Radiation Therapy for Vulvar Carcinoma

Carcinoma of the vulva is an uncommon neoplasm accounting for between 4% and 5% of all gynecologic primary tumors.¹ It is primarily a disease of older women (median age, 61 years) with a peak incidence in the 7th decade of life. However, it is being diagnosed with increasing frequency among younger premenopausal women. Approximately 15% of patients are diagnosed before age 40. Although tumors with several different histologic features can arise within the vulva or its associated adnexal structures, the preponderance of primary carcinomas of the vulva are squamous cell and arise from the epithelium.² The most frequent sites of involvement are the labia minora, clitoris, fourchette, perineal body, and the medial labia majora.

The etiology of vulvar cancer is unknown and likely multifactorial. Vulvar intraepithelial neoplasia (VIN) is found adjacent to these tumors in approximately 30% to 65% of cases.^3,4 Although most VIN lesions never become invasive, as many as one third of VIN I/II lesions persist, progress, or recur after a short remission and approximately 20% of carcinoma in situ lesions harbor an invasive component.^5,6 The direct progression of VIN to cancer is difficult to document. In addition, the natural history of these lesions appears to have age-related factors. For instance, Basta and colleagues found not only was carcinoma in situ and stage I carcinoma increasing in frequency in women under age 45 but also that this cohort was more likely to have multifocal disease (63% versus 32%).⁵ In addition, the younger cohort was more likely to have lesions associated with human papillomavirus infection (62% versus 18%). Treatment strategies for vulvar cancer are guided by observations of natural history and patterns of local and metastatic spread. This section examines the role radiotherapy plays in treating primary vulvar malignancy.

The classic pattern of growth and spread is local extension and lymphovascular embolization to either regional or distant sites. With the exception of vulvar melanoma, the pattern of spread is relatively predictable. Local growth of the primary tumor generally occurs in a radial pattern as well as invasion deep into the underlying subcutaneous tissues. Objective measures of this growth have been standardized with creation of the tumor-node-metastasis (TNM) staging system.⁷ In this manner, tumor is described in reference to size (T1 is 2 cm or smaller; T2 is larger than 2 cm), involvement of the distal midline structures (T3: urethra, vagina, anus) and invasion of the midline organs (T4: bladder mucosa, upper urethra, rectal mucosa) or bony fixation.

After invasion has occurred, the tumor can spread through the dermal lymphatics to the regional lymph nodes following the pattern described in other chapters. There appears to be a threshold with regard to the likelihood of groin metastatic disease based on the depth of invasion. Kelley and colleagues reported that the frequency of metastatic disease among 24 patients with 1 mm or less of invasion (measured as depth of deepest invasion from the most proximal adjacent rete peg) was 0%.⁸ This observation led to a subclassification of stage I disease with the clinical implication being that nodal dissection may safely be dismissed in these patients.⁹

Careful inspection of en bloc tumor resections and biopsies by Cherry and Glücksmann demonstrated that lymphatic tumor emboli were present in only 19% of cases.¹⁰ Clinically, recurrences in the retained skin bridges among patients undergoing separate incision removal of the primary lesion and the regional lymphatics are uncommon. This observation would support the hypothesis that lymphatic metastases occur as a result of embolization and most often without interval deposition of viable tumor.

Clinical experience of lymphatic metastatic disease suggests the descriptions by Parry-Jones of normal vulvar lymphatic drainage also hold for invasive disease. Lymphatic metastases from lateralized lesions rarely involve the contralateral groin in the absence of ipsilateral metastatic disease and spread from midline lesions can be identified in either or both groins.¹¹ Treatment paradigms are founded on these described principles of local invasion and potential metastatic spread sites.

The primary clinical treatment of vulvar carcinoma is surgical extirpation and it is discussed in detail elsewhere. However, the risk of local and regional recurrence and occasionally, a lack of surgical candidacy have ensured a dominant role of radiotherapy in ultimate curative treatment programs. This section elucidates the role radiotherapy plays in various clinical situations.

Postoperative Radiation Therapy

ADJUVANT TREATMENT OF THE GROIN AND PELVIS.

