Assessment of survival benefit from sentinel node biopsy for melanoma: a systematic review and meta-analysis
Menée à partir d'une revue systématique de la littérature publiée jusqu'en janvier 2025 (13 études incluant au total 40 287 participants), cette méta-analyse évalue le bénéfice en termes de survie d'une biopsie du ganglion sentinelle chez les patients atteints d'un mélanome
Background : It is unproven whether sentinel node biopsy (SNB) for people with cutaneous melanoma improves survival. This study aimed to establish whether there is a reduction in the risk of death from melanoma after SNB.
Methods : This systematic review and meta-analysis involved searches of Medline, Embase, Cochrane CENTRAL, and ClinicalTrials.gov (up to Jan 8, 2025) using terms for melanoma, SNB, and survival. Studies were included if they reported survival in adults (ie, aged ≥18 years) with melanoma who underwent SNB relative to those who did not, and excluded studies without these data, because survival risk in people with melanoma could not be calculated. Duplicate title review (by GJW and CAG) and data extraction (by GJW and MBW) were performed. The primary outcome was death from melanoma. Risk of bias was assessed using the Newcastle-Ottawa and Cochrane Collaboration tools. This study was registered with PROSPERO (number CRD4203494674).
Findings : From 1560 screened records, 60 studies were eligible for analysis. After the exclusion of one study with high selection bias meta-analysis, 13 studies (40 287 participants) reporting an adjusted risk were analysed and showed a significantly reduced risk of death from melanoma for people who underwent SNB: hazard ratio (HR) 0·86 (95% CI 0·81–0·92, p<0·0001), with low heterogeneity (I2 16%). Sensitivity analyses confirmed that this result was not dependent on any single study or on the use of systemic therapy. A similar result was obtained from the analysis of the five studies (27 540 participants) that reported 5-year risk of death from melanoma: HR 0·84 (95% CI 0·78–0·90, p<0·0001) with low heterogeneity (I2 15%). A 10-year risk of death from melanoma was reported in two studies with three estimates: 0·87 (0·71–1·06; p=0·17) and with greater heterogeneity (I2 41%). From nine studies (11 625 participants) reporting an adjusted risk of recurrence, the HR was 0·71 (95% CI 0·66–0·76, p<0·0001) with low heterogeneity (I2 23%).
Interpretation : This meta-analysis showed that people with melanoma who underwent SNB had a significantly reduced risk of death from melanoma and recurrence compared with those who did not. These findings are consistent with the only published randomised controlled trial and were robust on sensitivity analyses, indicating that SNB confers true survival and recurrence benefits.
Funding : None.
The Lancet Oncology , article en libre accès, 2026