Prognosis
of Thick Cutaneous Melanoma of the Trunk and Extremity
Daniel Coit,
MD; Paul Sauven, MD, FRCS; Murray Brennan, MD
\s=b\ The records of 129 patients with thick cutaneous melanoma of the trunk or extremity treated at Memorial Sloan-Kettering Cancer Center, New York, NY, between 1974 and 1984 were reviewed with the aim of defining prognostic variables. All primary lesions invaded subcutaneous fat, were Clark level V, or of a Breslow thickness of 4.0 mm or greater. Treatment in all cases was by wide excision with or without split-thickness skin graft; all patients underwent regional lymph node dissection. Overall survival rate for the group was 47% at 5 years and 36% at 10 years. Factors independently predictive of survival were pathologic negative nodes (71% at 5 years compared with 28% for pathologic positive nodes) and extremity site (58% at 5 years compared with 33% for truncal site). Patients with node-negative thick cutaneous melanoma of the extremity had a 5-year survival rate of 82%. Patients with node-positive truncal thick cutaneous melanoma had a 5-year survival rate of only 8%. There was no difference between the 5-year survival rate of patients with nodenegative truncal thick cutaneous melanoma, 52%, and patients with node-positive thick cutaneous melanoma of the extremity, 42%. Nearly half of the patients with thick cutaneous melanoma of the extremity and trunk present with locoregional disease, at a stage when an aggressive surgical approach is warranted. Prognostic variables of pathologic nodal status and site identify patients at risk for early systemic failure.
(Arch Surg. 1990;125:322-326)
frequently It (TCM) métastases,
stated that patients with thick cutaneous is melanoma have not only a high risk of regional nodal but also a high risk of occult systemic disease at the time of presentation.1'3 The concept that these patients have such a poor overall prognosis has led many surgeons to consider treatment of these patients to be primarily
palliative. The objective of this study was to reevaluate the outcome of patients with TCM of the extremity or trunk undergoing wide excision with or without split-thickness skin graft (WE ± STSG) and regional lymph node dissection (RLND), to define prognostic variables and patterns of failure, and to identify subsets of patients with low, intermediate, or high risk of recurrence who would be suitable candidates for accru¬ al to postoperative adjuvant therapy trials. PATIENTS AND METHODS Methods Between 1974 and 1984, patients with malignant melanoma underwent WE ± STSG and axillary or inguinal lymphadenectomy by the staff of the Department of Surgery at Memorial Sloan-Kettering Cancer Center, New York, NY. Of this group, 129 patients (13%) had TCM of the trunk or extremity, defined as melanoma invading subcu¬ taneous fat (Clark level V) or with a Breslow thickness of 4 mm or 1019
Accepted for publication December 30,1989. From the Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, NY. Read before the annual meeting of the Society of Surgical Oncology, San Francisco, Calif, May 23,1989. Reprint requests to Department of Surgery, Memorial Sloan-Kettering Cancer Center, 1275 York Ave, New York, NY 10021 (Dr Coit).
greater. These patients form the basis of this review.
