat physical examination or pelvic
ultrasonography is necessary to ensure adequate drainage.
Cervical Carcinoma After Extrafascial Hysterectomy
When invasive cervical cancer is found after simple hysterectomy, further
treatment is predicated on the extent of disease. Microinvasive disease in
patients at low risk for lymph node metastasis does not require further treatment.
2507Invasive disease may be treated with radiotherapy or reoperation involving a
pelvic lymphadenectomy and radical excision of parametrial tissue, cardinal
ligaments, and the vaginal stump (210).
Reoperation
Reoperation is indicated for a young patient who has a small lesion and in
whom preservation of ovarian function is desirable. It is not indicated for
patients who have positive margins or obvious residual disease (210). Survival
rates after radical reoperation are similar to those after radical hysterectomy for
stage I disease.
Concurrent cisplatin-based chemoradiation is recommended for gross residual
disease, positive imaging, disease in the lymph nodes or parametrium, or a
positive surgical margin; individualized brachytherapy is clearly indicated for a
positive vaginal margin (211).
Radiation Therapy
Survival after radiotherapy depends on the volume of disease, the status of
the surgical margins, and the length of delay from surgery to radiotherapy.
Patients with microscopic disease have a 95% to 100% 5-year survival rate;
the 5-year survival rate is 82% to 84% in those with macroscopic disease and
free margins, 38% to 87% in those with microscopically positive margins,
and 20% to 47% in those with obvious residual cancer (212–214). A delay in
treatment of more than 6 months is associated with a 20% survival rate (214).
Acute Hemorrhage
Occasionally, a large lesion can produce life-threatening hemorrhage. A biopsy of
the lesion should be performed to verify neoplasia, and a vaginal pack soaked in
Monsel’s solution (ferric subsulfate) should be packed tightly against the cervix.
After proper evaluation, external radiation therapy can be started with the
expectation that control of bleeding may require 8 to 10 daily treatments at
180 to 200 cGy per day. Broad-spectrum antibiotics should be used to reduce the
incidence of infection. If the patient becomes febrile, the pack should be removed.
Rapid replacement of the pack may be necessary, and a fresh pack should be
immediately available. This approach to management of hemorrhage in patients
previously untreated is preferable to exploration and vascular ligation. Vascular
embolization under fluoroscopic control may be required in severe cases, and
this procedure may obviate a laparotomy. However, vascular occlusion
ultimately may lead to decreased blood flow and oxygenation of the tumor,
compromising the effectiveness of subsequent radiotherapy.
Ureteral Obstruction
2508Treatment of bilateral ureteral obstruction and uremia in previously
untreated patients should be determined on an individual basis. Transvesical
or percutaneous ureteral catheters should be placed in patients with no evidence
of distant disease, and radiotherapy with curative intent should be instituted.
Patients with metastatic disease beyond curative treatment fields should be
presented with the options of ureteral stenting, palliative radiotherapy, and
chemotherapy. With aggressive management, a median survival rate of 17 months
may be achieved for these patients (215).
Barrel-Shaped Cervix
The expansion of the upper endocervix and lower uterine segment by tumor is
referred to as a barrel-shaped cervix. Patients with tumors larger than 6 cm in
diameter have a 17.5% central failure rate when treated with radiotherapy
alone because the tumor at the periphery of the lower uterine segment is too
far from the standard intracavitary source to receive an adequate
tumoricidal dose (216). Attempts were made to overcome this problem
radiotherapeutically by means of interstitial implants into the tumor with a
perineal template, but high central failure rates were reported with this technique
(217).
One approach is to use a combination of radiotherapy and surgery for
treatment of patients with a barrel-shaped cervix. An extrafascial
hysterectomy is performed 2 to 3 months after the completion of radiation therapy
in an effort to resect a small, centrally persistent tumor. The dose of external
radiotherapy is reduced to 4,000 cGy, and a single intracavitary treatment is
given, which is followed by an extrafascial hysterectomy (218,219). This method
appears to result in a lower rate of central failure (2%), although it is not clear that
the OS rate is improved. There is disagreement concerning the need for
extrafascial hysterectomy, and the GOG is undertaking a randomized study to
compare adjuvant hysterectomy with radiotherapy alone in patients who have no
evidence of occult metastases in the para-aortic nodes (see stages IB and IIA
discussion).
