Neville F. Hacker
Vulvar cancer is uncommon, representing approximately 4% of malignancies of the female genital tract. There were estimated to be 3,580 new cases of vulvar cancer diagnosed in the United States in 2009 and 900 deaths (1). Squamous cell carcinomas account for approximately 90% of the cases, whereas melanomas, adenocarcinomas, basal cell carcinomas, and sarcomas are much less common.
There has been a significant increase in the incidence of vulvar intraepithelial neoplasia (VIN) and VIN-related cancer in young women in recent decades (2,3,4), and this relates to changing sexual behavior, human papilloma virus (HPV) infection, and cigarette smoking (5). Judson et al. reviewed 13,176 in situ and invasive vulvar carcinomas from the SEER database over a 28-year period (1973-2000); 57% of the cases were in situ (2). There was a 411% increase in the incidence of in situ carcinoma from 1973 to 2000, while the incidence of invasive carcinoma increased 20% during the same period. The incidence of in situ disease increased until the age of 40 to 49 years, and then decreased, whereas the invasive cancer risk increased with age, and increased more rapidly after age 50.
It may be anticipated that the recent implementation of prophylactic HPV vaccination in young females will ultimately result in a significant decrease in the incidence of HPV-related in situ and invasive vulvar cancer, particularly in younger women.
In the early part of the 20th century, patients commonly presented with advanced disease, and surgical techniques were poorly developed; thus, the 5-year survival rate for vulvar cancer was 20% to 25% (6,7). Basset (8), in France, was the first to suggest an en bloc dissection of the vulva, groin, and iliac lymph nodes, although he performed the operation only on cadavers. Taussig (9), in the United States, and Way (10), in Great Britain, pioneered the radical en bloc dissection for vulvar cancer and reported 5-year survival rates of 60% to 70%. Postoperative morbidity was high after these procedures, with wound breakdown, infection, and prolonged hospitalization the norm. For patients with disease involving the anus, rectum, or proximal urethra, pelvic exenteration was often combined with radical vulvectomy.
Since approximately 1980, there has been a paradigm shift in the approach to vulvar cancer (11). The most significant advances have included:
Innovations that are currently being investigated include the sentinel node procedure to obviate the need for complete groin dissection in patients with early vulvar cancer (26,27,28,29) and chemoradiation for patients with advanced vulvar cancer (30,31,32,33,34,35). Resection of bulky positive nodes without complete groin dissection to decrease the risk of lymphedema prior to pelvic and groin radiation is another initiative aimed at decreasing morbidity without compromising survival (36).
This paradigm shift in the management philosophy of vulvar cancer has been well exemplified in retrospective reviews of the experience at the University of Miami (37) and the Mayo Clinic (38). Both centers reported a trend toward a more conservative approach, and both reported decreased postoperative morbidity, without compromised survival.
No specific etiologic factor has been identified for vulvar cancer, and the relationship of the invasive disease to vulvar dystrophy and to vulvar intraepithelial neoplasia (VIN) is controversial. Chronic pruritus is usually an important antecedent phenomenon in patients with invasive vulvar cancer (39).
VIN has traditionally been considered to have a low malignant potential, with progression to invasive disease most likely in the elderly or immunosuppressed (40). This concept has been challenged by Jones et al. (41), who reported a series of 405 cases of VIN 2-3 seen between 1962 and 2003 in Auckland, New Zealand. Progression to malignant disease occurred in 10 of 16 (62.5%) untreated patients with persistent VIN. Progression occurred between 1.1 and 7.3 years (mean 3.9 years). Invasive vulvar, perianal, or urethral cancer occurred in 17 women (3.8%) after treatment (mean age 42 years). Nine (2%) of the cases represented treated failure, with a median treatment-to-invasion interval of 2.4 years, while eight (1.8%) represented new “field” carcinomas, with a median treatment-to-invasion interval of 13.5 years. Spontaneous regression of VIN occurred in 47 women (11.6%). These women were young (mean age 24.6 years), and the median interval to complete regression was 9.5 months.
Using data on 2,685 patients with invasive vulvar cancer from the National Cancer Institute's Surveillance Epidemiology and End Results program (SEER), Sturgeon et al. (42) reportedan increased risk of a subsequent cancer of 1.3 fold. Most of the second cancers were smoking related (i.e., cancers of the lung, buccal cavity, pharynx, nasal cavity, and larynx) or related to infection with human papillomavirus (HPV; e.g., cervix, vagina, and anus).
The common association between cervical, vaginal, and vulvar cancer suggests a common pathogen, and the case-control study by Brinton et al. (43) found a significantly increased risk in association with multiple sexual partners, a history of genital warts, and smoking. HPV DNA has been reported in 20% to 60% of patients with invasive vulvar cancer (44). Hording et al. (45) reported HPV subtypes 16 or 33 in only 2 of 51 (4%) invasive keratinizing vulvar carcinomas, whereas one or the other was demonstrated in 12 of 17 (71%) invasive warty carcinomas and 10 of 10 (100%) invasive basaloid carcinomas. The HPV-positive group has been characterized by a younger mean age, greater tobacco use, and the presence of VIN in association with the invasive component (45,46,47,48).
These studies suggest two different etiologic types of vulvar cancer. One type is seen mainly in younger patients, is related to HPV infection and smoking, and is commonly associated with basaloid or warty VIN. The more common type is seen mainly in elderly patients, is unrelated to smoking or HPV infection, and concurrent VIN is uncommon, but there is a high incidence of dystrophic lesions, including lichen sclerosus, adjacent to the tumor. If VIN is present, it is of the differentiated type, and is situated adjacent to the invasive cancer (49).
Other diseases known to be occasionally associated with vulvar cancer include syphilis and nonleutic granulomatous venereal disease, particularly lymphogranuloma venereum and granuloma inguinale (Donovanosis). Such diseases are not seen commonly in Western countries.
Nonneoplastic Epithelial Disorders
The most recent recommendation on classification of vulvar diseases from the International Society for the Study of Vulvar Disease (ISSVD) is shown in Table 13.1. The new classification of squamous vulvar intraepithelial neoplasia (VIN) was introduced in 2004 (50). The term VIN 1 is no longer used, and VIN 2 and 3 are simply called VIN. Diagnosis in all cases requires biopsy of suspicious lesions, which are best detected by careful inspection of the vulva in a bright light, aided if necessary by a magnifying glass.
The older classification of VIN 1, 2, and 3 was based on the degree of histologic abnormality. However, there is neither evidence that the VIN 1-3 morphologic spectrum reflects a biologic continuum, nor that VIN 1 is a cancer precursor (49,50). The high-grade VIN lesions (2 and 3) include 2 types: (i) VIN usual type (warty, basaloid, and mixed), which is HPV related in most cases. Invasive squamous cancer of the warty or basaloid type is associated with VIN, usual type, (ii) VIN differentiated type, which is seen particularly in older women with lichen sclerosus and/or squamous cell hyperplasia. Neither VIN differentiated type nor associated keratinizing squamous cell carcinoma is HPV related.
The malignant potential of these nonneoplastic epithelial disorders is low, particularly now that the lesions with atypia are classified as VIN. However, patients with lichen sclerosus and concomitant hyperplasia may be at particular risk. Rodke and colleagues (51) reported the development of vulvar carcinoma in 3 of 18 such cases (17%), postulating that the areas of hyperplasia were superimposed on a background of lichen sclerosus because of chronic irritation and trauma.
Table 13.1 Classification of Vulvar Diseases
Table 13.2 Classification of Paget's Disease of the Vulva
Carli and colleagues from the Vulvar Clinic at the University of Florence, Italy, reported an association with lichen sclerosus in 32% of their cases of vulvar cancer that were not HPV related (52). They felt that the existence of accessory conditions necessary to promote the progression from lichen sclerosus to cancer remained to be established.
Scurry also supports an association between lichen sclerosus and vulvar cancer (53). He believes it contributes to a vicious cycle of itching and scratching which leads to superimposed lichen simplex chronicus, squamous cell hyperplasia, and ultimately carcinoma.
Vulvar Intraepithelial Neoplasia
The management of VIN is discussed in Chapter 8.
Paget's Disease of the Vulva
The original description of Paget's disease was of a breast lesion (Paget, 1874), in which the appearance of the nipple heralded an underlying carcinoma. Invasive carcinoma underlying Paget's disease of the vulvar is much less common.
Paget's disease of the vulva has been the subject of two recent classifications and that proposed by Wilkinson and Brown in 2002 is shown in Table 13.2 (54). In this classification,Type 1 Paget's disease is of primary cutaneous origin, and is divided into Type 1a, primary intraepithelial neoplasia; Type 1b, intraepithelial neoplasia with underlying invasion; and Type 1c, a manifestation of an underlying adenocarcinoma of a skin appendage or of vulvar glandular origin. Type 2 Paget's disease is of noncutaneous origin.
In the Kurman classification, Type 2 Paget's disease is a manifestation of an associated primary and/or rectal adenocarcinoma, and Type 3 is a manifestation of a urothelial neoplasm (55).
Type 1 Paget's Disease
In a review of Type 1 cases from the English literature, Niikura et al. reported on 565 cases of type 1 Paget's disease, including their own series of 22 cases (56). There were 425 patients (75%) with type 1a disease; 89 (16%) with type 1b; and 51 (9%) with type 1c. MacLean et al. reported that 6 of 76 patients (8%) on a British registry had an underlying carcinoma (type 1c) (57). In the British series, 14 patients (18.4%) had a systemic cancer. The primary was in the breast in six; bladder in three; and colorectum, cervix, uterus, and ovary in one each. One patient had a melanoma. Only four of these tumors were a synchronous cancer. Niikura et al. determined that if only synchronous neoplasms or those occurring within 12 months of the diagnosis were considered, only 8% of patients (44 of 534) with primary Paget's disease had a nonvulvar malignancy (56).
Fanning et al. reported on a combined series of 100 patients with Paget's disease of the vulva (58). Their median age was 70 years. There was a 12% prevalence of invasive vulvar Paget's disease, and a 4% prevalence of associated vulvar adenocarcinoma. Thirty-four percent of patients experienced a recurrence at a median of 3 years,
Clinical Features The disease predominantly affects postmenopausal white women, and the presenting symptoms are usually pruritus and vulvar soreness. The lesion has an eczematoid appearance macroscopically and usually begins on the hair-bearing portions of the vulva. It may extend to involve the mons pubis, thighs, and buttocks. Extension to involve the mucosa of the rectum, vagina, or urinary tract also has been described (59). The more extensive lesions are usually raised and velvety in appearance and may weep persistently.
