Propensity Score–Matching Analysis Shows No Disease-Specific Survival Benefit of Sentinel Lymph Node Biopsy vs Observation in Head/Neck Melanoma
In a Surveillance, Epidemiology, and End Results (SEER) database analysis reported in JAMA Otolaryngology-Head & Neck Surgery, Sperry et al found that sentinel lymph node biopsy was not associated with improved disease-specific survival in propensity score–matched patients with head and neck melanoma.
Study Details
The study involved data from 7,266 patients with head and neck melanoma treated with sentinel lymph node biopsy with or without neck dissection (54%) or no sentinel lymph node biopsy or neck dissection (nodal observation, 46%) between 2004 and 2011. The two groups differed significantly with regard to every patient and disease characteristic considered in the study.
The cohorts were matched using propensity scores modeled on 10 covariates known to be associated with sentinel lymph node biopsy treatment or melanoma survival, resulting in 2,551 matched pairs. These groups were stratified by thin (> 0.75–1.00 mm), intermediate (> 1.00–4.00 mm), and thick (> 4.00 mm) Breslow thickness and matched within five age categories.
No Difference in Thickness Cohorts
In the intermediate-thickness cohort (n = 2,808, 1,404 matched pairs), estimated 5-year disease-specific survival was 89% in patients receiving sentinel lymph node biopsy and 88% for those with nodal observation (P = .30). There was also no significant difference between sentinel lymph node biopsy and no sentinel lymph node biopsy groups in the thin-lesion cohort (552 matched pairs; P = .25) or the thick-lesion cohort (354 matched pairs; P = .22). Overall, estimated 5-year disease-specific survival was 96% in the thin-lesion cohort and 70% in the thick-lesion cohort.
Cox proportional hazards analysis showed that hazard ratios for death from melanoma for the sentinel lymph node biopsy vs no sentinel lymph node biopsy groups were 0.94 (P = .57) among all matched pairs, 0.87 (P = .31) in the intermediate-lesion cohort, 1.53 (P = .24) in the thin-lesion cohort, and 0.80 (P = .23) in the thick-lesion cohort.
The investigators concluded: “This SEER cohort analysis demonstrates no significant association between [sentinel lymph node biopsy] and improved disease-specific survival for patients with [head and neck melanoma].”
Nitin A. Pagedar, MD, MPH, of University of Iowa, Iowa City, is the corresponding author for the JAMA Otolaryngology-Head & Neck Surgery article.
The authors reported no conflicts of interest.
The content in this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO®) and does not necessarily reflect the ideas and opinions of ASCO®.
