Etiology and Epidemiology

No universal cause for vulvar cancer has been identified. The etiology is likely multifactorial, but some risk factors have been identified. Many patients with vulvar carcinoma have had previous malignant or premalignant lesions of the genital tract.^9–11 Between 2% and 5% of vulvar intraepithelial neoplasia (VIN) 3 lesions (severe dysplasia or carcinoma in situ) will progress to invasive carcinoma.¹² The risk factors for progression to invasive cancer are poorly understood but do include advanced age. One study observed that the relative risk for vulvar or vaginal carcinoma after treatment of cervical carcinoma was 2.9% regardless of whether the patient received radiation for treatment of the cervical cancer. Compromised immune status, whether iatrogenic in the setting of organ transplantation, congenital, or acquired based on human immunodeficiency virus (HIV) infection, leads to an increased risk for cancers in the lower genital tract, including the vulva and vagina. Improvement in immune function with highly active antiretroviral therapy (HAART) may decrease the risk in HIV-infected patients,^13–22 although longer survival with HAART may permit emergence of papillomavirus-associated anogenital malignancies in long-term survivors of HIV infection.²³

The human papillomavirus (HPV) has been implicated as a cause of vulvar cancer. Brinton and colleagues²⁴ established a relative risk of vulvar cancer of 14.6% when there is a history of condylomata acuminata. HPV has been detected widely in sexually active women. The type of HPV is related to the risk for benign versus malignant tumors of the vulva. In patients with condylomata acuminata, the detected HPV is usually type 6 or 11, whereas for vulvar in-situ neoplasia or vulvar carcinoma, HPV 16 and 18 are more common.^25,26 Invasive vulvar cancer and condyloma may be present synchronously in the same patient, delaying diagnosis of malignancy in some patients. Multiple studies have investigated the presence of HPV in vulvar carcinoma.^27,28 It may be detected in both primary tumors and nodal metastases. Review of published data²⁹ employing polymerase chain reaction or hybrid capture assays for HPV DNA detection in 2790 patients with invasive vulvar cancer—1320 patients with VIN 2 and 3, and 71 patients with VIN 1—reveals HPV prevalence in invasive cancer to be 40.1% in invasive lesions, 80.4% in VIN 2 and 3, and 77.7% in VIN 1. HPV 16 serotype was the most common, detected in 29.3% of invasive cancers and 71.2% of VIN 2 and 3 lesions.

Smoking is an additional risk factor for vulvar carcinoma as it is with cervical cancer, and the combination of smoking with a history of genital warts results in a strong increase in the relative risk for vulvar cancer.^11,30 It is possible that smoking and HPV infection have a synergistic effect on the genesis of vulvar cancer, with smoking functioning as a promoter.^31,32

Data are conflicting as to the etiologic role of vulvar dystrophies (lichen sclerosus and squamous cell hyperplasia) in the cause of vulvar cancer. These epithelial alterations may be seen in approximately half of patients with vulvar cancer.^33–35 Patients with vulvar cancer and associated lichen sclerosus are, on average, significantly older than patients with newly diagnosed lichen sclerosus with or without squamous hyperplasia.³⁶ However, less than 5% of patients with these epithelial alterations will progress to invasive vulvar malignancy.

It has been observed that there are two, broadly different groups of vulvar cancer patients. The first consists of younger patients who more frequently smoke and are at higher risk for HPV infection and whose invasive cancers are commonly associated with VIN. The second, and more common, group consists of elderly patients whose disease may be unrelated to smoking or HPV infection and in whom concurrent VIN is unusual but lichen sclerosus and squamous hyperplasia are common associated findings.³⁷

Investigators have examined a number of molecular and genetic parameters in an effort to define prognostic factors in patients with invasive cancer, as well as to identify patients with lichen sclerosus, squamous hyperplasia, or VIN who are at highest risk for development of invasive malignancy.38–45,46 However, at this time, the use of molecular markers has not become a part of routine clinical practice.

Prevention and Early Detection

Because most vulvar malignancies arise from cutaneous surfaces readily accessible to examination, both visibly and palpably, invasive lesions should be diagnosed when small and easily treated. Unfortunately, most advanced vulvar cancers are a result of patient delay or neglect or physician failure to diagnose. Some elderly patients may be mentally impaired. Modesty, anxiety, or frank denial may lead a patient to ignore or minimize symptoms of pruritus, oozing, bleeding, the presence of a mass, and pain. A physician should perform a thorough examination when patients have such symptoms.

No known measures to prevent invasive vulvar cancer exist except for the appropriate treatment of VIN when it is detected. Whether treatment of lichen sclerosus and squamous hyperplasia reduces the risk of subsequent development of invasive cancer is unknown. Pooled data from three prospective clinical trials with mean follow-up of 42 months after initiation of vaccination with quadrivalent HPV vaccine (HPV types 6/11/16/18) show 100% efficacy in prevention of VIN 2 and 3 and vaginal intraepithelial neoplasia (VAIN) 2 and 3 among patients treated per protocol and 79% in an intent-to-treat population (HPV exposed and unexposed).⁴⁷ Whether this will ultimately translate into a clinically meaningful reduction in the incidence of preinvasive and invasive vulvar and vaginal malignancies remains specultative.⁴⁸ At this time, the target population for whom efficacy would seem most likely consists of sexually naive/HPV unexposed adolescents.

A history of smoking tobacco products is associated both with VIN and invasive vulvar malignancy. Patients who smoke should be encouraged to quit, because the statistical risk of anogenital malignancy associated with former smoking is observed to be substantially less than that associated with current smoking and diminishes with increasing time since cessation of smoking.³² Additionally, women with vulvar or vaginal cancer appear to have a fourfold increased risk of subsequently developing a second primary cancer of the lung.⁴⁹

Because VIN 3 may be associated with an occult invasive lesion in as many as 20% of symptomatic patients,⁵⁰ care must be taken when employing treatment techniques that do not include full-thickness histologic evaluation. Excisional techniques, provided that there is minimal loss of functionality and cosmesis, are preferable to ablative techniques without histologic material for study. Core biopsy or excisional biopsy of smaller lesions is suggested for most vulvar abnormalities, if necessary under colposcopic guidance. In circumstances in which there exists a strong clinical suspicion of an invasive cancer, a core biopsy is preferable to an excisional biopsy, in order to preserve the diagnostic accuracy of a subsequent sentinel node procedure. Even if vulvar lesions appear to have a warty appearance and are believed to be benign, confluent warts should be sampled to exclude an underlying diagnosis of squamous cell carcinoma.

Because of the risk for multifocal disease throughout the lower genital tract and anus where vulvar carcinoma is detected, it is appropriate to perform colposcopy of the cervix, vagina, and vulva and digital rectal examination or anoscopy to evaluate for the presence of synchronous lesions at other sites.

Pathology and Pathways of Spread

Pathology

Squamous cancer is the most common malignant lesion of the vulva and constitutes 80% to 90% of vulvar tumors. It may be unifocal or multifocal. Variants of squamous cancer include cancers with discontiguous, infiltrative borders permeating lymphatic spaces with a so-called spray pattern of local invasion^51,52 and verrucous cancers with broad, pushing borders^53,54,55,56 (Fig. 58-1). Sarcomatoid differentiation is seen in a minority of squamous cancers and can be associated with an aggressive course.

Figure 58-1 Variants of squamous cell cancer of the vulva. A, Infiltrative pattern of growth with discontinuous satellites of malignant cells, sometimes called a spray pattern of growth. B, Verrucous cancer with a bulbous, pushing front of invasion into the surrounding stroma.

Photomicrographs courtesy Dr. Richard Oi, Department of Pathology, University of California at Davis.

Poorly differentiated cancers with a trabecular growth pattern manifesting finger-like projections and an infiltrative border with discontiguous islands and nests of malignant cells may spread to regional nodes, even when small, and frequently recur beyond the margins of local excision within dermal lymphatic channels. Recognition of this pattern of growth should prompt consideration of adjuvant radiotherapy.

Verrucous cancers may clinically manifest a warty appearance resembling a cauliflower floret, with exuberant hyperkeratosis microscopically. Broad, bulbous borders characterize these tumors, which rarely metastasize to regional nodes. Morphologic and biologic studies on 10 cases of verrucous carcinoma of the vulva support the theory that verrucous cancers constitute a discrete clinicopathologic subset of vulvar cancer.⁵⁷ Treatment is radical local excision. Despite a tendency for these tumors to recur locally, the role of radiotherapy as adjuvant treatment for this variant remains controversial because of anecdotal reports of clinical radioresistance.

Other, much less common tumors found in the vulva include malignant melanoma, basal cell carcinoma, Merkel cell tumors, carcinoid, transitional cell carcinomas, adenoid cystic carcinoma, vulvar Paget’s disease, and a variety of sarcomas. Tumors metastasizing from other sites may become manifest in the vulva.⁵⁸

Melanomas constitute less than 10% of primary tumors but are the second most common malignancy of the vulva. Even clinicians at tertiary referral centers will see only one or two patients with melanoma of the vulva annually. Melanomas may arise de novo or from preexisting junctional or compound nevi. Rare, amelanotic vulvar melanomas may occur. Diagnostic assessments, staging, and treatment broadly parallel the that for cutaneous melanoma arising at other sites. Prognosis correlates with depth of invasion^59–61 and with the presence or absence of spread to regional nodes. Locoregional therapy is primarily surgical.