The narrow upper vagina of older patients may preclude the use of an
intracavitary source of radiation. Because of the significant OS advantage
experienced by patients who successfully receive brachytherapy boosts,
placement of brachytherapy applicators with smaller geometry (mini-ovoids, or
tandem and cylinder, or tandem alone) should be strongly considered prior to
switching to an external beam-only regimen. Patients receiving their entire course
of therapy from external beam radiation have a higher central failure rate and
more significant bowel and bladder morbidity. If stage I disease is present in such
a patient, a radical hysterectomy with pelvic lymphadenectomy is preferable, if
2509the patient’s medical condition allows this approach. There may be a role for
IMRT and/or stereotactic radiation boosts in the management of such tumors.
Recurrent Cervical Cancer
Treatment of recurrent cervical cancer depends on the mode of primary
therapy and the site of recurrence. Patients who were treated initially with
surgery should be considered for radiation therapy, and those who had
radiation therapy should be considered for surgical treatment.
Chemotherapy is palliative only and is reserved for patients who are not
considered curable by either surgery or radiation therapy.
Radiotherapy for recurrence after surgery consists primarily of external
treatment. Vaginal ovoids may be placed in patients with isolated vaginal cuff
recurrences. Patients with a regional recurrence may require interstitial
implantation with a Syed type of template in addition to external therapy. A 25%
survival rate can be expected in patients treated with radiation for a postsurgical
recurrence (198).
Radiation Retreatment
Retreatment of recurrent pelvic disease by means of radiotherapy with
curative intent is confined to patients who had suboptimal or incomplete
primary therapy. This may allow the radiation oncologist to deliver curative
doses of radiation to the tumor. The proximity of the bladder and rectum to the
cancer and the relative sensitivity of these organs to radiation injury are the major
deterrents to retreatment with radiation. The insertion of multiple interstitial
radiation sources into locally recurrent cancer through a perineal template or nextgeneration combined intracavitary/interstitial applicator may help overcome these
dosimetric considerations (210,220). The fistula rates can be high, and those
consequences must be considered before interstitial therapy is initiated. For
patients considered curable with interstitial implant therapy, pelvic exenteration is
a better treatment choice. Palliative radiotherapy can be given to patients with
localized metastatic lesions that are deemed incurable. Painful bony metastases,
central nervous system lesions, and severe urologic or vena caval obstructions are
specific indications.
Surgical Therapy
Surgical therapy for postirradiation recurrence is limited to patients with
central pelvic disease. A few carefully selected patients with small-volume
disease limited to the cervix may be treated with an extrafascial or radical
hysterectomy. However, the difficulty of assessing tumor volume and the 30% to
251050% rate of serious urinary complications in these previously irradiated patients
have led most gynecologic oncologists to recommend pelvic exenteration as a last
chance for cure (221,222).
Exenteration
There are three types of exenterative procedures: (i) an anterior exenteration
(removal of the bladder, vagina, cervix, and uterus), (ii) a posterior
exenteration (removal of the rectum, vagina, cervix, and uterus), and (iii) a
total exenteration (removal of bladder and rectum with the vagina, cervix,
and uterus) (Fig. 38-9). A total exenteration that includes a large perineal phase
removes the entire rectum and leaves the patient with a permanent colostomy and
a urinary conduit (infralevator). In selected patients, a total exenteration may take
place above the levator muscle (supralevator), leaving a rectal stump that may be
anastomosed to the sigmoid, thus avoiding a permanent colostomy.
FIGURE 38-9 Pelvic exenteration specimen.
Preoperative Evaluation and Patient Selection
2511It is imperative to search for metastatic disease before the patient undergoes
an exenteration. The presence of metastatic disease in this setting is
considered a contraindication to exenterative procedures. Physical
examination includes careful palpation of the peripheral lymph nodes with FNA
cytologic sampling of any nodes that appear suspicious. A random biopsy of
nonsuspicious supraclavicular lymph nodes is advocated by some clinicians but is
not routinely practiced (174,223). A PET/CT scan of the chest and abdomen and
pelvis CT help in the detection of liver metastases and enlarged nodes. Cytologic
study of any abnormality should be undertaken with CT-guided FNA. If a
positive cytologic diagnosis is obtained, it will obviate the need for exploratory
laparotomy.
