Multicenter Adoption and Outcomes of Nodal Observation for Patients with Melanoma and Sentinel Lymph Node Metastases

BackgroundFollowing publication of the MSLT-II trial showing no survival benefit of completion lymphadenectomy (CLND) in patients with melanoma sentinel lymph node (SLN) metastases, it is expected that practice patterns have changed. The purpose of this study is to understand real-world practices and outcomes after publication of this landmark trial.Patients and MethodsPatients with truncal/extremity melanoma SLN metastases diagnosed between 2013 and 2019 at four academic cancer centers were included in this retrospective cohort study. Descriptive statistics, Cox proportional hazards model, and multivariable regression were used to characterize the cohort and identify predictors of CLND, harboring non-SLN (NSLN) metastases, and survival.ResultsResults of 1176 patients undergoing SLN biopsy, 183 had SLN metastases. The number of patients who underwent CLND before versus after trial publication was 75.7.% versus 20.5% (HR 0.16, 95% CI 0.09–0.28). Of those undergoing nodal observation (NO), 92% had a first nodal-basin ultrasound, while 63% of patients had a fourth. In exploratory multivariable analyses, age ≥ 50 years was associated with lower rate of CLND (HR 0.58, 95% CI 0.36–0.92) and larger SLN deposit (> 1.0 mm) with increased rate of CLND (HR 1.87, 95% CI 1.17–3.00) in the complete cohort. Extracapsular extension was associated with increased risk of NSLN metastases (HR 12.43, 95% CI 2.48–62.31). Adjusted survival analysis demonstrated no difference in recurrence or mortality between patients treated with CLND versus NO at median 2.2-year follow-up.ConclusionNodal observation was rapidly adopted into practice in patients with melanoma SLN metastases at four centers in Canada. Younger age and higher nodal burden were associated with increased use of CLND after trial publication. Ultrasound (US) surveillance decreased with time from SLNB. In our study, CLND was not associated with a decreased risk of recurrence or mortality.