Primary malignant tumors of the vulva may arise in the Bartholin complex, and some of these are variants of adenocarcinomas. Adenoid cystic carcinomas uncommonly disseminate to regional nodes and may have an indolent natural history with late local and hematogenous relapse.⁶² Perineural invasion is common. Tumors in the Bartholin complex also may be squamous and are thought to arise in ductal squamous epithelium.

Bartholin’s tumors tend to be deeply infiltrative and solid, often with the overlying skin remaining intact. This clinical presentation may be confused with benign cysts or infection, causing delay in diagnosis and difficulty in accomplishing surgical clearance with wide, negative margins because of proximity to the anorectum and the ischial ramus. Observational outcomes data suggest that patients with Bartholin’s tumors that are more locally extensive are less likely to manifest local failure when postoperative adjuvant RT is administered compared with patients with more favorable disease treated with surgery as monotherapy.⁶²

The International Society for the Study of Vulvar Disease (ISSVD) recommended abandoning the term microinvasion when applied to carcinoma of the vulva and proposed the current use of stage IA, which incorporates criteria for both tumor size and depth. The depth of invasion is defined as the distance from the epithelial stromal junction of the most superficial adjacent dermal papillae to the deepest point of invasion. Stage IA is defined as a solitary squamous cancer of the vulva measuring less than 2 cm in diameter, with clinically negative nodes in which the depth of invasion is 1 mm or less.^63,64 This definition encompasses the previous understanding of microinvasion, because the risk of nodal involvement with invasion less than 1 mm is close to 0%. In addition to tumor stage and depth, other pathologic features of the primary tumor should be noted, including presence or absence of vascular space invasion, tumor grade, and tumor growth pattern.⁶⁵

Pathways of Spread

Vulvar cancer spreads by direct extension to adjacent structures, including the vagina, urethra, perineum, anus, and ischial rami. Tumor may spread by embolization through the lymphatics to the regional nodes. Permeation of in-transit regional lymphatic vessels is uncommon but may account for the occasional patient with recurrence in the skin bridge conserved when vulvar cancer is treated surgically employing separate incisions for the primary tumor and the regional node dissection. This is rare in the absence of extensive groin node metastases.⁶⁶

The lymphatics of the vulva (Fig. 58-2) consist of a network that covers the entire labia minora, fourchette, prepuce, and distal vagina below the hymenal membrane. Lymphatics coalesce anteriorly, forming larger trunks, which run lateral to the clitoris to the mons veneris, acquiring tributaries from the lymphatics of the labia majora, which run in a parallel fashion anteriorly from the perineal body. At the mons veneris, the vulvar lymphatic trunks diverge laterally to the inguinal nodes. Study of the localization of dye or radiolabeled tracer in regional lymph nodes after focal injection of discrete sites in the vulva and on the perineum reveals that the lymphatic drainage of the perineum, clitoris, and anterior labia minora is often bilateral whereas the lymph flow from well-lateralized sites (>2 cm from midline structures) in the vulva is predominantly to the ipsilateral groin.⁶⁷

Figure 58-2 Lymphatic drainage of the vulva and perineum. Arrows indicate the flow patterns.

Adapted from Russell AH: Cancer of the vulva. In Leibel SA, Phillips TL (eds): Textbook of Radiation Oncology, 2nd ed. Philadelphia, WB Saunders, 2004, p 1179.

The vulvar lymphatics run through the vulva and do not traverse the labiocrural fold. Perineal lymphatics course lateral to the labiocrural fold through the superficial tissues of the upper medial thigh. In the radiation treatment of patients with advanced vulvar cancer that extends to the perineal skin, these lateral channels should be taken into consideration. Advanced vulvar cancer extending up the vagina proximal to the hymenal ring may spread through vaginal lymphatics directly to pelvic nodes.

Primary lymphatic drainage is usually to the superficial inguinal nodes with secondary lymphatic drainage through the cribriform fascia to the femoral nodes, with subsequent tertiary flow under the inguinal ligaments to the external iliac nodes. However, metastases have been reported to the femoral lymph nodes without involvement of the superficial inguinal lymph nodes,^68,69 especially from carcinomas of the clitoris and Bartholin’s gland. Sentinel node studies have demonstrated primary lymphatic flow to deep nodes in as many as 16% of cases,⁷⁰ and a Gynecologic Oncology Group (GOG) study also found an unexpectedly high incidence of ipsilateral groin recurrences secondary to presumed involvement of lymph nodes deep to the cribriform fascia despite prior negative superficial inguinal lymphadenectomy.⁷¹ The actuarial risk of groin recurrence at 5 years was 16% (19/119) among patients with vulvar cancer treated with groin node dissection alone at M.D. Anderson Cancer Center, of whom 111 had superficial inguinal node dissections. Of these patients, 117 had groin specimens histopathologically negative for metastases.⁷² Groin recurrence after superficial and deep groin dissection retrieving histologically negative nodes is 2% or less.^73–75

Clinical assessment of groin node status by palpation is notoriously inaccurate, with approximately 20% of groins judged clinically nonsuspicious harboring occult node metastases and approximately 20% of groins considered clinically suspicious for metastases containing only reactive or inflammatory nodes* (Table 58-1). The frequency of lymph node metastases to the inguinal-femoral nodes is related to the depth of stromal invasion^52,81–85 (Table 58-2), defined as the distance from the epithelial stromal junction of the most superficial adjacent dermal papillae to the deepest point of invasion. In Table 58-3 the incidence of occult nodal involvement in clinically nonsuspicious groin nodes treated by complete lymphadenectomy is cross-tabulated by primary tumor size.⁸⁶

TABLE 58-1 Clinical Assessment of Inguinofemoral Nodes by Palpation

Node Status	Histologically Negative	Histologically Positive
Clinically negative (451 patients)	363 (80.5%)	88 (19.5%)
Clinically positive (243 patients)	53 (21.8%)	190 (78.2%)

Pooled data from six institutions; see references 7, 76, 77, 78, 79, 80.

TABLE 58-2 Incidence of Groin Node Metastasis Correlated with Depth of Invasion for Primary Tumors 2 cm or Smaller

TABLE 58-3 Probability of Clinically Occult Metastasis to Inguinofemoral Lymph Nodes Correlated with Primary Tumor Size in Operable Vulvar Cancer

With lateralized primary tumors, metastases to the contralateral groin in the absence of ipsilateral groin metastases may be seen in up to 15% of patients with very advanced primary disease.⁸⁷ Metastases to contralateral groin nodes in the setting of uninvolved ipsilateral groins have been reported in patients with Bartholin’s gland primary disease⁸⁸ and tumors of the anterior labia minora.⁸³ In the absence of spread to ipsilateral nodes, discrete (≤2 cm diameter), well-lateralized primary cancers limited to the vulva and not approaching midline structures manifest spread to contralateral groin nodes in less than 1% of patients undergoing bilateral lymphadenectomy.^{82–85,89–91} However, 5 patients (2.6%) of 192 pooled patients with T1 lateralized cancers managed with ipsilateral groin dissection with no evidence of nodal disease have manifested contralateral groin recurrence.^{71,85,92–96} A theoretic explanation for this discrepancy is the possibility that dissection of negative groin nodes may be therapeutic in a small number of patients in whom routine sectioning of groin nodes (as opposed to pathologic ultrastaging) may fail to detect minute micrometastases consequent to undersampling.

The overall incidence of metastases to the pelvic lymph nodes is 5%. Lymphadenectomy will detect metastatic involvement in pelvic nodes in 15% to 28% of patients with metastases to inguinal nodes.* Pelvic node metastasis is extremely rare in the absence of groin node metastases and very uncommon in the context of occult, microscopic involvement of a single groin node.

Hematogenous spread, usually to lung and bone, is unusual in the absence of prior inguinofemoral lymph node involvement and generally occurs late in the course of the disease. Most patients who die of disease do so with uncontrolled locoregional disease. However, in patients with three or more positive lymph nodes the ultimate risk of hematogenous spread is 66%. In contrast, patients with fewer than three positive lymph nodes have only a 4% risk of hematogenous spread.^102,103

Biology and prognostic factors

In multivariate analysis the presence or absence of lymph node involvement remains the single most important prognostic factor in outcomes of treatment for vulvar cancer.¹⁰⁴ For those with negative lymph nodes, the average 5-year overall survival (OS) from nine large literature series is 91% (range, 83% to 100%).† For those with involved lymph nodes treated with curative intent, the average 5-year OS is 52% (range, 38% to 61%).‡ Ninety-five percent of relapses after primary surgical treatment occur in the vulva, perineum, or groins.^{73,82,109–112} The extent of lymphatic involvement is prognostic and includes volume of tumor in the involved nodes, extracapsular penetration, number of positive nodes, and level of metastatic disease in the nodal chain.^{3,4,28,105,113}

Because of the dominant negative impact of inguinal node metastasis, some investigators only examine primary tumor-related characteristics in relation to their correlation with nodal involvement rather than their impact on risk of recurrence of disease at the primary site.