Extension of the tumor to the pelvic sidewall is a contraindication to
exenteration; however, this may be difficult for even the most experienced
examiner to determine because of radiation fibrosis. If any question of
resectability arises, exploratory laparotomy and parametrial biopsies should be
offered (224–227). The clinical triad of unilateral leg edema, sciatic pain, and
ureteral obstruction is nearly always pathognomonic of unresectable disease
on the pelvic sidewall. Preoperatively, the patient should be prepared for a major
operation. Total parenteral nutrition may be necessary to place the patient in an
anabolic state for optimal healing. A bowel preparation, preoperative antibiotic
administration, and prophylaxis for deep venous thrombosis with low-dose
heparin or pneumatic calf compression should be undertaken (228). Surgical
mortality increases with age, and the operation should rarely be considered in a
patient who is older than 70 years. Other medical illnesses should be taken into
account. When life expectancy is limited, exenterative surgery is unwise.
Anterior Exenteration
Candidates for anterior exenteration are those in whom the disease is limited
to the cervix and anterior portion of the upper vagina. Proctoscopic
examination should be performed because a positive finding would mandate a
total exenteration. However, a negative proctoscopic examination finding does
not exclude disease in the rectal muscularis, and findings at laparotomy still must
be considered. The presence of disease in the posterior vaginal mucosa directly
over the rectum mandates removal of the underlying rectum.
Posterior Exenteration
A posterior exenteration is rarely performed for recurrent cervical cancer. It
is indicated, however, for the patient with an isolated posterior vaginal recurrence
in which dissection of the ureters through the cardinal ligaments will not be
necessary.
2512Total Exenteration
Total exenteration with a large perineal phase is indicated when the disease
extends to the lower part of the vagina (Fig. 38-9). Because distal vaginal
lymphatics may empty into the nodal basins of the inguinal region, these nodes
should be carefully evaluated preoperatively. A supralevator total exenteration
with low rectal anastomosis is indicated in the patient whose disease is confined
to the upper vagina and cervix (229,230). Samples from margins of the rectal
edge should be obtained for frozen-section evaluation because occult metastases
to the muscularis may occur.
The development of techniques to establish continent urinary diversion
helps improve a woman’s physical appearance after exenteration (231–233).
When both a rectal anastomosis and a continent diversion are performed, the
patient will not have a permanent external appliance. Associated psychological
trauma in such cases may be avoided. Every effort should be made to create a
neovagina simultaneously with the exenteration (234). This procedure helps in
the reconstruction of the pelvic floor after extirpation of the pelvic viscera.
Regardless of whether a neovagina is constructed, it is desirable to mobilize the
omentum on the left gastroepiploic artery to create a new pelvic floor.
Surgical mortality from exenterative procedures has steadily decreased to less
than 10%. Common causes of postoperative death are sepsis, pulmonary
thromboembolism, and hemorrhage. Fistulas of the gastrointestinal and
genitourinary tract are serious surgical complications, with a 30% to 40%
mortality rate despite attempts at surgical repair. The risk for fistula formation is
decreased if nonirradiated segments of bowel are used for formation of the
urinary conduit (228). The 5-year survival rate is 33% to 60% for patients
undergoing anterior exenteration and 20% to 46% for those undergoing
total exenteration (224–234). Survival rates are worse for patients with recurrent
disease (larger than 3 cm), invasion into the bladder, positive pelvic lymph nodes,
and recurrence diagnosed within 1 year after radiotherapy (227). The 5-year
survival rate of patients with positive pelvic lymph nodes is less than 5%; thus,
the performance of an extensive lymphadenectomy in the irradiated field is not
warranted. Discontinuation of the procedure is advisable if any nodes are positive
for metastatic cancer. Patients who have any disease in the peritoneal cavity have
no chance of survival.
Laterally Extended Endopelvic Resection
Locally recurrent cervical cancer in a previously irradiated field is associated with
a dismal prognosis. Exenterative therapy traditionally was reserved for the highly
select patient with centrally recurrent disease, a selection criteria that excludes
most patients with recurrence. A technique called the laterally extended
2513endopelvic resection (LEER) procedure was described, which offers a surgical
treatment option for patients with recurrent disease involving the pelvic sidewall.
The LEER procedure involves extending the lateral resection plane of the
traditional pelvic exenteration to include resection of the internal iliac vessels; the
endopelvic portion of the obturator internus muscle; and the coccygeus,
iliococcygeus, and pubococcygeus muscles. Extension of the surgical plane
allows for resection of lateral tumors with a negative margin. Experience with it is
limited to one center, which reports as high as a 62% recurrence-free survival, but
as high as 70% moderate-to-severe morbidity (235).
