Liquid Biopsy Assay Proves Ability to Predict Response to Immune Checkpoint Inhibitor

John Simmons, PhD

John Simmons, PhD

In May 2017, the FDA granted an approval of immune checkpoint inhibitor pembrolizumab (Keytruda) for thetreatment of patients with unresectable or metastatic tumors that tested high for microsatellite instability high (MSI-H) or mismatch repair deficiency (dMMR), which can be detected with tissue biopsy. However, a liquid biopsy test has been developed by Personal Genome Diagnostics to detect MSI, as well as tumor mutational burden (TMB) in a noninvasive procedure.

In a recent study, 61 patients with metastatic cancer were analyzed, including 163 plasma samples. A 98 kb pan-cancer 58-gene panel was used to identify MSI frameshift alleles in cell-free DNA (cfDNA). This liquid biopsy assay led to a specificity of greater than 99% and a sensitivity of 78% in the study subjects.

Of the 23 patients in the study that were identified as MSI-H by tissue analysis, 18 were also identified as MSI from plasma, and the cases identified as  microsatellite stable (MSS) in tissue patients were all correctly classified by the ctDNA test. Additionally, researchers analyzed the test’s ability to detect TMB. The liquid biopsy assay also led to a specificity of greater than 99% in its detection of TMB with a sensitivity rate of 67%.

“Assessment of the efficacy of response to immune checkpoint inhibition has proven challenging utilizing imaging-based methodologies…,” the study authors wrote in their report published inClinical Cancer Research. “Therefore, cfDNA-based approaches for comprehensive genome-profiling may be useful for the rapid determination of patients who ultimately may benefit from immune checkpoint blockade.”

For patients with MSI and high levels of TMB, researchers found that the assay was able to predict the progression-free survival prior to treatment with an immune checkpoint inhibitor. If a liquid biopsy assay could predict MSI and TMB status in patients with cancer, this may provide physicians a less invasive method for predicting response to pembrolizumab.

In an interview withTargeted Oncology, John Simmons, PhD, vice president of Translational Medicine at Personal Genome Diagnostics, discussed the potential role of liquid biopsy assays in predicting response to immune checkpoint inhibitor therapy for patients with cancer. He also highlighted other liquid biopsy assays that are in development for detecting MSI.

TARGETED ONCOLOGY: What was the rationale for evaluating MSI detection with a liquid biopsy assay?

Simmons:MSI has appeared over the last few years as a very promising biomarker for prediction of response to the immune checkpoint inhibitors. Work that had begun in 2013 and 2014 found that you could actually stratify patients by their MSI status across tumor types. Regardless of tumor types, patients who had MSI would have a higher probability of response to treatment with the immune checkpoint inhibitors like pembrolizumab and nivolumab (Opdivo). This work led to the approval a few years ago of pembrolizumab, and for the first time, that was with a pan-cancer tumor-agnostic biomarker of MSI or dMMR.

Traditionally, testing for MSI has been through an analysis of tumor tissue, often compared to DNA from normal cells from that individual to learn if they were MSI-H. Over the years, we have been quite interested in MSI-H and its ability to predict immune therapy response, so we developed 1 of the first next-generation sequencing (NGS)—based tissue assays that we ran in our lab in Baltimore, Maryland. A few years later, we developed a plasma-based lab test to detect MSI-H from the circulating tumor DNA (ctDNA) of patients with cancer.

The FDA has now approved the indication for pembrolizumab for patients who have MSI-H tumors, and we know that in many cases, tissue may not be readily available or performing a biopsy may not be clinically appropriate. We wanted to continue our work to develop noninvasive tests that would allow those patients to have an opportunity for screening for this very important biomarker. That was the rationale for this study, to determine if detection of MSI-H status in ctDNA was possible and to demonstrate that in the clinical setting, it was predictive of immunotherapy response similar to how MSI-H found in tissue was predictive.

TARGETED ONCOLOGY: How was this study designed?

Simmons: In this study, we had 29 patients who enrolled in a clinical trial to evaluate monotherapy treatment with pembrolizumab. Patients were evaluated and enrolled based on their tissue status. There were 23 MSI-H patients and 6 MSS patients who had plasma that was evaluated for this study.

TARGETED ONCOLOGY: What were the findings in these patients?

Simmons:For our findings, we found that 18 of the 23 MSI-H patients were detected by plasma across tumor types, so that is an agreement of about 78%. We further evaluated correlations with the level of TMB by looking at the single nucleotide variant output from this assay, and we found that we could identify high levels of TMB, which is known to be associated with MSI. That was helpful.

Quite interestingly, we found that there is good promise for identifying patients who, while MSI-H, were actually responding to the immune checkpoint inhibitor monotherapy by looking at the change in ctDNA levels. Patients who had ctDNA levels dropping after treatment had begun were more likely to experience durable benefit from that therapy. It’s a promising area of continued investigation for using ctDNA for monitoring, and hopefully 1 day, this will inform treatment decision.

TARGETED ONCOLOGY: What should oncologists take away from this research?

Simmons:The takeaway, from my perspective, is MSI-H is a very important pan-cancer biomarker linked to drugs that have very good benefit in that patient population. Oncologists should note that there are tests available and in development that can allow for detection of MSI and in cases where tissue is not available, through the use of these noninvasive ctDNA assays.

TARGETED ONCOLOGY: Is there anything unique to note about this liquid biopsy assay in particular?

Simmons:The test in the study is a comprehensive profiling assay that we detect all the clinically relevant biomarkers in solid tumors from plasma. It is a test that we developed in our lab and was the first ctDNA test to determine MSI status.

TARGETED ONCOLOGY: Do you have any next steps planned for this research?

Simmons:We do not [have next steps] for this exact test that was used in this study. It was a laboratory-developed test in our lab in Baltimore. However, a few years ago, we began the process of trying to democratize our NGS tests through the development of in vitro diagnostic kits that can be placed in labs globally. We are developing an NGS-based kitted liquid biopsy solution called elio plasma resolve. That test has been CE-marked earlier this year, and last year we received an FDA breakthrough designation for the MSI function of that test.

Our work has really been focused on further developing these lab tests into kitted solutions that can be placed in labs globally to drive NGS testing and personalized medicine to patients around the globe.

TARGETED ONCOLOGY: What is important to note about the elio plasma resolve test?

Simmons:In our elio plasma resolve product, the great thing about that is we took that core knowledge and development from our lab in Baltimore and developed that as a kitted solution with automated bioinformatics. When approved, molecular labs globally can bring that test  into their lab and have the same validated high-quality results. We are moving away from the lab-developed test paradigm.

TARGETED ONCOLOGY: Looking ahead, how do you see the role of liquid biopsy impacting treatment decisions in patients with cancer?

Simmons:We will continue to see more and more clinical utility and clinical evidence being demonstrated. I believe we will see FDA approvals for ctDNA-based liquid biopsy assays. As I mentioned, this year we have CE-marked our liquid biopsy ctDNA test, and I believe we will see these tests used more broadly for late-stage cancer profiling and to increase the amount of late-stage patients who can have comprehensive genomic profiling performed and used to guide their therapy. I also think we will really begin to see a lot more evidence and utility for using these tests for longitudinal monitoring applications and ultimately driving better treatment outcomes by using  information from ctDNA tests to guide the patient care journey.

Reference:

Georgiadis A, Durham JN, Keefer LA, et al. Noninvasive Detection of Microsatellite Instability and High Tumor Mutation Burden in Cancer Patients Treated with PD-1 Blockade. [published online September 10, 2019].Clin Cancer Res.doi: 10.1158/1078-0432.CCR-19-1372.