A number of primary tumor-related characteristics, identified on multivariate analysis, predict for vulvar relapse regardless of nodal involvement, among them tumor size and depth of invasion.⁷⁷ The latter correlates not only with locoregional recurrence but also with the risk of node metastasis. Heaps and colleagues⁶⁵ identified stage, width of surgical margin, depth of invasion, tumor thickness, growth pattern (infiltrative as opposed to pushing), and presence of vascular space invasion as factors predictive of local recurrence after primary surgery.

Three publications^65,114,115 have reported a clear association between risk of vulvar recurrence and width of surgical resection margins. When microscopic margins are 8 mm or less in formalin-fixed tissue, local recurrence has been observed in 43 of 145 patients (30%)^65,114,115 (Table 58-4). These findings have not, however, been consistently observed in all pathology laboratories.¹¹⁶ Allowing for estimated fixation shrinkage artifact between 25% and 45%, this 8-mm pathologic margin correlates with a minimal clinical free space of 1 cm of uninvolved normal tissue, providing a practical guideline for identifying patients who are likely to be advised to undergo postoperative local irradiation for inadequate margins if treated by initial surgery. Under such circumstances, preoperative chemoradiation may be preferable to secure better margins and possibly to reduce the scope of subsequent conservative surgery.

Clinical Manifestations, Patient Evaluation, and Staging

Surgical Trends Including Sentinal Node Studies

Radical surgery for vulvar cancer has moved away from the classic radical vulvectomy with en bloc bilateral inguinal lymph node dissection toward less extensive vulvar and groin surgery, particularly in patients with early-stage lesions. Indicators of this trend include the use of separate vulvar and groin incisions (so-called triple incision) with preservation of the intervening skin bridges; use of less comprehensive groin node surgery coupled with irradiation in patients with gross inguinofemoral node metastases; and pelvic irradiation in lieu of pelvic node dissection superior to the inguinal ligament in patients with confirmed groin node metastases.84,90,139,140 The definition of radical vulvar surgery has changed with the realization that the efficacy of radical surgery may best be defined by the closest resection margin^{65,114,115,141} rather than by achievement of total organ ablation (vulvectomy).

In patients with limited disease and clinically negative groin nodes or occult groin node metastases, utilization of separate incisions for the primary and groins is associated with decreased acute and chronic morbidity without compromise of disease-specific (DSS) or OS compared with en bloc resection.¹⁴² However, when corrected for other prognostic variables in a multivariate analysis, the type of surgical treatment (triple incision vs. en bloc) was an independent predictor for vulvar recurrence but not for inguinal/pelvic relapse.¹⁴³ The lack of impact on DSS and OS may be attributed to high salvage rates for isolated vulvar recurrence reported in some clinical series,^111,144 although 2-year actuarial survival after vulvar recurrence has been reported to be as low as 25% and uncontrolled local recurrence was a contributing cause of death in 15 of 16 patients dying of cancer in one series.¹⁴¹ Risk of recurrence in preserved skin bridges appears related to trapped dermal lymphatic emboli in patients with extensive groin adenopathy and is rare in the absence of groin adenopathy because lymphatic dissemination in vulvar cancer is characteristically embolic rather than permeative.

Radical excision for invasive tumors extends to the level of the deep perineal fascia usually 1 to 2 cm below the skin. Although there is some uncertainty about the optimal extent of surgical margin after wide local excision, data from three clinical series reveal high local control rates for cases in which the surgical margin was greater than 8 mm in the fixed state postoperatively (see Table 58-4). This suggests that, in the nonfixed state, tumor margins of at least 1.0 to 1.5 cm in all dimensions are desirable to achieve high local control rates. Many clinicians endorse a margin of 1.5 to 2 cm.

Although radical local excision is widely accepted in the treatment of small lesions, there is a growing trend to extend its use to lesions larger than 2 cm. Unless preservation of the clitoris and sexual function are priority patient concerns, radical vulvectomy is generally the best surgical option for patients with multifocal invasive disease, patients with invasive cancer associated with extensive VIN, and patients with cancer associated with symptomatic vulvar dystrophy unresponsive to topical therapies.

Effective management of the inguinal nodes is essential because approximately 95% of patients who experience relapse in regional lymph nodes after unsuccessful initial therapy will ultimately die of cancer.* Patients whose primary lesion in the vulva invades to a depth less than 1 mm and is smaller than 2 cm (stage IA) have a negligible risk of groin node metastasis and do not require exploration and management of the inguinal region^52,81–85 (see Table 58-2). In every patient with a primary tumor larger or more deeply invasive (≥ stage IB), surgical treatment must include the ipsilateral groin dissection unless a successful sentinel node study confirms at least one negative sentinel node. For patients with well-lateralized primary tumors, preferably 1.5 cm or more from midline structures, treatment of the contralateral groin may be omitted unless the ipsilateral groin is contaminated.

The move toward more conservative, tailored surgical management has led to changes in the extent of groin dissection and the use of unilateral groin dissection for selected patients. The standard radical bilateral inguinofemoral lymph node dissection, including both superficial and deep nodes, results in as much as a 50% incidence of acute wound breakdown, lymphangitis, or lymphocyst formation and a 10% to 25% incidence of chronic lower extremity lymphedema.^{77,97,134,148} The surgical morbidity associated with limited, superficial groin dissections is generally acknowledged to be less than that associated with superficial and deep dissections.¹³⁹ The rationale for limiting the extent or depth of groin dissection relates to the somewhat orderly sequence of metastatic progression in the groin nodal chain and the observation that involvement of the deep femoral nodes without involvement of the superficial nodes is uncommon. However, the actuarial risk of groin recurrence at 5 years was 16% among vulvar cancer patients treated with superficial groin node dissection alone at M.D. Anderson Cancer Center⁷² and anatomic study of sentinel node location has shown that sentinel nodes will include deep femoral nodes in as many as 16% of patients.⁷⁰ Surgical results might be improved by extending the superficial groin dissection to remove the medial portion of the cribriform fascia and underlying femoral nodes en bloc with the superficial inguinal fat pad.¹⁴⁹ Routine study of sentinel nodes by combined injection of blue dye and lymphoscintigraphy might further reduce the risk of groin failure by identifying the minority of patients with direct lymphatic pathways to deep nodes.^70,121–126

If careful study of sentinel nodes shows no evidence of tumor contamination, full therapeutic groin dissection may be sensibly omitted in patients with discrete, unifocal tumors less than 4 cm in diameter because subsequent groin relapse will be rare. A large international observational study (GROINSS-V, GROningen INternational Study on Sentinel nodes in Vulvar cancer) on sentinel node detection using radioactive tracer and blue dye was conducted in 15 institutions. Team skill verification in sentinel node biopsy technique was required before patient enrollment. Patients with a unifocal primary tumor confined to the vulva and less than 4 cm in diameter with one or more sentinel nodes identified and histologically negative by ultrastaging had their disease treated by vulvar surgery without groin dissection. A total of 623 groins were studied in 403 assessable patients. Among 259 patients meeting criteria for groin observation, six groin recurrences were subsequently diagnosed (2.3%). Short-term morbidity was decreased in patients after sentinel node dissection only, compared with patients who had a positive sentinel node biopsy who underwent complete inguinofemoral lymphadenectomy. Acutely, groin wound breakdown was seen in 11.7% of sentinel node biopsy–only patients versus 34.0% of patients undergoing complete groin dissection. Cellulitis was observed in 4.5% versus 21.3% of patients, respectively. Chronic morbidity also was less frequently observed after removal of only the sentinel node compared with sentinel node removal and inguinofemoral lymphadenectomy. Recurrent erysipelas was seen in 0.4% versus 16.2% of patients, respectively, and lymphedema of the legs was observed in 1.9% of patients undergoing sentinel node biopsy versus 25.2% of patients undergoing full groin dissection. The 3-year DSS rate for patients with unifocal disease and negative sentinel nodes was 97%.¹⁵⁰

A parallel multi-institutional study of the diagnostic accuracy of the sentinel node procedure was conducted in the United States by the GOG (GOG 173), which enrolled 510 patients from 47 institutions beginning in December 1999. Importantly, skill verification in sentinel node biopsy technique was not required before study participation, and patients with primary cancers 2 to 6 cm in diameter were included. Initially the node identification was accomplished by blue dye technique with lymphoscintigraphy optional, but in 2001 the protocol was amended to require preoperative lymphoscintigraphy with intraoperative radiolocalization. After sentinel node retrieval, groin dissection was performed to define the sensitivity (true positive/true positive + false negative), negative predictive value (true negative/true negative + false negative), and false-negative predictive value (1 − negative predictive value) associated with this technique. In preliminary analysis of 411 fully reviewed and assessable patients of whom 129 had groin node metastases, the sentinel node biopsy sensitivity was 89.9%, the negative predictive value was 95.6%, and the false-negative predictive value was 4.4%.¹⁵¹ The false-negative rate was 6.9% for patients with primary tumors less than 4 cm in diameter and 14% in patients with tumors of 4 to 6 cm.¹⁵¹

If one or more sentinel nodes were involved with tumor in the GROINSS-V study and other sentinel node studies, surgery would generally proceed to complete the full therapeutic groin node dissection. A limited retrospective outcome analysis of conservative node debulking followed by RT compared with full groin dissection in patients with gross nodal metastases¹⁵² has shown no obvious differences in the rates of groin relapse or survival. This provides grounds for cautious optimism that limiting groin treatment to removal of only microscopically positive sentinel nodes followed by RT or chemoradiation may provide equivalent control in the affected groin compared with historical results obtained by more comprehensive groin node dissection.