Previously, pelvic lymph node dissection was routinely performed in patients with a node-positive groin dissection. Because survival is markedly reduced in patients with pathologically involved inguinal lymph nodes, controversy surrounded the appropriate extent of lymph node dissection. Although the risk of pelvic lymph node involvement with the pathologically negative groin was reported to be negligible, it was postulated that in as many as 20% of patients, the pelvic nodes may harbor subclinical disease. Pelvic lymphadenectomy was supported by some even when the groin dissection is negative.^{12,13,14,15,16,17}

Hacker and Berek showed that pelvic lymph node dissection does not contribute to survival when the groin nodes are pathologically negative. In their review of 79 patients with negative groin nodes and 16 patients with only one involved node, omission of pelvic lymphadenectomy did not affect survival.¹⁵

The addition of pelvic lymphadenectomy may be associated with increased complications.¹⁸ GOG Protocol 37 randomly assigned 114 patients with invasive vulvar cancer and positive groin nodes after radical vulvectomy and bilateral groin lymphadenectomy to receive pelvic lymph node dissection or 45 to 50 Gy of radiation therapy to the groin and pelvis. The primary site was not radiated. The estimated 2-year survival rate was 68% in the radiation group and 54% in the pelvic dissection group (p = .03). Pelvic recurrence was 6.8% in the radiation group and 1.8% in the pelvic node resection group. More significantly, groin recurrence was 5.1% in the radiation group and 23.6% in the node resection group, supporting the strong association between prevention of groin relapse and survival with radiation. The survival benefit with radiation in this study was shown for patients with clinically fixed nodes, two or more positive groin nodes, or both. Early and late morbidity was similar in both groups.⁹

The population sample of GOG Protocol 37 was too small to conclude that radiation was of benefit for patients with only one positive groin node. Mariana and coworkers published a review of 58 patients following radical vulvectomy and bilateral inguino-femoral groin dissection. Twenty patients had positive groin nodes and most of those who received treatment (9 of 16) had only one positive node. Sixteen patients received 45 to 50 Gy to the dissected groin with cobalt-60 (Co-60). The 64% survival rate seen in the radiated patients was comparable with results of GOG 37 and better than historical results with surgery alone.²⁰

The need for whole pelvic radiation in patients with a single positive groin node has not been prospectively established and treatment in these cases may be individualized; however, in the presence of clinically fixed or at least two positive groin nodes, adjuvant groin and elective whole pelvic radiation is the standard of care.

ADJUVANT TREATMENT OF THE PERINEUM.

Local and nodal recurrence characterizes most failures after surgical treatment of vulvar cancer. Indications for radiation therapy of the perineum have been identified following analysis of GOG studies and by histopathologic review of patients treated with surgery alone. Heaps and coworkers concluded that margins smaller than 8 mm were associated with a higher risk of local failure.²¹ Review of 184 patients from GOG Protocol 36 noted that presence of tumor size larger than 4 cm and lymphovascular invasion was associated with a 21% risk of local recurrence, whereas absence of these risk factors reduced local failure to 9%.²²

Faul and partners retrospectively analyzed the impact of adjuvant radiation to the perineum in 62 patients with vulvar cancer following radical vulvectomy (52 of 62) and wide local resection (10 of 62) with either positive or close (smaller than 8-mm) margins. The type of surgery was equally divided between the radiation and observation groups. The 31 patients who received post-operative radiation had 16% local recurrence whereas the 31 patients treated with surgery alone had 58% local recurrence (p < .05). The 2-year actuarial survival rate after local recurrence was 25%. The benefit of radiation was seen whether margins were close or positive. Subset analysis showed that local failure with positive margins was reduced from 75% to 42.5% after a mean dose of 57 Gy was delivered to the perineum. Local failure with close margins was reduced from 40% to 6% after a mean dose of 49 Gy was delivered to the perineum.²³

The value of adjuvant perineal radiation was confirmed at the Mallinckrodt Institute of Radiology where 25 patients treated with wide local resection or partial vulvectomy received adjuvant radiation to the perineum. Local control following wide local excision was reduced from 100% to 75% if the radiation dose was less than 50 Gy, whereas there appeared to be no dose-response above 50 Gy for patients treated with partial vulvectomy. Advanced lesions (T3, T4) appeared to benefit from doses higher than 60 Gy as local control increased from 62% to 80%^24,25 (p = NS).

Current recommendations for postoperative radiation treatment of the perineum include close or positive margins, depth of invasion more than 5 mm, lymphovascular invasion, or an infiltrative pattern of growth.

ELECTIVE TREATMENT OF THE GROIN.

Radiation therapy may be given to the dissected groin in patients with positive inguinofemoral lymph nodes or to the undissected groin following wide local tumor excision when the risk of groin involvement is considered high. The GOG prospectively evaluated 588 patients with squamous cell carcinoma of the vulva reporting five independent factors predicting positive groin nodes: high tumor grade, suspicious groin nodes on examination, lymphovascular invasion, advanced age, and tumor thickness. Inaccuracy of clinical node assessment was illustrated by one fourth of patients with clinically negative groins (N0, N1) were subsequently found to be positive on pathologic assessment.⁹ Because groin recurrence is nearly uniformly fatal, prophylactic irradiation in appropriately selected patients potentially reduces both local failure and improves survival.

