Blood Test Appears to Enable Immunotherapy Selection, Monitoring

Expert Critique

FROM THE ASCO Reading Room

I. Alex Bowman, MD Assistant Professor, Division of Hematology/Oncology University of Texas Southwestern Medical Center

Non-invasive testing of biomarkers to guide the selection of anti-cancer therapy shows tremendous promise. Already companion diagnostic tests utilizing peripheral blood are approved for the detection of targetable kinase alterations or mutations in lung and ovarian cancer. An ideal test is both noninvasive and comprehensive, testing for all targetable mutations as well as markers of response to immunotherapy such as PD-L1, microsatellite instability, and tumor mutational burden (TMB). This can be accomplished through a variety of assays, often utilizing circulating tumor cells (CTCs) and circulating tumor (ct)DNA. Imaging tests utilizing biomarker-specific tracers are also under investigation. Concerns persist about the accuracy of these testing modalities compared with tissue biopsy. Tissue biopsy, however, is not without limitations: It is invasive, typically yields limited tissue for analysis, and is able to evaluate only a small area of disease at a single point in time. This may fail to reproduce the complex and heterogeneous tumor microenvironment, which can change over time.

Presented here are several recent related abstracts. Gandara et al present an exploratory analysis of data from the OAK and POPLAR trials (which studied the anti-PD-L1 checkpoint inhibitor atezolizumab in the second-line treatment of lung cancer), assessing TMB in ctDNA from plasma samples of patients treated on those trials. This demonstrates that this appears to be a viable avenue for testing, although it is limited by the lack of comparison to TMB in tissue samples from these patients.

Hsieh et al and Gupta et al report on assays from CellMax Life testing for both molecular and immunotherapy biomarkers with CTCs and ctDNA in non-small cell lung cancer patients, and a 73-gene panel testing ctDNA in colorectal cancer patients versus healthy controls, respectively. Both represent a promising field of inquiry, and further study is planned.

Full Critique

A new blood test appears to detect expression of the programmed death-ligand 1 (PD-L1) protein in circulating tumor cells (CTCs) with high sensitivity, enabling repeat testing to monitor a patient's response to immunotherapy throughout the course of treatment.

The test is able to capture CTCs for PD-L1 testing in blood in about 90% of patients, the researchers said. About half of the CTC-tested patients were PD-L1 positive, which is consistent with previously reported PD-L1-positivity rates in clinical trials.

"PD-L1 testing by immunohistochemistry is a standard of care as a predictive biomarker for checkpoint immunotherapy, and blood-based testing offers potential advantages," said David Gandara, MD , director of the Thoracic Oncology Program at the University of California Davis Comprehensive Cancer Center. "First, about 25% of patients have inadequate tumor tissue upfront to test for both molecular biomarkers and PD-L1. Second, PD-L1 is a dynamic biomarker, which can change over time, so using tumor from a prior biopsy may be misleading. PD-L1 can also be unevenly distributed in tissue, leading to a false negative and denying patients the opportunity to receive immunotherapy."

At the European Society for Medical Oncology 2017 congress, Gandara reported on a novel blood-based assay to measure tumor mutational burden, which appears to predict the clinical efficacy of the anti-PD-L1 agent atezolizumab in patients with advanced non-small cell lung cancer (NSCLC). The retrospective analysis of two large studies using atezolizumab to analyze blood for tumor mutational burden demonstrated for the first time that a liquid biopsy can measure tumor mutational burden with a high degree of precision and accuracy from single-agent checkpoint inhibition, and may also provide a non-invasive biomarker to identify patients who derive clinical benefit from specific immunotherapies.

This year, at the American Association for Cancer Research (AACR) annual meeting, two abstracts reported data on a novel blood test for detecting both PD-L1 and microsatellite insufficiency (MSI), combining CTC analysis for PD-L1 and circulating tumor DNA (ctDNA) analysis for MSI from the same blood sample.

Both PD-L1 expression and MSI are biomarkers approved to select patients for immunotherapy. Patients with high PD-L1 expression in more than 50% of tissue cells expressing PD-L1 are eligible for first-line pembrolizumab immunotherapy in NSCLC. MSI-high or -deficient mismatch repair patients with any solid tumor are also eligible for pembrolizumab therapy upon recurrence.

The standard-of-care testing for both tests is tissue-based, Gandara noted. For PD-L1, that means testing by immunohistochemistry and for MSI testing by polymerase chain reaction or immunohistochemistry expression. However, tissue testing can result in false negatives due to tumor heterogeneity, denying patients the opportunity to receive immunotherapy.

Most patients do not respond to immunotherapy despite having high PD-L1 expression in tissue, and about 20% of responders are not high expressors of PD-L1 in tissue, he explained.

Blood-based testing offers several advantages over tissue testing -- for example, providing a more global view of the tumor, overcoming tumor heterogeneity. There is also a need for biomarkers that predict response to immunotherapy during the early stages of treatment to identify non-responders to spare them the side effects of treatment, and to lessen the costs of care. Tissue tests require invasive repeat biopsies and may not be practical for monitoring treatment effectiveness.

The combination of DNA and CTC analysis makes the new liquid biopsy (developed by CellMax Life) the first blood test that includes all mutation classes recommended in the guidelines of the National Comprehensive Cancer Network (NCCN) for immunotherapy and targeted therapy selection in patients with advanced solid tumors. The first study, reported at AACR18, showed that this non-invasive assay could accurately detect all guideline-recommended markers for NSCLC treatment selection and thereby enable optimal treatment selection. Up to half of advanced lung cancer patients have insufficient or unavailable tissue for comprehensive genomic profiling.

In the second study, a 73-gene panel covering NCCN recommended actionable variants for solid tumors in 60 reference ctDNA samples with known variants was analyzed to establish the limit of detection, sensitivity, specificity, accuracy, and reproducibility of the platform. For clinical validation, 36 patients with metastatic colorectal cancer (mCRC) and 34 healthy controls from the Chang Gung Memorial Hospital and 227 patients diagnosed with solid tumors from Taiwan were tested.

The platform successfully detected variants at low mutant allele fraction, and somatic variants were identified in 35 of 36 mCRC patient samples (97.2%). No false positives were observed within the targeted region for all 34 healthy controls tested. In paired samples, the platform showed 89.7% concordance with tissue biopsy. The observed gene mutation profiles from ctDNA were consistent with published tissue biopsy data. Actionable variants were detected in 170 of the 227 (75%) solid tumor patients.

Gandara explained that liquid biopsy tests will make it possible for oncologists to choose between three treatment modalities for their patients using a single test -- immunotherapy, targeted therapy, or chemotherapy. "The advantage of a blood-based biomarker is that the oncologist can do repeated tests for monitoring or test before and after immunotherapy to see changes."

Looking into the future, he said, "blood-based testing provides a means for repeat testing and timely monitoring, which would not be possible otherwise. A PD-L1 blood test can overcome these issues and would be an attractive alternative to tissue testing."