Is Low Tumor Mutational Burden Predictive of Response to Oncolytic Polio Virus Therapy in Recurrent Glioblastoma?

Duke University, Durham NC

Investigators

Linked publications & trials

Abstract

Recurrent WHO grade IV malignant glioma (rGBM) is a uniformly lethal disease refractory to all currently approved therapies. We have developed a recombinant poliovirus, PVSRIPO, that has shown promise in a completed phase I clinical trial for rGBM; with a 21% survival rate 3 years after therapy. However, only a subset of patients experienced such responses, indicating a need for biomarkers that predict therapy success and elucidation of mechanisms of therapy resistance. More broadly, the understanding of the immunological landscape of rGBM and how it may be targeted by cancer immunotherapy strategies remains crude. Tumor mutation burden (TMB) is an emerging biomarker for immune checkpoint immunotherapy (ICI), with higher TMB-carrying tumors being more responsive. Rather than engaging distinct receptors on adaptive immune cells like ICI, PVSRIPO engages innate, antiviral signaling in tumors to prime adaptive anti-tumor immunity and reverse immunosuppression. Whether TMB may hold value as a biomarker for such innate simulating cancer immunotherapy strategies has not been explored. Remarkably, in contrast to observations for ICI, patients responding to PVSRIPO with long-term survival had lower tumor mutation burden. Stratifying patients from the TCGA GBM cohort by TMB did not reveal such a survival benefit for patients with lower TMB; suggesting that low TMB may be a predictive biomarker for response to PVSRIPO rather than a general prognostic marker for GBM. Additionally, prior alkylating chemotherapy exposure [Temozolomide (Temo)] and the associated Temo-mutation signature percentage were also lower in patients that responded to PVSRIPO therapy. Suggesting a link between TMB, Temo-induced mutations, and the immunological status of tumors: cytolytic score, an RNA based marker for immune effector presence and activity, was higher in patients with low TMB. Collectively, these findings suggest that TMB may serve as a predictive biomarker for response to PVSRIPO therapy; and that prior Temo therapy and the associated genetic mutations may preclude PVSRIPO mediated efficacy in rGBM patients. Therefore, we hypothesize that low levels of TMB is an indicator of an immunological state in rGBM required for response to oncolytic poliovirus therapy. Towards this end we will (i) Determine the utility of low TMB as a predictive biomarker for survival benefit from PVSRIPO therapy and (ii) Explore the molecular and cellular link(s) between TMB, Temo related mutations and the tumor immune landscape. These studies will substantiate and advance our novel observation that low TMB is a predictive biomarker for response to PVSRIPO. They will inform clinical administration of PVSRIPO, begin to reveal both tumor intrinsic and immune mechanisms mediating primary resistance to PVSRIPO, and may inform other innate- stimulating caner immunotherapy modalities.