Nodal debulking followed by adjuvant RT or chemoradiation may carry a lower risk of chronic lymphedema and erysipelas than completion of a therapeutic superficial and deep inguinofemoral lymphadenectomy, particularly if complete groin dissection is followed by adjuvant RT because multiple nodes are involved. This alternative is currently the subject of an ongoing observational study (GROINSS-V II). Eligible patients are limited to those with unifocal squamous cancers confined to the vulva that are less than 4 cm in diameter without encroachment on the urethra, vagina, or anus and with clinically and radiographically negative groin nodes. Patients who have successful identification of negative sentinel node(s) undergo no further therapy directed to the groins. Patients with positive sentinel node(s) undergo groin RT or chemoradiation as an alternative to conventional completion of a therapeutic superficial and deep inguinofemoral lymphadenectomy.

A study by Rob and associates⁷⁰ is informative for the determination of radiation target volumes and the design of radiation ports for patients with clinically and radiographically negative groins undergoing elective or prophylactic groin irradiation. Four regions were defined: the superficial medial group located above and medial to the femoral vein and medial to the saphenous vein; the superficial intermediate group located in the vicinity of but superior and lateral to the saphenous and femoral veins; the superficial lateral group located in the outer third of the groin; and the profundum nodes located deep and medially along the femoral vein. Of the 118 sentinel lymph nodes found in 82 groins in this study, 84% were in the superficial medial region or in the superficial intermediate region, and 16% were in the profundum group. Importantly, no sentinel lymph nodes were found in the superficial lateral region. Covering only tissues overlying and medial to the femoral vessels allows substantial reduction in radiation dose to the femoral necks and metaphyseal regions while ensuring that first echelon nodes are comprehensively included. Because insufficiency fractures of the femur constitute a rare but serious complication of groin node irradiation, tailored treatment volumes based on understanding of the relative risk of different node groups within the groin may be helpful in reducing the risk of this serious cause of delayed treatment morbidity.

For patients with small lateralized lesions and no involvement of the ipsilateral nodes, the incidence of contralateral groin node involvement is extremely low and routine dissection of the contralateral nodes is omitted.67,90 Bilateral dissections are recommended for patients whose tumors involve or approach within 1 cm of midline structures or involve the anterior labia minora.

When nodal disease is fixed or ulcerating, the local tumor burden is large and unlikely to be controlled by either surgery or irradiation alone. In such patients, combined therapy with chemoradiation and surgery may offer the best chance of local control.¹⁵³

Radiation Therapy

Radiation therapy for vulvar cancer was pioneered in Europe in the early 20th century in an era when surgical options for advanced vulvar cancer were limited. As radical surgery became more feasible and surgical results improved owing to improvements in anesthesia, postoperative nutritional support, blood transfusion, antibiotics, and refinements in reconstructive surgery and wound care that evolved in response to the carnage of two world wars, primary RT was largely abandoned. Because of the acute desquamation accompanying radical irradiation of the vulva and the severe late effects of radiation on the vulva in long-term survivors, the use of RT was largely restricted to palliative cases and patients with anatomic extent of disease or medical comorbidities precluding operative intervention.^154–157 Radiotherapy equipment and techniques were primitive in comparison to current technology, and radiation fractionation and total dose were naive with respect to potential late normal tissue effects. The perception that RT was a poor second choice as a curative modality stemmed, in part, from the low survival rates reported in early series for patients selected for RT who were unsuitable for primary surgery because of poor general medical condition, advanced disease, or both.

Despite these concerns, the use of irradiation (alone or in concert with concurrent radiopotentiating chemotherapy) in the treatment of vulvar cancer has increased dramatically in the past three decades. Commonly administered as adjuvant therapy before or after conservative surgery, chemoradiation has also emerged as a rational alternative treatment for patients whose extent of disease would require exenterative procedures or other extensively deforming surgical interventions. Two important prospective clinical trials of the GOG (i.e., GOG 37, GOG 101)^98,158 helped form the foundation for these changes in management. Additionally, two randomized prospective trials^159,160 demonstrated the superiority of chemoradiation to RT alone in the treatment of cancers of the anal canal. Extension of chemoradiation to the treatment of patients with vulvar cancer was a natural extrapolation from encouraging reports of successful treatment of anal canal cancer patients with anatomic and functional preservation of the anorectum.

Preoperative Chemoradiation

Several single-institution studies reported encouraging results employing preoperative chemoradiation before conservative resection in patients with locally advanced or locally recurrent vulvar cancer.^167,168 These reports, along with reported treatment successes using chemoradiation for anal and esophageal cancer, provided much of the foundation for protocol work beginning in 1989 in the GOG.

GOG protocol 101 utilized preoperative chemoradiation consisting of 47.6 Gy in once- or twice-daily fractions of 1.7 Gy synchronously with two courses of infusional 5-fluorouracil (5-FU) and cisplatin as initial therapy for 73 patients with extensive squamous cancer of the vulva (1969 FIGO stages III and IVA, American Joint Committee on Cancer [AJCC] 1969 T3/T4, N0/N1), followed by surgical excision of residual primary tumor plus bilateral inguinofemoral lymph node dissection.¹⁵⁸ The dose-fractionation schedule for chemoirradiation employed in GOG 101 is schematically depicted in Figure 58-5, and the target volume irradiated is shown in Figure 58-6. The planned 1.5- to 2.5-week treatment interruption was intended to mitigate the severity of acute radiation dermatitis that can progress to confluent moist desquamation, as well as to permit hematologic recovery. Twice-daily irradiation was employed to optimize the radiation drug synergism. Seven patients did not undergo a post-treatment surgical procedure because of deteriorating medical condition (2 patients); other medical condition (1 patient); unresectable residual tumor (2 patients); and patient refusal (2 patients). After chemoradiation, 33 of 71 (46.5%) patients had no visible vulvar cancer at the time of planned surgery; in the 31 of 33 who proceeded to surgical resection, a pathologic complete response was found in 22 (70%). Of the 38 patients who had gross residual cancer before surgery (53.5%), 5 had positive resection margins and 4 of 5 had either further RT to the vulva (3 patients) or wide local excision and vaginectomy (1 patient). Only 2 of 71 patients (2.8%) ultimately had residual unresectable disease. Urinary or fecal continence was not preserved in only 3 patients. The authors¹⁵⁸ concluded that preoperative chemoradiation is feasible and may reduce the need for more radical surgery, including primary pelvic exenteration.

Figure 58-5 Gynecologic Oncology Group (GOG) protocol 101 for treating advanced vulvar cancer with preoperative chemoradiation. Patients who are judged to have unresectable disease after completion of 47.6-Gy preoperative chemoradiation receive an addition 20 Gy in fractions (Fx) of 1.7 to 2.0 Gy, with a reduced treatment volume encompassing gross residual disease, or they may receive an additional radiation dose by brachytherapy. A third cycle of chemotherapy is recommended if teletherapy is employed. 0, a day without radiation treatment; F, 5-fluorouracil given by continuous intravenous infusion (1000 mg/m²/24 hr for 96 hours on days 1 to 4 of each cycle); P, short infusion of cisplatin (50 mg/m² on day 1 of each cycle); R, a fraction of external radiation encompassing the primary site and regional extensions; a 4-hour minimum intertreatment interval is mandatory, but 6 hours or more is suggested when feasible.

Adapted from Moore DH, Thomas GM, Montana GS, et al: Preoperative chemoradiation for advanced vulvar cancer. A phase II study of the Gynecologic Oncology Group. Int J Radiat Oncol Biol Phys 42:79-85, 1998.

Figure 58-6 Volumes irradiated in GOG protocol 101.¹⁵⁸ Small rectangular volume (dashed lines) was irradiated in patients with AJCC 1969 T3-4, N0-1 disease with clinically uninvolved groin nodes. Larger volume (solid lines) was irradiated in patients with grossly involved groin nodes (AJCC 1969 TX, N2-3).

Adapted from Moore DH, Thomas GM, Montana GS, et al: Preoperative chemoradiation for advanced vulvar cancer. A phase II study of the Gynecologic Oncology Group. Int J Radiat Oncol Biol Phys 42:79-85, 1998.

