Conservative therapy for melanoma of the vulva

Conservative therapy for melanoma of the vulva Peter G. Rose, MD, M. Steven Piver, MD, Yoshiaki Tsukada, MD, and Taishing Lau, PhD Buffalo, New York Recent cooperative studies have demonstrated that less radical local resection of cutaneous melanomas is equally effective as a traditional radical approach. A retrospective review of vulvar melanoma was undertaken to determine if mode of therapy affected recurrence. Survival correlated independently with depth of invasion and age (p = 0.05 and p < 0.02, respectively). In the comparison of radical vulvectomy with local excision, no patient differences in age or histopathologic variables were determined (nodal disease status, histology, mitotic count, lymphocytic infiltration, or ulceration). Radical vulvectomy did not improve survival over local therapy (p > 0.2). Six of eight patients whose melanoma had <2 mm of invasion treated with local therapy are disease free after a median of 127 months (range 6 to 300 months). For local excision, recurrences were more frequent when margins were <2 cm, but this was not statistically significant in this small sample. Although the current series is small and retrospective, its findings suggest that treatment recommendations of large cutaneous nonvulvar melanoma studies are applicable to vulvar melanoma. A prospective randomized study of radical versus conservative surgery for vulvar melanoma will be necessary to confirm these treatment recommendations. (AM J OssrET GYNECOL 1988;159:52-5.)

Key words: Melanoma of the vulva, radical versus local resection, recurrence

Recent reports have demonstrated that less radical surgical resection of melanoma is equally as effective as a traditional radical approach.' This seems to be true for both early lesions without nodal involvement and advanced lesions for which metastatic nodal or distant disease precludes benefit from extensive local resection. Progress in understanding the biology of early melanoma, its histology, and its invasive and metastatic potential have allowed pathologic variables to be identified' and therapies to be more rationally based. The most significant of these pathologic variables is the depth of tumor invasion, which is highly correlated with lymph node metastasis." The presence of regional nodal metastasis dramatically decreases 5-year survival from 76% to 31 o/c:.' The depth of tumor invasion correlates highly with nodal and local recurrence.'· 5 A subgroup of patients with <0. 76 mm of invasion were reported by Breslow and Macht 6 to have no nodal metastasis of recurrence after 5 years of follow-up. Despite this progress in disease concept, radical vulvectomy and bilateral lymphadenectomy remain standard therapy for vulvar melanoma. 1 · 10 A major deterrent to change has been the rarity of this lesion and resultant small surgical experience of any single institution. At Roswell Park Memorial Institute patients From the Departmenl> of Gynecology and Pathology, Roswell Park Memorial Institute, and the Department of Statistics, State University of New York at Buffalo. Received for publication August 17, 1987; revised December 7, 1987; accepted January 8, 1988. Reprint request>: M. Steven Piver, MD, Department of Gynecologic Oncology, Roswell Park Memorial Institute, 666 Elm St., Buffalo, NY 14263.

52

with favorable prognostic variables have been treated by more conservative surgery. The current report details their pathologic findings and current disease status.

Material and methods The medical records and pathologic status of patients with primary vulvar melanoma treated at Roswell Park Memorial Institute from 1927 to 1986 were reviewed. Pathologic slides were reviewed and evaluated for melanoma histology, depth of invasion according to Breslow," mitotic count per IO high-power fields, vascular invasion, lymphocytic infiltration, and the presence of ulceration. Patients were classified as having received local therapy or radical therapy. Based on the findings of the World Health Organization study, 12 local therapy was further defined as adequate if there was a 2 cm margin of normal tissue from the primary lesion as described in the operative report. Reports of excisions with a margin <2 cm or when the margin was not described were termed inadequate. Radical therapy consisted of classic Basset's vulvectomy. Recurrences were classified as local, regional nodal, distant, or none. Survival based on age, depth of invasion, and treatment was analyzed by the BMDP 2L Cox regression model. Categoric data were analyzed by Fisher's exact test.

Results Thirty-six patients with primary vulvar melanoma were studied. Ten patients were excluded, eight of

Conservative therapy for melanoma of vulva

Volume 159 Number I

53

Table I. Patient characteristics

Patient No.

