1. Home>
2. Specialties>
3. Dermatology>
4. Summary and Comment

Excisional Margins in High-Risk Melanoma: The MSG Trial

Issues cloud the clinical relevance of these findings.

The much-anticipated U.K. Melanoma Study Group (MSG) prospective surgical trial presents interesting but perplexing results. Nine hundred patients with melanomas thicker than 2 mm were randomly assigned to 1-cm- or 3-cm-wide resection margins. The original enrollment was 600 patients, and time to local or in-transit recurrence and regional or distant metastasis were the primary endpoints. However, fewer events than anticipated occurred by 3 years, so the study was reconfigured to enroll 900 patients, and local, in-transit, and regional recurrences were lumped together as "locoregional relapse."

There were 37 and 25 local recurrences; 10 and 7 in-transit recurrences; and 135 and 118 nodal recurrences in the 1-cm and 3-cm groups, respectively. Collectively, there were 168 and 142 locoregional relapses, respectively (P=0.05). Median overall survival after locoregional recurrence was 27.9 months and 18.5 months, respectively (P=0.05), suggesting more aggressive disease in the 3-cm group.

Comment: This trial raises many issues. First, there is the unknown effect of the biostatistical reconfiguration. The original endpoints were clumped together halfway through the study, and 300 patients were added to compensate for fewer than anticipated events.

Second, the notion that wider margins lead to fewer recurrences is intuitively appealing, but the increase was in regional rather than local recurrence. Inadequate excision is traditionally linked with local recurrence. The possibility that cells left after narrow resection complete the migration to lymph nodes over time, without local or in-transit stops, is difficult to understand biologically. Nevertheless, these results may have serious medicolegal implications.

Third, had patients undergone sentinel lymph node biopsy, the pattern of recurrence would have been different, but distribution and extent are impossible to predict. The literature does not support "slow migration", because same-basin nodal relapse after negative sentinel lymph node biopsy is probably related to failure of detection rather than absence of tumor in the original node.

Fourth, this is the only prospective surgical trial to include thick melanomas and to show benefit from wide margins. However, it is difficult to apply these results clinically: Patients with tumors thicker than 2 mm usually receive 2-cm margins, an option not evaluated here. Other researchers have found no difference between 4-cm and 2-cm margins in patients with lesions of intermediate thickness (2-4 mm), suggesting that 3-cm margins would convey no benefit.

These findings will undoubtedly generate much discussion. The long-term clinical impact remains to be seen, especially in light of the frequent use of sentinel lymph node biopsies in the U.S.

— Hensin Tsao, MD, PhD

Published in Journal Watch Dermatology March 16, 2004

Citation(s):

Thomas JM et al. Excision margins in high-risk malignant melanoma. N Engl J Med 2004 Feb 19; 350:757-66.

• Original article (Subscription may be required)
• Medline abstract (Free)