During the same period, we identified 19 patients with TCM of the trunk or extremity who did not undergo regional lymphadenectomy. While the reasons for nonoperative treatment of the nodes was not always apparent from a review of the medical record, these patients were excluded from the current analysis as they did not undergo pathologic staging with respect to the nodes. Furthermore, these patients were too few and too selected to compare as a group with the patients presented in our study. Prognostic features studied included patient age and sex, primary tumor site, level, depth, ulcération, clinical and pathologic nodal status, macroscopic nodal appearance, number of positive nodes, and status of the highest node dissected. Staging was performed according to the 1984 recommendations of the American Joint Committee on Cancer (AJCC). The primary melanoma was classified as follows: TX, not assessable; TO, Clark level I or in situ; Tl, Clark level II («0.75 mm); T2, Clark level III (0.75 to 1.50 mm); T3, Clark level IV (1.51 to 4.0 mm); and T4, Clark level V (>4.0 mm) or satellites within 2 cm of the primary lesion. The nodes were classified as follows: NX, not assessable; NO, none; Nl, one regional node station involved, nodes mobile and less than 5 cm in diameter, or less than five intransit métastases and negative nodes; and N2, more than one regional nodal station involved, or nodes fixed or greater than 5 cm in diameter, or more than five intransit métasta¬ ses, or any intransit métastases and positive nodes. Métastases were classified as follows: MX, not assessable; MO, none; Ml, skin and/or subcutaneous métastases; and M2, visceral métastases. The stages were grouped as follows: IA, patients showing characteristics Tl, NO, and MO; IB, patients with T2, NO, and MO; IIA, patients with T3, NO, and MO; IIB, patients with T4, NO, and MO; III, patients with any T, Nl, and MO; and IV, patients with any T, N2, and MO, or any T, any N, Ml or M2. According to this system all node-negative patients with no intransit metastasis in our study were classified as having stage II disease. Patients with one nodal station involved, or patients with less than five intransit lesions and no nodal involvement, were classified as having stage III disease. Patients with more than one nodal station involved (eg, bilateral axillae, bilateral groins, either axilla and either groin, or both superficial and deep groin on the same side were all equivalent to N2), nodal and intransit métastases (also N2), or any systemic métastases (Ml or M2) were classified as having stage IV disease. The patterns of first and all recurrences after lymphadenectomy were recorded. Recurrence was defined as local if occurring within 5 cm of the primary; nodal, if occurring within the primarily dissected lymph node basin; and regional, if occurring as intransit disease between the primary site and the dissected lymph node basin. Sys¬ temic recurrence was defined as extraregional métastases, including remote nodal, soft-tissue, and visceral sites. Follow-up was determined from time of diagnosis to time of death or last follow-up. Mean follow-up was 54 months for the entire group, 30 months for patients who died of the disease, and 85 months for patients free of disease at last follow-up. Survival analysis was by the method of Kaplan and Meier,4 with comparison of survival distributions by the log rank method of Peto et al." Multivariate analysis of variables found significant by univariate analysis was performed using the Cox model of proportional hazards.6 Differences were considered significant at the P
of Thick Cutaneous Melanoma of the Trunk and Extremity
Daniel Coit,
MD; Paul Sauven, MD, FRCS; Murray Brennan, MD
\s=b\ The records of 129 patients with thick cutaneous melanoma of the trunk or extremity treated at Memorial Sloan-Kettering Cancer Center, New York, NY, between 1974 and 1984 were reviewed with the aim of defining prognostic variables. All primary lesions invaded subcutaneous fat, were Clark level V, or of a Breslow thickness of 4.0 mm or greater. Treatment in all cases was by wide excision with or without split-thickness skin graft; all patients underwent regional lymph node dissection. Overall survival rate for the group was 47% at 5 years and 36% at 10 years. Factors independently predictive of survival were pathologic negative nodes (71% at 5 years compared with 28% for pathologic positive nodes) and extremity site (58% at 5 years compared with 33% for truncal site). Patients with node-negative thick cutaneous melanoma of the extremity had a 5-year survival rate of 82%. Patients with node-positive truncal thick cutaneous melanoma had a 5-year survival rate of only 8%. There was no difference between the 5-year survival rate of patients with nodenegative truncal thick cutaneous melanoma, 52%, and patients with node-positive thick cutaneous melanoma of the extremity, 42%. Nearly half of the patients with thick cutaneous melanoma of the extremity and trunk present with locoregional disease, at a stage when an aggressive surgical approach is warranted. Prognostic variables of pathologic nodal status and site identify patients at risk for early systemic failure.
