LAUSR

Dear Editor, I have read with interest the article entitled “Sentinel lymph node biopsy vs. observation in thick melanoma: a multicenter propensity score matching study.” by Boada et al. who evaluated whether sentinel lymph node (SLN) biopsy increases survival in patients with thick cutaneous melanoma (Breslow> 4 mm). The authors also investigated correlations between survival and lymph node status. By median followup of 40 months, adjusted hazard ratio (AHR) (95% confidence interval [CI]) of the SLN biopsy group against observation group for disease-free survival and for overall survival were 0.74 (0.61–0.90) and 0.75 (0.60–0.94), respectively. Although SLN-negative patients had better 5and 10-year melanoma-specific survival (MSS) compared to SLN-positive patients, there was no significant difference of MSS between SLN biopsy group and observation group. I have some concerns about their study. First, the authors presented an advantage of survival by conducting SLN biopsy except MSS. They handled enough number of samples and propensity score matching procedure was an advantage for stable risk assessment. As there was a significant difference in the mean value of age between SLN biopsy group and observation group after propensity score matching, caution should be paid for the adjustment of age. Sinnamon et al. conducted a retrospective cohort study and odds ratios (95% CIs) of patients with thin melanoma younger than 40 years and patients aged from 40 to 64 years against patients of 65 years or older for lymph node metastasis were 2.04 (1.44–2.90) and 1.59 (1.19–2.11), respectively. Although the clinical stage of melanoma is different between two studies, the adverse effect of age on the prognosis of MSS would partly be related to no difference of MSS between SLN biopsy group and observation group. Second, Boada et al. conducted a retrospective cohort study, and random allocation on SLN biopsy could not be conducted. This means that selection bias exists although the authors adopted propensity score matching procedure for the adjustment of confounders. Study design sometimes influences the main results, and advantage of SLN biopsy for disease-free survival and for overall survival should be confirmed by further study. Finally, there are significant differences of AHR for survival in different centers and data from Sevilla showed superiority in disease-free survival, MSS, and overall survival. Contributing factors of AHR for survival in the different center should be specified.