Standard treatment of vulvar cancer with a clinically negative groin has been radical vulvectomy or wide local tumor excision with bilateral inguinofemoral node dissection. Given that 70% to 80% of N0 patients have pathologically negative nodes, most patients undergoing groin dissection do not benefit from the procedure, which is associated with significant morbidity. A reasonable alternative has been elective groin irradiation. An early report from the University of Florida treated 6 patients with N0-1 nodes with elective bilateral groin irradiation to 45 Gy in 5 weeks and observed no groin failures or treatment complications.²⁶ This experience led to elective inguinal lymph node irradiation for multiple pelvic carcinomas associated with inguinal metastasis. The most recent update of this experience included 18 patients with vulvar cancer treated to 45 to 50 Gy to the clinically negative groin; 81% had control of disease in the groin.^27,28

Following publication of the results of GOG Protocol 37, pelvic radiation therapy and irradiation of the involved groin have become the standard of care for patients with positive groin dissections. These results and encouraging retrospective data led to an effort to omit elective groin dissection. GOG Protocol 88 prospectively randomized 58 patients with N0-1 groins postradical vulvectomy to receive groin radiation to 50 Gy or groin dissection. There was an 18.5% groin relapse rate in the radiation arm and no groin failures in the dissection arm prompting early termination of the study. Both progression-free interval and survival were significantly better in the groin dissection arm. The conclusion of the study was that radiation of the undissected groin was significantly inferior to groin dissection.²⁹

Two explanations for the poor results in the radiation group have been examined. First, the dose of radiation was prescribed to a depth of 3 cm in all patients although inguinal lymph node depth can vary widely, contributing to underdosage of the groin at risk. Second, 20% to 25% of the patients in the radiation group could be expected to have occult groin involvement so that 50 Gy would be considered an inadequate dose. Nonetheless, elective groin radiation for the N0-1 vulvar cancer is not routinely given unless groin dissection is otherwise contraindicated.

Manavi and associates reviewed 135 patients with clinically negative groins who had been treated with simple vulvectomy. No radiation therapy was given to the primary site. Sixty-five patients received an average of 60 Gy to the undissected groin with Co-60. Seventy remaining patients were observed without radiation. Inguinal relapse was 4.6% in the radiation group and 10% in the observation group. Total survival time was similar in both groups. Significant complications occurred in 7.7% of the radiation group and 2.9% of the observation group. The study had insufficient power to evaluate the distribution of prognostic factors in the two groups but suggested that groin irradiation may be omitted in low risk patients, that is, primaries with no central location, no lymphovascular invasion, thickness of 2 mm or less, and low and intermediate grade tumors.³⁰ With adequate dose delivery to the inguinal lymph nodes based on careful planning using computed tomography (CT), radiation may be an appropriate and less morbid substitute for bilateral groin dissection in low-risk patients but cannot be strongly advocated at this time without further prospective evaluation.

Preoperative Radiation Therapy and Organ Preservation

PREOPERATIVE RADIATION FOLLOWED BY ORGAN SPARING SURGERY: THE EARLY EXPERIENCE.

Pelvic exenteration to adequately resect locally advanced vulvar cancer is associated with major complications including wound infections, fistulas, and 10% postoperative mortality. Exenteration includes removal of the bladder and rectum. Efforts to decrease the extent of surgery, preserve rectal and bladder function, and avoid stomas led to attempted combinations of external beam radiation, brachytherapy, and surgery in selected patients.^4,12,31 Berven in pioneering work in the 1940s presented evidence demonstrating the feasibility of preoperative radiation therapy with radium in 286 patients with locally advanced disease. Electrocoagulation of the lesion and teletherapy followed by en bloc resection resulted in a 38% 5-year cure rate.⁴ Boronow in 1973 presented 9 patients with advanced lesions treated with brachytherapy and external beam radiation followed by radical vulvectomyin place of pelvic exenteration. Only 1 patient recurred locally and survival was similar to that seen in exenteration series. Vaginal stenosis was the commonest complication.⁶ He updated his experience in 1982 with 26 primary cases and 7 recurrences. The whole pelvis was treated to 20 Gy followed by 20 Gy to the groin and pelvic lymph nodes blocking the midline structures. Finally, two radium applications to the vulvar primary were delivered. Disease-free survival rate with 1- to 11-year follow-ups was 65%. The disease-free survival rate for the recurrent cases was 71% with 1- to 3-year follow-up.⁵

Further and more recent evidence that a conservative approach to advanced disease can be offered without compromising survival was presented by Hacker and colleagues who reviewed their experience with 8 patients who needed pelvic exenteration. All patients received 44 to 54 Gy to the primary lesion and regional lymph nodes. Conservative excision was possible in 87.5%, 50% had a complete pathologic response, and 5 patients were alive from 15 months to 10 years following treatment. One patient developed bilateral hip fractures. Half these patients suffered uncomplicated moist desquamation.³²

Rotmensch and colleagues in 1990 described 16 patients with stage III or IV tumors treated with 40 Gy to the primary site and 45 Gy to the groin and pelvis, then resection with radical vulvectomy and groin dissection 1 month later. The 5-year survival rate was 45%, the recurrence rate was 46%, and the visceral preservation rate was 63%.³³ These early results supported preoperative radiation therapy and radical vulvectomy as a reasonable alternative to pelvic exenteration.