Chemotherapy for Recurrent Cervical Cancer
Recurrent cervical cancer is not considered curable with chemotherapy
(236–238). Several clinical trials with various drugs (e.g., carboplatin, cisplatin,
and paclitaxel) showed response rates of up to 45%. Most responses are partial;
complete responses are unusual and limited to patients with chest metastases in
whom the dose of drug delivered to the disease is better than that delivered to the
fibrotic postirradiated pelvis (237,238). The uptake of chemotherapy to recurrent
tumor in a previously irradiated field may be compromised because of altered
blood supply caused by that radiation.
Carboplatin and paclitaxel combination chemotherapy is as effective and
less toxic than the cisplatin and paclitaxel combination (185). Combination
chemotherapy with carboplatin– paclitaxel–bevacizumab may produce better
survival, but it is more toxic than doublet chemotherapy (186). Single-agent
carboplatin is an acceptable alternative in these patients whose disease cannot
be cured.
Palliative Therapy
Palliative therapy for patients with incurable disease consists of radiation or
chemotherapy or both. Palliative radiation therapy is intended to relieve
symptoms of pain or bleeding associated with advanced disease and may be
administered as either external beam therapy (teletherapy) or
brachytherapy. Special care should be given to previously irradiated sites
because additional radiation therapy may be associated with unacceptable
morbidity. Single or multiagent palliative chemotherapy may be used with
variable response rates. Symptomatic recurrent disease within previously
irradiated fields may not respond well to palliative chemotherapy.
VAGINAL CARCINOMA
[6] Primary vaginal cancer is a relatively uncommon tumor, representing
2514only 2% to 3% of malignant neoplasms of the female genital tract. In the
United States, it is estimated that there were 1,312 new cases in 2014, and 430
deaths from the disease (1). Squamous histology accounts for 80% (239).
Primary vaginal cancer should be differentiated from cancers metastatic to
the vagina, which constitute the majority of cancers found in the vagina
(84%) (240).
Table 38-6 FIGO Staging of Vaginal Cancer
Stage I The carcinoma is limited to the vaginal wall
Stage II The carcinoma has involved the subvaginal tissue but has not extended to the
pelvic wall
Stage
III
The carcinoma has extended to the pelvic wall
Stage
IV
The carcinoma has extended beyond the true pelvis or has involved the
mucosa of the bladder or rectum; bullous edema as such does not permit a
case to be allotted to stage IV
IVA Tumor invades bladder and/or rectal mucosa and/or direct extension beyond
the true pelvis
IVB Spread to distant organs
FIGO, International Federation of Gynecology and Obstetrics.
From Beller U, Benedet JL, Creasman WT, et al. Carcinoma of the vagina. FIGO 26th
Annual Report on the Results of Treatment in Gynecological Cancer. Int J Gynecol Obstet
2006;95:S29–S42 and FIGO Committee on Gynecologic Oncology. Current FIGO staging
for cancer of the vagina, fallopian tube, ovary, and gestational trophoblastic neoplasia. Int
J Gynecol Obstet 2009;105:3–4.
Staging
The FIGO staging of vaginal cancer dictates that a tumor extending to the
vagina from the cervix be regarded as a cancer of the cervix, whereas a
tumor involving both the vulva and the vagina should be classified as a
cancer of the vulva.
The FIGO staging for vaginal carcinoma is shown in Table 38-6. Staging is
performed by clinical examination and, if indicated, cystoscopy, proctoscopy,
and chest and skeletal radiography. Information derived from lymphangiography,
CT, MRI, or PET cannot be used to change the FIGO stage; however, it can be
2515used for planning treatment. Less than 30% of vaginal cancers present at stage I
(241–243).
Surgical staging and resection of enlarged lymph nodes may be indicated
in selected patients. FIGO staging does not include a category for microinvasive
disease. Because vaginal cancer is rare and treatment is by radiotherapy, there is
very little information concerning the spread of disease in relation to depth of
invasion, lymph–vascular space invasion, and size of the lesion.
Etiology
The association of cervical cancer with HPV suggests that vaginal cancer has a
similar association (244). A study of 341 cases revealed that in younger patients,
the disease seemed to be related to HPV infection, while in older patients, there
was no association (243). As many as 30% of women with vaginal cancer have
a history of cervical cancer treated within the previous 5 years (186,235–
247). As with cervical cancer, there appears to be a premalignant phase called
vaginal intraepithelial neoplasia (VAIN) (see Chapter 16). The exact incidence
of progression to invasive vaginal cancer from VAIN is not known; however,
there are documented cases of invasive disease occurring despite adequate
treatment of VAIN (248,249).