Age (yr)

Local therapy l 21 2 23 3 31 4 5 6 7 8

44 58 64 69 71

9 76 Local therapy and IO 20 11 43 12 51 Radical therapy 13 42

14 15

62 73

16 77 Radical and nodal 17 31 18 35 19 38 20 45 21 53 22 58 23 24 25 26

65 68 71 78

Depth of mvaswn (mm)

Vascular lymphatic space

Mitotic count JO HPF

Lymplwcytic infiltration

Inguinal nodes

Histology

ND ND ND

SSM SSM NA

0.55 None

Absent Absent


++ +

Absent Absent

ND ND ND ND ND

SSM SSM SSM SSM NA

0.40 I.IO l.80 0.73

Absent Absent Absent Absent

1-2 8
++ ++ ++ ++

Absent Absent Present Absent

Absent

5

4.00 1.60 l.90

Absent Absent Absent


0.30

Absent

ND NM nodal dissection Negative NM Negative SSM Negative SSM

>IO

Ulceration

Recu1Tence

Absent

None 74 mo None 299 mo Regional 5 mo None 37 mo None 300 mo None 181 mo Local 20 mo Regional 116 mo Local 22 mo

++ + +

Present Present Absent

None 62 mo None 6 mo Local 53 mo


++

Absent

Died of extragenital melanoma 19 mo Local 20 mo Local 67 mo Distant 72 mo Local 34 mo

ND

LM

ND ND

NA NA

ND dissection Negative Negative Negative Negative Negative Negative

NM

>6.00

Absent

IO

+

Present

SSM SSM NM SSM SSM LM

1.03 0.40 >IO 2.70 None 1.50

Absent Absent Absent Absent Absent Absent


++ ++ + ++ ++

Absent Absent Present Present Absent Absent

Negative Positive Positive Negative

SSM SSM NM NM

0.86 >10 6.00 3.80

Absent Absent Absent Absent


+ + + +

Absent Present Present Present

None 54 mo None 69 mo Distant 52 mo Distant 24 mo None 187 mo Distant and regional 32 mo None 136 mo Distant 3 mo Distant 10 mo Local 8 mo

HPF, High-power fields; ND, not determined; SSM, superficial spreading melanoma; NA, slides not available for study; NM, nodular melanoma; LM, lentigo melanoma.

whom were treated with radiation therapy in combination with local therapy, one who had stage IV disease with lung metastasis at presentation, and one was not treated because of advanced age (94 years old). Twelve patients received local therapy consisting of local excision only (nine patients) and local excision and inguinal lymphadenectomy (three patients). Fourteen patients received radical therapy consisting of radical vulvectomy alone (four patients) and radical vulvectomy and inguinal lymphadenectomy (IO patients). The patient's age, nodal status when disease was determined, histologic findings, and months to recurrence are listed in Table I. The slides of four patients, of whom two received local therapy and two received radical therapy, were not available for our review. These pathology slides had previously been reviewed

and diagnoses confirmed at Roswell Park Memorial Institute. For all patients in this study survival correlated with depth of invasion (p = 0.05); Table II). Younger patient age resulted in improved patient survival (p < 0.02) but was not dependent on depth of invasion (p = 0.3). Survival was not worsened by mode of therapy when local and radical treatment groups were compared (p > 0.2). In comparing patients receiving radical therapy with those receiving local therapy, no difference in age or histopathologic variables were determined: age (p = 0.24), nodal status (p = 0.28), histology (p = 0.31), depth of invasion (p = 0.62), vascular and lymphatic space permeation (p = 1.0), mitotic count (p = 0.14), lymphocytic infiltration (p = 0.20), or ulceration (p = 0.30).

54

July 1988 Am J Obstet Gynecol

Rose et al.

Table II. Correlations between survival and depth of invasion Depth of Patient No.


7 23 1-2 mm 17 5 22 II 6 12 >2 mm 20 26 JO 25 16 24 9 19

Recurrence

invasion (mm)

None None 0.30 0.40 0.40 0.55 0.73 0.86 1.03 J.10 1.50 1.60 1.80 1.90 2.7 3.8 4.0 6.0 >6.0 >JO >JO >10

% of patients with recurrence

None 299 mo None 187 mo None None None Local None

69 mo 37 mo 74 mo 20 mo 136 mo

None 54 mo None 300 mo Distant and regional 32 mo None 6 mo None 181 mo Local 533 mo Distant 24 mo Local 8 mo None 6 mo Distant JO mo Local 34 mo Distant 3 mo Local 22 mo Distant 52 mo

14.3%

33.3%

87.5%

*Eliminated from patient recurrence calculation because of death due to second primary melanoma.

