Addition of MPDL3280A to Bevacizumab Induces Responses in mRCC

Mario Sznol, MD

MPDL3280A, an investigational antibody that targets programmed death-ligand 1 (PD-L1), in combination with bevacizumab had strong antitumor activity and induced responses in 4 of 10 patients with metastatic renal cell carcinoma (mRCC) in an open-label phase Ib study. The agent was well tolerated without exacerbating bevacizumab-associated adverse events.

With promising preliminary clinical activity and documented immune modulation of the tumor microenvironment observed, MPDL3280A alone and in combination with bevacizumab is currently being investigated in a phase II study, comparing these regimens with sunitinib in patients with previously untreated locally advanced or mRCC, said Mario Sznol, MD, at the 2015 Genitourinary Cancers Symposium.

MPDL3280A is a humanized monoclonal anti-PD-L1 antibody that inhibits PD-L1 signaling. It is engineered to remove Fc-effector function to prevent PD-L1 binding to the inhibitory receptors PD-1 and B7.1 on activated T cells, to restore tumor-specific T-cell immunity. It has demonstrated durable antitumor effects as a single agent in several cancers, including RCC.

Bevacizumab with interferon is an approved regimen for the treatment of mRCC. “A lot of people don’t use that combination because it’s harder to give than just giving a pill,” said Sznol, associate professor of internal medicine, Yale Cancer Center, New Haven, Connecticut. “So outside of a clinical trial we don’t use a lot of bevacizumab in metastatic kidney cancer by itself. Most people use the small molecule oral agents, but this [combination] could bring it back.”

Bevacizumab increases chemokine expression in the tumor microenvironment, which theoretically recruits T cells because T cells recognize their antigen and get activated, he explained. “PD-L1 blockade keeps those T cells active, so it’s a really nice combination,” he said. “The bottom line is the antitumor effects.”

The study had multiple arms of patients with various solid tumors (colorectal cancer, cutaneous lesions, or mRCC); there were 10 patients with mRCC with clear cell histology who had pretreatment tumor specimens available, and data from this cohort were reported here.

Bevacizumab was given for one 21-day cycle prior to a first biopsy, and MPDL3280A was added for the second cycle and continued. A second biopsy was performed after the third cycle of treatment. The median duration of treatment with MPDL3280A was 288 days.

There were four objective responses (40%) by Response Evaluation Criteria in Solid Tumors (RECIST) criteria, all of which were partial responses. Five other patients (50%) had stable disease as their best response. Four of these five had prolonged (≥24 weeks) stable disease.

In comparison, the objective response rate for MPDL3280A monotherapy in a previous trial in RCC was 15%, and the objective response rate for bevacizumab monotherapy in this setting is 10%.

Six grade-3 or -4 adverse events occurred, but none were attributable to MPDL3280A. Adverse events possibly related to MPDL3280A, as judged by the investigator, included fatigue (40%), decreased appetite (30%), and diarrhea (30%).

Increases in CD8+ cell infiltration and decreases in CD31 expression were observed after treatment with MPDL3280A plus bevacizumab. The increase in CD8+ cells with bevacizumab was greatly enhanced with the combination.

Expression of some chemokines in tumors increased after bevacizumab administration. Further increases in expression of CCL5 (a chemoattractant for T cells, eosinophils, and basophils), CCR5 (the receptor for CCL5), CX3CL1 (a potent chemoattractant for T cells and monocytes), CCR7 (chemoattractant for T cells and stimulator of dendritic cell maturation), and CXCL10 (chemoattractant for immune cells) were observed after the addition of MPDL3280A. These increases in chemokine expression suggest that the two agents together can enhance the antitumor immune response.