This phase II trial tests the addition of BMS-986016 (relatlimab) to the usual immunotherapy after initial treatment for nasopharyngeal cancer that has come back after a period of improvement (recurrent) or that has spread from where it first started (primary site) to other places in the body (metastatic). Relatlimab is a monoclonal antibody that may interfere with the ability of tumor cells to grow and spread. The usual approach of treatment is initial treatment with chemotherapy such as the combination of cisplatin (or carboplatin) and gemcitabine, along with immunotherapy such as nivolumab. After the initial treatment is finished, patients may continue to receive additional immunotherapy. Carboplatin is in a class of medications known as platinum-containing compounds. It works in a way similar to the anticancer drug cisplatin, but may be better tolerated than cisplatin. Carboplatin works by killing, stopping or slowing the growth of tumor cells. Immunotherapy with monoclonal antibodies, such as nivolumab, may help the body's immune system attack the cancer, and may interfere with the ability of tumor cells to grow and spread. Gemcitabine is a chemotherapy drug that blocks the cells from making deoxyribonucleic acid (DNA) and may kill cancer cells. Giving BMS-986016 in addition to the usual immunotherapy after initial treatment may extend the time without the tumor cells growing or spreading longer than the usual approach in patients with recurrent or metastatic nasopharyngeal cancer.
Tumor showing (histological/cytological) Epstein-Barr encoded ribonucleic acid (EBER)-positivity (e.g., In situ hybridization, immunohistochemistry) or
A known history of detectable plasma EBV DNA (via a polymerase chain reaction [PCR]-based assay) at any time point since the initial diagnosis of NPC.
Measurable disease as defined by RECIST 1.1 criteria. Lesion(s) that have been irradiated previously can be counted as measurable as long as radiological progression after the prior radiation therapy has been demonstrated.
Contrast enhanced CT scan of the chest. The contrast enhanced CT component of a whole-body PET-CT is also acceptable. The plain (non-contrast) CT component of a PET-CT is not acceptable.
None
Sequential
Biospecimen Collection
Procedure/SurgeryUndergo blood sample collection
Bone Scan
Procedure/SurgeryUndergo bone scan
Carboplatin
DrugGiven IV
Cisplatin
DrugGiven IV
Computed Tomography
Procedure/SurgeryUndergo CT or PET/CT
Gemcitabine
DrugGiven IV
Magnetic Resonance Imaging
Procedure/SurgeryUndergo MRI
Nivolumab
Biological/VaccineGiven IV
Positron Emission Tomography
Procedure/SurgeryUndergo PET/CT
Relatlimab
Biological/VaccineGiven IV
Progression-free survival (PFS)
Will be assessed in both treatment arms and will be estimated using the Kaplan-Meier method. The comparison of PFS distributions between treatment arms will be performed using the log-rank test. Additional analyses of treatment effect will be performed using Cox models with the stratification factors included as a fixed covariates, as well as other tumor and patient characteristics, listed below. Median PFS and the corresponding 95% confidence intervals (CIs) for each arm will be estimated with the Brookmeyer-Crowley method. Multivariable analysis will be performed using a Cox proportional hazards model and relevant patient and tumor characteristics (country \[Asian sites versus (vs.) Non-Asian sites\], keratinizing squamous carcinoma, sex, age, and baseline Epstein-Barr virus (EBV) deoxyribonucleic acid \[DNA\]). Hazard ratios and their respective 85% lower confidence bound, and 95% confidence intervals will be provided.
Overall survival (OS)
Will be estimated using the Kaplan-Meier method and between-arm differences will be assessed using the log-rank test. Median OS and the corresponding 95% CIs using log-log transformation to survival function for each arm will be estimated with the Brookmeyer-Crowley method.
Tumor response
Will be assessed using Response Evaluation Criteria in Solid Tumors version 1.1. Cox proportional hazards models will be used to determine hazard ratios (cause-specific hazard ratios in the case of endpoints with competing risks) and to assess the effects of covariates of interest. Fine-Gray sub-distribution hazards Cox model will be used to complement the competing risk analyses.
Locoregional failure
Cox proportional hazards models will be used to determine hazard ratios (cause-specific hazard ratios in the case of endpoints with competing risks) and to assess the effects of covariates of interest. Fine-Gray sub-distribution hazards Cox model will be used to complement the competing risk analyses.
Distant metastasis
Cox proportional hazards models will be used to determine hazard ratios (cause-specific hazard ratios in the case of endpoints with competing risks) and to assess the effects of covariates of interest. Fine-Gray sub-distribution hazards Cox model will be used to complement the competing risk analyses.
Post-induction EBV DNA as a predictive biomarker
Will be assessed of the predictive role of post-induction EBV DNA on PFS and OS will be performed using the binary biomarker. Cox proportional hazards model with treatment arm, binary biomarker, and its interaction will be used for these analyses. Treatment effect hazard ratios and 95% confidence intervals will be reported. Adjusted HR estimates will be obtained after adding key patient and tumor characteristics.
Click on the lead sponsor to view all of their trials.