Six of 12 patients underwent inadequate local excision because the margin described was <2 cm (n = 4) or the margin size was not specified (n = 2). Of four recurrences, three occurred in patients with inadequate excision (50%) in contrast to one recurrence among six patients with adequate excision (16%). This difference was not statistically significant (p = 0.25).

Comment The rarity of vulvar melanoma has hindered its study. Earlier reports 7 · •."established that survival correlated closely with nodal disease status. Chung et al. 9 were the first to apply the principle of tumor penetration depth to vulvar melanomas. Because papillary dermis is not well defined in vulvar skin, levels in that report were similar to those previously described by Breslow." Although the 5-year survival of 33 patients reported by Chung et al. was only 30.3%, survival was 100% in eight patients with level II invasion (< 1 mm). Three of these eight patients were treated with wide local excision only. Podratz et al. ' 0 also noted statistically improved survival with less tumor invasion. Survivorship correlated with histologic findings; patients with superficial spreading and nodular melanomas had 5-year survival rates of 71 % and 38%, respectively. However, nodular melanomas penetrated deeper in the dermis (median 7 mm) than superficial spreading melanomas (median 2 mm). Only 2% of the 48 patients who were treated

underwent wide local excision, and survival as a function of therapy mode was not reported. Jaramillo et al. 11 reported 15 patients treated with radical vulvectomy and one patient treated with local excision. Survival was 100% when the depth of penetration was < 1.49 mm but decreased to 29% with invasion > 1.50 mm. Among the four patients with nodal metastasis, depth of penetration of the primary lesion ranged from 3.0 to 9.5 mm. Beller et al.," reporting 14 patients treated with radical surgery, again demonstrated the improved survival with < 1.5 mm of invasion. Phillips et al. 16 reported two patients and reviewed IO other patients previously reported by Chung et al. 9 and Cleophax et al. 17 with <0. 76 mm of invasion, all of whom were without evidence of disease after either local or radical therapy. Recently Davidson et al.'" reported 32 patients with vulvar melanoma treated with local excision (n = 14), simple vulvectomy (n = 7), or radical resection (n = 11). No mode of therapy demonstrated statistically superior survival. However, only one patient had a lesion with <0.75 mm of invasion. Two patients had distant metastasis and five patients clinically had regional nodal metastasis at presentation. The overall survival was 25% at 5 years and 3% at IO years. The issue of surgical margins has been extensively addressed in the surgical literature. Historically, margins of 5 to 15 cm have been advocated for surgical therapy. Several studies have opposed this dictum,

Volume 159 Number I

demonstrating that smaller margins have not been associated with worsened survival. 1. 11 • 19 Aitkin et al.' correlated margin size with both depth of invasion and diameter of lesion. For lesions with <2.0 mm of invasion and <10 mm in width, margins of 2 cm were adequate. For deeper or larger lesions, survival improved with a 3 cm margin. However, margins >3 cm did not improve survival. The World Health Organization collaborative study" reported on 281 patients with lesions <2 mm of invasion. That local recurrence rate was 1.8% and was not effected by margin sizes of I to 5 cm. The current series presents eight patients who received conservative therapy via local excision (with 2 cm margins) for early lesions (<2 mm) and have a high disease-free survival rate (75%). Survival for vulvar melanoma in reported series varies from 30% to 54%. This poor survival seems to be related to deep invasion and regional nodal metastasis at initial diagnosis. Furthermore, the radical vulvectomy may fail even for stage I disease because of inadequate medial margins. Podratz et al. ' 0 demonstrated differential IO-year survival rates for lateral vulvar lesions (61 %) when compared with medial lesions (37%). As the authors emphasize, radical vulvectomy is a treatment applied to a disease rather than a disease determining the treatment. In this series the important relationship between depth of invasion and survival is again demonstrated. Age was statistically correlated with increased survival but not with depth of invasion. Larger studies have demonstrated younger age is associated with more frequent superficial spreading melanoma and overall less melanoma invasion. 20 Melanoma continues to increase on a worldwide basis, doubling every 10 to 17 years. Whether the incidence of vulvar melanoma conforms to this trend is not known. Although it occupies only 2% of the body surface area, some authors have stated the vulva is predisposed to melanoma because it accounts for 3% to 5% of all melanomas!' Public self-examination education programs such as those sponsored by the American Cancer Society, along with greater physician awareness, may account for the increased detection of cutaneous melanoma in early stage disease. Histopathologic variables are critical to appropriate surgical therapy. Based on multiple retrospective and prospective studies of nonvulvar melanoma and the accumulating, but limited, data of vulvar melanomas, the recommendations advanced for treatment of melanomas at other sites appear applicable to treatment of