PREOPERATIVE RADIATION AND CONSERVATIVE SURGERY: THE ERA OF CHEMOTHERAPY.

Chemotherapy is directly cytotoxic to cancer cells and sensitizes tumor cells to the lethal effects of radiation. Chemosensitization of tumor cells was the basis of encouraging clinical work in the treatment of anal canal carcinoma in the United States and in Europe. The demonstrated ability of chemoradiation to preserve organ function in anal canal carcinoma patients, as well as the modest success with preoperative radiation alone to spare exenterative procedures in vulvar cancer, led to several efforts to improve functional outcome without compromising survival by administering concurrent preoperative chemotherapy and radiation.⁹ To this end, during the past 10 years a substantial body of evidence has accumulated using various chemotherapeutic agents and radiation fractionation schemes. With improved radiation-planning techniques and ever-expanding experience with chemotherapy, organ preservation in vulvar carcinoma continues to be an exciting frontier in the management of gynecologic cancers.

The addition of chemotherapy to radiation has the theoretical advantage of allowing for increased acute tumoricidal effects while reducing risk to normal tissue of late toxicity, which increases with higher doses of radiation. As tumor bulk increases as typically seen in locally advanced disease, higher total doses (more than 60 Gy) of radiation are required for disease control. Before sequencing, dosing, or timing guidelines had been established, Thomas and colleagues in the 1980s treated 33 patients at Princess Margaret Hospital with radiation therapy and concurrent 5-fluorouracil (5-FU) and mitomycin C to examine the efficacy and toxicity of combined modality therapy for advanced primary or recurrent vulvar cancer. Several clinical situations were studied that produced various radiation doses and fractionation schemes. Radiation dose to the vulva did not exceed 60 Gy. 5-FU was given as a continuous infusion over 4 or 5 days for one or two courses in most patients. Mitomycin C was also given in one or two injections. Although this approach was used in preoperative, postoperative, and recurrent settings and also in place of surgery, combined modality therapy appeared to be well tolerated without requirement of treatment breaks and allowed for less extensive surgery in selected cases.³⁴

Substantiating and extending the findings of Thomas and coworkers, Berek and colleagues at the University of California at Los Angeles reported the results of a phase II trial exploring pathologic response and toxicity with the use of preoperative chemoradiation. Cisplatin (50 to 100 mg/m²) and 5-FU (1000 mg/m²) as a continuous infusion over 4 to 5 days were given at days 1 and 29 of radiation to 12 patients with stage III or IV disease. The pelvic lymph nodes, groin, and vulva were treated to a dose of 44 to 54 Gy. A complete pathologic response was observed in 67% of the patients and no further treatment was offered. A partial response was seen in 3 patients and no response was seen in 1 patient, all of whom had radical vulvectomy or exenteration. With a median follow-up of 37 months, 10 of 12 patients had no evidence of disease. Most patients suffered only grade 2 toxicity and there was no grade 4 or 5 toxicity.³ The strong association between a complete pathologic response and long-term local control has also been demonstrated in several other reports.^{35,36,37,38,39}

The combination of mitomycin C and 5-FU in the preoperative setting was more recently reviewed at Loma Linda University where 19 patients with advanced vulvar lesions received concurrent radiation to 45 to 50 Gy to the primary, pelvis, and groin with a 96-hour infusion of 5-FU (1000 mg/m²) in weeks 1 and 5. The last few patients in the series also received mitomycin C (10 mg/m²) on day 1. With a median follow-up of 34 months, a 53% complete pathologic response was noted. Radical vulvectomy or exenteration was only required in five treatment failures. Overall, local control for the entire series was 95%. Only 3 patients required treatment breaks, all secondary to skin reaction, and there was no other significant observed toxicity.⁴⁰

