Abstract
METHODS
Data were collected from PRADO case report forms. Group differences were assessed using Fisher's exact, chi-square, or Mann-Whitney U tests. Logistic regression identified predictors of TLND, with a subgroup analysis restricted to pathology-positive TLND. Model performance was assessed using (cross-validated) area under the receiver operating characteristic curve.
CONCLUSIONS
Age younger than 40 years and unfavorable radiological response to neoadjuvant immunotherapy were associated with TLND, suggesting these parameters may identify patients at risk of residual nodal disease and subsequent need for TLND.
RESULTS
Ninety-nine patients were included; 36 underwent TLND: 20 (56%) with pathological nonresponse, eight (22%) with partial response, and eight (22%) for other indications (nodal recurrence after MPR, n = 3; extranodal extension, n = 1; suspicious nodes on imaging, n = 2; or infeasible ILN resection due to immune-related adverse events, n = 2). Age < 40 years (odds ratio 7.91, 95% confidence interval 1.35-46.30; P = 0.022) and radiographic increase in tumor size (odds ratio 1.15; 95% confidence interval 1.07-1.24; P < 0.001) were associated with TLND.
BACKGROUND
In the PRADO trial, a multicenter phase II study, patients with macroscopic stage III melanoma received neoadjuvant ipilimumab and nivolumab followed by resection of the marked index lymph node (ILN). Patients without a major pathological response (MPR) underwent therapeutic lymph node dissection (TLND), although some required TLND for other indications. We investigated whether baseline clinicopathological and preoperative response parameters identify patients benefiting from immediate TLND rather than ILN resection followed by potential secondary TLND.