Oncology Central

Researchers pioneer new methodology to predict patient response to ‘breakthrough’ melanoma drug

University of California, Los Angeles (UCLA) Jonsson Comprehensive Cancer Center (CA, USA) researchers have spearheaded a new technique to predict patient responses to the recent breakthrough drug pembrolizumab. The results were reported recently in the journal Nature.

T cells are prevented from attacking tumor cells by a protein termed PD-1. Pembrolizumab binds to PD-1, a cell surface receptor, resulting in the activation of T-cell mediated immune responses against cancer cells. Patients with advanced melanoma will respond well or not at all when treated with the drug, and this study could result in more effective use of the drug for melanoma as well as other cancers.

Paul Tumeh (UCLA) explained the problem: “We’ve had amazing clinical success treating patients battling advanced melanoma with pembrolizumab. The challenge is that it only works in approximately 30% of patients with melanoma. To address this challenge, we developed an approach that can select for patients that are likely to respond to this therapeutic class.”

The 2-year study involved 46 patients with advanced melanoma who had undergone biopsies before and during treatment with pembrolizumab. The team analyzed these tissue samples and grouped them according to patients who had responded to treatment and those who had not. This information allowed the researchers to design an algorithm that predicts the likelihood of treatment success in individual patients.

Antoni Ribas (UCLA) highlighted the clinical value of the work: “Our job was to figure out why some patients are predisposed to respond and others are not. Now, with these results, researchers can develop better drug combinations that are more effective, less costly and with fewer side effects.”

In order to test the validity of their algorithm, it was applied to 15 additional tumor samples from patients in Paris (France) previously given pembrolizumab. Though outcomes were not revealed to the team in advance, their algorithm correctly predicted 13 out of 15 patient outcomes.

According to Tumeh, this work has paved the way for further individualization of this treatment in melanoma and other cancers: “The next big step is to classify the different types of patients that do not respond to treatments so we could modulate the drug to target their tumors.”

Source: UCLA press release




News Headlines

Leave A Comment

Please wait...

Would you like access to more immuno-oncology content?

Register with Oncology Central to find out the latest news, opinions and journal articles published in the oncology field by leading experts.