Similar to the experience in anal canal carcinoma, increased toxicity has been observed with higher doses of chemotherapy. In particular, mitomycin C has a potent radiosensitizing effect both on tumor cells and normal tissue. Two prospective phase I/II studies from Milan employed concurrent 5-FU and mitomycin C to spare radical surgery in advanced primary and recurrent disease. Landoni and associates treated 41 primary tumors that were either unresectable or would have required radical vulvectomy or pelvic exenteration. Radiation was given with a planned 2-week break with 36 Gy delivered before the break and 18 Gy afterward, along with 5-FU (750 mg/m²) and mitomycin C (15 mg/m²) with each course. Median follow-up was 22 months. Ten patients had a complete pathologic response and had an 80% disease-free survival rate, whereas 8 patients who could not have surgery died of their disease. Of the 23 patients with a partial response, presence of a pathologically positive groin alone or with a primary was associated with decreasing long-term disease control. Patients with only a pathologically positive groin had the best outcome compared with those with positive primary sites. Although 89% of the patients in the Landoni and associates' study completed treatment and 72% were able to undergo surgery, moderate and severe complications were common and no significant improvement in survival was noted over management with surgery alone.³⁷ Lupi and coworkers prospectively treated 31 patients with primary and recurrent tumors requiring exenteration with a similar split-course regimen (36 Gy before a 2-week break, 18 Gy after) concurrent with 5-FU (750 mg/m²) and mitomycin C (15 mg/m²). Although the disease-free survival rate was 63% in the primary group, major postoperative complications were found in 65% of patients and there was a 14% treatment-related mortality.³⁹ The dosage of mitomycin C in most institutions in the United States is limited to 10 mg/m², which appears to be better tolerated. Higher doses of mitomycin C as used in the Milan studies require planned or unplanned breaks from radiation therapy, which allow for tumor repopulation and may produce an adverse impact on the efficacy of treatment.

Various chemotherapy combinations in addition to 5-FU, cisplatin, and mitomycin C have been explored in the preoperative setting based on in vitro results, although clinical results have, in some cases, been disappointing. Jaakkola and colleagues demonstrated an additive effect of paclitaxel and radiation therapy in four different cell lines of vulvar squamous carcinoma.⁴¹ Clinical results with paclitaxel in this setting awaits further investigation. Preoperative use of bleomycin (180 mg) both with and without radiation has been reported from Norway and Rome respectively. In the former, median survival was only 8 months in 20 patients who were treated for stages III and IV tumors and was only 6.4 months in 22 recurrent cases, notwithstanding the addition of 34 to 45 Gy to the tumor volume if it was thought to be inoperable after initially delivering 30 Gy.⁴² No significant benefit in mitigating radical surgeries was observed by Benedetti-Panici and coworkers with the preoperative use of bleomycin, cisplatin, and methotrexate without radiation.²

Massive tumors have also been successfully treated with concurrent radiation and chemotherapy. Twelve patients from M.D. Anderson Cancer Center with large vulvar cancers (mean size, 8.7 cm) were treated with concurrent cisplatin (4 mg/m² 96 hours), 5-FU (250 mg/m²/day) and 40 to 50 Gy to the groin, pelvis, and primary. Eight of these patients were able to have resection 6 weeks later and 4 of these 8 had a complete pathologic response, all of whom on follow-up had no evidence of disease. No treatment breaks were required and there was no significant hematologic toxicity suggesting that higher doses of chemotherapy could have been used.³⁶

The GOG has recently published results of the largest prospective study exploring the feasibility of organ preservation with concurrent chemotherapy and radiation. Seventy-three patients with T3 and T4 tumors received split-course radiation and concurrent cisplatin (50 mg/m²) on day 1 with 5-FU (1000 mg/m²) as a continuous infusion over the first 4 days. Radiation therapy was divided into 2380 cGy courses, separated by a 1- to 2-week planned break depending on the degree of perineal reaction. Treatments were given twice daily at 170 cGy per fraction. Inoperable groin nodes and the lower pelvic lymph nodes were included in the radiation portals. Surgical excision and bilateral inguinofemoral dissection was planned in all patients. No visible tumor was seen at surgery in 33 patients. Of the remaining 38 patients with gross residual disease, only 2 were unresectable and only 5 had positive margins. With a median follow-up of 50 months, 55% of the patients were alive and without evidence of disease. Toxicity was acceptable and was confined to expected perineal reactions and wound complications.⁴³

The most obvious benefit of organ preservation is improved functional outcome; however, the psychologic impact of cosmetic salvage cannot be overstated. Figure 1 shows a patient with a massive vulvar cancer. Often patients do not seek care until disease has reached an advanced stage because of embarrassment and fear. Figure 2 shows the same patient following combined chemotherapy and radiation treatment. Although preliminary data have been very encouraging, longer follow-up is needed. The clinical experience with organ preservation strategies during the past 10 years has shown efficacy and acceptable toxicity and should provide the impetus for future prospective trials.