By convention, any new vaginal carcinoma developing at least 5 years after
cervical cancer is considered a new primary lesion. There are three possible
mechanisms for the occurrence of vaginal cancer after cervical neoplasia:
1. Residual disease in the vaginal epithelium after treatment of the cervical
neoplasia
2. New primary disease arising in a patient with increased susceptibility to lower
genital tract carcinogenesis (the role of HPV in this setting is suspected)
3. Increased susceptibility to carcinogenesis caused by radiation therapy
Screening
Routine screening of all patients for vaginal cancer is inappropriate. For
women who had a cervical or vulvar neoplasm, the Pap test is an important part of
routine follow-up with each physician visit, because these patients are at an
increased lifetime risk for developing vaginal cancer. It is recommended that Pap
test surveillance for vaginal cancer be performed yearly after the patient has
completed surveillance for cancer of the cervix or vulva. For women who had a
hysterectomy for benign disease and have no antecedent history of CIN 2–3,
performance of Pap testing is unnecessary. If the patient has a history of
cervical dysplasia or cervical cancer, yearly screening is recommended. When
2516adjusted for age and prior cervical disease, the incidence of vaginal cancer is not
increased in women who had hysterectomy for benign disease (250). Because
primary vaginal tumors tend to be multicentric, the entire vaginal mucosa should
be considered at-risk. Therefore, in addition to screening cytology, careful
bimanual examination of the entire vagina and vulva is recommended for women
at high risk.
Symptoms
Painless vaginal bleeding and discharge are the most common symptoms of
vaginal cancer. With more advanced tumors, urinary retention, bladder spasm,
hematuria, and frequency of urination may occur. In a large series, 14% were
asymptomatic and diagnosis was made by routine examination or abnormal
cytology (243). Tumors developing on the posterior vaginal wall may produce
rectal symptoms, such as tenesmus, constipation, or blood in the stool.
Diagnosis
The diagnostic workup includes a complete history and physical examination,
careful speculum examination and palpation of the vagina, and bimanual pelvic
and rectal examinations. It is important to rotate the speculum to obtain a
careful view of the entire vagina because posterior wall lesions frequently
occur and may be overlooked.
In early squamous cell lesions, the diagnosis is often suggested by an abnormal
HPV or Pap test result; however, this is not true for clear-cell adenocarcinomas,
which are characterized by submucosal growth. In these cases, the diagnosis is
suggested by cytologic findings in only 33% of cases. Visually suspicious areas in
the vagina should be evaluated with a targeted biopsy using the same instruments
as those used for cervical biopsies. Careful palpation of the vagina may be helpful
in detecting submucosal irregularities. The most common site of vaginal cancer
is in the upper one-third of the vagina on the posterior wall. The developing
tumor may be missed during initial inspection because of obscured visualization
caused by the blades of the speculum (251). Colposcopy is valuable in evaluating
patients with abnormal HPV or Pap test results, unexplained vaginal bleeding, or
ulcerated erythematous patches in the upper vagina. A colposcopically targeted
biopsy may not allow a definitive diagnosis, and a partial vaginectomy may be
necessary to determine invasion. Occult invasive carcinoma may be detected by
such an excision, particularly in patients who have a history of prior hysterectomy
in whom the vaginal vault closure may bury some of the vaginal epithelium at
risk for cancer (252).
2517Pathology
Cancer of the vagina spreads most often by direct extension into the pelvic
soft tissues and adjacent organs. Metastases to the pelvic and para-aortic lymph
nodes may occur in advanced disease. Lesions in the lower one-third of the
vagina may spread directly to the inguinal femoral lymph nodes and the pelvic
nodes (253). Hematogenous dissemination to the lungs, liver, or bone may occur
as a late phenomenon.
Squamous cell carcinomas are the most common form of vaginal cancer,
occurring in 80% to 90% of cases. These tumors most commonly occur in the
upper one-third, posterior wall of the vagina. The mean age of patients with
squamous cell cancer is 60 years (254,255).
Malignant melanoma is the second most common cancer of the vagina,
accounting for 2.8% to 5% of all vaginal neoplasms (256–258). Other histologic
subtypes include adenocarcinoma and sarcoma.
