Blog June 14, 2017

Towards a Universal Immuno-oncology Assay Predictive of Drug Response

We have written frequently about the growing importance of biomarkers and how diagnostics will increasingly play a dominant role in the way that therapeutics are prescribed and reimbursed. Immuno-oncology (IO) treatments in particular are expected to benefit from the availability of truly predictive biomarkers, based on the dramatic treatment gains they have achieved for a limited number of patients and the high price of such drugs. Clearly the ability to determine who is most likely to respond to treatment with such therapies will play a key role in demonstrating their value to third party payers.

 

Recent developments suggest that the first universal IO clinical assays for predicting who should receive PD-1/PD-L1 inhibitors may be on the horizon.

 

In February, Merck and NanoString announced an expanded collaboration, based on earlier joint research, aimed at the development of a diagnostic assay to predict response to Keytruda (pembrolizumab), Merck’s PD-1 inhibitor. NanoString will be responsible for seeking regulatory approval and commercialization of the assay, which looks at a specific, complex gene expression signature predictive of PD-1 response regardless of tumor type. The new test will run on the NanoString nCounter Dx Analysis System, which uses a digital color-coded barcode technology that enables the multiplexed measurement of hundreds of targets in a single reaction from as little as 300ng of DNA.

 

But Merck and NanoString aren’t the only group moving forward to develop a better, more universal way of determining who should be treated with an IO drug.  Foundation Medicine has a broad strategic collaboration with Roche (Roche acquired a majority interest in Foundation in 2015), and part of that collaboration is aimed at identifying biomarkers for Roche’s PD-L1, Tecentriq (atezolizumab).  Just prior to the American Academy for Cancer Research meeting at the start of April, Bristol-Myers Squibb and Foundation Medicine announced their own collaboration to leverage Foundation Medicine’s Molecular Information Platform to identify predictive biomarkers across multiple tumor types and immunotherapy agents.

 

Foundation’s approach is based on providing a validated, quantitative read-out of Tumor Mutational Burden (TMB) and other genomic alterations.  BMS and Foundation plan to examine biomarkers, including TMB, from patients receiving any of more than a dozen investigational or approved I/O therapies in 35 tumor types. Those patients who are identified as having a high mutational load may then be assigned to receive Opdivo. While published clinical data has suggested a correlation between high TMB and response to IO in bladder cancer, lung cancer and melanoma patients, this study could provide important evidence that high TMB can actually predict IO response.

 

Foundation Medicine’s FoundationOne test is currently under parallel review by the U.S. Food and Drug Administration and the Centers for Medicare and Medicaid Services as part of FDA’s Expedited Access Pathway for Break-through Devices. If approved, FoundationOne could be the first FDA-approved genomic profiling assay to incorporate multiple companion diagnostics across diverse range of solid tumors. Obtaining a Medicare National Coverage Determination (NCD) from CMSS at the same time as FDA approval would allow FoundationOne to be offered as a covered benefit under Medicare, and to avoid the significant time lag between FDA approval and uncertainty regarding Medicare reimbursement that has too often plagued medical devices and diagnostics. A decision by the two agencies is expected in the second half of this year.

 

As drug developers pursue various biomarker strategies for their IO therapies (PD-L1, MMR/MSI, all-comers), it is becoming increasingly likely that there may be many factors which determine whether a patient will respond to IO therapy.  It is hoped that a universal test will eliminate the need for multiple disparate biomarker tests, and provide a simple yet comprehensive way to select the appropriate therapy regimen out of the increasing number of options and combinations.