In the elderly, aggressive surgery is often medically contraindicated. Patients who could not be offered any surgical treatment for vulvar cancer have been successfully treated with radiation therapy alone. Several sizable series from the United States and Europe have noted long-term survival rates of between 20% and 45% in advanced cases. Although surgery remains the treatment of choice in vulvar cancer, observed survival rates in the inoperable setting with acceptable morbidity strongly support attempting definitive radiation therapy alone if necessary.^44,45,46,47

Recurrent disease usually carries a grave prognosis. Treatment is often difficult because of previous surgery and consequent tissue effects. These cases can be salvaged occasionally with external beam radiation, sometimes in combination with radiation implants. Failure in the groin is almost never salvaged whereas perineal failures can be successfully managed depending on the extent of disease.⁴⁸ Prempree and Amornmarn identified five factors in 21 recurrent cases predicting the most favorable candidates for salvage radiation. The size of the primary recurrence (5 cm or smaller), size of the groin recurrence (2 2 cm or smaller), lack of perineal skin involvement, lack of tissue necrosis, and higher radiation doses were prognostic for a better chance of cure.⁴⁹ Recently confirming the poor prognosis associated with recurrence while highlighting the potential for salvage in selected cases, Hruby and partners reviewed 26 women with recurrent vulvar cancer. Combinations of surgery and radiation were attempted. In a few cases, chemotherapy was combined with radiation. The overall survival rate was 22% at 5 years. Although the survival rate was 46% if disease was localized, no patients survived if disease extended beyond the vulva.⁵⁰ Aggressive treatment of recurrent vulvar cancer should be reserved for patients with localized disease and with a reasonable performance status. Otherwise, the morbidity of perineal radiation and additional surgery is not justified in the face of almost uniformly poor outcomes.

Radiation kills both cancer and normal cells primarily by two mechanisms. First, the indirect effect involves interaction of ionizing rays with water molecules to produce oxygen free radicals that in turn disrupt cellular DNA. Cancer cells are far less capable of repairing radiation damage than normal cells. This difference is exploited by dividing the total prescribed radiation dose into several fractions usually as one treatment a day or conventional fractionation. The interval between treatments allows tumor cells to repopulate. One strategy to counteract repopulation is hyperfractionation which refers to two or three radiation treatments a day separated by 4 to 6 hours to allow for normal tissue recovery. The second mechanism of radiation damage to cells is the direct effect or the interaction of energy beams with the target molecules. The first goal of radiation treatment planning is delivering a sufficiently high dose to tumor volume to cure disease or effect durable local control potentially, while minimizing both acute toxicity to normal tissues and reducing the risk of long-term complications. Treatment planning is a major challenge because the dose required to eradicate disease commonly approaches the dose associated with major complications. Dose-response curves for normal tissue complications are steep after crossing the threshold at which meaningful tumor response can be achieved. Given that both normal tissue and cancer cells are in the path of the radiation beam, the treating physician must pay careful attention to details of patient setup and to plan verification.

In the past, external beam radiation treatments were delivered with Co-60 teletherapy machines. These devices had an active Co-60 source in the head of the machine that emitted gamma rays directed at the patient lying on the table below. Co-60 rays are not so penetrant in tissue as high energy megavolt photon and electron beams from linear accelerators that are widely available and are standard equipment in most cancer centers today. Co-60 teletherapy had been the standard means of delivering radiation in the United States until the early to middle 1980s when use of linear accelerators became widespread. High energy photon beams deposit ionizing radiation in tissue that attenuates the dose more gradually than with the less penetrant Co-60 rays. This beam characteristic allows target volumes that may be several centimeters below the skin surface to receive the full prescribed dose while keeping the skin dose relatively low. Co-60 treatments necessitated high skin doses to deliver dose to deeper targets adequately and was associated with far more morbid perineal reactions. Despite the advent of linear accelerators, the sensitivity of the perineum to radiation damage still frequently requires treatment breaks to allow completion of therapy. Nonetheless, modern equipment, computers, and the rapid development of increasingly sophisticated treatment planning software has enabled today's patients to receive radiation therapy more accurately, effectively, and with fewer side effects than ever before.

Before a single radiation treatment is given, patient setup and field borders are arranged on a simulator table. Although not a treatment machine, the simulator reproduces the distances from the radiation beam to the treatment table as well as the angle and rotation of the head of the machine that delivers the energy beam. Patients rest in the supine position or with the legs raised and knees bent (frog-leg position). These are the best positions for targeting the perineum, groin and pelvis. Plain film radiographs are obtained in the simulator after review under fluoroscopy to confirm patient and field border setup. These simulation films are then compared with weekly port films that document actual radiation treatments. Adjustments are made as necessary so that the planned treatment is accurate and reproducible.