In parallel with the study of preoperative chemoradiation for patients with locally advanced primary vulvar cancer, the GOG studied patients presenting with advanced disease in the inguinofemoral nodes (1969 nodal classification N2/N3), including some patients with matted, fixed, or ulcerated groin nodes not judged resectable by initial surgery.¹⁵³ The identical preoperative chemoradiation schedule was employed to treat 46 patients except the treatment volume was increased to encompass the groins and caudal pelvic nodes as depicted in Figure 58-6. Three patients died during treatment, and 1 refused to complete the therapy. Four patients who completed chemoradiation did not undergo planned surgery—2 patients dying of non–cancer-related causes and 2 patients with disease that remained unresectable. After preoperative chemoradiation, the disease in the lymph nodes was judged resectable in 38 of 40 patients. Two patients who completed chemoradiation did not undergo surgery because of pulmonary metastasis. One underwent radical vulvectomy and unilateral node dissection and the other had only radical vulvectomy. The operative lymph node specimen was histologically negative in 15 of 37 patients. Nineteen patients developed recurrent and/or metastatic disease. Local control of the disease in the lymph nodes was achieved in 36 of 37 and in the primary area in 29 of 38 patients. Two patients died of treatment-related complications. The authors¹⁵³ concluded that resectability and local control of advanced nodal disease may be obtained with a high probability after preoperative chemoradiation in patients with carcinoma of the vulva with extensive groin adenopathy.

Acute toxicity in GOG 101 was primarily in the irradiated skin, with moist desquamation reported in “many” patients, although this was probably underreported. Major hematologic toxicity was rare. When initial radiation target volume includes the groins and pelvic nodes (see Fig. 58-6) because of confirmed node metastasis, hematologic and gastrointestinal toxicities can be expected to be significantly more severe.

Evaluation for potential surgical resection of residual disease may be performed 2 to 3 weeks after completion of preoperative chemoradiation. If resection can be carried out without compromising functionally important midline structures, this may be preferred management to avoid the chronic sequelae of high-dose radiation on the vulva and perineum. In patients with initially negative groin nodes (palpably and by diagnostic imaging assessment) electively irradiated groin nodes would not be routinely dissected. In patients with initial gross but resectable groin node metastases, limited groin node dissection removing only grossly contaminated nodes may be carried out before initiation of chemoradiation to a volume encompassing the groins. Alternatively, residual palpable nodes may be removed surgically after modest dose (45 to 50 Gy) preoperative chemoradiation, independent of the ultimate decision regarding surgical intervention for the vulvar primary tumor. For patients with residual, unresectable node metastases, limited-volume boost radiation is used to carry grossly involved nodes to approximately 65 Gy with additional concurrent chemotherapy. Similarly, reduced volume boost treatment is administered to gross residual primary disease that has not become resectable by conservative excision. Often this is accomplished through the use of an appositional shaped perineal port employing an electron beam with choice of energy tailored to the expected depth of the tumor bed or low-energy photons to minimize the volume of tissue carried to high dose.

Possible options for further management of patients with a clinical complete response to “preoperative chemoradiation” include generous biopsy to confirm pathologic complete response, conservative surgery to remove palpable scar tissue and the tumor bed, or a modest dose of radiation consolidation (8.5 to 9.0 Gy/5 fractions/week).

Preoperative neoadjuvant chemotherapy (without irradiation) has been used in the treatment of locally advanced vulvar cancer.^169–171 Response rates have been quite variable; and although complete clinical or pathologic responses have been recorded,^170,171 these are very rare. This approach has not gained broad acceptance, because response to combined chemoradiation is more reliable and predictable.

Treatment Algorithms and Approaches

The initial management decision for vulvar cancer is based on clinical features, including the extent of the primary tumor, presence or absence of detectable node metastases, as well as the age, comorbidities, and preferences of the patient. The need for adjunctive therapies is determined by histopathologic findings. Management of the nodes and the primary tumor is largely based on the clinical characteristics of each entity considered independently.

Because there exist no precise definitions of what constitutes “favorable” or “early” disease versus locally advanced disease, clinical definitions are provided to outline the principles that guide therapeutic choices.

Early, Favorable or Low-Risk Disease

A practical definition of early, favorable (low risk) disease is disease that can be reasonably expected to be curatively managed by conservative surgery alone. The extent and location of the primary tumor are amenable to radical excision without sacrificing the function of the urethra, the anorectum, the clitoris, and vagina. Generally this will include patients without palpable or imaging evidence of inguinal node metastasis. For low-risk cancers that are unifocal and well lateralized, radical wide local excision of the primary lesion in the vulva is recommended for most patients. The reported results from observational studies of radical wide local excision for T1 or T2 tumors suggest an average survival rate of approximately 90%. The criteria proposed for the selection of patients for such conservative surgical resection vary but in general include patients with tumors less than 2 cm in diameter and depth of invasion less than 5 mm.^{71,90,92,139,199}

The role and indications for postoperative adjuvant vulvar irradiation are ill defined. Although data have clearly demonstrated that surgical margins less than 8 mm, increasing depth of invasion, presence of capillary lymphatic space involvement, and increasing size of primary cancer portend increased risk for vulvar recurrence after excision of any extent, it is still unclear to what extent postoperative adjuvant irradiation can overcome these risks.

Although wider excision, if feasible, may for some patients eliminate the risk of close margins, it may not be feasible or advisable for medial or deep margins and will not overcome other risk factors. One retrospective series of 62 patients with positive or close (<8 mm) margins showed, on multivariate analysis, that adjuvant irradiation significantly reduced local recurrence rates and significantly improved survival rates for the positive margin group.¹⁴¹ The role of adjuvant RT after radical wide local excision of vulvar cancer was being explored in a Gynecology Oncology Group randomized clinical trial, but the trial had to be abandoned because of lack of accrual.

Historically, standard management of the inguinal regions for all patients with invasive disease of larger than 2 cm or greater than 1 mm depth of invasion (FIGO stage IA) has been groin dissection. Results from the GROINSS-V study suggest that in selected patients with unifocal primary tumors less than 4 cm without encroachment on the urethra, vagina, or anus and with successful retrieval of sentinel node(s) that are negative by pathologic ultrastaging, the acute and chronic morbidities of groin dissection may be avoided. The risk of subsequent groin relapse is low, and, with careful observation, isolated node recurrence may rarely (anecdotally) be salvaged by secondary treatment of an undissected groin. However, the majority of patients with relapse of the disease in the groin will die of vulvar cancer.* If this approach is adopted, pretreatment nodal assessment by CT, MRI, or ultrasonography should be performed to exclude the presence of bulky metastatic nodes and minimize the risk of false-negative sentinel node studies.

Postoperative RT to the inguinal and pelvic nodes is recommended for patients with involvement of two or more nodes and for those with a single node with macroscopic metastasis (>3 mm) or extranodal extension. In patients with diagnostic imaging (preferably PET/CT) that shows no evidence of pelvic adenopathy, the superior border of the treatment volume should probably extend no more cephalad than the inferior or middle of the sacroiliac joints in order to limit acute and chronic enteric morbidity. The randomized study from which these recommendations are largely drawn (GOG 37) administered adjuvant radiation to both groins even when involvement was unilateral.⁹⁸ The benefit of adjuvant RT to the contralateral negative groin is uncertain, and treatment of that volume may be unnecessary. Two issues remain controversial. Whether residual vulva and tumor bed should routinely be included in the treatment volume when groins and pelvic nodes require irradiation remains unclear, although histopathologic features associated with the primary tumor may suggest that this is prudent in many if not all patients. Whether concurrent chemotherapy should be administered with purely adjuvant treatment to further enhance the regional efficacy of irradiation is also the substance of conjecture and debate. Hypothetically, the increase in acute toxicity could be offset by improved regional control, possibly with a reduced dose of irradiation that might translate into diminished late radiation effects. This hypothesis remains untested. Algorithms for management of patients with early, lateralized disease found in Figure 58-3 and Figure 58-7 are premised on the presence of a team skilled in the performance of sentinel node biopsies or the availability of PET/CT. Absent either capability, radical wide local excision and at least ipsilateral inguinofemoral lymphadenectomy is recommended. Predicated on histopathologic findings from study of the excised groin nodes and the primary tumor, subsequent management is suggested by the algorithms in Figures 58-7 and 58-8.

Figure 58-7 Algorithm for groin management of limited, unifocal primary cancers of the vulva less than 4 cm with clinically and radiographically uninvolved groin nodes. Experienced sentinel node team available.

Figure 58-8 Algorithm for management of limited, unifocal, lateralized primary tumors of the vulva less than 4 cm and ≥1.5 cm from midline structures.

Small Primary Tumors Encroaching on Midline Structures

Although surgical excision remains the historical standard treatment for patients with small, midline primary tumors, it is not unreasonable to consider treating a primary midline lesion, regardless of size, with a localized field of primary irradiation or chemoradiation if surgical excision would result in predictable loss of cosmesis or function. Although such patients can be cured with surgery as monotherapy, the functional and cosmetic impairments can be psychologically devastating. In many patients, clinical shrinkage after a modest dose (45 to 54 Gy) of irradiation or chemoradiation will permit conservative excision of residual tumor with excellent preservation of function and cosmesis (Fig. 58-9).

Figure 58-9 Clinical results after combined-modality therapy for a small, T2N0, anterior right labia minora primary tumor invading the clitoris. The images were obtained at a follow-up examination 18 months after treatment that consisted of 45 Gy given in 25 fractions to a volume encompassing the primary tumor and electively including the groin nodes bilaterally with two courses of concurrent infusional 5-fluorouracil and bolus cisplatin. Conservative excision of residual tissue spared the clitoris and most of the vulva. Images were obtained with the labia apposed (A) and with the right labium majorum retracted, revealing the excision scar (B).