(Arch Surg. 1990;125:322-326)
frequently It (TCM) métastases,
stated that patients with thick cutaneous is melanoma have not only a high risk of regional nodal but also a high risk of occult systemic disease at the time of presentation.1'3 The concept that these patients have such a poor overall prognosis has led many surgeons to consider treatment of these patients to be primarily
palliative. The objective of this study was to reevaluate the outcome of patients with TCM of the extremity or trunk undergoing wide excision with or without split-thickness skin graft (WE ± STSG) and regional lymph node dissection (RLND), to define prognostic variables and patterns of failure, and to identify subsets of patients with low, intermediate, or high risk of recurrence who would be suitable candidates for accru¬ al to postoperative adjuvant therapy trials. PATIENTS AND METHODS Methods Between 1974 and 1984, patients with malignant melanoma underwent WE ± STSG and axillary or inguinal lymphadenectomy by the staff of the Department of Surgery at Memorial Sloan-Kettering Cancer Center, New York, NY. Of this group, 129 patients (13%) had TCM of the trunk or extremity, defined as melanoma invading subcu¬ taneous fat (Clark level V) or with a Breslow thickness of 4 mm or 1019
Accepted for publication December 30,1989. From the Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, NY. Read before the annual meeting of the Society of Surgical Oncology, San Francisco, Calif, May 23,1989. Reprint requests to Department of Surgery, Memorial Sloan-Kettering Cancer Center, 1275 York Ave, New York, NY 10021 (Dr Coit).
greater. These patients form the basis of this review.
During the same period, we identified 19 patients with TCM of the trunk or extremity who did not undergo regional lymphadenectomy. While the reasons for nonoperative treatment of the nodes was not always apparent from a review of the medical record, these patients were excluded from the current analysis as they did not undergo pathologic staging with respect to the nodes. Furthermore, these patients were too few and too selected to compare as a group with the patients presented in our study. Prognostic features studied included patient age and sex, primary tumor site, level, depth, ulcération, clinical and pathologic nodal status, macroscopic nodal appearance, number of positive nodes, and status of the highest node dissected. Staging was performed according to the 1984 recommendations of the American Joint Committee on Cancer (AJCC). The primary melanoma was classified as follows: TX, not assessable; TO, Clark level I or in situ; Tl, Clark level II («0.75 mm); T2, Clark level III (0.75 to 1.50 mm); T3, Clark level IV (1.51 to 4.0 mm); and T4, Clark level V (>4.0 mm) or satellites within 2 cm of the primary lesion. The nodes were classified as follows: NX, not assessable; NO, none; Nl, one regional node station involved, nodes mobile and less than 5 cm in diameter, or less than five intransit métastases and negative nodes; and N2, more than one regional nodal station involved, or nodes fixed or greater than 5 cm in diameter, or more than five intransit métasta¬ ses, or any intransit métastases and positive nodes. Métastases were classified as follows: MX, not assessable; MO, none; Ml, skin and/or subcutaneous métastases; and M2, visceral métastases. The stages were grouped as follows: IA, patients showing characteristics Tl, NO, and MO; IB, patients with T2, NO, and MO; IIA, patients with T3, NO, and MO; IIB, patients with T4, NO, and MO; III, patients with any T, Nl, and MO; and IV, patients with any T, N2, and MO, or any T, any N, Ml or M2. According to this system all node-negative patients with no intransit metastasis in our study were classified as having stage II disease. Patients with one nodal station involved, or patients with less than five intransit lesions and no nodal involvement, were classified as having stage III disease. Patients with more than one nodal station involved (eg, bilateral axillae, bilateral groins, either axilla and either groin, or both superficial and deep groin on the same side were all equivalent to N2), nodal and intransit métastases (also N2), or any systemic métastases (Ml or M2) were classified as having stage IV disease. The patterns of first and all recurrences after lymphadenectomy were recorded. Recurrence was defined as local if occurring within 5 cm of the primary; nodal, if occurring within the primarily dissected lymph node basin; and regional, if occurring as intransit disease between the primary site and the dissected lymph node basin. Sys¬ temic recurrence was defined as extraregional métastases, including remote nodal, soft-tissue, and visceral sites. Follow-up was determined from time of diagnosis to time of death or last follow-up. Mean follow-up was 54 months for the entire group, 30 months for patients who died of the disease, and 85 months for patients free of disease at last follow-up. Survival analysis was by the method of Kaplan and Meier,4 with comparison of survival distributions by the log rank method of Peto et al." Multivariate analysis of variables found significant by univariate analysis was performed using the Cox model of proportional hazards.6 Differences were considered significant at the P