Primary adenocarcinoma of the vagina is rare, constituting 9% of primary
tumors of the vagina. The most common adenocarcinoma of the vagina is
metastatic, originating from the colon, endometrium, ovary, or, rarely,
pancreas and stomach. In general, adenocarcinomas affect a younger population
of women, regardless of whether they were exposed to diethylstilbestrol (DES) in
utero (259). Adenocarcinomas may arise in wolffian rest elements, periurethral
glands, and foci of endometriosis (260). In women exposed to DES in utero,
adenocarcinoma may develop in vaginal adenosis.
DES was used in the United States from 1940 until 1971 to maintain high-risk
pregnancies in women with a history of spontaneous abortions. In 1970, seven
young women were reported with clear-cell adenocarcinoma of the vagina (Fig.
38-10); later, an association between this cancer and maternal ingestion of DES
during pregnancy was identified (261). Subsequently, more than 500 cases of
clear-cell cancer of the vagina and cervix were reported to the Registry for
Research on Hormonal Transplacental Carcinogenesis.
2518FIGURE 38-10 Vaginal clear-cell carcinoma. Note the formation of tubules with hobnail
cells lining the lumen. These cells are characterized by nuclear protrusion into the apical
cytoplasm.
The estimated risk for developing clear-cell adenocarcinoma for an
exposed offspring is 1 in 1,000 or less. The mean age of diagnosis is 19 years
(262). Because the use of DES in pregnant women was discontinued in 1971,
most of these tumors probably have been discovered. Clear-cell adenocarcinoma
in women with a history of in utero exposure to DES typically presented in the
exocervix or anterior, upper one-third of the vagina. These tumors varied greatly
in size and were most frequently exophytic and superficially invasive. Stage is the
most important prognostic factor. Other statistically significant factors included a
tubulocystic growth pattern, size less than 3 cm2, and less than 3 mm of stromal
invasion. It is uncertain, however, what will happen to this cohort of women as
they move into their fifth, sixth, and seventh decades of life. Continued
surveillance of these women is indicated.
Ninety-seven percent of cases of vaginal clear-cell adenocarcinoma are
associated with adenosis. Adenosis is characterized by the presence of persistent
2519müllerian-type glandular epithelium. Although adenosis is the most common
histologic abnormality in women exposed to DES in utero, adenosis can be found
in women without a history of exposure. Adenosis typically appears as red,
grapelike clusters in the vagina.
Malignant melanoma of the vagina is rare and lethal, occurring most often
in white women. The average age of these patients is 58 years (263). Most
lesions are deeply invasive, corresponding to a Clark level IV when compared
with the staging for vulvar melanomas. The most common location of these
tumors is in the lower one-third of the vagina (264). Melanomas have a wide
variety of size, color, and growth patterns (262,265). Radical excision
(vaginectomy, hysterectomy, and pelvic lymphadenectomy) is the mainstay of
treatment. The goal of treatment is to avoid local (vaginal) recurrence, which is
the most common site of recurrence (263,264). The need to dissect regional
lymph nodes is uncertain. Because the disease is deeply invasive, hematogenous
spread is the most common lethal recurrence. There is no difference in OS of
patients with local as opposed to radical excision (263). The survival rate is
approximately 10% at 5 years.
The most common benign and malignant mesenchymal tumors of the
vagina in adult women are smooth muscle tumors (266). Vaginal sarcomas are
usually fibrosarcomas or leiomyosarcomas and are extremely rare. Radical local
excision, followed by adjuvant chemotherapy or radiation therapy, is the indicated
treatment.
The most common malignant mesenchymal tumor of the vagina in
children and infants is botryoid rhabdomyosarcoma (Fig. 38-11). Botryoid
sarcoma is usually found in the vagina during infancy and early childhood, in the
cervix during the reproductive years, and in the corpus uteri during
postmenopausal years. Preoperative chemotherapy with vincristine, actinomycin
D, and cyclophosphamide, followed by conservative surgery or radiation, has
improved survival.
2520FIGURE 38-11 Embryonal rhabdomyosarcoma of the vagina (botryoid sarcoma). This
lesion consists of primitive mesenchymal cells and rhabdomyoblasts, which have abundant
eosinophilic cytoplasm. With further differentiation, cross-striations may become evident.