Investigations All patients with Paget's disease of the vulva should be screened for any associated malignancy. These investigations should include mammography, computed tomography (CT) scan of the pelvis and abdomen, transvaginal ultrasonography, and cervical cytology. If the lesions involve the anus, colonoscopy should be undertaken, while if the urethra is involved, cystoscopy is indicated.
Treatment The mainstay of treatment is wide superficial resection of the gross disease. Underlying adenocarcinomas usually are clinically apparent, but this is not invariable. Paget cells may invade the underlying dermis, which should be removed for adequate histologic evaluation. For this reason, laser therapy is unsatisfactory for primary Paget's disease. The surgical defect can usually be closed primarily, but sometimes a split thickness skin graft may be required to cover an extensive defect.
Unlike squamous cell carcinoma in situ, in which the histologic extent of disease usually correlates reasonably with the macroscopic lesion, Paget's disease usually extends well beyond the gross lesion (60). This results in frequent positive surgical margins. The group at Memorial Sloan-Kettering Cancer Center reported positive margins in 20 of 28 patients (71%) (61). Of the 20 patients with microscopically positive margins, 14 (70%) developed recurrent disease, while of the 8 patients with negative margins, 3 (38%) developed a recurrence. With a median follow-up of 49 months (range 3-186 months), there was no correlation between disease recurrence and margin status (p = 0.20).
Surgical margins may be checked with frozen sections (62), but these can be misleading when compared to the definitive histology (63). Resection of the entire gross lesion will control symptoms and exclude invasive disease.
The role of radiotherapy for vulvar Paget's cancer has been reviewed by Brown et al. (64). It may be most useful when the disease involves the anus or urethra, and surgery would involve diversion and stoma formation.
If an underlying invasive carcinoma is present, it should be treated in the same manner as a squamous vulvar cancer. This may require radical vulvectomy and at least an ipsilateral inguinofemoral lymphadenectomy. From their literature review, Niikura et al. reported positive nodes in 30% of patients (23 of 70) with types 1b or 1c Paget's disease of the vulva (56).
Lymph node metastases have been reported in patients with very superficially invasive tumors. Fine et al. reported a patient who had invasion no greater than 1 mm over a maximum length of 10 mm at her primary operation. Three weeks later, she was noted to have a palpable groin node, and at groin dissection, six positive groin nodes were resected from each groin (65). Ewing et al reported a positive sentinel node in a patient whose excised Paget's disease had scattered foci of superficial dermal invasion to a maximum depth of 0.7 mm (66).
Minimally invasive Paget's disease is rare, and in both cases the surgical margins were positive for intraepithelial disease. It would seem prudent to at least monitor the groins carefully, preferably with ultrasound, if the nodes are not dissected in a patient with superficially invasive Paget's disease.
The disease is characterized by local recurrences over many years (61,57). Recurrent lesions are usually in situ, although 5 of 76 patients (6.7%) in the British registry showed progression to invasive disease between 1 and 21 years after the initial diagnosis (57). The age at the time of diagnosis of the invasive recurrence ranged from 63 to 88 years (mean 76.2 years). Investigators at the Norwegian Radium Hospital have demonstrated that nondiploid tumors have an increased risk of recurrence regardless of surgical radicality (67). In general, recurrent lesions should be treated by further surgical resection, although laser therapy may occasionally be useful, particularly for perianal disease.
Invasive Vulvar Cancer
Squamous Cell Carcinoma
Squamous cell carcinoma of the vulva is predominantly a disease of postmenopausal women, with a mean age at diagnosis of approximately 65 years.
Most patients present with a vulvar lump or mass, although there is often a long history of pruritus, usually associated with a vulvar dystrophy. Less common presenting symptoms include vulvar bleeding, discharge, or dysuria. Occasionally a large metastatic mass in the groin may be the initial presenting symptom, although this is much less common than in the past because women are now more likely to present with earlier-stage disease.
On physical examination, the lesion is usually raised and may be fleshy, ulcerated, leukoplakic, or warty in appearance. Warty lesions are often initially diagnosed as condylomata acuminata.
Most squamous carcinomas of the vulva occur on the labia majora, but the labia minora, clitoris, and perineum also may be primary sites. A recent study from one University Hospital in Germany reported that the tumor localization had changed significantly from the labia to the area between the clitoris and the urethra (3). Approximately 10% of the cases are too extensive to determine a site of origin, and approximately 5% of the cases are multifocal.
As part of the clinical assessment, the groin lymph nodes should be evaluated carefully and a complete pelvic examination performed. A Papanicolaou smear should be taken from the cervix, and colposcopy of the cervix and vagina should be performed because of the common association with other squamous intraepithelial neoplasms of the lower genital tract.
Diagnosis requires a wedge or a Keyes biopsy specimen, which usually can be taken in the office under local anesthesia. The biopsy specimen must include some underlying dermis and connective tissue so that the pathologist can adequately evaluate the depth and nature of the stromal invasion. It is preferable to leave the primary lesion in situ, if possible, to allow the treating surgeon to fashion adequate surgical margins.
Figure 13.1 Inguinal-femoral lymph nodes. (Reproduced from Hacker NF. Vulvar cancer. In: Hacker NF, Moore JG. Essentials of obstetrics and gynecology, 5th ed. Philadelphia: Elsevier Saunders, 2008, with permission.)
Table 13.3 Incidence of Lymph Node Metastases in Operable Vulvar Cancer
Physician delay is a common problem in the diagnosis of vulvar cancer, particularly if the lesion has a warty appearance. Although isolated condylomata do not require histologic confirmation for diagnosis, any confluent warty lesion should be biopsied adequately before medical or ablative therapy is initiated.
Routes of Spread
Vulvar cancer spreads by the following routes:
Lymphatic metastases may occur early in the disease. Initially, spread is usually to the inguinal lymph nodes, which are located between Camper's fascia and the fascia lata. From these superficial groin nodes, the disease spreads to the femoral nodes, which are located medial to the femoral vein (Fig. 13.1). Cloquet's node, situated beneath the inguinal ligament, is the most cephalad of the femoral node group. Metastases to the femoral nodes without involvement of the inguinal nodes have been reported (68,69,70,71). In addition, Gordinier and colleagues reported groin recurrence in 9 of 104 patients (8.7%) treated by superficial inguinal lymphadenectomy at the M. D. Anderson Cancer Center (72). The median number of lymph nodes removed per groin was 7, and the median time to recurrence was 22 months.
From the inguinofemoral nodes, the cancer spreads to the pelvic nodes, particularly the external iliac group. Although direct lymphatic pathways from the clitoris and Bartholin gland to the pelvic nodes have been described, these channels seem to be of minimal clinical significance (17,73,74).
Since 1970, the overall incidence of lymph node metastases is reported to be approximately 30% (Table 13.3). The incidence in relation to clinical stage of disease is shown in Table 13.4, and that in relation to depth of invasion is shown in Table 13.5.
Table 13.4 Incidence of Lymph Node Metastases in Relation to Clinical Stage of Disease
Table 13.5 Nodal Status in T1 Squamous Cell Carcinoma of the Vulva Versus Depth of Stromal Invasion
Metastases to pelvic nodes are uncommon, the overall reported frequency being approximately 9%. Approximately 20% of patients with positive groin nodes have positive pelvic nodes (90). Pelvic nodal metastases are rare in the absence of clinically suspicious (N2) groin nodes (18) and three or more positive groin nodes (17,18,79,82). Recent Mayo Clinic data suggested that in addition to having 3 or more positive groin nodes, all patients with positive pelvic nodes had a tumor with invasion >4 mm (82). Systematic pelvic lymphadenectomy is no longer performed in the management of patients with vulvar cancer.
Hematogenous spread usually occurs late in the course of vulvar cancer and is rare in the absence of lymph node metastases. Hematogenous spread is uncommon in patients with one or two positive groin nodes, but is more common in patients with three or more positive nodes (18).
A clinical staging system based on the TNM classification was adopted by the International Federation of Gynecology and Obstetrics (FIGO) in 1969 (Table 13.6). The staging was based on a clinical evaluation of the primary tumor and regional lymph nodes and a limited search for distant metastases.
Clinical evaluation of the groin lymph nodes is inaccurate in approximately 25% to 30% of the cases (10,19,91). Microscopic metastases may be present in nodes that are not clinically suspicious, and suspicious nodes may be enlarged because of inflammation only. Compared with surgical staging of vulvar cancer, the percentage of error in clinical staging increases from 18% for stage I disease to 44% for stage IV disease (92).
These factors led the Cancer Committee of FIGO to introduce a surgical staging for vulvar cancer in 1988. For updated FIGO surgical staging tables (2009) see pages 665-668. Various modifications have been made, with a subdivision of stage I in 1994. The current FIGO staging is shown in Table 13.7. Most available data are still based on the 1969 FIGO staging, which is appropriate because the new staging system requires further modification.
There are two major problems with the staging system as currently proposed. First, patients with negative lymph nodes have a very good prognosis, regardless of the size of the primary tumor (18,78,81,92), so the survival rate for both stages I and II should be better than 80%. Second, survival depends on the number of positive lymph nodes (17,18,92,93).Therefore, stage III represents a very heterogeneous group of patients, ranging from those with negative nodes and involvement of the distal urethra or vagina, who should have an excellent prognosis, to those with multiple positive groin nodes, who have a very poor prognosis. Rouzier et al. reported a cohort of 895 patients with FIGO Stage III vulvar cancer who had been registered with the SEER database from 1988 through 2004. The 5-year overall survival for patients with regional metastatic nodal disease (39%) was significantly worse than that of patients with locally advanced tumors but negative nodes (62%, p <0.0001) (94).
After the pioneering work of Taussig (9) in the United States and Way (7,10) in Great Britain, en bloc radical vulvectomy and bilateral dissection of the groin and pelvic nodes became the standard treatment for most patients with operable vulvar cancer. If the disease involved the anus, rectovaginal septum, or proximal urethra, some type of pelvic exenteration was combined with this dissection.
