Read the full EBM Focus and earn CME credit.
Reference – MSLT-II trial (N Engl J Med 2017 Jun 8;376(23):2211) (level 1 [likely reliable] evidence)
- Immediate complete lymph node dissection (CLND) is thought to help control regional disease in patients with melanoma and sentinel lymph node metastases, but its effect on overall survival is unclear.
- In the MSLT-II trial, 1,939 adults (median age 54 years) with localized cutaneous melanoma (mean Breslow thickness 2.7 mm) and sentinel lymph node metastases were randomized to CLND within 140 days of diagnostic biopsy vs. nodal observation with ultrasonography.
- There was no difference in melanoma-specific survival which was 86% in both groups. Lymphedema occurred in 24.1% with CLND vs. 6.3% with nodal observation (p < 0.001, NNH 5).
CLND is recommended by the American Society of Clinical Oncology and Society of Surgical Oncology for patients with melanoma and positive sentinel lymph node biopsy to help control regional disease, but its effect on overall survival is unclear (J Clin Oncol 2012 Aug 10;30(23):2912). Although the previous DeCOG-SLT trial of 483 patients found no difference in survival comparing CLND vs. observation, definitive conclusions could not be drawn because the trial was underpowered (Lancet Oncol 2016 Jun;17(6):757). To address this question in a larger trial, the MSLT-II trial evaluated survival until death from melanoma following immediate CLND in 1,939 adults (median age 54 years) with localized cutaneous melanoma (mean Breslow thickness 2.7 mm) and sentinel lymph node metastases. The patients were randomized to CLND within 140 days of diagnostic biopsy vs. nodal observation with ultrasonography. All patients had a clinical exam every 4 months during years 1-2, every 6 months during years 3-5, and annually thereafter. The observation group had ultrasound assessment of the sentinel-node basin at every visit during years 1-5. The use of ultrasound to predict nodal metastases based on specific characteristics of nodal appearance has been previously shown to have a diagnostic sensitivity of 100% and specificity of 96% (J Am Acad Dermatol 1999 Nov;41(5 Pt 1):703).
Sentinel node metastases were determined by standard pathology (in 88%) or quantitative RT-PCR assay. At the third interim analysis, the data and safety monitoring board recommended release of the melanoma-specific survival data due to a low likelihood of a significant survival difference between the groups. The assigned intervention was declined in 14.4% of patients with CLND and in 3% with nodal observation. The presented data reflect the per protocol analysis in 91% of the randomized patients. Approximately 70% of the patients had just one pathologically detected sentinel node metastasis, 19% had more than one, and 11% had metastases by RT-PCR assay only. Comparing CLND vs. nodal observation at 3 years, there was no difference in melanoma-specific survival which was 86% in both groups with consistent results obtained in the intention-to-treat analysis. There was also no difference in distant metastasis-free survival. However, CLND was associated with increased nodal recurrence-free survival (in 92% vs. 77%, p < 0.001, NNT 7) and increased disease-free survival (in 68% vs. 63%, p = 0.048, NNT 20). Lymphedema occurred in 24.1% with CLND vs. 6.3% with CLND (p < 0.001, NNH 5). Nonsentinel lymph node metastases were present in 11.5% of the CLND group and their presence was associated with an increased risk of melanoma-related death (hazard ratio 1.78, 95% CI 1.19-2.67).
The results of the MSLT-II trial indicate that immediate (< 140 days) CLND does not improve 3-year melanoma-specific survival compared to nodal observation with ultrasound and that immediate CLND is associated with an increased risk of lymphedema. Although a benefit with immediate CLND was seen for nodal recurrence-free survival, it should be noted that it was not shown that the observation group had a poorer outcome. Another benefit of immediate CLND is that it provides information on the metastatic status of nonsentinel lymph nodes, a factor that was shown to be of prognostic value. Patients with metastatic nonsentinel lymph nodes were shown to have a poorer prognosis and this information may help guide choices for adjuvant therapy. Overall, the results indicate that immediate CLND does not increase melanoma-specific survival compared to clinical observation with ultrasound assessment using specific criteria, and that the high risk of lymphedema with CLND will have to be weighed against obtaining additional staging information.
For more information, see the Melanoma topic in DynaMed Plus. DynaMed users click here.