| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
EDITORIALS |
Department of Obstetrics & Gynecology, Division of Gynecologic Oncology, University of Texas Southwestern Medical Center, Dallas, Texas.
Correspondence: Address correspondence to: Robert L. Coleman, MD, Department of Obstetrics & Gynecology, Division of Gynecologic Oncology, University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd., J7.124, Dallas, TX 75390–9032; Fax: 214-648-8404; E-mail: Robert.Coleman{at}UTSouthwestern.edu
Understanding and planning efficacious management for a rare malignancy in which little specific pathobiological and clinical data are available is challenging and frequently calls for one’s best appraisal of the reported "trial and errors" of treatment frequently modeled on the disease in a common locale or histology. In some cases, such as epithelial squamous cell carcinoma of the ovary, the treatment strategy is largely commutable (i.e., maximal surgical effort and adjuvant chemotherapy); others, such as small carcinoma of the cervix or uterine papillary serous carcinoma, require more deductive imagination as the "standard" management program may not address the specific natural history characteristics of the histology in that locale (e.g., a greater propensity for intraabdominal or distant metastases).1–3 Unfortunately, meaningful randomized clinical trials of these entities are unlikely to be the methodology for outlining future treatment. Yet, for the patient, a treatment decision must be made—a decision that usually falls in the purview of empiricism. While tumor molecular profiling may help in modulating one’s expectations in this regard and in some cases (e.g., gastrointestinal stromal tumors) may identify novel therapeutic targets, the additional information frequently only highlights the complexity of the malignant process.4 When enough patients, treated and untreated, are reviewed, specific attention can be paid to important characteristics of metastatic spread, patterns of recurrence, responses to specific therapeutic modalities, and natural history. While far from perfect, repeated assessment becomes the process of developing treatment precision. Such is the case for vaginal melanoma therapy.
In the current issue of the Annals of Surgical Oncology, Miner and colleagues at the Memorial Sloan Kettering Hospital5 synthesize data from their largest-to-date cohort of vaginal melanoma patients (n = 35) in an effort to illicit prudent therapy recommendations. Patients were accrued over a 24-year period, underwent secondary pathological review, and were evaluated by primary treatment modality (surgery vs. radiation) for local control and survival. The authors confirm previously identified common clinical-pathological features of this population, namely, late presentation, deep invasion, and frequent ulceration, all which typically accompany the most frequent presenting clinical symptom, vaginal bleeding. Pathologically, ulceration, lack of a clear intraepithelial component, and absence of pigment can make pathological confirmation difficult; however, increasingly, use of specific immunostains (not outlined in the current series) such as S-100, HMB-45, MART-1, tyrosinase and Mitf aids in diagnostic accuracy.6 The authors also confirm this population’s poor overall survival, lack of prognostic association to clinical stage, and propensity for distant recurrence, even in those with long-term local control. So the challenge is, do we approach this disease like vaginal cancer or like mucosal melanoma?
The authors make a strong case for the latter, and opinion in the literature is becoming more uniform in this regard. The a priori approach taken by this group and others has been modeled after the approach for malignant melanoma in other sites, namely, wide surgical excision. Problematic in the vagina, as well as other mucosal locales, is the ability to gainfully achieve disease-free margins of resection. In the current series, few cases were margin-negative (n = 4), a factor limiting the evaluation of this feature on survival. Although the authors report that benign melanosis and melanoma in situ were identified at the margin in 83% of surgical specimens, there is little functional information about how these conditions impacted survival, if at all. Indeed, adjuvant radiotherapy was given to 10 of 17 patients with positive margins, and local failure was seen in just 6 of 18 surgical patients identified with a recurrence. It is not known whether adjuvant therapy reduced the risk of local recurrence (especially for melanoma in situ) or whether these conditions are low risk relative to the invasive primary. Nonetheless, gross tumor excision was achievable in 22 of 24 patients undergoing surgery.
Another challenge presented by the vaginal locale is the proximity of the bladder and rectum. Surgical procedures required to obtain wide and deep margins generally involve radical excision or exenterative operations. Some have advocated this approach, although most—current series included—have noted high rates of distant "first site of recurrence" even among these patients, tempering enthusiasm for en mass resection. The authors bolster their claim by demonstrating equivalent relapse-free survival curves for patients undergoing wide excision and radical excision. One could argue that the term "radical excision," applied here to the therapy of those patients undergoing hysterectomy as part of their wide excision (n = 10) as well as those undergoing exenteration (n = 2), is generously applied and may underestimate the impact of negative histological margins. It is interesting that one of two patients undergoing exenteration survived for more than 130 months yet died of distant disease following recurrence at 10 years. Still, long-term survivorship is a rarity and has been observed among patients undergoing less radical excisional procedures. In their literature review and analysis of 17 patients surviving more than 5 years, Buchanan et al. noted that just 6 patients (35%) had undergone "radical" resection (defined as total colpectomy or exenteration) and that this frequency was no different from the 13 of 39 patients surviving less than 5 years.7 These findings suggest that local disease failure even among those with histologically positive margins is an important but lesser problem and would suggest that ultimately, successful systemic therapy will need to be developed to readdress this issue.8
Metastatic disease to the regional lymph nodes is a poor prognostic factor and often heralds risk for distant disease. However, routine pretreatment lymphatic dissection is impractical, given the rich anastomotic nature of the vaginal lymphatics and the imperfect prediction of a negative limited sampling. Sentinel node identification has been advanced as one strategy to address ambiguous lymphatic drainage and has been met with success in cutaneous melanoma as well as epithelial and melanotic lesions of the lower genital tract, including the vagina. Select cases of vaginal melanoma may be appropriate for sentinel node evaluation but would remain largely prognostic, given the lack of effective systemic therapy and short survival of these patients.
Radiotherapy represents a hallmark of therapy for most cases of advanced vaginal epithelial cancers and is administered as a combination of external teletherapy and vaginal brachytherapy. Five-year stage-dependent survival rates approach 19% to 68% in this otherwise elderly population.9 Few authors have reported success with primary radiotherapy in vaginal melanoma.10 Such treatment is generally reserved as adjuvant therapy or palliation. Although radiotherapy in the current series was found to be inferior to surgical therapy (see figure 1 in the article by Minor et al.5), small sample size, unknown treatment intent (palliation vs. cure), indeterminate dose and type of radiation, and selection bias cloud effective comparison and render this comparison implausible. High dose fractions achievable with vaginal brachytherapy, combined with surgical excision, are likely able to achieve local control as effectively as radical excision, although data to this effect are limited.
The authors are to be commended for their attempt to wrestle these issues in a cohort where seemingly no two patients are enough alike to a make clear conclusion. Treatment strategies for vaginal melanoma most appropriately follow those for melanoma in other sites, and conservative gross tumor excision appears to provide as much benefit as radical therapy. Surgical issues aside, however, the well-illustrated problem with distant and remote disease failure represents the real target with this and other noncutaneous lesions.
Received for publication November 5, 2003. Accepted for publication November 12, 2003.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