[6] Treatment
Treatment selection is based on the clinical examination, CT scan results, chest
radiography results, age, and condition of the patient. PET scans may give more
accurate information about disease spread than MRI or CT scan alone (267). Most
tumors are treated by radiation therapy. Surgery is limited to selected cases. These
are as follows:
1. Women with stage I disease involving the upper posterior vagina may be
treated by radical vaginectomy and pelvic lymphadenectomy. If the uterus
is in situ, it is removed as a radical hysterectomy specimen. When margins are
clear and lymph nodes are negative, no additional therapy is necessary.
2. Patients with stage IV disease with either rectovaginal or vesicovaginal
fistula may be candidates for primary pelvic exenteration with pelvic and
para-aortic lymphadenectomy (262). Low rectal anastomosis, continent
2521urinary diversion, and vaginal reconstruction are indicated and are more
successful in these nonirradiated patients than in patients who received prior
radiation therapy.
3. Women with central pelvic recurrence after radiation therapy are
candidates for pelvic exenteration similar to that used for cervical cancer.
4. Surgical staging with resection of enlarged lymph nodes followed by
radiation therapy may improve the control of pelvic disease.
Radiation therapy is the treatment of choice for all patients except those
described previously. Small superficial lesions may be treated with
intracavitary radiation alone (262). Larger, thicker lesions should be treated
first with external teletherapy to decrease tumor volume and to treat the
regional pelvic nodes, followed by intracavitary and interstitial therapy to
deliver a high dose to the primary tumor (255,263). If the uterus is intact and
the lesion involves the upper vagina, an intrauterine tandem and ovoids can be
used. If the uterus was previously removed, a vaginal cylinder may be used for
superficial irradiation. When brachytherapy is used, high- and low-dose rate
techniques were described. If the lesion is more than 0.5-cm thick, interstitial
radiation techniques can improve the dose distribution to the primary tumor.
Surgical exploration or laparoscopy at the time of insertion of Syed interstitial
implants defines more precisely the placement of the needles and ensures that
needles do not pass into adherent loops of bowel. Extended-field radiation may be
used for vaginal cancer in a manner similar to its use for cervical carcinoma,
although there is no experience reported with the use of this technique in the
treatment of vaginal cancer. Likewise, there is little reported experience with
combination chemoradiation treatment (268). Although there will never be
enough patients for a proper randomized control trial, concurrent use of 5-FU and
cisplatin was highly successful in anal and cervical cancer and thus should be
considered for treatment of vaginal cancer.
Sequelae
The proximity of the rectum, bladder, and urethra leads to a major
complication rate of 10% to 15% for surgery and radiation treatment. For
large tumors, the risk of bladder or bowel fistula is significant. Radiation cystitis
and proctitis are common, as are rectal strictures or ulcerations. Radiation
necrosis of the vagina occasionally occurs, requiring debridement, and often leads
to fistula formation. Vaginal fibrosis, stenosis, and stricture are common after
radiation therapy. Use of vaginal dilators and resumption of regular sexual
relations should be encouraged, along with the use of topical estrogen to
maintain adequate vaginal function.
2522Survival
The overall 5-year survival rate for patients with vaginal cancer is 52%
(Table 38-7). This reflects the difficulties of treatment and the fact the disease
presents at late stage. For patients with stage I disease, the 5-year survival rate
is 74%. Most recurrences are in the pelvis, either from enlarged regional nodes or
from large central tumors. Radiation techniques, including interstitial implants
with Syed template and combination chemoradiation, are the mainstay of therapy.
Careful evaluation of patients who receive radiation therapy to detect central
recurrence may allow some patients to be saved by pelvic exenteration. Because
of the rarity of vaginal cancer, these patients should be treated in a center that is
familiar with the complexity of treatment and modalities of therapy.
Table 38-7 Primary Vaginal Carcinoma: 5-Year Survival
Stage No. of Patients No. Surviving 5 Years Percentage
I
509 378
74.3
II
622 333
53.5
III
377 128
34.0
IV
163 24
15.3
Total
1,671 864
51.7
Data compiled from Pride et al. (269); Houghton and Iversen (266); Benedet et al. (242);
Rubin et al. (244); Kucera et al. (265); Eddy et al. (270); Kirkbride et al. (271); Perez et al.
(272); Tewari et al. (273); Otton et al. (274); Frank et al. (275); Hellman et al. (241); Tran
et al. (276).
Nhận xét
Đăng nhận xét