By John McCleery
This year, more than 1.8 million people living in the United States will receive a cancer diagnosis, and more than 600,000 deaths will be attributed to the disease.1 Although that number may seem high, it’s a tremendous decrease in the number of deaths associated with cancer over the last twenty-five years. Early detection and screening are a major factor for this decline of cancer deaths as are innovations in cancer therapies.
Liquid biopsy screening approaches are becoming a key application of early detection of cancer and are being explored in numerous cancer trials for early detection, prediction of recurrence, and evaluation of resistance to therapy. As noted in recent oncology conferences, such as ASCO GU 2020, the prospect of liquid biopsies showed favorability in renal cell carcinoma, bladder cancer, and prostate cancer. The techniques used ran the gamut and included circulating tumor cells, cell-free DNA (cfDNA), circulating tumor DNA (ctDNA) as did the applications explored including, detecting biomarkers, selecting the right patient for treatment, monitoring response, and predicting survival.2
However, development of liquid biopsy assays in cancer screening and early detection presents unique regulatory challenges as many of the new tests do not have sufficient clinical and technical validation.3 When reviewing in vitro diagnostic devices and robust liquid biopsy tests, the U.S. Food and Drug Administration (FDA) looks at the study’s design for evaluating the performance of tests. The review process includes sample size and study endpoints, as well as pre-analytical and analytical performance.4
To address and clarify these issues, the FDA hosted a public workshop in March of this year titled, “Detecting Circulating Tumor DNA for Cancer Screening.” The all-day workshop took place over three sessions and covered ctDNA cancer screening and early detection; the benefit/risk of multi-site tumor tests vs single-site tumor tests; and explained studies to evaluate pre-analytical and analytical performance.
Moderated by Soma Ghosh, PhD, Branch Chief, Division of Molecular Genetics and Pathology, FDA, the first session titled, “State of the Science: Technical Considerations and Translation of Results into Clinical Actionability ctDNA” discussed the study designs needed to establish the safety and efficacy of ctDNA liquid biopsy screenings to achieve regulatory approval. These tests are used to isolate specific deviations in circulating tumor cells and in ctDNA cells in asymptomatic subjects through blood samples obtained via liquid biopsies.
ctDNA Cancer Screening and Early Detection
According to Dr. Ghosh, a key advantage of liquid biopsies is that they are a faster, simple, noninvasive alternative screening tool to tissue biopsies. They come with low-risk complications and minimal pain. Other appealing advantages of liquid biopsy include their ability to capture the complexities of intra-tumoral heterogeneity of cancer cells, measuring the impact of the tumor microenvironment, and capturing signals from multiple different clones that a single tissue biopsy may not be able to capture.
In diagnosed patients with cancer, liquid biopsies were able to capture acquired mutations during the course of therapy. Moreover, when there isn’t enough tissue available to obtain a sample, liquid biopsy tests provide an alternative.
“Liquid biopsy tests are fast and associated with less risk and complications compared to single tissue biopsy tests,” said Dr. Ghosh.
While the advantages of liquid biopsy testing are many, as with other assay designs, they also come with their own set of limitations. “The primary limitation of liquid biopsy tests is the variability in the shedding rates of tumor-derived materials, which include proteins and other metabolites,” she said. Other limitations include the size and location of the tumor and clonal hematopoiesis of indeterminate potential, or CHIP, mutations, which may lead to false-positive results.
Ghosh indicated that false positive rates of liquid biopsy tests may be impacted by the development of a CHIP mutation. In a recent article, she said studies have shown that CHIP may lead to the detection of mutations in healthy, aging individuals, which may result in a false positive in a liquid biopsy test. “Therefore, in order to obtain reliable and accurate ctDNA-based liquid biopsy test results, one should consider sequencing paired white blood cells, followed by analyzing, and comparing with cfDNA to rule out false positives due to CHIP.”5
She encourages assay developers to keep the potential for CHIP mutation-induced false positives front-of-mind when developing their assays.
The FDA review process of in vitro diagnostic devices, including liquid biopsy assays, typically include the intended use: what does the device measure, what types of specimen are needed, who is the intended population, what are the clinical indications it is intended for, is the test performed in a lab or is it distributed, and what instrument platform does it use.4
Key pre-analytical considerations include evaluating factors that influence quality and quantity of cfDNA/ctDNA for reproducible and accurate results. Types of studies depend on the technology and the intended use claims. Study endpoints, sensitivity, specificity as well as positive and negative predictive values also need to be determined as does the nature and magnitude of the benefit/risk balance, including the level of uncertainty associated with these data.4
“Clinical validation is key,” Ghosh said, adding that “the liquid biopsy community has made several efforts to standardize testing.”