To use primary RT or chemoradiation to anatomically spare the clitoris remains controversial, not because of a lack of efficacy for irradiation (which has been proven valuable in more locally advanced disease) but rather because no systematic studies of subsequent sexual function and morbidity have been performed. Anecdotally, clitoral function can be preserved in some sexually active patients after treatment with 45 Gy combined with 5-FU–based chemotherapy supplemented by clitoris-sparing local excision. Such patients are sexually active and orgasmic before treatment, are sexually active and orgasmic 12 or more months after completion of treatment, and subjectively report intact sensation, vascular engorgement, and erectile clitoral function. Whether this is also feasible after higher doses of irradiation is unknown.

Locally Advanced Nodal Disease

When groin nodes are ulcerating or fixed, initial treatment of the inguinal regions consists of primary RT or chemoradiation. Because it is unclear whether chemoradiation in tolerable doses can predictably eradicate bulky nodal disease, if there is substantial regression of nodal disease it is reasonable to subsequently operate to remove residual adenopathy.

Patients with initially bulky resectable involved nodes may undergo either initial surgical removal of the bulky nodes followed by postoperative RT or chemoradiation or initial RT or chemoradiation followed by operative removal of any residuum. The optimal sequencing of therapies remains controversial, because surgery will delay initiation of RT and because preoperative irradiation may compromise wound healing. Commonly, patients with bulky inguinal nodes will also have locally advanced primary tumors, in which case initial RT or chemoradiation is a logical sequence choice. Conversely, initial gross adenopathy may be difficult to reliably identify and excise after preoperative chemoradiation, and a good case can be made for initial surgical removal as depicted in the algorithm in Figure 58-10. If there is clinical uncertainty whether all gross adenopathy can be removed from the groins, imaging with ultrasonography, CT, or MRI may aid in elucidating the spatial relationships between blood vessels, nerves, bone, and metastatic adenopathy. These imaging modalities may be particularly useful in defining disease extent in large patients.

Regardless of the sequencing of therapies, groin surgery should be limited to removal of gross adenopathy rather than an en bloc removal of all groin lymphatic tissue. It should be accomplished through the smallest feasible incision to accomplish this objective. The intent of such surgery is to reduce groin cancer volume to a microscopic, occult burden that can be controlled by modest doses of fractionated radiation. Conversely, radiation dose administered either before or after surgical cytoreduction should generally be restricted to an effective dose for controlling microscopic or occult disease. By limiting the extent of groin dissection in anticipation of subsequent radiation, and by limiting radiation dose in anticipation of surgical removal of residual, bulky disease, some acute complications (wound dehiscence, infection) and late sequelae (lymphedema, femoral fracture, arterial occlusion) might be reduced in severity or avoided altogether. One retrospective series describing outcomes after debulking groin surgery and radiation compared with complete groin dissection and RT reported no difference in groin recurrence-free survival or OS.¹⁵²

Locally Recurrent or Metastatic Cancer

The outcome for patients who experience relapse of vulvar cancer after primary therapy is markedly dependent on the site of relapse. Although salvage rates for isolated vulvar recurrences are substantial, relapses in groins or pelvic nodes or hematogenous metastases are almost uniformly fatal, regardless of salvage modalities employed. For patients with distant metastases, systemic chemotherapy with agents active against squamous cancer is often offered but response rates are disappointing and response durations are generally measurable in a few months.^{111,202–211}

Agents that may be useful when given concurrently with radiation (5-FU, cisplatin, mitomycin-C) have limited efficacy when given alone. Inferential evidence for this statement includes patients with regionally extensive vulvar cancer treated by irradiation and synchronous cytotoxic chemotherapy who manifest prompt and marked response within the irradiated volume while simultaneously developing overt metastatic disease outside the irradiated volume. An example is illustrated in Figure 58-11.

Figure 58-11 Progression of cancer as biopsy-confirmed dermal metastases outside the target volume during chemoradiation in a patient with extensive locoregional cancer of the vulva responding favorably at the primary site and groin nodes.

Inguinal recurrences may be amenable to palliative management with irradiation or chemoradiation to provide local control. Few patients, however, survive long term.^{73,111,112,146,147}

Local vulvar recurrence appears to be related to the previously described risk factors and, in particular, to surgical resections with margins less than 8 mm pathologically (in formalin-fixed tissue). Several reports have suggested that up to 75% of patients with vulvar recurrence of cancer may be cured, usually when treated with surgical re-excision.^73,212 Boronow and colleagues¹⁶³ reported a 5-year OS of 63% with modest dose radiation. However, Faul and associates¹⁴¹ reported only a 25% 2-year actuarial survival after vulvar recurrence. Radiation therapy or chemoradiation for local recurrence is an option, particularly when the lesion is not amenable to repeat resection without predictably inadequate surgical margins or compromise of functionally important midline structures.^168,184

Because vulvar neoplasia is often a field defect, new primary cancers often develop in sites remote from the initial tumor and may be inappropriately labeled vulvar recurrences. It is important, if possible, to differentiate a new primary tumor from a recurrence; the latter would attribute a treatment failure to the original treatment and lead to an underestimate of its efficacy.

Groin and Nodal Irradiation

GOG protocol 37,⁹⁸ which established the usefulness of postoperative adjuvant groin and pelvic irradiation in patients with metastases to inguinofemoral nodes, used external beam radiation therapy (EBRT) bilaterally to the inguinal and femoral dissection beds and pelvic nodes even when node involvement was unilateral. The specific radiation doses in the study (45 to 50 Gy in 1.8- to 2.0-Gy fractions) are in the range commonly prescribed to control occult microscopic disease. The technique utilized an anterior and posterior parallel, opposed pair of fields. The inferior border of the volume recommended was determined clinically using a margin lying parallel and 6 to 8 cm inferior to the inguinal skin crease or ligament. The GOG study used a superior border somewhat arbitrarily placed at the L5-S1 interspace, although the value of extending the field to this level superiorly is uncertain because the observed survival gains appear to be related to improved control of the inguinal regions and not to improved control of the pelvic nodes.¹⁴⁰ Treating this volume with this technique necessarily will limit the radiation dose to 45 to 50 Gy because of normal tissue dose limitations (bowel, femoral necks, pelvic girdle).

Involved nodes located high in the pelvis or in the common iliac region connote a high probability of extrapelvic failure; it is unclear that irradiation of higher nodes or a conventional, entire pelvic volume would significantly contribute to cancer control or survival. To reduce morbidity, it seems reasonable to define nodal target volumes based on the principle of electively irradiating the next proximal echelon of lymph nodes to those clinically involved. For patients with unilateral node metastases to inguinofemoral nodes without clinical or radiographic evidence of pelvic node metastases, the volume would be limited to obturator and external iliac node groups on the affected (ipsilateral) side. Irradiation of hypogastric nodes and the contralateral external and obturator areas is more likely to add morbidity than clinical benefit and rarely will be indicated unless there is significant extension of the primary cancer proximal to the hymenal ring in the vagina or invasion of the urethra or rectum.

The patient can be treated in the supine position with the feet apart or in the “frogleg” position with leg supports to minimize skin reaction in the medial thigh. The central residual vulvar tissue may or may not be included, depending on risk factors associated with the primary tumor and institutional philosophy. The lymphadenectomy scar usually lies within the limits normally used for the lateral margins of this volume; however, the necessity for specifically encompassing the inguinal lymphadenectomy scar is unclear because the incidence of scar recurrence is unknown. Depending on beam energies employed, bolus may be applied to lymphadenectomy scars. The medial two thirds of distance between the origin and insertion of the inguinal ligament (i.e., between the pubic tubercle and the anteroinferior iliac crest) should be included unless surgical clips indicate more lateral nodal involvement.

For patients receiving nodal irradiation only, with no indications for vulvar irradiation, midline structures may be spared superfluous treatment. It is important to confirm clinically that the lateral margins of the excluded central tissue do not encroach on the medial aspects of the inguinal node bed and obturator nodes, resulting in underdosage. In general, the shielded midline tissue should be no wider than the distance between the pubic tubercles, because it is intended only to reduce the dose to the distal urethra, vagina, and anorectum and residual vulva. Use of a midline block has been reported to be associated with a high risk of local recurrence in the shielded midline. In a series of 26 patients administered radiation postoperatively for positive groin nodes who had midline shielding, 13 patients (50%) had local recurrence.¹⁶² A midline block or other techniques to exclude midline structures should be used only if risk factors for local recurrence are minimal.