Table 13.6 Clinical Staging of Carcinoma of the Vulva
Although the survival rate improved markedly with this aggressive surgical approach, several factors have led to modifications of this “standard” treatment plan during the past 25 years. These factors may be summarized as follows:
Table 13.7 FIGO Staging for Vulvar Cancer (1994)
Modern management of vulvar cancer requires an experienced, multidisciplinary team approach, which is available only in tertiary referral centers. The shortcomings of treatment in nonreferral units were highlighted in two community-based European studies.
In the British study, investigators retrospectively reviewed the records of 411 patients with squamous cell carcinoma who had been notified to the Central Intelligence Unit of the West Midlands during two 3-year periods; 1980 to 1982 and 1986 to 1988 (95). The women were treated at 35 different hospitals, 16 of which averaged 1 case or less per year.
Fifteen different operations were used, the most common of which were simple vulvectomy (35%) and radical vulvectomy with bilateral inguinal lymphadenectomy (34%). Hemivulvectomy was performed in only five patients (1.2%). Management of the lymph nodes was equally inappropriate. Only 190 of the 411 patients (46%) had a lymphadenectomy performed, and a unilateral dissection was performed in only 9 patients (2.1%).
Survival data for all FIGO stages compared unfavorably with the Gynecologic Oncology Group (GOG) data from tertiary units in the United States (92): 78% versus 98% for stage I disease; 53% versus 85% for stage II; 27% versus 74% for stage III; and 13% versus 31% for stage IV. Omission of lymphadenectomy was the single most important prognostic factor, but treatment in a hospital with less than 20 cases in total was a poor prognostic factor in univariate analysis.
A similar experience was reported from the Netherlands (96). As in the British study, older patients tended not to be referred to gynecologic oncology units, and 80% of patients in the community hospitals had omission of groin node dissection.
Management of Early Vulvar Cancer (T1 or T2 N0 or N1)
The modern approach to the management of patients with carcinoma confined to the vulva should be individualized (12,13,15,16,97,98,99). There is no “standard” operation applicable to every patient, and emphasis is on performing the most conservative operation consistent with cure of the disease.
In considering the appropriate operation, it is necessary to determine independently the appropriate management of:
Before any surgery, all patients should have colposcopy of the cervix, vagina, and vulva because preinvasive (and rarely invasive) lesions may be present at other sites along the lower genital tract.
Management of the Primary Lesion
The two factors to take into account in determining the management of the primary tumor are:
Although radical vulvectomy has been regarded as the standard treatment for the primary vulvar lesion, this operation is associated with significant disturbances of sexual function and body image. DiSaia et al. (14) regarded psychosexual disturbance as the major long-term morbidity associated with the treatment of vulvar cancer. Andersen and Hacker (100) reported that, when compared with healthy adult women, sexual arousal was reduced to the eighth percentile and body image to the fourth percentile in women who had undergone vulvectomy.
Since the early 1980s, several investigators have advocated a radical local excision rather than a radical vulvectomy for the primary lesion in patients with T1 and T2 tumors(9,10,11,12,13,16,90,97,98,101,102). Regardless of whether a radical vulvectomy or a radical local excision is performed, the surgical margins adjacent to the tumor are the same, and an analysis of the available literature indicates that the incidence of local invasive recurrence is low if the histopathologic margin is at least 8 mm (Table 13.8). Allowing for 20% tissue shrinkage with formalin fixation, this translates to a surgical margin of at least 1 cm. De Hulla et al. reported that in 50% of patients, the histologic margins were 8 mm or less in spite of intentional macroscopic margins of 1 cm (102), but in the author's experience, this will happen only if the skin is placed under tension prior to making the incisions.
When vulvar cancer arises in the presence of VIN or some nonneoplastic epithelial disorder, treatment is influenced by the patient's age. Elderly patients who have often had many years of chronic itching are not usually disturbed by the prospect of a radical vulvectomy. In younger women, it is desirable to conserve as much of the vulva as possible; thus, radical local excision should be performed for the invasive disease, and the associated disease should be treated in the most appropriate manner. For example, topical steroids may be required for squamous hyperplasia or lichen sclerosus, whereas VIN should be treated by superficial local excision and primary closure or split thickness skin grafting.
Table 13.8 Invasive Vulvar Recurrence versus Histopathologic Resection Margins
Radical local excision is most appropriate for lesions on the lateral or posterior aspects of the vulva (Fig. 13.2), where preservation of the clitoris is feasible. For anterior lesions that involve the clitoris or are close to it, any type of surgical excision has psychosexual consequences, particularly in younger patients. Chan et al. identified 41 patients with squamous carcinoma of the anterior vulva not involving the clitoris (105). Thirteen patients (32%) had clitoral sparing modified radical vulvectomy and 28 (68%) had radical vulvectomy. The 13 patients who had clitoral sparing surgery included 8 with stage I, 2 with stage II, 2 with stage III, and 1 with stage IV disease. After a median follow-up of 59 months, none of the 13 patients having conservative surgery had locoregional failure.
Figure 13.2 Small (T1) vulvar carcinoma at the posterior fourchette.
In young patients with actual involvement of the clitoris or in whom surgical margins would be <5 mm, consideration should be given to treating the primary lesion with a small field of radiation therapy. Small vulvar lesions respond very well to approximately 5,000 cGy external radiation, and biopsy can be performed after therapy to confirm the absence of any residual disease (106).
Two recent papers have looked at a single institutional experience with T1 and T2 squamous cell carcinoma of the vulva (97,98).
In the study from Kentucky, 61 patients with a lateral T1 lesion and 61 patients with a lateral T2 lesion were seen from 1963 to 2003 (97). Radical vulvectomy was performed on 60 patients (49%) and radical hemivulvectomy on 62 (51%). Ipsilateral inguinal node metastases were present in 11% of patients (7 of 61) with a T1 lesion, and 31% (19 of 61) of patients with a T2 lesion. Disease-free survival of patients with T1 and T2 lesions was 98% and 93% respectively at 5 years. Local or distant recurrence was not more common in patients treated by radical vulvectomy or radical hemivulvectomy.
Our experience at the Royal Hospital for Women in Sydney with FIGO stages I and II vulvar cancer has recently been reported (98). There were 121 cases managed from 1987 through 2005, which represented 37.7% of the 339 patients with invasive vulvar cancer seen during this period. Radical local excision was performed in 116 patients (95.9%). Only 5 patients underwent radical vulvectomy, in all cases for tumor multifocality. With a median follow-up of 84 months, the overall survival at 5 years was 96.4%.
Technique for Radical Local Excision
Radical local excision implies a wide and deep excision of the primary tumor. The surgical margins should be at least 1 cm, and should be drawn using a marking pen with the vulva in its natural state. The incision should be carried down to the inferior fascia of the urogenital diaphragm, which is coplanar with the fascia lata and the fascia over the pubic symphysis. The surgical defect is closed in two layers. For perineal lesions, proximity to the anus may preclude adequate surgical margins, and consideration should be given to preoperative or postoperative radiation in such cases. For periurethral lesions, the distal half of the urethra may be resected without loss of continence. Figure 13.3 shows the satisfactory cosmetic result achieved in the treatment of the lesion shown in Fig. 13.2.
Figure 13.3 Satisfactory cosmetic result after radical local excision and bilateral groin dissection (for the small posterior vulvar carcinoma shown in Fig. 13.2).
Table 13.9 Death From Recurrence in an Undissected Groin
Management of the Groin Lymph Nodes
Appropriate management of the regional lymph nodes is the single most important factor in decreasing the mortality from early vulvar cancer. With an increasing number of reports in the literature, two facts have become apparent:
All patients with a T1 tumor with more than 1 mm of stromal invasion, and all patients with a T2 tumor require inguinofemoral lymphadenectomy. A wedge or Keyes biopsy of the primary tumor should be obtained, and the depth of invasion determined. If it is less than 1 mm on the biopsy specimen, and the lesion is 2 cm or less in diameter, the entire lesion should be locally excised and analyzed histologically to determine the depth of invasion. If there is still no invasive focus deeper than 1 mm, groin dissection may be omitted. Although an occasional patient with a T1 tumor and less than 1 mm of stromal invasion has had documented groin node metastases (108,109,110), the incidence is so low that it is of no practical significance.
If groin dissection is indicated in patients with early vulvar cancer, it should be a thorough inguinofemoral lymphadenectomy. The GOG reported six groin recurrences among 121 patients with T1N0 or N1 tumors after a superficial (inguinal) dissection, even though the inguinal nodes were reported as negative (111). Whether all these recurrences were in the femoral nodes is unclear, but this large, multi-institutional study does indicate that modification of the groin dissection increases groin recurrences and, therefore, mortality.
From the accumulated experience now available in the literature, it is clear that it is not necessary to perform a bilateral groin dissection if the primary lesion is unilateral and the ipsilateral nodes are negative (Table 13.10). A recent series from the Mayo Clinic reported 48 positive groin nodes (29.4%) among 163 patients with a unilateral vulvar cancer. Only three patients (1.8%) had positive contralateral nodes with negative ipsilateral nodes and none had T1 tumors. The only independent risk factor for bilateral nodal involvement was the total number of positive groin nodes, with an OR of 1.84 (CI: 1.30, 2.59). With each additional positive node, the possibility of having bilateral groin node involvement increased by 84%.
Lesions involving the anterior labia minora should have bilateral dissection because of the more frequent contralateral lymph flow from this region (114).
Measurement of Depth of Invasion
The Nomenclature Committee of the International Society of Gynecological Pathologists has recommended that depth of invasion should be measured from the most superficial dermal papilla adjacent to the tumor to the deepest focus of invasion. This method was originally proposed by Wilkinson et al. (84). Tumor thickness is also commonly measured (83,115), and Fu (116) estimated that the average difference between tumor thickness and depth of invasion as determined by the Wilkinson method was 0.3 mm.