Thus far, the FDA has approved two liquid biopsy ctDNA-based companion diagnostics (CDx). Roche’s Cobas EGFR Mutation Test v.2 detects EGFR mutations associated with patients with non-small cell lung cancer and QIAGEN’s therascreen PIK3CA RGQ PCR Kit detects PIK3CA mutations associated with patients with advanced breast cancer.5
Ghosh expects the number of liquid biopsy test approvals to increase as demand grows. For both of these FDA approved liquid biopsy CDx tests, clinical evidence showed that the drug efficacy was preserved in patients identified using the plasma samples relative to the efficacy based on the tissue-based enrollment test.5
At ClinicalTrials.gov, when inputting the keywords “liquid biopsy” 91 studies were retrieved—many of which are actively recruiting, ongoing, or recently completed. Studies were either observational or prospective.
Looking forward, circulating tumor DNA and other blood-based assays may have an opportunity to offer real impact on improving the long-term prospects for cancer screening in populations, driven largely in part by other techniques that are either expensive, inconvenient, underutilized or not yet effective in their current state.
- American Cancer Society. Cancer Facts & Figures 2020. 2020.
- Bennett C. Liquid Biopsy Gains at GU 2020. OBR. 2020;11(3).
- Babayan A, Pantel K. Advances in liquid biopsy approaches for early detection and monitoring of cancer. Genome Med 2018;10(21).
- FDA.gov. Public Workshop on use of ctDNA in Cancer Screening. https://www.fda.gov/medical-devices/workshops-conferences-medical-devices/public-workshop-detecting-circulating-tumor-dna-cancer-screening-03092020-03092020. Published 2020. Updated March 9, 2020. Accessed Accessed April 15, 2020.
- Tsui D, Blumenthal GM, Philip R, et al. Development, Validation, and Regulatory Considerations for a Liquid Biopsy Test Chemical Chemistry. 2020;66(3):408-414.
Liquid Biopsies Show Clinical Utility at AACR Virtual I
By Christina Bennett, MS
|Many studies are assessing various blood tests for cancer applications, but three high-profile studies—Circulating Cell-free Genome Atlas (CCGA), DETECT-A, and TRACERx—stole the show at the 2020 AACR Virtual Annual Meeting I.
A subgroup analysis of the prospective CCGA study, which includes more than 15,000 individuals, showed that GRAIL’s cell-free DNA test could detect cancer in asymptomatic individuals with a high suspicion of cancer (Abstract CT021). However, study presenter David Thiel, MD, Chair, Mayo Clinic Florida Department of Urology, cautioned, “This test is intended to be used alongside guideline-recommended screening and not replace it.”
Among 303 patients with a high clinical suspicion of cancer, the blood test showed a 100% specificity in both the training (49 patients) and validation (15 patients) sets, meaning all patients with a positive result indeed had cancer. The specificity was comparable to the more than 99% specificity seen in the preplanned CCGA sub-study cohort 2.
Although the specificity was high, the sensitivity was low, indicating that many patients with a negative result actually had cancer. The sensitivity was only 40.2% for the training set and 46.7% for the validation set, which rose to 50.4% and 59.3%, respectively, when kidney cancer was excluded. The tissue of origin prediction accuracy was 85.5% for the training set and 97.1% for the validation set.
Dr. Thiel said the sensitivity and tissue of origin prediction accuracy for the subgroup analysis were “comparable” to that seen in the CCGA sub-study cohort 2.
Two studies—STRIVE and SUMMIT—are ongoing to validate the performance of the test. Also ongoing, the PATHFINDER study is monitoring the return of test results to physicians and patients and the subsequent diagnostic journey after a positive test result, Dr. Thiel said.
An interim analysis of the prospective DETECT-A study showed that a multi-cancer, multi-analyte blood test that uses DNA and protein markers could detect cancer in women with no history of cancer (n=10,000), especially when combined with standard-of-care imaging screening (Abstract CT022).
“People were not discouraged from undergoing standard-of-care screening based on surveys conducted in individuals within the study population,” said study presenter Nick Papadopoulos, PhD, Johns Hopkins School of Medicine.
Specifically, the positive predictive value of the blood test alone was 19%, and this value increased to 41% when the blood test was coupled with PET-CT screening. Also, 65% of cancers detected with the blood test were localized or regional, “which have a higher chance of successful treatment with intent to cure,” he said.
Only 1% of individuals with positive results that did not have cancer later received an unnecessary non-invasive or minimally invasive procedure. In addition, three individuals had unnecessary surgeries.
The results of the TRACERx study highlighted the potential of using circulating tumor DNA (ctDNA) as a biomarker for minimal residual disease (MRD) after surgery in early-stage non-small cell lung cancer (Abstract CT023).
Among 53 patients who had disease relapse, the study showed that being a ctDNA shedder (ie, having detectable ctDNA before surgery) was associated with a longer MRD lead-time and shorter disease-free survival after surgery compared with being a non-ctDNA shedder (ie, having no detectable ctDNA before surgery).
The study also showed that 10 patients with an MRD-positive result had surveillance scans that indicated no evidence of relapse, yet 9 eventually had relapse. In addition, 16 patients with an MRD-positive result had surveillance scans that indicated relapse, inflammation, or a non-specific finding and 15 eventually had relapse.
“MRD surveillance can lead to detection of relapse in advanced standard-of-care imaging,” concluded study presenter Chris Abbosh, MD, UCL Cancer Institute.