Biomarkers for Nivolumab

Immunotherapy

Nivolumab appears to be an effective immunotherapy, improving overall survival (OS) in patients with metastatic renal cell carcinoma (mRCC) compared with currently available therapies, according to a prospective biomarker study. Moreover, the study found that nivolumab was effective in both programmed death-1 ligand 1 (PD-L1)-positive (PD-L1+) and PD-L1–negative (PD-L1–) patients. These findings suggest that better biomarkers are needed to predict which patients will not benefit from this immunotherapy.

In this study, PD-L1 status was assessed by a test from Dako, and ?5% staining in ?1 biopsy specimens was defined as PD-L1+.

“Consistent with the randomized phase 2 study of nivolumab in mRCC, overall survival appears longer in PD-L1+ patients, but it is promising in both PD-L1+ and PD-L1– patients, especially those who are treatment naive,” stated lead author Toni K. Choueiri, MD, Dana-Farber Cancer Institute, Boston, MA.

The study included 67 previously treated patients and 24 treatment-naive patients (N = 91, mostly males with prior nephrectomy). Those who were previously treated were randomized to 3 dose levels of nivolumab, and the treatment-naive patients received standard-dose nivolumab.

One of the objectives of this study was to assess the association of survival with several biomarkers, including PD-L1 expression, serum cytokines, gene expression, and T-cell receptor sequencing in archival tissue obtained at baseline and in a fresh biopsy after cycle 2 of therapy.

Median OS was 16.4 months at the lowest dose of nivolumab to 25 months or higher for patients who received 10 mg/kg (standard dose).

Two-year survival exceeded 60%, Choueiri told listeners.

When OS was analyzed according to PD-L1 expression, median OS was not reached (NR) in patients whose tumors expressed at least 5% of staining versus 23.4 months in PD-L1– patients. Looking at previously treated patients, OS was NR in PD-L1+ patients and 22.3 months in PD-L1– patients. Among treatment-naive patients, OS was NR in PD-L1+ patients versus 33.3 months in PD-L1– patients; however, 12-month survival was similar (71%) regardless of the level of PD-L1 expression in this study. Two-year survival trended toward better survival in PD-L1+ patients: 64% versus 48%, respectively.

“This is in the context of median overall survival of less than 16.5 months for currently available therapies in previously treated mRCC,” Choueiri said.

“We also tried to identify biomarkers beyond PD-L1,” he continued.

None of the changes in 4 serum-soluble factors were associated with survival. CXCL9 and ferritin levels trended toward an association with improved survival.

Analysis of immune checkpoint molecules and their association with <20% reduction in tumor burden suggests that responders had significantly higher expression of cytotoxic T-lymphocyte antigen-4 (CTLA-4), TIGIT, and PD-L2.

Looking at T-cell receptor sequencing, a higher frequency of T cells in the tumor was associated with longer survival; clonality of T cells in the tumor and blood trended in the right direction for an association with survival but did not reach statistical significance.

“These results suggest that adaptive immune activity may mediate responses to nivolumab in mRCC patients. The significant upregulation of CTLA-4 and PD-L2 underscores the potential of nivolumab and ipilimumab combination therapy in mRCC. Additional randomized controlled trials are needed to investigate this,” Choueiri concluded.

Noah M. Hahn, MD, Johns Hopkins University School of Medicine, Baltimore, MD, discussant of this abstract, said, “The clear message is that PD-L1 status does seem to be associated with increased tumor responses in a wide array of tumors, but there are several gaps.”

There are many problems with PD-L1 testing. PD-L1 status has been studied with different tests using different antibodies and different cutoffs for positivity, and these cutoffs are different even when the same antibody is used. Also, it is not clear whether to study tumor cells or immune cells.

“Right now, I would say the antibody used for the test matters, and this should be evaluated independently. Ideally, the cut point for positivity should be consistent with a given antibody. Also, the disease may matter. PD-L1+ patients have higher response rates in most tumors,” he noted.

Hahn stated that PD-L1 is a dynamic marker, so recent tumor tissue is better to study. “Consider enrichment for PD-L1 status in clinical trials only if you have a stable antibody and consistent cut points. It is more important to look at PD-L1 status in earlier-stage trials, and this may not be needed in combination therapy trials,” he noted.

Hahn found the data on the association of response with T-cell receptor signaling “interesting” and deserving of further study.

“We are at the end of the beginning of a new era for immunotherapy. We are so past the era of cytotoxic chemotherapy trials in nonselected patients that we can’t even see it in the rearview window anymore,” he commented.

Related Articles