Response to PD-1 Inhibitors in Metastatic Melanoma Predicted With Bim Protein

Frequency of the cell death-regulating protein Bim (BCL-2-interacting mediator of cell death) may predict patient response to PD-1 inhibitor immunotherapy.

Roxana Dronca, MD

Frequency of the cell death-regulating protein Bim (BCL-2-interacting mediator of cell death) may predict patient response to PD-1 (programmed death protein-1) inhibitor immunotherapy, according to researchers at the Mayo Clinic. The results were presented in an abstract at the International Cancer Immunotherapy Conference,on September 16, 2015, in New York City.1

“The discovery of biomarkers of sensitivity are vital not only for informing clinical decisions, but also to help identify which patients with melanoma, and possibly other malignancies, are most likely to benefit from PD-1 blockade," said Roxana Dronca, MD, a hematologist at Mayo Clinic and lead author of the abstract, in a recent press release.2

Interaction between the PD-1 receptor on tumor-targeting CD8 T cells and its ligand PD-L1 activates Bim on the T cell to induce T-cell death. Many tumors overexpress PD-L1 to destroy the tumor-targeting T cells and prevent the individual’s immune system from recognizing and eliminating the tumor. Therefore, researchers have aimed to develop cancer therapies that re-establish T-cell and immune system function by inhibiting the interaction between PD-1 and PD-L1.

PD-1 inhibitors such as pembrolizumab have shown promising results for several types of treatment-refractory cancers, particularly metastatic melanoma.3However, PD-1 blockade yields variable clinical responses among patients. A sensitivity analysis of three PD-1/PD-L1 blockade agents (pembrolizumab, nivolumab, and MPDL3280A) showed a greater overall response rate.4However, a recent review5indicated that tumor expression of PD-L1 is likely an inadequate biomarker on its own to predict response to PD-1 blockade therapy.

A previous study6by the Mayo Clinic researchers showed that soluble B7-H1 was present in the serum of patients with renal cell carcinoma and, like Bim, induced apoptosis in tumor-targeting T cells due to retention of the PD-1 binding domain. The researchers had also observed that patients who responded to PD-1 inhibitor therapy had higher levels of soluble PD-L1 in the blood, suggesting that PD-L1—expressing tumor tissue may trigger release of biologically active molecules that could be measured in the systemic circulation. According to Dronca, measuring and monitoring the levels of these molecules could identify patients who could benefit from PD-1 blockade or combination therapies.

“If I know that a patient has a very high likelihood of responding to anti—PD-1 therapy, I’m going to be more inclined to recommend that treatment and feel better about the choice,” said Dronca.

To investigate potential biomarkers, Dronca et al studied blood samples of patients with unresectable metastatic melanoma who received pembrolizumab. Patients who responded to pembrolizumab had higher serum levels of Bim- and PD-1-expressing tumor-targeting T cells prior to therapy than non-responders. In addition, levels of Bim- and PD-1-expressing tumor-reactive T cells decreased significantly more in responders than in nonresponders after the first 3 months of treatment. According to Dronca, the decrease suggests reduced PD-1 engagement with tumor-associated PD-L1 due to regression of the tumor or redistribution of the PD-1-containing T cells into the tumor sites.

Dronca also indicated two patients who experienced pseudo-progression at the first tumor assessment had decreased frequency of Bim- and PD-1—expressing tumor-reactive T cells, suggesting Bim could be used to monitor responses during treatment.

“[Identifying those who would benefit from PD-1 therapy] will allow us to expose fewer patients to inadequate treatments and their associated toxicities and costs,” said Dronca.

Responders also had higher serum levels of soluble PD-L1 than nonresponders, indicating that PD-1 inhibitor therapy is most effective when the interaction between PD-l and PD-L1 plays a major role in the disease. To further understand the ability of Bim to predict response to PD-1 inhibitor therapy, Dronca and her colleagues plan to investigate the relationship between Bim levels in the circulation and PD-L1 expression in tumor tissue, as well as mechanistic studies of the PD-1 signaling pathway.

Identifying serum, rather than tissue, biomarkers that predict response to therapy also has added benefits, as blood testing is more convenient and less invasive than repeated tissue biopsies.

“A great advantage of our approach lies in the ease of serial peripheral blood testing compared with repeated invasive tissue biopsies,” said Dronca.

According to Dronca, modulation of Bim is an area of active research both in immunotherapies and combination with chemotherapeutic agents in a variety of cancers. The Mayo Clinic researchers are currently validating their findings in a larger prospective cohort of patients with melanoma and lung cancer and also plan to test levels of Bim in response to nivolumab, another PD-1 inhibitor.


1. Dronca RS, Markovic SN, Kottschade LA, Nevala W, Dong H. Bim as a predictive T cell biomarker for response to anti-PD-1 therapy in metastatic melanoma (MM). Poster presented at: CRI-CIMT-EATI-AACR - The Inaugural International Cancer Immunotherapy Conference: Translating Science into Survival. September 16-19, New York, NY, Abstract A007.

2. Mayo Clinic News Network. Identifying protein that may predict response to PD-1 immunotherapy for melanoma. Available at Accessed September 21, 2015.