For patients with clinically and radiographically negative groin nodes in whom elective nodal irradiation or chemoradiation is to be used without subsequent surgical dissection, the techniques and volume of radiation are similar to those described for adjuvant therapy except that the superior border of the target volume will be selected to provide secure coverage of the nodes lying underneath the inguinal ligament. Pelvic nodes will not be included unless the primary tumor invades the vagina, urethra, or anorectum. In general, elective groin irradiation will be done when preoperative irradiation or chemoradiation is needed for a locally advanced vulvar primary tumor, and thus both groins should be included. The nodal volume of interest including the deep femoral node depth should be confirmed by CT or ultrasonography. In general, such patients will have radiation treatment planned with CT simulation. Conventionally, the prescription depth for the inguinofemoral nodes is defined as the depth of the femoral artery as it passes beneath the inguinal ligament. Pelvic nodes obviously lie at greater depth. The tumor dose delivered must be substantially increased for patients with fixed or ulcerating (unresectable) lymph nodes. Although the optimal dose is not defined, a dose of 64 to 66 Gy in combination with concurrent chemotherapy may be necessary to provide the highest probability of local control while remaining within the tolerance of adjacent normal tissues.

A recognized complication of groin irradiation is femoral neck fracture.^72,213 Particularly at risk are older patients who may have some degree of osteoporosis and are already vulnerable to falls and fractures. Techniques for reducing femoral neck dose include partial central transmission block technique in which dose to the deep groin nodes is prescribed through a large anterior photon field employing low-energy photons of 4 to 6 MV. Various beam arrangements may accomplish dose delivery to deeper pelvic nodes and the vulva (e.g., use of a smaller, opposed posterior pelvic field using photons of higher energy). A central partial transmission block is placed over the anterior field corresponding to the narrower posterior field that attenuates the contribution of the anterior field to the deep pelvic nodes and vulva, allowing better dose homogeneity. If inguinal nodes are superficial, the dose may be further supplemented by administering a portion of the treatment with anterior appositional high-energy electrons. When abutting photon fields or photon and electron fields to encompass nodal tissue overlying the femoral necks there is considerable potential for “hot” and “cold” spots. Hot spots may injure the femoral arteries as well as the femurs, and cold spots hazard disease recurrence within the clinical target volume. Moving the field abutments once or twice during a course of treatment will serve to diminish these hazards.

Technical alternatives include the use of intensity-modulated radiation therapy (IMRT)176,214 or proton beam therapy. Patients appropriate for the very conformal dose distributions that can be achieved with these alternatives should preferably be cooperative and amenable to very precise and reproducible daily setup because the potential for marginal miss is amplified by the tight tolerances required by these modalities and the somewhat longer treatment times associated with some IMRT techniques. Depending on extent of tumor both at the primary site and at the groins, the clinical target volume may be specified to either exclude, or intentionally include, skin of the vulva, adjacent perineum, and overlying the groin nodes. With IMRT, a somewhat greater degree of dose inhomogeneity within the treatment volume is part of the price paid for more conformal therapy. Use of either IMRT or protons will result in marked reduction of normal tissue volumes carried to radiation doses that can contribute to both acute side effects as well as late complications, including small and large bowel, bladder, femoral neck/head, and skin. Essential to successful application of these modalities are both equipment and a sophisticated medical dosimetry and medical physics team functioning with rigorous quality assurance. CT simulation is a critical element in defining both target volumes and avoidance structures. An illustrative IMRT axial dose distribution for a patient with a locally extensive vulvovaginal cancer in whom groin nodes were electively irradiated is seen in Figure 58-12.

Figure 58-12 IMRT dose distribution for a patient with a posterior vulvovaginal cancer encroaching on the anal sphincter for whom the target volume included elective treatment of clinically and radiographically uninvolved inguinofemoral nodes.

Termed a modified segmental boost technique, use of right and left anterior oblique, partially beam-split photon fields to supplement groin node dose is a technique suitable for any department with an isocentric linear accelerator with appropriate beam energies.215,216 With this technique, the oblique fields are angled so the medial border of each anterior oblique field corresponds to the beam divergence of the lateral border of the posterior field, minimizing the potential for hot and cold spots. The tumor dose must be brought to the skin surface if there is evidence of cutaneous involvement. Bolus material may be required. Axial dose profiles employing various techniques encompassing groin nodes are depicted in Figure 58-13. The magnitude of hot spots at the boundaries of treatment fields is shown in Figure 58-14.

Figure 58-13 Treatment technique for irradiation of the pelvis and inguinal nodes. Beam arrangements for two techniques: segmental boost technique (A) and modified segmental boost technique (B). AP, anterior-posterior; LA, left anterior; LAO, left anterior oblique; PA, posterior-anterior; RA, right anterior; RAO, right anterior oblique.

Data from Moran M, Lund MW, Ahmad M, et al: Improved treatment of pelvis and inguinal nodes using modified segmental boost technique. Dosimetric evaluation. Int J Radiat Oncol Biol Phys 59:1523-1530, 2004.

Figure 58-14 Comparative dose profiles for various techniques employed to treat the pelvis and groin nodes. Color rendering of axial dose distribution generated from vertical film irradiation in a plastic-water (PW) phantom using five different techniques for treating pelvis and inguinal nodes. High-dose areas are visible in field overlap regions. A, Segmental boost technique. B, Modified segmental boost technique. C, Partial transmission block technique. D, Photon with electron tag technique. E, Photon with electron boost technique.

Data from Moran M, Lund MW, Ahmad M, et al: Improved treatment of pelvis and inguinal nodes using modified segmental boost technique. Dosimetric evaluation. Int J Radiat Oncol Biol Phys 59:1523-1530, 2004.

It should be understood that selection of target volumes, choice of treatment technique, and prescription of radiation dose and fractionation should be individualized based on tumor factors, patient comorbidities, availability of equipment, and sophistication of medical dosimetry support. An example of individualization of target volume, portal design, and technique is depicted in Figure 58-15.

Figure 58-15 Simulation digitally reconstructed radiograph for treatment of a patient with a lateralized T3N1 squamous carcinoma of the left labium majorum and minorum with metastases in two medial left groin nodes based on fused PET/CT. The volume encompasses the clinically negative contralateral groin nodes medial to the femoral vessels, which have been reconstructed from CT images. On the involved side, the treatment volume is extended cephalad to encompass the obturator and caudal external iliac nodes while shielding central pelvic soft tissues. The left lateral groin has a portion of the treatment delivered by an anterior oblique field using modified segmental boost technique, with the medial beam border corresponding to the left lateral border of the posterior field.

Adjuvant Postoperative Irradiation of the Vulva

When adjuvant irradiation to the vulva is indicated because of risk for local recurrence, an appositional, en-face perineal portal may be used with the patient in the dorsal lithotomy position. Although often requiring daily physician verification of treatment setup, this permits precisely tailored volumes to be treated reproducibly. Just as surgical concepts are changing to limit radical excision to the site of the primary cancer rather than extirpation of the entire vulva, radiation volumes are also being modified. A localized field of radiation may be used to encompass the vulvar scar with an appropriate margin. This allows maximal sparing of uninvolved vulvar tissues from acute and late radiation reactions. When therapy is adjuvant, irradiation alone employing doses of 45 to 54 Gy is recommended, depending in part on whether a margin is narrow or grossly contaminated, as well as on the health of the surrounding tissues. The role of chemoradiation in this adjuvant setting is unestablished. The beam energy should be tailored to the estimated depth or thickness of the original lesion from the vulvar surface, with appropriate use of bolus to deliver the prescribed tumor dose to the skin of the vulva. Electron beam or a low-energy photon beam is employed. If the target volume in the vulva is localized and can be encompassed using a perineal vulvar port, even if there is also an indication to treat the regional nodes, the use of separate fields to the groin region and vulva may spare a significant volume of the vulva from the radiation effects. Care must be exercised to establish the location of the exit beam from the perineal port and ensure safety with respect to possible overlap with the nodal irradiation volumes. Alternatively, anterior and posterior photon fields may be sensible when insecure margins are treated with less than precise definition of where the at-risk tissue lies.

Preoperative or Definitive Vulvar Irradiation

When initial irradiation is selected as primary management of a small or early midline vulvar primary, a technique may be used similar to that described for adjuvant treatment (i.e., a perineal port tailored to the depth and thickness and size of the lesion) if the groin nodes will not be treated. For advanced large tumors, the vulvar primary tumor may be treated with an anterior-posterior parallel opposed pair of fields. More commonly, the target volume will include the groin nodes bilaterally. The majority of vulvar cancers are radioresponsive, and visible response to radiation is often apparent after a tumor dose of 30 to 36 Gy in fractions of 1.8 to 2.0 Gy.

Because available dose-control data for vulvar cancer are scant and seldom correlated with the volume of disease treated, and because it is unproven whether the addition of concurrent chemotherapy will modify the total dose required to sterilize disease, dosage guidelines are imprecise. In general, concurrent chemotherapy (weekly cisplatin, infusional 5-FU with or without cisplatin or mitomycin C) is recommended for definitive treatment. If primary tumors shrink and become resectable, a preoperative dose of 45 to 57.6 Gy in fractions of 1.8 Gy should suffice. For medically inoperable patients or primary tumors that remain unresectable by conservative surgery, tumor doses should be escalated to 63 to 72 Gy. Total dose will be contingent on the fractionation employed, the elapsed time to complete treatment, and whether concurrent chemotherapy is employed.

