NEW YORK (GenomeWeb) – A team of researchers from Columbia University, Northwestern University School of Medicine, and elsewhere has identified tumor and microenvironment features in glioblastoma (GBM) brain tumors that are linked to response to immune checkpoint immunotherapy targeting PD-1.
"Our study shows that clinical response to anti-PD-1 immunotherapy in GBM is associated with specific molecular alterations, immune expression signatures, and immune infiltration that reflect the tumor's clonal evolution during treatment," co-corresponding authors Raul Rabadan, Adam Sonabend, and Fabio Iwamoto and their colleagues wrote.
As they reported today in Nature Medicine, the researchers retrospectively analyzed longitudinal samples from 66 individuals with GBM who were treated with anti-PD-1 drugs such as nivolumab or pembrolizumab at disease recurrence after prior chemo- and radiotherapies, including 17 GBM patients with long-term immunotherapy responses.
Using gene panel sequencing, exome sequencing, and transcriptome sequencing, together with tissue imaging and clinical data, the team tracked down non-response-related mutations in the PTEN gene, for example, and response-linked alterations affecting components of the MAP kinase pathway. The data also highlighted tumor evolution, T cell diversity, and tumor microenvironment features with potential ties to GBM patient response to anti-PD-1 treatment.
"While overall PD-1 inhibitors do not provide a survival benefit for patients with GBM, our study showed that a sub-group of patients might benefit from this therapy, suggesting a molecular, personalized approach for refining patient selection for immunotherapy," the authors wrote. "Although this approach requires further validation, it might provide a means for the effective application of therapy for glioblastoma."
In clinical trials so far, the team noted, less than 10 percent of individuals with GBM showed long-term response to checkpoint blockade immunotherapy, though the molecular features that forecast such responses remain largely unknown.
To address that puzzle, the researchers sequenced the exomes of 58 matched tumor-normal samples collected longitudinally from 17 GBM patients, as well as the transcriptomes of 38 of the tumor-normal pairs. Along with gene panel sequencing on dozens more GBM cases, the data revealed features that distinguished cases with anti-PD-1 drug response.
The team's analysis suggested that the number of non-synonymous mutations did not correlate with anti-PD-1 response, for example. The tumors they studied contained a median of 47 non-synonymous somatic mutations each.
Median survival times were stretched out to more than 14 months in 13 GBM patients who responded to the immune checkpoint treatment, compared to just over 10 months in 12 non-responders.
When they compared immunotherapy response in relation to IDH1 mutation status, the researchers saw anti-PD-1-response in four of 11 IDH1-mutated tumors, while PTEN mutations were far more common in non-responder tumors with wild-type IDH1 than in responding tumors with wild-type IDH. On the other hand, MAP kinase mutations more frequently turned up in checkpoint blockade responders, they reported.
With single-cell RNA sequencing data for some 9,000 cells from three GBM tumors, the team also saw hints that PTEN-mutated tumor cells may cluster together, influencing how well immune cells can infiltrate the tumor after the PD-1 immune checkpoint is eased with the checkpoint blockade treatment.
Likewise, the team's analysis unearthed distinct tumor evolution routes in responders and non-responders, including branched clonal evolution features in tumors from individuals who did respond and linear evolution in tumors from those who did not. In addition, T cell antigen receptor and immunoglobulin RNA sequences from seven GBM patients pointed to increased T cell diversity being correlated with anti-PD-1 non-response.
For two related studies in Nature Medicine, a team led by investigators at the University of California. Los Angeles reported on immune features and survival patterns in 35 GBM patient randomized to receive the anti-PD-1 treatment pembrolizumab in combination with neoadjuvant or adjuvant therapy, while researchers from the US and Spain tracked tumor microenvironment features in 30 GBM patients who received neoadjuvant nivolumab for a single-arm, phase 2 trial.