Table 13.10 Incidence of Positive Contralateral Nodes in Patients With Lateral T1 Squamous Cell Vulvar Carcinomas and Negative Ipsilateral Nodes
Technique for Groin Dissection
A linear incision is made along the medial four-fifths of a line drawn between the anterior superior iliac spine and the pubic tubercle. The incision is best made about 1 cm above the groin crease (Fig. 13.4). Studies of bipedal lymphangiograms have demonstrated that there are no lymph nodes adjacent to the anterior superior iliac spine (117). On the basis of embryological and anatomical studies, Micheletti et al. have proposed that the superficial circumflex iliac vessels could represent the lateral surgical landmark (118). The incision is carried through the subcutaneous tissues to the superficial fascia. The latter is incised and grasped with artery forceps to place it on traction, and the fatty tissue between it and the fascia lata is removed over the femoral triangle (Fig. 13.5). The dissection is carried 2 cm above the inguinal ligament to include all the inguinal nodes.
Figure 13.4 Skin incision for groin dissection through a separate incision. The incision is made along the medial four fifths of a line drawn between the anterior superior iliac spine and the pubic tubercle.
Figure 13.5 Camper's fascia kept on traction with forceps while the underlying node-bearing fatty tissue is dissected out of the femoral triangle. Note the preservation of the subcutaneous tissue above the superficial fascia. This ensures that skin necrosis will not occur.
The saphenous vein is usually tied off at the apex of the femoral triangle and at its point of entry into the femoral vein. Some authors have suggested that saphenous vein sparing may decrease postoperative morbidity (119,120), although in a study of 64 patients, 31 of whom underwent saphenous sparing, Zhang et al. reported no difference in the incidence of postoperative fever, acute cellulitis, seroma, or lymphocyst formation (120). To avoid skin necrosis, all subcutaneous tissue above the superficial fascia must be preserved.
The fatty tissue containing the femoral lymph nodes is removed from within the fossa ovalis. There are only one to three femoral lymph nodes, and they are always situated medial to the femoral vein in the opening of the fossa ovalis (121). Hence, there is no need to remove the fascia lata lateral to the femoral vessels and no need to perform a sartorius muscle transposition. Cloquet's node is not consistently present but should be checked for by retraction of the inguinal ligament cephalad over the femoral canal. At the conclusion of the dissection, a suction drain is placed in the groin and the wound is closed in two layers.
In spite of the age and general medical condition of most patients with vulvar cancer, the surgery is usually remarkably well tolerated. However, a postoperative mortality rate of 1% to 2% can be expected, usually as a result of pulmonary embolism or myocardial infarction. Patients should be able to commence a low-residue diet on the first postoperative day. Bed rest is advisable for 2 to 3 days to allow immobilization of the wounds to foster healing. Pneumatic calf compression and subcutaneous heparin should be used to help prevent deep venous thrombosis, and active, non-weight-bearing leg movements are to be encouraged. Perineal swabs are given until the patient is fully mobilized, at which stage sitz baths or whirlpool therapy are helpful, followed by drying of the perineum with a hair dryer. Suction drainage of each groin is continued for approximately 7 to 10 days to help decrease the incidence of groin seromas. A Foley catheter is left in the bladder until the patient is ambulatory.
Early Postoperative Complications
The major immediate morbidity is related to the groin dissection. With the separate incision approach, and sparing of the subcutaneous fat in both the superior and inferior flap, the incidence of wound breakdown is now very low. The most common problem is lymphocyst formation, which occurs in about 40% of cases (122). These seem to have become more common since introduction of the practice of leaving the fascia lata over the muscles in the floor of the femoral triangle. If large, they are usually best managed by making a linear incision approximately 1 cm long to allow adequate drainage. Drainage must be maintained by placing a narrow gauze wick in the incision and changing it twice daily until the skin flaps seal to the underlying tissues. Early mobilization and long walks immediately after discharge from hospital seem to increase the incidence of lymphocysts.
Other early postoperative complications include cellulitis, urinary tract infection, deep venous thrombosis, pulmonary embolism, myocardial infarction, hemorrhage, and, rarely, osteitis pubis.
The major late complication is chronic leg edema, which has been reported in up to 69% of patients (93). In our experience at the Royal Hospital for Women, the incidence of lymphedema after groin dissection is 62% (123). In about 50% of patients, the onset of lymphedema occurs within 3 months, while about 85% experience the onset within 12 months (123). Lymphedema is significantly related to the occurrence of early complications (122), particularly cellulitis. Recurrent lymphangitis or cellulitis of the leg occurs in approximately 10% of patients and usually responds to erythromycin tablets or flucloxacillin. Urinary stress incontinence, with or without genital prolapse, occurs in approximately 10% of patients and may require corrective surgery. Introital stenosis can lead to dyspareunia and may require a vertical relaxing incision, which is sutured transversely. An uncommon late complication is femoral hernia, which can usually be prevented during surgery by closure of the femoral canal with a suture from the inguinal ligament to Cooper's ligament. Pubic osteomyelitis and rectovaginal or rectoperineal fistulae are rare late complications.
Management of a Patient With Positive Groin Nodes
Traditionally, patients with positive groin nodes had a pelvic lymphadenectomy, but in 1977, the GOG initiated a prospective trial in which patients with positive groin nodes were randomized to either ipsilateral pelvic node dissection or bilateral pelvic plus groin irradiation (21). Radiation therapy consisted of 4,500 to 5,000 cGy to the midplane of the pelvis at a rate of 180 to 200 cGy/day. The survival rate for the radiation group (68% at 2 years) was significantly better than that for the pelvic lymphadenectomy group (54% at 2 years; p= 0.03). The survival advantage was limited to patients with clinically evident groin nodes or more than one positive groin node. Groin recurrence occurred in 3 of 59 patients (5.1%) treated with radiation, compared with 13 of 55 (23.6%) treated with lymphadenectomy (p = 0.02). Four patients who received radiation had a pelvic recurrence, compared with one who had lymphadenectomy. These data highlight the value of prophylactic groin irradiation in preventing groin recurrence in patients with multiple positive groin nodes.
In the 1990s, several investigators demonstrated that the morphology of the positive groin nodes was also of prognostic significance, allowing further discrimination among patients with positive nodes. Origoni et al. (124) demonstrated that for patients with positive lymph nodes, there was a significant difference in survival, depending on the size of the involved nodes and the presence or absence of extracapsular spread. Patients whose involved nodes were less than 5 mm in diameter had a 5-year survival rate of 90.9%, compared with 41.6% for nodes 5 to 15 mm in diameter and 20.6% for nodes larger than 15 mm diameter (p = 0.001). Similarly, if nodal involvement remained intracapsular, the 5-year survival rate was 85.7%, compared with 25% if there was extracapsular spread (p = 0.001).
Similar results were obtained by the group at Gateshead, who reported that in a multivariate analysis, the only significant variables were FIGO stage (III, IVA, or IVB) and the presence or absence of extracapsular spread (125). Van der Velden et al. (126) demonstrated that even for patients with one positive node, the presence of extracapsular spread decreased the survival rate from 88% (14 of 16 patients) to 44% (7 of 16 patients). Raspagliesi et al. have recently reported a 10-year survival of 55% in lymph node positive patients with <50% of nodal replacements, compared to 34.3% in lymph node positive patients with >50% nodal replacement (p <0.01) (81).
From the foregoing observations, our recommendations for the management of patients with positive groin nodes are as follows:
The major morbidity associated with the modern management of vulvar cancer is chronic lymphedema, which occurs in about 60% of patients and is a lifelong affliction. Hence, there is significant interest in eliminating or modifying the groin dissection for patients with negative nodes.
Several noninvasive methods for detecting lymph node metastases from vulvar cancer have been disappointing, including positron emission tomography (128) and computerized tomographic scanning (129). Ultrasonic scanning, particularly when combined with fine needle aspiration cytology, shows more promise, but false negatives and false positives still occur (129,130).
For the past decade considerable investigation has been undertaken of the role of sentinel node identification in patients with vulvar cancer. This concept was initially introduced by Cabanas for the management of patients with penile cancer (131), and subsequently for the management of melanomas by Morton et al. (132). The hypothesis is that if the sentinel node is negative, all other nodes will be negative, so the patient can be spared the morbidity of full groin dissection.
The sentinel node (or nodes) is identified by the injection of intradermal isosulfan blue dye around the primary vulvar lesion, either alone or in combination with intradermal radioactive 99mTc-labeled sulfur colloid (26,28). After the injections, the node(s) is isolated in the groin by dissection and gamma counting.
In a review of the literature in 2001, Makar et al. reported successful identification of sentinel node(s) in 85 of 103 patients (82.5%) using the blue dye technique, and 128 of 128 patients (100%) using lymphoscintigraphy (29). False-negative sentinel nodes have been reported (132,133), although the incidence appears to be low. In the presence of palpably suspicious nodes, or nodes replaced by cancer, the incidence of false negatives is higher, presumably because metastatic disease obstructs flow to those nodes (134,135). The incidence of false-negative sentinel nodes can be reduced by ultrastaging, using either serial sectioning alone (136) or in combination with immunohistochemical staining for cytokeratin (137). Molpus et al. reported that 2 of 18 (11%) negative sentinel nodes had micrometastases (<0.2 mm) upon serial sectioning and immunohistochemical staining (138).
Results of a European multicenter observation study on sentinel node detection in vulvar cancer was recently reported by van der Zee et al. (26). Both radiotracer and blue dye were used, and eligible patients were those with T1 or T2 squamous cell carcinomas <4 cm diameter. If the sentinel node was negative at pathologic ultrastaging, groin dissection was omitted and the patient was observed clinically every 2 months for 2 years.
From March 2000 to June 2006, 403 assessable patients were recruited to the study, and they underwent 623 groin dissections. Metastatic sentinel nodes were found in 163 groins (26.2%). Routine pathologic examination detected 95 (58.3%) positive sentinel nodes and ultrastaging detected 68 (41.7%). In eight of 276 patients in the observational study, groin recurrence was observed after a negative sentinel node procedure. The actuarial groin recurrence rate after 2 years was 3% (95%, CI 1%, to 6%). All patients with a groin recurrence underwent bilateral inguinofemoral lymphadenectomy and adjuvant (chemo)radiation. Six of the eight patients died of disease, while two remained disease-free at 6 and 50 months after recurrence. The median time to recurrence was 12 months (range 5-16 months). As expected, both short-term and long-term morbidity were significantly decreased.
While sentinel node biopsy is clearly superior to earlier attempts to decrease postoperative morbidity by selective omission (69,112) or modification (14) of the groin dissection, the key issue really is to determine the false-negative rate of the procedure, not just from the best centers, but in the hands of the average operator. The high mortality rate from recurrence in an undissected groin is vitally important to a patient undergoing this procedure, and proper informed consent will be crucial.