Because most patients are thicker in the low pelvis where the central axis dose is frequently calculated when parallel opposed fields are treated, dose per fraction may be significantly higher at the vulva owing to reduced tissue attenuation of the beams. This will also depend significantly on beam energies. Wedges and compensators may be employed to minimize the impact of this effect. Thermoluminescent dosimeters placed on the tumor and in the cleft between the labia are a simple mechanism to confirm actual absorbed dose at the vulva. This permits tailoring of compensatory measures should the dose inhomogeneity be excessive, which is sometimes the case in large patients.

Although planned split-course irradiation is believed to be undesirable because of the concern for increased tumor proliferation with protracted overall treatment times, most patients require a short rest to allow epithelial regeneration before completing a planned definitive course of chemoradiation to the vulva. The treatment break begins when confluent moist desquamation with its accompanying pain and discomfort occurs (usually after 25 to 35 Gy). The therapy may be restarted where epithelial islands are visible; healing is then usually rapid, and repeated desquamation is rare. Twice-daily fractionation during reduced-volume boost irradiation to gross disease is an attractive way to reduce overall duration of treatment when treatment interruption has been necessitated by acute skin toxicity. Because volumes are small, this can be accomplished without unacceptable acute radiation toxicity.

Interstitial radiation implants have been used by some, particularly for boosting radiation to bulky disease or when there is extension to the vagina.^217,218 No randomized data are available to evaluate the relative benefits and toxicities of interstitial versus external boost irradiation. A potential pitfall of interstitial techniques is difficulty in achieving dose homogeneity and potential increased late effects consequent to the unavoidable “hot spots” in tissues in direct contact with interstitial sources.

Sequelae of Irradiation

When adjuvant or definitive vulvar irradiation is used, the full tumor dose needs to be delivered to the skin and mucosal surfaces of the vaginal introitus. Even when doses of 45 to 54 Gy are given in the adjuvant setting, it is expected that almost all patients will experience some degree of desquamation. This may vary from patchy, moist desquamation to confluent, moist desquamation over the entire treatment volume. In general, this reaction heals in the 2 to 3 weeks after completion of radiation therapy. The reaction is usually managed with symptomatic measures, including frequent sitz baths, exposure to air, application of topical ointments such as those used to treat superficial burns, and pain medication. In some situations, particularly in elderly, frail patients, hospitalization for skin care may be required. Depending on the volume treated, acute effects may include urethritis, cystitis, and proctitis. Watery diarrhea is unusual unless a full pelvic volume is irradiated, which is rarely appropriate.

The late effects of RT may include skin changes of atrophy, fibrosis, or telangiectasia, which, in small volumes, are not usually functionally distressing. The development of mucosal or perineal ulcers is rare. Late effects on the vulva appear to be minimized by limiting the radiation fraction size to 1.6 to 1.8 Gy and the volume of reduced target boost therapy. Inclusion of the anorectum in boost volume treatment may result in chronic rectal bleeding (radiation proctitis) or fibrosis and stricture of the anal canal.

Irradiation of the node-bearing areas may result in similar acute effects if full doses are required at the skin because of skin infiltration or ulceration or when superficial groin nodes lie only millimeters below the skin in very thin patients. Acute effects may vary from erythema to moist desquamation, particularly along skinfolds.

The contribution of RT to the problem of leg edema induced by groin surgery is unknown. One series suggested that, after concurrent chemotherapy and irradiation, 4 of 14 surviving patients treated for advanced primary or recurrent disease experienced leg edema. In 2 patients this edema was minimal and did not require treatment. Two patients did require treatment with compression and diuretics. Both had an inguinal lymphadenectomy before chemoradiation.¹⁸⁴ Clearly, the magnitude of the surgical intervention in the groins (sentinel node biopsy vs. complete groin dissection) has a major impact on the risk of lymphedema (GROINSS-V).¹⁵⁰ The extent of lymphadenectomy (gross total removal vs. en bloc complete lymphadenectomy) does not appear to have a major impact on either disease control in the groin or survival when coupled with groin radiation.¹⁵² In the interests of minimizing the risk and severity of subsequent lymphedema and cellulitis, it seems prudent to limit groin dissection to removal of only grossly contaminated nodes when planned RT or chemoradiation will follow.

Radiation therapy to the groins may induce femoral neck fracture.^72,213 The risk of hip fracture may not correlate well with the dose delivered to the hips and may be more frequent in frail, elderly women with underlying osteoporosis. Based on the anatomic study of sentinel node location by Rob and colleagues,⁷⁰ a potential way to minimize the risk of this complication is to limit the groin target volumes to tissues directly overlying the femoral vessels and extending medially to the pubic tubercles when elective groin radiation is performed in patients with clinically and radiographically negative groins. By excluding lateral inguinal nodes, the substantial majority of the femoral neck and femoral metaphysis can be excluded from receiving the full prescribed tumor dose of radiation in most patients. Use of IMRT or proton beam techniques to irradiate inguinofemoral nodes should minimize the risk of subsequent femoral neck fractures.

Fractures through the ischial rami may occasionally be seen in patients carried to higher radiation dose for unresectable primary disease. Rarely, groin radiation may induce hemodynamically significant femoral artery stenosis.

Administered concurrently with radiation, chemotherapy appears to enhance radiation effects in all cycling cell systems, both neoplastic and normal, and acute complications can be severe.^219,220,221 Treatment interruption necessitated by severe acute radiation dermatitis can delay treatment longer than a planned, elective interruption as was required in GOG 101. Whether late effects are impacted to the same degree is unknown, but because the cell systems thought most important for delayed (chronic) radiation effects are infrequently cycling, the impact of concurrent drug administration is thought to be less. Chronic normal tissue sequelae of chemoradiation are thought to be most directly related to the size of the radiation target volume, fraction size, and total dose of radiation administered.

Melanoma

Malignant melanomas of the vulva are rare and constitute about 10% of all vulvar malignancies. Even large, tertiary cancer referral centers will treat only one or two patients annually.222,223,224 Malignant melanoma occurs predominantly in postmenopausal white women, with a peak incidence between the sixth and seventh decades. These lesions may arise de novo or from preexisting junctional or compound nevi.

Three basic histologic types exist: (1) superficial spreading melanoma is most common and tends to remain superficial in its early course; (2) lentigo malignant melanoma is a “flat freckle” that has a tendency to remain superficial; and (3) nodular melanoma is the most aggressive variety, carries the worst prognosis, and tends to invade deeply early.¹⁰⁷ Prognosis and staging are based on measurements of tumor depth of penetration as described by several melanoma staging systems.⁶¹ The FIGO staging system for the vulva is not usefully applied to melanoma. In a study of 71 surgically treated patients with primary malignant melanoma of the vulva, the AJCC 1992 clinical stage was the factor with the highest significant correlation with recurrence-free interval. Absent AJCC staging, Breslow’s depth of invasion was the most prognostically important factor.²²⁵ The AJCC 2002 melanoma TNM staging formalism is presented in web-only Table 58-1, available on the Expert Consult website for this chapter. The prognosis of melanoma of the vulva approximately parallels prognosis by stage of cutaneous and mucosal melanomas arising at other sites.

Web-only TABLE 58-1 AJCC Pathologic Classification of Cutaneous Melanoma

AJCC Cancer Staging Manual, 7th edition. Springer Science + Business Media LLC. © 2010 American Joint Committee on Cancer.

* Micrometastases are diagnosed after sentinel lymph node biopsy and completion lymphadenectomy (if performed).

† Macrometastases are defined as clinically detectable nodal metastases confirmed by therapeutic lymphadenectomy or when nodal metastasis exhibits gross extracapsular extension.

Verrucous Carcinoma

Verrucous carcinoma of the vulva represents a variant of squamous cell carcinoma. These lesions were originally described as occurring in the oral cavity but have also been described involving the vagina, cervix, and vulva. Lesions grossly may resemble cauliflower. Generally large, pearly gray to white, this cancer characteristically invades with a sharply circumscribed pushing margin. Clinically, these tumors are usually slow growing. Because metastases to regional lymph nodes are rare, radical local excision is the standard treatment.53,54,55,56 If there are suspicious groin nodes, fine-needle aspiration or excisional biopsy should be performed. Usually, enlarged nodes will be caused by inflammatory hypertrophy, but if they do contain metastases, groin node dissections are indicated. Verrucous carcinoma has a reputation of being radioresistant, and reports of favorable outcomes after irradiation are uncommon. Surgery remains the treatment of choice when feasible, with the use of irradiation limited to circumstances in which potentially curative surgery cannot be performed.

Sarcomas

Primary sarcomas of the vulva are extremely rare in adults. Characteristically, these will arise subcutaneously and involve overlying skin secondarily (Fig. 58-16). Because tumors are often bulky, treatment by conservative primary excision is not feasible. Preoperative RT or chemoradiation is favored to permit ultimate surgical treatment with less than conventional margins employed for sarcomas in other anatomic locations. Other than the special problems associated with treating tumors in proximity to multiple functionally important structures, there is nothing that distinguishes treatment of vulvar sarcomas from the treatment of soft tissue sarcomas at other sites.

Figure 58-16 Subcutaneous epithelioid sarcoma of the vulva with secondary satellite lesions (red arrows) invading overlying skin and prominent tumor neovascularity (black arrows).