Vulvar tumors require treatment of the primary site as well as draining lymph nodes. Areas at risk for harboring microscopic disease typically receive doses between 45 and 50 Gy, whereas gross disease usually requires doses above 60 Gy. Opposed anteroposterior (AP/PA) treatment fields encompass the pelvic lymph nodes, inguinofemoral lymph nodes and perineum. High-energy beams (10 mV or more) delivered from opposing sides of the target yield a homogeneous dose distribution across wide treatment depths. This characteristic is important because the target volume often lies across several centimeters of tissue, different targets lie at different depths, that is, the groin versus the pelvic lymph nodes, and because the body has a heterogeneous surface. Simulation films used for the treatment of a typical vulvar cancer patient are shown in Figures 3 and 4. The radiation beam can be pictured passing through the plane of the page. The grid marks the central rays of the photon beam and the straight lines at the corners represent the area in the beam's path, which is blocked by Cerrobend, a metal that is liquefied when heated, poured and then allowed to cool in the shape of the area of desired protection. Thus, 1-cm Cerrobend attenuates the radiation beam by 50%. Five to six centimeters of block are required to diminish the radiation dose to less than 5% at the surface of the skin. In the example shown, bowel and soft tissue that are not overlying or underlying targets of interest are blocked out. One structure of concern for radiation damage is the head of the femur. The posterior treatment field shown in Figure 4 does not pass through the femoral heads and therefore excludes the groin, which would be underdosed without a make-up dose from the anterior side. Because the inguinal lymph nodes lie close to the skin surface, treatment of this area with separate electron beams, which deposit more energy near the skin surface compared with photons, allows coverage of the groin while limiting the total dose to the femoral heads. After 45 to 50 Gy is delivered to areas at risk for microscopic disease, the treatment field is reduced to cover only that area of gross tumor that can be marked at the time of initial simulation with a radiopaque wire. Treatment can then proceed to the necessary dose.

A different technique to cover the groin without overdosing the femoral heads involves the use of only photon fields. The anterior and posterior fields are identical to those shown in Figures 3 and 4, but no separate electron fields are used. Instead, a central partial transmission block is placed in the path of the anterior beam such that the primary site and pelvic nodes receive more dose contribution from the posterior field whereas the extended portion of the anterior field that covers the groin is unattenuated. The “wing field” technique has been well described and analyzed by Kalend and colleagues who found that the divergence of the posterior beam creates a 30% hot spot in the groin, which can be greatly reduced by tapering the edges of the anterior transmission block increasing dose contribution to the wing portion of the field from the anterior beam.⁵¹

Appropriate selection of target volume, that is, tumor and nodal groups at risk for disease recurrence, has previously been discussed in detail. In general, preoperative radiation treatment portals encompass the groin, perineum, and low pelvic lymph nodes, whereas postoperative treatment portals are tailored based on surgical findings. In the postoperative setting, efforts to reduce acute perineal reactions that frequently mandate treatment breaks have included use of a central block to permit treatment of the groin and pelvis only. Long-term results with this technique have been reported at the University of Minnesota where 27 patients were found to have a central recurrence rate of 48% following radical vulvectomy. Although this figure is far higher than had been seen by others using this technique, the use of midline blocks is discouraged.⁵²

Because failure in the groin is disastrous, adequate dosage of this area is critical to successful radiation therapy. Groin nodes are not midplane structures and vary in depth among different patients. In Protocol 88, the GOG attempted to answer the question of whether radiation therapy to the clinically negative groin could replace groin dissection. This study was appropriately terminated when it was noted that patients in the radiation study arm experienced a comparatively high groin recurrence rate, but it has been roundly criticized because of poor study design with respect to defining the target volume and ensuring adequate dose delivery to the inguinofemoral nodes.^29,53 With a combination of electrons and photons, GOG Protocol 88 prescribed dosage to a depth of 3 cm from the anterior skin surface in all patients. This depth may cover superficial inguinal nodes adequately in most cases but may not cover deep inguinal nodes that are also at risk even with early lesions. Analysis of pretreatment CT scans of the 5 patients in the radiation study arm who suffered groin relapse has shown that all groins received less than the prescribed dose and 3 patients were underdosed by at least 30%.⁵⁴ Use of opposed photon fields alone to encompass the groin was reviewed by Petereit and colleagues. After radical vulvectomy, 48 patients fell into two groups, one receiving 50 Gy of inguinofemoral radiation and the other undergoing lymphadenectomy. Actuarial nodal control was almost equivalent with no difference in survival noted. Infection and wound separation was much higher in the groin dissection group whereas lymphedema was occurred in only 9%. No avascular necrosis of the head of the femur was reported.⁵⁵ The experience of GOG Protocol 88 illustrates the importance of careful and accurate treatment planning.