In this latter regard, a recent study by de Hulla et al. is important (139). They sent structured questionnaires both to patients who had been treated for vulvar cancer and to gynecologists. The response rate among patients was 91% (107 of 118), 40% of whom had experienced lower limb cellulitis and 49% of whom still experienced severe pain and/or lymphedema in the legs. Sixty percent of the patients preferred complete lymphadenectomy in preference to a 5% false-negative rate of the sentinel node procedure. Their preference was not related to age or the side effects they had experienced. The response among gynecologists was 80% (80 of 100), of whom 60% were willing to accept a 5% to 20% false-negative rate for the sentinel node procedure. The authors concluded that although gynecologists may consider this a promising approach, the majority of vulvar cancer patients would not advise its introduction because they are not prepared to take any risk of missing a lymph node metastasis.
Exactly similar sentiments have been expressed about the sentinel node procedure by women with breast cancer (140), even though most patients with breast cancer receive adjuvant treatment after surgery, and the nodal recurrence risk is much lower (0.1% to 0.3%) (141). A recent study from our own institution of 60 patients with vulvar cancer and clinically negative lymph nodes revealed similar findings. Although 73% of these women reported lymphedema, 80% indicated they would choose complete lymphadenectomy rather than take a 5% risk of a false-negative sentinel node procedure (142).
If the sentinel lymph node procedure is performed after proper informed consent, it is important to undertake complete inguinofemoral lymphadenectomy if a sentinel node is not detected. This is particularly likely to occur for lesions close to the midline, when bilateral groin dissection would normally be required (28,135).
In the author's opinion, an undissected groin should be followed with ultrasonography every 3 months for the first 12 months to allow early detection of any enlarging lymph nodes.
Vulvar cancer may be considered to be advanced on the basis of a T3 or a T4 primary tumor or the presence of bulky, positive groin nodes. Advanced vulvar cancer is uncommon in developed countries, and most data derive from single institutional experience or single-arm multiinstitutional reports.
Management should be individualized, and a multidisciplinary team approach is desirable. As with early stage disease, it is advantageous to independently determine the most appropriate treatment for (i) the primary tumor and (ii) groin and pelvic lymph nodes.
Management of the Groin and Pelvic Lymph Nodes
The author's preference is to initially determine the status of the groin and pelvic lymph nodes, using clinical examination and a CT scan of the groin, pelvis, and abdomen. Patients can then be triaged into 3 groups, as follows:
o All enlarged groin nodes are removed through a separate incision approach and sent for frozen-section diagnosis. If metastatic disease is confirmed, full lymphadenectomy is not carried out.
o Any enlarged pelvic nodes seen on CT scan are removed by an extraperitoneal approach.
o Full pelvic and groin irradiation is given as soon as the groin incisions are healed, which is usually approximately 2 weeks.
o If the frozen section reveals no metastatic disease in the removed nodes, full groin dissection is performed.
We have reported our experience with resection of bulky positive nodes rather than full groin dissection for patients with advanced vulvar cancer (36). Seventeen patients treated by nodal debulking in Australia were compared with 23 similar patients treated by full groin dissection at the Academic Medical Hospital in Amsterdam. Both groups of patients received groin and pelvic radiation post-operatively. Both disease-specific survival and groin recurrence-free intervals were superior in the group having nodal debulking although with the small numbers in both series, the differences were not statistically significant.
An algorithm for the management of patients with advanced vulvar cancer is shown in Fig. 13.7.
Management of the Primary Tumor
If the tumor involves the distal vagina and/or urethral orifice, and can be resected without need for a stoma, primary surgical resection is the best option. Radical vulvectomy often will be required, although a modified radical vulvectomy to allow adequate clearance around the lesion while preserving some normal vulva may also be appropriate.
Figure 13.6 A fixed ulcerated lymph node in the right groin.
Figure 13.7 Algorithm for the management of patients with locally advanced vulvar cancer, in whom surgical resection of the primary tumor would necessitate a stoma.
Two basic surgical approaches can be used:
Figure 13.8 Incision used for en bloc radical vulvectomy and bilateral groin dissection.
Technique for En Bloc Radical Vulvectomy and Groin Dissection
The operation is usually performed with the patient in the low lithotomy position, and groin and vulvar dissections can proceed simultaneously with two teams of surgeons if appropriate. The skin incision has been significantly modified from the original Stanley Way technique to allow primary skin closure. The groin dissection is accomplished initially, with the abdominal incision carried down to the aponeurosis of the external oblique muscle, approximately 2 cm above the inguinal ligament. A skin flap is raised over the femoral triangle, with preservation of the subcutaneous fat above the superficial (Camper's) fascia. The technique for groin dissection has been described earlier.
The vulvar incision is carried posteriorly along each labiocrural fold, or within a 1-cm margin of the primary lesion. The technique for vulvectomy is described in the next section.
Technique for Radical Vulvectomy
If the radical vulvectomy is performed through a separate incision, the lateral incision is basically elliptical. Each lateral incision should commence on the mons pubis anteriorly and extend through the fat and superficial fascia to the fascia over the pubic symphysis. It is then easy to develop bluntly the plane immediately above the pubic symphysis and fascia lata. The skin incision is extended posteriorly along the labiocrural folds to the perianal area and carried down to the fascia lata. The medial incision is placed to clear the tumor with margins of at least 1 cm. If necessary, the distal half of the urethra may be resected without compromising continence. If the tumor is involving the urethra or the vagina, dissection around the tumor is facilitated by transection of the vulva, thereby improving exposure of the involved area.
The specimen includes the bulbocavernosus muscles and the vestibular bulb. Because of the vascularity, it is desirable to perform most of the dissection by diathermy after the initial skin incision. In addition, the vessels supplying the clitoris should be clamped and tied, as should the internal pudendal vessels posterolaterally.
Closure of Large Defects
It is usually possible to close the vulvar defect without tension. However, if a more extensive dissection has been required because of a large primary lesion, a number of options are available to repair the defect. These include the following:
The technique for these grafts is discussed in Chapter 20.
When the primary disease involves the anus, rectum, rectovaginal septum, or proximal urethra, adequate surgical clearance of the primary tumor is possible only by pelvic exenteration combined with radical vulvectomy and bilateral groin dissection. Such radical surgery is often inappropriate for these elderly patients, and even in suitable surgical candidates, psychological morbidity is high (100,150), and postoperative morbidity is significant. Nevertheless, a 5-year survival rate of approximately 50% can be expected with this approach (151,152). Surgery alone is rarely curative for patients with fixed or ulcerated (N3) groin nodes.
Boronow (24) was the first to suggest a combined radiosurgical approach as an alternative to pelvic exenteration for patients with advanced vulvar cancer. In his initial report, he recommended intracavitary radium, with or without external irradiation, to eliminate the internal genital disease, and subsequent surgery, usually radical vulvectomy and bilateral groin dissection, to treat the external genital disease.
Figure 13.9 Unilateral gracilis myocutaneous graft used to cover a large lateral vulvar defect.
Figure 13.10 Advanced squamous cell carcinoma of the vulva involving the anal canal. A primary surgical approach would have necessitated radical vulvectomy, anoproctectomy, and permanent colostomy.
Figure 13.11 Advanced vulvar cancer shown in Figure 13.10 after 50.4 cGy of externalbeam radiation therapy. Resection of the tumor bed showed microscopic residual disease. The radiation therapy prevented the need for a permanent stoma.
In 1984, Hacker et al. (25) reported the use of preoperative teletherapy in patients with advanced vulvar cancer; brachytherapy was reserved for patients with persistent disease that would otherwise necessitate exenteration (Figs. 13.10, 13.11). Rather than performing radical vulvectomy for all patients, only the tumor bed was resected, on the assumption that any microscopic foci originally present in the vulva would have been sterilized by the radiation.
In specimens from one-half of the patients, there was no residual disease. Long-term morbidity was low with the predominant use of teletherapy, and no patient developed a fistula. Two patients whose primary tumor was fixed to bone were long-term survivors (25).
In 1987, Boronow et al. (153) updated their experience with preoperative radiation for locally advanced vulvovaginal cancer, reporting 37 primary cases and 11 cases of recurrent disease. The 5-year survival rate for the primary cases was 75.6%, whereas the recurrent cases had a 5-year survival rate of 62.6%. Seventeen of 40 vulvectomy specimens (42.5%) contained no residual disease. Eight patients (16.7%) had a local recurrence, and five patients (10.4%) developed a fistula.
This second report had three major refinements: (i) the use of external-beam therapy for all cases, with more selective use of brachytherapy; (ii) more conservative vulvar surgery; and (iii) resection of bulky N2 and N3 nodes without full groin dissection to minimize lymphedema.
In 1989, Thomas et al. (154) reported on the use of radiation with concurrent infusional 5-fluorouracil (5-FU), with or without mitomycin C who received primary chemoradiation; six had an initial complete response in the vulva, but three of the six subsequently had a local recurrence.
Several subsequent studies have reported on the use of chemoradiation followed by wide excision of the tumor bed. Italian investigators reported 31 patients with locoregionally advanced vulvar cancer who were treated with a combination of mitomycin C and 5-FU in combination with radiation to the vulva, groins, and pelvis (31). A total of 54 Gy was given, with a 2-week break after 36 Gy. The pathologic complete response rate was 36% in the vulva and 55% in the groin. The 5-year survival rate was 55% for patients treated for primary lesions and 57% for those with recurrent disease. A second similar Italian study of 58 patients reported a pathologic complete response rate of 31% in both the vulva and the groin (32).
The pathologic complete response rate in the vulva with these radiation doses is not greater than that seen with radiation alone, but the local acute toxicity is much greater with the addition of chemotherapy, invariably necessitating at least a 1-week break in therapy.
Cunningham et al. (34) used radiation therapy in combination with cisplatin (50 mg/m2 on day 1) and 5-FU (1,000 mg/m2/24 hours × 96 hours) during the first and last weeks of therapy. Radiation doses to the vulva and groins ranged from 50 to 65 Gy. Nine of 14 patients (64%) had a complete clinical response, and surgical excision of the primary site was not performed in these 9 patients. Only one recurrence was noted with a mean follow-up of 36 months (range, 7 to 81 months).
