Biomarker analyses from the phase III CheckMate 214 trial evaluating nivolumab plus ipilimumab in patients with advanced renal cell carcinoma: the search continues
Despite the fact that nivolumab in combination with ipilimumab (NIVO+IPI) is an established first-line treatment option for patients with advanced renal cell carcinoma, no biomarkers exist to identify those patients that will most likely respond to treatment. Biomarker analyses from the CheckMate 214 trial failed to show an advantage of the PD-L1 combined positivity score over tumour cell PD-L1 expression. In addition, the evaluated genomic biomarkers were not predictive for progression-free or overall survival in patients treated with NIVO+IPI. Also, immune-related gene expression signatures, previously derived from RCC patients who were treated with an immune checkpoint inhibitor (either as monotherapy or in combination with a vascular endothelial growth factor inhibitor), did not prove to be predictive for PFS or OS with NIVO+IPI. In contrast, the presented biomarker analysis did reveal that differences in the inflammatory response and the tumour microenvironment might be associated with a longer PFS under NIVO+IPI.
Nivolumab combined with ipilimumab (NIVO+IPI), among other immune checkpoint inhibitors (ICIs), is an established first-line treatment for patients with advanced renal cell carcinoma (aRCC). To date, however, there is an unmet need for biomarkers that allow the identification of patients with aRCC who are most likely to respond to this treatment. Efficacy of ICIs, alone or in combination with vascular endothelial growth factor (VEGF) inhibitors, has previously been associated with the expression of signatures indicative of T-effector function in myeloid inflammation. The predictive value of tumour cell programmed death ligand 1 (PD-L1) expression and tumour mutational burden (TMB) for ICI therapy is under investigation. During ASCO 2020, Motzer et al. presented an analysis of potential response biomarkers and studied the underlying biology of patients responsive to treatment with NIVO+IPI.
To this end, formalin-fixed, paraffin-embedded aRCC samples were collected from nearly 1,000 patients enrolled in the randomised, phase III CheckMate 214 trial. In this study, NIVO+IPI was compared to sunitinib as a first-line treatment for aRCC. Nearly 500 tumours were studied for selected genomic characteristics, gene expression signatures, and a gene set enrichment analysis. The two latter methodologies were performed on 213 and 171 patients, respectively. For this analysis, PD-L1 expression was assessed by immunohistochemistry according to percent positive observed in tumour cells (TC PD-L1), and categorised as less than 1% versus greater or equal to 1%. In addition to this, also the combined positivity score (CPS) was used in which the PD-L1 expression is quantified on both tumour and immune cells.
As reported earlier, the overall survival (OS) was improved with NIVO+IPI compared to sunitinib. This OS benefit was seen regardless of the PD-L1 expression and irrespective of the quantification method that was used (CPS or tumour cell expression). Similar findings were found with respect to PFS. The investigators did not see a benefit of using CPS over a PD-L1 assessment using the TC PD-L1 system.
In a second step, whole genome sequencing (WES) was performed to identify the genomic characteristics of the different samples. Neither of the four WES-derived biomarkers that were previously reported to be of importance in predicting the outcome to ICI therapy in RCC or other malignancies (high TMB, high tumour indel burden, PBRM1 mutation or HLA heterozygosity) were found to be associated with improved OS in the NIVO+IPI arm. In contrast, a longer PFS and OS was associated with high tumour indel burden in the sunitinib arm (p< 0.01 and p= 0.05, respectively).
RNA sequencing was used to compare the survival of patients within each treatment arm by previously published gene expression signatures. The Angio signature score from IMmotion 150 was studied in 213 patients in the CheckMate 214 trial. Sunitinib treated patients with a higher angiogenesis (Angio) gene signature score had a longer PFS than those with a lower score (p= 0.02), as well as a trend to a more favourable overall survival (39.7 versus 33.6 months, p=0.17). However, the high versus low Angio score did not predict OS in patients treated with NIVO+IPI (p=0.35 for PFS and p=0.71 for OS). No association between PFS or OS for NIVO+IPI was observed according to T-effector and JAVELIN immune signatures. One hypothesis is that IPI may enhance T-cell infiltration into tumours, perhaps bypassing a need for T-effector cells in the TME prior to therapy.
Finally, a gene set enrichment analysis was performed using the molecular signatures database Hallmark gene sets collection of 50 well-defined biological processes. Genes were identified that were differentially expressed in patients with an exceptionally prolonged PFS with the NIVO+IPI combination, relative to patients who definitely progressed by 18 months or less. Prolonged PFS with NIVO+IPI (≥ 18 months, N=27) was associated with a higher expression of Hallmark inflammatory response (TNFα signalling via NFκβ) and Hallmark epithelial mesenchymal transition (EMT) gene sets (both adjusted p=0.002).
PD-L1 CPS showed no advantage over the evaluation of PD-L1 expression on tumour cells alone, and the genomic biomarkers evaluated were not predictive for PFS or OS with NIVO+IPI. Immune-related gene expression signatures previously derived from patients with advanced RCC treated with ICI monotherapy, or ICI plus VEGF inhibitor combinations, did not predict PFS or OS with NIVO+IPI. One hypothesis is that NIVO+IPI may enhance T-cell infiltration in tumours, bypassing a need for T-effector cells in the tumour microenvironment prior to therapy. The OS with NIVO+IPI was similar in patients with high versus low Angio scores. PFS with sunitinib was improved in patients with high versus low Angio scores. Prolonged PFS with NIVO+IPI may be linked to differences in inflammatory response in the tumour microenvironment, but further validation of these exploratory analyses are necessary to make any firm conclusions.
Motzer RJ, Choueiri TK, McDermott DF, et al. Biomarker analyses from the phase III CheckMate 214 trial of nivolumab plus ipilimumab (N+I) or sunitinib (S) in advanced renal cell carcinoma (aRRC). Presented at ASCO 2020; Abstract 5009.