3493 - Mixture-cure modeling for resected stage III/IV melanoma in the phase 3 CheckMate 238 trial

Background

A goal of adjuvant therapy is to prolong disease-free survival, with the risk of mortality approaching that of the general population. From the literature, recurrences in stage III melanoma are most likely to occur within 3 years after surgery. Mixture-cure models (MCMs) capture survival heterogeneity of cancer patients by assuming they are either “cured” or “uncured”, where cured patients are defined as having a low risk of relapse.

Methods

We applied MCMs to 3-year recurrence-free survival (RFS) outcomes in patients with resected melanoma treated with adjuvant nivolumab (NIVO) or ipilimumab (IPI) in the double-blind, phase III CheckMate 238 trial (NCT02388906) and assumed that the cured fraction has mortality risks similar to that of the general population. Patient age, sex, and geographic region information from CheckMate 238 and mortality rates from the World Health Organization were used to derive a cohort-level background survival distribution representative of “cured” or “disease-free” patients. Parametric functions were used to model uncured population outcomes. Parameters of the MCM were estimated by means of maximum likelihood methods. Statistical goodness-of-fit metrics and visual inspection showed that generalized gamma and log-logistic distributions were best suited for survival analyses of cured and uncured patients, respectively.

Results

The 3-year RFS rates were explained by the fraction of patients with a low risk of relapse in the NIVO arm at 55.1% (95% CI, 49.2–60.1) and IPI arm at 39.9% (95% CI, 33.6–46.1). Sensitivity analyses indicated that the estimated fraction of patients with a low risk of relapse was > 50% in the NIVO arm. Estimated RFS distributions of the uncured patients exhibited similar patterns for both arms, implying the differences in RFS rates between treatment arms can be explained with the elicited differences in proportions of patients at high vs low risk of relapse.

Conclusions

MCM analyses from CheckMate 238 suggested that adjuvant treatment with NIVO leads to a higher proportion of patients with low risk of relapse compared with IPI. Validation of the results from MCMs will include analyses of longitudinal RFS and overall survival data from the EORTC 18071 trial.

Clinical trial identification

NCT02388906.

Editorial acknowledgement

Kakoli Parai, PhD, and Andrea Lockett at StemScientific, an Ashfield Company, funded by Bristol-Myers Squibb.

Legal entity responsible for the study

Bristol-Myers Squibb.

Funding

Bristol-Myers Squibb.

Disclosure

J.S. Weber: Honoraria (self), Advisory / Consultancy: AbbVie; Honoraria (self), Advisory / Consultancy, Shareholder / Stockholder / Stock options: AltorBioScience; Honoraria (self), Advisory / Consultancy, Travel / Accommodation / Expenses: Amgen; Research grant / Funding (institution): Astellas Pharma; Honoraria (self), Advisory / Consultancy, Travel / Accommodation / Expenses: AstraZeneca; Licensing / Royalties, Named on a patent for a PD-1 antibody biomarker: Biodesix; Shareholder / Stockholder / Stock options: Biond; Honoraria (self), Advisory / Consultancy, Research grant / Funding (institution), Travel / Accommodation / Expenses: Bristol-Myers Squibb; Honoraria (self), Advisory / Consultancy, Travel / Accommodation / Expenses: cCam Biotherapeutics; Honoraria (self), Advisory / Consultancy, Travel / Accommodation / Expenses: Celldex; Honoraria (self), Advisory / Consultancy, Travel / Accommodation / Expenses: Daiichi Sankyo; Honoraria (self), Advisory / Consultancy: Eisai; Honoraria (self), Advisory / Consultancy, Research grant / Funding (institution), Travel / Accommodation / Expenses: Genentech; Honoraria (self), Advisory / Consultancy, Research grant / Funding (institution), Travel / Accommodation / Expenses: GlaxoSmithKline; Honoraria (self), Advisory / Consultancy: Ichor Medical Systems; Research grant / Funding (institution): Incyte; Honoraria (self), Advisory / Consultancy, Travel / Accommodation / Expenses: Lion Biotechnologies; Honoraria (self), Advisory / Consultancy: Medivation; Honoraria (self), Advisory / Consultancy, Research grant / Funding (institution), Travel / Accommodation / Expenses: Merck; Licensing / Royalties, Named on patent for an ipilimumab biomarker: Moffitt Cancer Center; Honoraria (self), Advisory / Consultancy: Nektar; Honoraria (self), Advisory / Consultancy, Travel / Accommodation / Expenses: Pieris Pharmaceuticals; Honoraria (self), Advisory / Consultancy, Research grant / Funding (institution), Travel / Accommodation / Expenses: Roche; Honoraria (self), Advisory / Consultancy: Sellas Life Sciences; Honoraria (self), Advisory / Consultancy: WindMIL; Honoraria (self), Advisory / Consultancy, Shareholder / Stockholder / Stock options: CytomX Therapeutics; Honoraria (self), Advisory / Consultancy, Research grant / Funding (institution), Travel / Accommodation / Expenses: Novartis. M. Kurt: Shareholder / Stockholder / Stock options, Full / Part-time employment: Bristol-Myers Squibb. M. Edmondson-Jones: Advisory / Consultancy: Bristol-Myers Squibb. A. Amadi: Shareholder / Stockholder / Stock options, Full / Part-time employment: Bristol-Myers Squibb. M. Lobo: Full / Part-time employment: Bristol-Myers Squibb. A. Moshyk: Shareholder / Stockholder / Stock options, Full / Part-time employment: Bristol-Myers Squibb. S. Kotapati: Shareholder / Stockholder / Stock options, Full / Part-time employment: Bristol-Myers Squibb. P. Mohr: Honoraria (self), Advisory / Consultancy: Pierre Fabre; Honoraria (self), Advisory / Consultancy: GlaxoSmithKline; Honoraria (self), Advisory / Consultancy: Merck, Sharp & Dohme; Honoraria (self), Advisory / Consultancy: Merck Germany; Honoraria (self), Advisory / Consultancy: Roche; Honoraria (self), Advisory / Consultancy: Bristol-Myers Squibb; Honoraria (self), Advisory / Consultancy: Novartis; Honoraria (self), Advisory / Consultancy: Sanofi.