Leiserowitz et al. (35) omitted groin dissection after preoperative chemoradiation with 5-FU, with or without cisplatin, in 23 patients. No patient failed in the groins, but with a median radiation dose to the groins of only 36 Gy, it is difficult to believe that these results could be duplicated.
Beriwal et al. reported 18 patients having preoperative intensity-modulated radiotherapy and chemotherapy for locally advanced vulvar cancer (30). Fourteen patients had surgery performed with a pathological complete response in 9 patients (64%) and a partial response in 5 (36%).
With the experience now accrued, preoperative radiation, with or without concurrent chemotherapy, should be regarded as the treatment of first choice for patients with advanced vulvar cancer who would otherwise require some type of pelvic exenteration. Chemoradiation is associated with more acute and chronic toxicity, particularly if no break is given during therapy.
Role of Radiation
Radiation therapy, with or without the addition of concurrent chemotherapy, is playing an increasingly important role in the management of patients with vulvar cancer. The indications for radiation therapy in patients with this disease are still evolving. At present, radiation seems to be clearly indicated in the following situations:
Bilateral groin and pelvic radiation is recommended (21) although workers in North England have suggested that unilateral groin and pelvic radiation may be appropriate for patients with unilaterally positive groin nodes. This recommendation (155) was based on experience with 20 patients, nine (45%) of whom recurred. None of the recurrences were in the contralateral (nonirradiated groin).
It has also been suggested that adjuvant radiation may benefit patients with a single positive node if a less extensive groin node dissection has been performed (156).
Possible roles for radiation therapy include the following:
Groin irradiation has been proposed as an alternative to groin dissection in patients with N0 lymph nodes. The GOG reported the results of a phase III trial in which patients with T1, T2, or T3 tumors and N0 or N1 groin nodes were randomized between surgical resection (and postoperative irradiation for patients with positive groin nodes) and primary groin irradiation (160). Patients with N1 nodes were allowed fine-needle aspiration cytologic analysis of the nodes and exclusion from the trial if findings were positive. The study was closed prematurely because 5 of 26 patients in the groin irradiation arm of the study had recurrences in the groin. Of 23 patients undergoing groin dissection, 5 showed groin node metastases, but no groin recurrences occurred after postoperative irradiation. The dose of radiation was 5,000 cGy given in daily 200-cGy fractions to a depth of 3 cm below the anterior skin surface.
Subsequently, Koh et al. (161) reviewed pretreatment CT scans of 50 patients with gynecologic cancer to determine the distance of each femoral vessel beneath the overlying skin surface. Femoral vessel depths in these patients ranged from 2.0 to 18.5 cm, with an average depth of 6.1 cm. It is apparent that many patients in the GOG study would have been underdosed because CT scanning was not used to define the target.
In 2002, van der Velden and Ansink published a Cochrane review of primary groin irradiation versus primary groin dissection for patients with early vulvar cancer (162). Only three studies met the minimum criteria for inclusion in the review, but it was concluded that although groin irradiation is less morbid, it is associated with a higher risk of groin recurrence. Large positive nodes, not likely to be controlled by radiation, may be quite inapparent clinically, even in relatively slim patients, so this finding is not surprising.
Recurrent Vulvar Cancer
Most recurrences from vulvar cancer occur on the vulva, but distant recurrences do occur, particularly in the presence of multiple lymph node metastases (18,163).
Rouzier et al. from France identified three patterns of local recurrence with very different prognoses: (i) primary tumor site recurrence (up to and including 2 cm from the vulvectomy scar); (ii) remote vulvar recurrence (>2 cm from the primary tumor site); and (iii) skin bridge recurrence (80). Their study included 215 patients, and the local relapse-free survival was 78.6% at 5 years. Patients with positive margins who did not receive radiotherapy and patients with greater than 1 mm stromal invasion who did not have a groin dissection were excluded from analysis.
Local recurrence at a site distant from the primary tumor (which could be considered a new primary lesion), had a good prognosis, 66.7% of patients surviving 3 years. By contrast, survival after recurrence at the primary tumor site was poor, only 15.4% of patients surviving 3 years. None of seven patients with a skin bridge recurrence was alive at 1 year.
Review of our own data from the Royal Hospital for Women in Sydney has confirmed that these three patterns of local recurrence are distinct entities (98). In our experience,primary site recurrences occurred at a median interval of 21 months, and were more commonly associated with surgical margins <8 mm. By contrast, remote site vulvar recurrences occurred at a median interval of 69 months, and were commonly associated with lichen sclerosus or VIN. In contrast to Rouzier, our patients with both primary and remote site recurrence had an excellent prognosis.
Local vulvar recurrences are usually amenable to further surgical resection (22,80,164). A variety of plastic surgical techniques may facilitate adequate surgical resection, particularly for larger recurrences. Myocutaneous grafts which may be used include the gluteus thigh flap, the rectus abdominus flap, the gracilis flap, and the tensor fascia lata flap (Fig. 13.12)(165).
Radiation therapy, particularly a combination of external-beam therapy plus interstitial needles, also has been used to treat vulvar recurrences. Hoffman et al. reported on 10 patients treated in this manner, and 9 were still alive with a mean follow-up of 28 months (166). However, 6 of the 10 had severe radionecrosis at a median of 8.5 months after radiation, and the authors concluded that although this treatment was highly effective, it was also highly morbid.
Regional and distant recurrences are difficult to manage (157). Radiation therapy may be used with surgery for groin recurrence, whereas chemotherapeutic agents that have activity against squamous carcinomas may be offered for distant metastases. The most active agents are cisplatin, methotrexate, cyclophosphamide (Cytoxan), bleomycin, andmitomycin C, but response rates are low and the duration of response is usually disappointing (167). Long-term survival is very uncommon with regional or distant recurrence (157).
With appropriate management, the prognosis for vulvar cancer is generally good, the overall 5-year survival rate in operable cases being approximately 70%. Survival correlates with the FIGO clinical stage of disease (Table 13.11) and also with lymph node status. In the 26th FIGO annual report, patients with negative lymph nodes had a 5-year survival rate of 80.7%, which fell to 13.3% for patients with 4 or more positive nodes (Table 13.12).
Figure 13.12 Tensor fascia lata myocutaneous graft used to cover the left groin following resection of a recurrence in an irradiated field.
Table 13.11 Five-Year Survival Rate Versus Stage for Patients Treated With Curative Intent
Table 13.12 Five-Year Survival Versus Lymph Node Status for Squamous Cell Carcinoma of the Vulva
The GOG staged 588 patients with vulvar cancer by the new surgical staging criteria and reported 5-year survival rates of 98%, 85%, 74%, and 31% for stages I, II, III, and IV, respectively (92).
The number of positive groin nodes is the single most important prognostic variable (18,20,21,93,171). Patients with one microscopically positive node have a good prognosis, regardless of the stage of disease (18,20), but patients with three or more positive nodes have a poor prognosis (18,103). Because the number of positive nodes correlates with the clinical status of the groin nodes (18), survival also correlates significantly with this variable. In the GOG study, patients with N0 or N1 nodes had a 2-year survival rate of 78%, compared with 52% for patients with N2 nodes and 33% for patients with N3 nodes (p = 0.01) (21). Extracapsular spread is a poor prognostic factor (124,125,126). The survival rate for patients with positive pelvic nodes is approximately 11% (90).
Workers at the Norwegian Radium Hospital evaluated DNA ploidy for its prognostic significance in 118 squamous cell carcinomas of the vulva (171). The 5-year crude survival rate was 62% for the diploid and 23% for the aneuploid tumors (p <0.001). Aneuploid tumors without lymph node metastases had a 5-year cancer-related survival rate of 44%, compared with 58% for the diploid tumors with lymph node metastases. In a multivariate Cox regression analysis, the most important independent prognostic parameters were:
Figure 13.13 Melanoma of the vulva involving the right labium minus.
A more recent paper from the Norwegian Radiation Hospital evaluated the prognostic significance of aberrant expression of the cell cycle kinase inhibitors p16, p21, and p27 among 224 patients with squamous cell carcinoma of the vulva (173). A low level of p16 protein and a high level of p21 protein were associated with a shorter disease-related survival.
Dutch workers studied 75 patients age 80 years or older, 57 (76%) of whom had standard treatment (174). When preoperatively available parameters of all patients were assessed in relation to survival in the total group, Eastern Cooperative Oncology Group (ECOG) performance status was the only independent prognostic variable. When all clinical and histopathological variables were assessed in the subgroup that had standard treatment, both ECOG performance status and extracapsular lymph node involvement were independent prognostic variables for overall survival. Age was not a significant prognostic variable.
Vulvar melanomas are rare, although they are the second most common vulvar malignancy. Most arise de novo (175), but they may arise from a preexisting junctional nevus. They occur predominantly in postmenopausal white women, most commonly on the labia minoris or the clitoris (Fig. 13.13). The incidence of cutaneous melanomas worldwide is increasing significantly.
Most patients with a vulvar melanoma have no symptoms except for the presence of a pigmented lesion that may be enlarging. Some patients have itching or bleeding, and a few present with a groin mass. Amelanotic varieties occasionally occur. Any pigmented lesion on the vulva should be excised or biopsied, unless it is known to have been present and unchanged for some years.
There are three basic histologic types: (i) the superficial spreading melanoma, which tends to remain relatively superficial early in its development; (ii) the mucosal lentiginous melanoma, a flat freckle, which may become quite extensive but also tends to remain superficial; and (iii) the nodular melanoma, which is a raised lesion that penetrates deeply and may metastasize widely. A Swedish study of 219 cases reported that the mucosal lentiginous melanoma was the most frequent type (57%) (176).
The FIGO staging used for squamous lesions is not applicable for melanomas, because these lesions are usually much smaller and the prognosis is related to the depth of penetration rather than to the diameter of the lesion (177,178,179). The leveling system established by Clark et al. (180) for cutaneous melanomas is less readily applicable to vulvar lesions because of the different skin morphology. Chung et al. (174) proposed a modified system that retained Clark's definitions for levels I and V but arbitrarily defined levels II, III, and IV, using measurements in millimeters. Breslow (181) measured the thickest portion of the melanoma from the surface of intact epithelium to the deepest point of invasion. A comparison of these systems is shown in Table 13.13.