Conservative therapy for melanoma of vulva

55

melanomas of the vulva. A prospective randomized study of radical versus conservative surgery for vulvar melanoma will be necessary to confirm these treatment recommendations. REFERENCES I. Aitkin DR, Clausen K, Klein JP, et al. The extent of primary melanoma excision-A re-evaluation-How wide is wide? Ann Surg 1983;198:634-41. 2. Balch CM, Murad TM, Soong SJ, et al. A multifactorial analysis of melanoma: Prognostic histopathological features comparing Clark's and Breslow's staging methods. Ann Surg 1978;188:732-42. 3. Lee YN. Diagnosis, treatment and prognosis of early melanoma: The importance of depth of microinvasion. Ann Surg 1980;191:87-97. 4. Mastrangelo MJ, Baker AR, Katz HR. Cutaneous melanoma. In: De Vita VT Jr, Hellman S, Rosenberg SA, eds. Cancer: Principles and practices of oncology. 2nd ed. Philadelphia: JB Lippincott, 1985: 1382. 5. Milton GW, Shaw HM, Farago GA, et al. Tumour thickness and the site and time of first recurrence of cutaneous malignant melanoma (stage I). Br J Surg 1980;67 :543-6. 6. Breslow A, Macht SD. Optimal size of resection margin for thin cutaneous melanoma. Surg Gynecol Obstet 1977; 145:691-2. 7. Yackel DB, Symmonds RE, Kempers RD. Melanoma of the vulva. Obstet Gynecol I 970;35:625-31. 8. Morrow CP, Rutledge FN. Melanoma of the vulva. Obstet Gynecol I 972;39:745-52. 9. Chung AF, Woodruff JM, Lewis JL. Malignant melanoma of the vulva: A report of 44 cases. Obstet Gynecol I 975;45:638-46. IO. Podratz KC, Gaffey TA, Simmonds RE, et al. Melanoma of the vulva: An update. Gynecol Oncol 1983; 16: 153-68. I I. Breslow A. Thickness, cross-sectional areas and depth of invasion in the prognosis of cutaneous melanoma. Ann Surg 1970;172:902-8. 12. Cascinelli N, van der Esch EP, Breslow A, et al. Stage I melanoma of the skin: The problem of resection margins. Eur J Cancer 1980;16:1079-85. 13. Karlen JR, Piver MS, Barlow JJ. Melanoma of the vulva. Obstet Gynecol 1975;45: 181-5. 14. Jaramillo BA, Ganjei P, Averette HE, et al. Malignant melanoma of the vulva. Obst et Gynecol I 985;66:398-40 I. 15. Beller U, Demopoulos RI, Beckman EM. Vulvovaginal melanoma: A clinicopathologic study. J Reprod Med 1986;3 I :315-9. 16. Phillips GL, Twiggs LB, Okagaki T. Vulvar melanoma: A microstaging study. Gynecol Oncol 1982;14:80-8. 17. Cleophax JP, Pilleron JP, Durand JC, et al. Le melanome de la vulve. Gynecologie I 977;27:333-9. 18. Davidson T, Kissin M, Westbury G. Vulvo-vaginal melanoma-should radical surgery be abandoned? Br J Obstet Gynecol 1987;94:473-6. 19. Day CL, Mihm MC, Sober AJ, et al. Narrower margins for clinical stage I malignant melanoma. N Engl J Med 1982;306:4 79-82. 20. Shaw M, McGovern VJ, Milton GW, et al. Malignant melanoma: Influence of site of lesion and age of patient in the female superiority in survival. Cancer 1980;46:27315. 21. Morrow CP, DiSaia PJ. Malignant melanoma of the female genitalia: A clinical analysis. Obstet Gynecol Surv 1976; 31 :233-71.