The value of planning treatment of the groin with CT scans was confirmed by McCall and coworkers at Loyola University Medical Center where 100 CT scans of women without groin adenopathy or prior surgery were analyzed. Using the guidelines of the GOG study, isodose curves were reconstructed for hypothetical groin irradiation. More than 50% of the women in the study would have received less than 60% of the prescribed dose across the groin and only 1 in 5 women had all groin nodes at a depth of 3 cm or less.⁵⁶ The likelihood of underdosage of the groin without CT planning was elaborated in a review of 31 patients with gynecologic cancers at risk for groin metastasis. Almost one third of superficial groin nodes are at a depth from the skin surface exceeding 3 cm and all deep inguinal nodes are outside this range.⁵⁷

Treatment planning CT scans as shown in Figure 5 can localize superficial and deep inguinal nodes that can be readily identified at the level of inguinofemoral vessels. At the time of simulation, fiducial markings correspond to the central rays of the radiation beam and are visible on the subsequent CT scan facilitating digital reconstruction of dose distribution in axial, coronal, and sagittal planes. Figure 6 shows a composite dose distribution in the axial plane of the mixed photon and electron fields of Figures 3 and 4. Each of the dose regions (or isodose curves) represented by the various lines is normalized to a reference at the central axis. The delivered dose is close to 100% of the prescribed dose across the entire region corresponding to the marked groin nodes on the planning CT from Figure 5. Optimization of dose delivery with treatment planning software and improved imaging should permit individualized radiation therapy. In-field failures potentially can be significantly reduced with careful attention to treatment technique and patients' individual anatomic considerations.

Brachytherapy

The practice of placing radiation sources within and adjacent to tumors is called brachytherapy. The earliest and still the largest published series describing the use of radiation implants in vulvar cancer patients dates back to 1949. Margaret Tod at the Holt Radium Institute in Manchester, England treated 116 patients with early and advanced lesions using radium needles inserted with freehand placement and kept in place for 7 days. Patients were selected on the basis of medical inoperability in some cases, but most frequently they had tumors near the urethra or vaginal orifice. The overall 5-year survival rate was 25% for the entire group but was 33% for patients with early lesions and 14% for those with advanced disease. Although not impressive compared with surgical series, these results established the feasibility of brachytherapy for patients in whom radical surgery was undesirable or not practicable.⁵⁸

Technologic advances in treatment planning have renewed confidence in the accuracy of not only external beam radiation but also radiation implants. In medical centers in the United States and Europe, computer assisted calculations of dose distribution over target volumes and novel source applicators were used in the treatment of recurrent and advanced vulvar cancers.^59,60,61 Figure 7 shows a commonly used applicator with holes at regular intervals to allow placement of Ir-192 ribbons. In this example, an inoperable tumor extending into the vaginal orifice receives an adequate dose by the contribution of numerous source positions. Although efficacy of these procedures has been clearly demonstrated, a high risk of severe tissue necrosis has also been observed. Most reports have included heterogeneous mixtures of patients treated sporadically with brachytherapy, but local control and survival rates appear to be slightly less than would be expected for comparable patients treated surgically, whereas the risk of local necrosis can be as high as 20%.^62,63

Even with improvements in the measurement of tumor and normal tissue dose-minimizing toxicity without compromising disease control, successful practice of brachytherapy demands a high level of clinical experience and confidence. Implantation of advanced tumors or attempted salvage of recurrent disease by this approach should be reserved for clinicians with expertise in this area. The role of brachytherapy in the management of advanced vulvar cases remains to be established.

Acute toxicity refers to side effects experienced during treatment or shortly after completion of radiation therapy, whereas late toxicity refers to chronic problems experienced after 6 months or longer. The specific incidence of toxicity in different reports has been previously noted. The timing and extent of side effects can be markedly influenced by patient related factors, that is, sensitive skin, poor hygiene, obesity, immunocompromised states, or autoimmune disorders. In addition, chemotherapy can hasten and exacerbate skin reactions. Finally, acute and late toxicity are directly related as a function of the total dose and volume treated. Figure 8 shows examples of desquamation during treatment. Dry desquamation is typically apparent 1.5 to 2.5 weeks into therapy, whereas wet (or moist) desquamation usually occurs in the third or fourth week. Virtually all patients experience some degree of moist desquamation not uncommonly requiring treatment breaks. Acute mucosal and skin reactions are managed with drying agents as required. Sitz baths and Domeboro soaks to the affected area usually mitigate the severity of symptoms, but meticulous skin care during and following the course of therapy is the best preventive measure.

Various potential toxicities been observed following radiation therapy to the groin and perineum. The commonest is vulvar dryness that results from scarring and fibrosis with or without antecedent surgery. Fibrosis of soft tissues in the groin can result in chronic leg edema. Figure 9 shows a woman with chronic lower limb edema associated with postirradiation cellulitis. Leg edema is the commonest serious consequence of radiation therapy to the groin and its incidence is usually reported between 10% and 20%. Complications requiring surgical intervention include hip fractures, bowel obstruction, and fistulas.^64,65,66,67 Fortunately, these are uncommon events. Careful attention to radiation technique and individualized treatment planning can reduce the likelihood of side effects, but effective treatment of life-threatening disease and successful organ preservation invariably entails risk of serious morbidity.