Table 13.13 Microstaging of Vulvar Melanomas
A revised American Joint Committee on Cancer (AJCC) Staging System for cutaneous melanomas came into effect in 2002 to reflect the new prognostic factors that have been found to be important in predicting survival (182). These factors include primary tumor thickness (replacing level of invasion), ulceration, number of metastatic lymph nodes, micrometastatic disease based on the sentinel lymph node biopsy technique or elective node dissection, the site(s) of distant metastatic disease, and serum lactate dehydrogenase (LDH) levels.
With better understanding of the prognostic significance of the microstage, some individualization of treatment has developed. Lesions with less than 1 mm of invasion may be treated with radical local excision alone (174,175). Traditionally, for more invasive lesions, en bloc radical vulvectomy and resection of regional groin nodes has been performed.
More conservative surgery for cutaneous melanomas commenced in the 1980s (183,184), and although vulvar melanomas seem to be biologically different and carry a much worse prognosis than cutaneous melanoma (185), this trend has been followed (186,187,188,189). In 1987, Davidson et al. reported on 32 patients with vulvar melanoma who underwent local excision (n = 14), simple vulvectomy (n = 7), or radical resection (n = 11) (184). No group had a superior survival, although the overall survival rate at 5 years was only 25%. Trimble et al. reported on 59 patients who underwent radical vulvectomy and 19 who underwent more conservative resections (188). Survival was not improved by the more radical approach, and they recommended radical local excision for the primary tumor, with groin dissection for tumors thicker than 1 mm. In 1994, the Gynecologic Oncology Group (GOG) conducted a prospective clinicopathologic study of 71 evaluable patients with melanoma of the vulva diagnosed between 1983 and 1990 (190). All patients were required to have a modified radical hemivulvectomy as minimal therapy. Seven of 37 patients (19%) having radical vulvectomy developed a local recurrence, compared with 3 of 34 (9%) having a hemivulvectomy.
The advisability of groin node dissection is controversial. The Intergroup Surgical Melanoma Program conducted a prospective, multiinstitutional, randomized trial of elective lymph node dissection versus observation for intermediate thickness cutaneous melanomas (1 to 4 mm) (191). There were 740 patients entered into the trial, and elective lymph node dissection resulted in a significantly better 5-year survival rate for the 522 patients 60 years of age or younger (88% vs. 81%; p <0.04), the 335 patients with tumors 1 to 2 mm thick(96% vs. 86%; p <0.02), the 403 patients without tumor ulceration (95% vs. 84%; p <0.01), and the 284 patients with tumors 1 to 2 mm thick and no ulceration (97% vs. 87%; p <0.005).
De Hulla et al. reported 9 patients with vulvar melanoma who underwent a sentinel node procedure (192). Two of nine (22%) patients developed a groin recurrence after having negative sentinel nodes, compared to 0 of 24 patients who were treated conventionally (p = 0.06). The authors postulated that the recurrences were due to in-transit metastases. In a study of 344 patients with cutaneous melanomas treated at the M.D. Anderson Hospital, 27 of 243 patients (11%) with a histologically negative sentinel node developed local, in transit, nodal, and/or distant metastases after a median follow-up of 35 months (193). Ten patients (4%) developed a nodal recurrence in the previously mapped basin, but ultrastaging demonstrated evidence of occult micrometastases in 80% of these 10 cases. The authors concluded that the data provided further support for sentinel node biopsy in patients with cutaneous melanomas.
As with sentinel nodes in squamous carcinoma of the vulva, sentinel node biopsy in patients with vulvar melanoma is a compromise operation, but in a well informed patient, in the hands of an experienced team, and with ultrastaging of the negative nodes, lymphatic mapping would seem to be significantly superior to no node dissection for patients with more than 1 mm of stromal invasion.
Pelvic node metastases do not occur in the absence of groin node metastases (194,195). In addition, the prognosis for patients with positive pelvic nodes is so poor that there appears to be no value in performing pelvic lymphadenectomy for this disease.
As melanomas commonly involve the clitoris and labia minora, the vaginourethral margin of resection is a common site of failure, and care should be taken to obtain an adequate “inner” resection margin. Podratz et al. (179) demonstrated a 10-year survival rate of 61% for lateral lesions, compared with 37% for medial lesions (p <0.027).
The author's current policy is to perform a radical local excision with 1-2 cm margins for the primary lesion. In patients with more than 1 mm of stromal invasion, at least an ipsilateral inguinofemoral lymphadenectomy is performed. Sentinel node biopsy is reserved for the few patients who don't want to take the 50% to 60% risk of developing lymphedema.
Interferon alpha-2b (IFN-α-2b) is the first agent to show significant value as an adjuvant for melanoma in a randomized controlled trial (196). The Eastern Cooperative Oncology Group entered 287 patients onto an adjuvant trial of high-dose IFN-α-2b after surgery for deep primary (>4 mm) or regionally metastatic melanoma. With a median follow-up of 6.9 years, there was a significant prolongation of relapse-free and overall survival for the group receiving interferon. The proportion of patients who remained disease free also improved from 26% to 37%.
The results were confirmed in a larger intergroup trial that compared the efficacy of high-dose interferon-α-2b for 1 year with vaccination using GM2 conjugated to keyhole limpet hemocyanin (197). Eight hundred and eighty patients were randomized, and the trial was closed after interim analysis indicated inferiority of the vaccination compared with high-dose interferon-α-2b.
High-dose interferon regimens cause significant morbidity, but should be considered standard therapy for all high-risk melanoma patients expected to be able to tolerate the interferon (198). Immunotherapy for melanoma includes a number of different strategies with vaccines utilizing whole cell tumors, peptides, cytokine-mediated dendritic cells, DNA and RNA, and antibodies. Although initial clinical trials are promising, these approaches remain experimental (199).
The behavior of vulvar melanomas can be quite unpredictable, but the overall prognosis is poor. The mean 5-year survival rate for reported cases of vulvar melanoma ranges from 21.7% (175) to 54% (176). Patients with lesions invading to 1 mm or less have an excellent prognosis, but as depth of invasion increases, prognosis worsens. Chung et al. (177) reported a corrected 5-year survival rate of 100% for patients with level II lesions, 40% for level III or IV lesions, and 20% for level V lesions. Tumor volume has been reported to correlate with prognosis, with patients whose lesion has a volume less than 100 mm3 having an excellent prognosis (195). DNA ploidy and angioinvasion have been shown to be independent prognostic factors for disease-free survival (200).
Bartholin Gland Carcinoma
Primary carcinoma of the Bartholin gland accounts for approximately 5% of vulvar malignancies. Because of its rarity, individual experience with the tumor is limited, and recommendations for management must be based on literature reviews (73,201).
The bilateral Bartholin glands are greater vestibular glands situated posterolaterally in the vulva. Their main duct is lined with stratified squamous epithelium, which changes to transitional epithelium as the terminal ducts are reached. Because tumors may arise from the gland or the duct, a variety of histologic types may occur, including adenocarcinomas, squamous carcinomas, and, rarely, transitional cell, adenosquamous, and adenoid cystic carcinomas. One case of small cell neuroendocrine cancer of the Bartholin gland has been reported (202).
Classification of a vulvar tumor as a Bartholin gland carcinoma has typically required that it fulfill criteria proposed by Honan in 1897. These criteria are:
Strict adherence to these criteria results in underdiagnosis of some cases. Large tumors may ulcerate through the overlying skin and obliterate the residual normal gland. Although transition between normal and malignant tissue is the best criterion, some cases are diagnosed on the basis of their histologic characteristics and anatomic location.
Bartholin gland carcinomas are often misdiagnosed initially as a Bartholin cyst or abscess. A study from Tampa reported that 8 of 11 cases had initially been treated for an infectious process before referral (203). Hence, delay of diagnosis is common, particularly in premenopausal patients. Other differential diagnoses of any pararectovaginal neoplasm should include cloacogenic carcinoma and secondary neoplasm (201).
The adenoid cystic variety accounts for approximately 10% of Bartholin gland carcinomas (204,205). The largest series has been reported from the University of Michigan, where 11 cases were seen over a 58-year period (201). It is a slow-growing tumor with a marked propensity for perineural and local invasion. The perineural infiltration is quite characteristic and may account for the pruritus and burning sensation that many patients experience long before a palpable mass is evident (206).
Although treatment has traditionally included radical vulvectomy and bilateral groin dissection, Copeland et al. (201) at the M. D. Anderson Hospital have reported good results with hemivulvectomy or radical local excision for the primary tumor. Because these lesions are deep in the vulva, extensive dissection is required in the ischiorectal fossa, and, this is facilitated by performing an en bloc resection of the primary lesion and the ipsilateral inguinofemoral lymph nodes (Fig. 13.14). Postoperative radiation to the vulva decreased the likelihood of local recurrence in Copeland's series from 27% (6 of 22) to 7% (1 of 14). If the ipsilateral groin nodes are positive, bilateral groin and pelvic radiation may be indicated, based on the same criteria as apply for squamous cell carcinomas.
Figure 13.14 En bloc resection of the right groin and right-posterior vulva for a Bartholin gland carcinoma. Note the preservation of the clitoris and right anterior labium minus.
Workers at the Massachusetts General Hospital reported 10 women with Bartholin's gland carcinoma who were treated with primary chemoradiation to the primary tumor and regional lymph nodes (207). There were four patients with stage I disease, one with stage II, three with stage III and two with stage IV. The 5-year survival was 66%, and the authors concluded that chemoradiation offered an effective alternative to surgery. Primary chemoradiation should certainly be used if the tumor is fixed to the inferior pubic ramus or involves adjacent structures, such as the anal sphincter or rectum, in order to avoid exenterative surgery.
Radical local excision, with or without ipsilateral inguinal-femoral lymphadenectomy, is also the treatment of choice for the primary lesion with adenoid cystic carcinomas, and adjuvant radiation is recommended for positive margins or perineural invasion.
Because of the deep location of the gland, cases tend to be more advanced than squamous carcinomas at the time of diagnosis, but stage for stage, the prognosis is similar.
