Sentinel node biopsy and survival in cutaneous melanoma: a nationwide population-based cohort study

Introduction
The therapeutic benefit of sentinel node biopsy (SNB) for melanoma patients remains controversial. We aimed to evaluate the association between SNB and survival outcomes in a nationwide population-based Danish cohort.
Materials and methods
This retrospective cohort study included patients diagnosed with cutaneous melanoma between 2010 and 2017 who were candidates for SNB according to contemporary national guidelines. Augmented inverse probability of treatment weighting (AIPTW) was used to estimate standardized absolute risks and risk differences (RD) for 5-year overall survival, melanoma-specific mortality, other-cause mortality, and recurrence outcomes under two hypothetical strategies: all patients undergoing SNB versus no patients undergoing SNB.
Results
Of 6952 eligible patients, 5679 (81.7%) underwent SNB. After AIPTW, the standardized 5-year overall survival was higher under the SNB strategy (82.5% vs 73.7%, RD -8.8 pp, 95% CI -12.0; -5.6), whereas there was no clear difference in melanoma-specific mortality between strategies (9.0% vs 9.8%, RD -0.8 pp, 95% CI -3.0; 1.5). Other-cause mortality was lower under the SNB strategy (8.5% vs 16.8%, RD -8.3 pp, 95% CI -11.1; -5.5). No difference was found for standardized 5-year risk of any recurrence (19.0% vs 18.9%; RD 0.1 pp, 95% CI -3.2; 3.4), nor for distant recurrence. Regional nodal recurrence was lower under the SNB strategy (4.6% vs 8.5%, RD -3.9 pp, 95% CI -6.3; -1.5).
Conclusion
Since SNB cannot plausibly influence other-cause mortality, our findings indicate residual confounding despite robust adjusted analyses. Observational data such as ours cannot reliably determine whether SNB confers a therapeutic effect.