Adenoid cystic tumors are less likely to metastasize to lymph nodes and carry a somewhat better prognosis. Late recurrences may occur in the lungs, liver, or bone, so 10- and 15-year survival rates are more appropriate when evaluating therapy (208). The slowly progressive nature of these tumors is reflected in the disparity between progression-free interval and survival curves (209).
Other Vulvar Adenocarcinomas
Adenocarcinomas of the vulva usually arise in a Bartholin gland or occur in association with Paget's disease. They may rarely arise from the skin appendages, paraurethral glands, minor vestibular glands, aberrant breast tissue, endometriosis, or a misplaced cloacal remnant (116).
A particularly aggressive type is the adenosquamous carcinoma. This tumor has a number of synonyms, including cylindroma, pseudoglandular squamous cell carcinoma, and adenoacanthoma of the sweat gland of Lever. The tumor has a propensity for perineural invasion, early lymph node metastasis, and local recurrence. Underwood et al. (210) reported a crude 5-year survival rate of 5.6% (1 of 18) for adenosquamous carcinoma of the vulva, compared with 62.3% (48 of 77) for patients with squamous cell carcinoma. Treatment should be by radical vulvectomy and bilateral groin dissection, and postoperative radiation may be appropriate.
Basal Cell Carcinoma
Basal cell carcinoma (BCC) is the most common human malignant neoplasm. As with melanomas, its incidence is strongly correlated with sun exposure, and the vast majority occurs in the head and neck region. Of 3,604 cases of BCC seen at the University of Florence, Italy, between 1995 and 2003, there were 63 cases (1.7%) arising on the vulva (211).
Basal cell carcinomas represent 2% to 4% of vulvar cancers. As with other basal cell carcinomas, vulvar lesions commonly appear as a “rodent ulcer” with rolled edges, although nodules and macules are other morphologic varieties. Most lesions are smaller than 2 cm in diameter and are usually situated on the anterior labia majora. Giant lesions occasionally occur (212).
Basal cell carcinomas usually affect postmenopausal white women, a Vancouver study reporting a mean age of 74 years (213). They are locally aggressive, and radical local excision usually is adequate treatment. They are moderately radiosensitive, so radiation may be useful in selected cases. Metastasis to regional lymph nodes has been reported but is rare(214,215), and there has been one reported case with hematogenous spread (212). The duration of symptoms prior to diagnosis is usually several years in patients with metastatic basal cell carcinomas (215). The local recurrence rate is 10% to 20% (215).
Approximately 3% to 5% of basal cell carcinomas contain a malignant squamous component, the so-called basosquamous carcinoma. These lesions are more aggressive and should be treated as squamous carcinomas (214). Another subtype of basal cell carcinoma is the adenoid basal cell carcinoma, which must be differentiated from the more aggressive adenoid cystic carcinoma arising in a Bartholin gland or the skin (215).
Verrucous carcinomas are most commonly found in the oral cavity, but may be found on any moist membrane composed of squamous epithelium (216). They are a distinct entity, with no association with human papillomavirus infection, and a peculiar distribution pattern of cytokeratins AE1 and AE3 on immunohistochemical staining (217).
Grossly, the tumors have a cauliflower-like appearance, and the diameter of reported lesions ranges from 1 to 15 cm (218). Microscopically, they contain multiple papillary fronds that lack the central connective tissue core that characterizes condylomata acuminata. The gross and microscopic features of a verrucous carcinoma are very similar to those of thegiant condyloma of Buschke-Loewenstein, and they probably represent the same disease entity (116). Adequate biopsy from the base of the lesion is required to differentiate a verrucous carcinoma from a benign condyloma acuminatum or a squamous cell carcinoma with a verrucous growth pattern.
Clinically, verrucous carcinomas usually occur in postmenopausal women, and they are slowly growing but locally destructive lesions. Even bone may be invaded. Metastasis to regional lymph nodes is rare but has been reported (219).
Radical local excision is the basic treatment, although if there are palpably suspicious groin nodes, these should be evaluated with fine-needle aspiration cytologic testing or excisional biopsy. Usually, enlarged nodes are due to inflammatory hypertrophy (220). If the nodes contain metastases, radical local excision and at least an ipsilateral inguinofemoral lymphadenectomy are indicated.
Vulvar intraepithelial neoplasia or invasive squamous cell carcinoma may be seen in association with verrucous carcinoma. A Greek study of 17 cases diagnosed over a 12-year period reported coexistence of verrucous and squamous carcinoma of the vulva in 6 cases (35%) (221).
Radiation therapy is contraindicated because it may induce anaplastic transformation with subsequent regional and distant metastasis (222). Japaze et al. (220) reported a corrected 5-year survival rate of 94% for 17 patients treated with surgery alone, compared with 42% for 7 patients treated with surgery and radiation. If there is a recurrence, further surgical excision is the treatment of choice. This may occasionally necessitate some type of exenteration.
Sarcomas represent 1% to 2% of vulvar malignancies and comprise a heterogenous group of tumors. Leiomyosarcomas are the most common, and other histologic types includefibrosarcomas, neurofibrosarcomas, liposarcomas, rhabdomyosarcomas, angiosarcomas, epithelioid sarcomas, and malignant schwannomas (116). A recent paper from Johns Hopkins reported seven cases of vulvar sarcoma among 453 patients with vulvar malignancies seen from 1977 to 1997, an incidence of 1.5% (223).
The primary treatment is wide surgical excision (221). Adjuvant radiation may be helpful for high-grade tumors and locally recurrent low-grade lesions (222). The overall survival rate is approximately 70%. There were no recurrences in the series from Johns Hopkins (220), with follow-up ranging from 60 to 172 months. Only one of their patients had groin dissection.
Leiomyosarcomas usually appear as enlarging, often painful masses, usually in the labium majus. In a review of 32 smooth-muscle tumors of the vulva, in 1979, Tavassoli and Norris (226) reported that recurrence was associated with three main determinants: diameter greater than 5 cm, infiltrating margins, and five or more mitotic figures per 10 high-power fields. The absence of one, or even all, of these features did not guarantee that recurrence would not occur (226). A more recent study from the Massachusetts General Hospital suggested that a vulvar smooth muscle tumor should be considered a sarcoma when three or all of the following four features were present: (i) over 5 cm in greatest dimension; (ii) infiltrate margins; (iii) 5 mitoses/10 HPF; and (iv) moderate to severe cytologic atypia (227).
Lymphatic metastases are uncommon, and radical local excision is the usual treatment.
Epithelioid sarcomas characteristically develop in the soft tissues of the extremities of young adults but may rarely occur on the vulva. Ulbright et al. (228) described two cases and reviewed three other reports. They concluded that these tumors may mimic a Bartholin cyst, thus leading to inadequate initial treatment. They also suggested that vulvar epithelioid sarcomas behave more aggressively than their extragenital counterparts, with four of the five patients dying of metastatic disease.
Epithelioid sarcomas in general have a propensity for extensive local disease at presentation, local recurrence, lymph node metastasis, and distant metastasis (229,230). Treatment consists of radical excision of the tumor, and at least ipsilateral groin dissection. Systemic therapy is ineffective.
Rhabdomyosarcomas are the most common soft tissue sarcomas in childhood, and 20% involve the pelvis or genitourinary tract (231). Dramatic gains have been made in the treatment of these tumors since the late 1970s. Previously, radical pelvic surgery was the standard approach, but results were poor. A multimodality approach has evolved, principally as a result of four successful protocols organized by the Intergroup Rhabdomyosarcoma Study Group (IRSG), and survival rates have improved significantly, with a corresponding decrease in morbidity (232).
Arndt et al. summarized the results of these four protocols in 2001. There were 151 patients entered on the studies, and the vulva was the least common primary site, there being only 20 (13%) vulvar rhabdomyosarcomas. Only 5 (25%) of the patients were 15 years or older, and the histologic subtypes were enbryonal 8 (40%); botryoid 3 (15%); and alveolar/undifferentiated 9 (45%). All were managed with chemotherapy (vincristine, dactinomycin ± cyclophosphamide ± doxorubicin), with or without radiation therapy. Wide local excision of the tumor, with or without inguinofemoral lymphadenectomy, was carried out before or after the chemotherapy.
Patients with local and/or regional rhabdomyosarcoma of the female genital tract have an excellent prognosis, with an estimated 5-year overall survival of 87% (232).
The genital tract may be involved primarily by malignant lymphomas, but involvement more commonly is a manifestation of systemic disease. In the lower genital tract, the cervix is most commonly involved, followed by the vulva and the vagina (116). Most patients are in their third to sixth decade of life, and approximately three-fourths of the cases involve diffuse large cell or histiocytic non-Hodgkin's lymphomas. The remainder are nodular or Burkitt's lymphomas (233). Treatment is by surgical excision followed by chemotherapy and/or radiation, and the overall 5-year survival rate is approximately 70% (233).
Endodermal Sinus Tumor
There have been eight case reports of endodermal sinus tumor of the vulva. Three of the eight for whom data were available died of distant metastases (116,234). Most patients were young adults with a median age of 21 years, unlike vaginal yolk sac tumors, which usually occur in infants. Unlike their ovarian counterpart, they are not always associated with elevated serum alpha fetoprotein titers. Spread to lymph nodes occurs early, so wide excision of the primary tumor, ipsilateral groin dissection, and platinum based chemotherapy appears to be the most appropriate management (234).
Merkel Cell Carcinoma
Merkel cell carcinomas are primary small cell carcinomas of the skin that resemble oat cell carcinomas of the lung. They metastasize widely and have a very poor prognosis (235,236). They should be locally excised and treated with cisplatin-based chemotherapy.
This is a rare, low-grade cutaneous malignancy of the dermal connective tissue that occasionally involves the vulva. It has a marked tendency for local recurrence but a low risk of systemic spread (237,238). Radical local excision should be sufficient treatment.
Five cases of malignant schwannoma in the vulvar region have been reported. The patients ranged in age from 25 to 45 years. Four of the five were free of tumor from 1 to 9 years after radical surgery, and the fifth patient died of multiple pulmonary metastases (116).
Secondary Vulvar Tumors
Eight percent of vulvar tumors are metastatic (116). The most common primary site is the cervix, followed by the endometrium, kidney, and urethra. Most patients in whom vulvar metastases develop have advanced primary tumors at presentation, and in approximately one-fourth of the patients, the primary lesion and the vulvar metastasis are diagnosed simultaneously (239).