Blood Test for Many Cancers: More False Than True Positives

Dr Deborah Schrag

PARIS — New results from a large prospective trial give a better idea of how a blood test that can detect multiple cancers performs in a “real-life” setting.

“As this technology develops, people must continue with their standard cancer screening, but this is a glimpse of what the future may hold,” commented study investigator Deborah Schrag, MD, MPH, chair, Department of Medicine, Memorial Sloan Kettering Cancer Center, New York City.

For the PATHFINDER study, the Galleri blood test (developed by Grail) was used in 6621 healthy individuals aged over 50, with or without additional cancer risk factors (such as history of smoking or genetic risk).

It found a positive cancer signal in 92 individuals (1.4%). 

None of the individuals who tested positive were known to have cancer at the time of testing. Subsequent work-up, which could include scans and/or biopsy, found cancer in 38% of those with a positive test.

“When the test was positive, the work-ups were typically done in less than 3 months,” Schrag commented, adding that “the blood test typically predicted the origin of the cancer.”

Schrag presented the findings here at the European Society for Medical Oncology (ESMO) Annual Meeting 2022.

Approached for comment, Anthony J. Olszanski, MD, RPh, vice chair of research at the Fox Chase Cancer Center, Philadelphia, Pennsylvania, noted that the use of a blood test to “find” cancer has long been on the minds of patients. “It is not uncommon to hear oncology patients ask: ‘Why didn’t my doctor find my cancer earlier, on blood tests?'”

As this study suggests, finding a malignancy before it becomes apparent on imaging or due to symptoms is one step closer to becoming a reality. “But although this is an important study, it must be noted that only about 40% of patients with a positive test result were actually found to have cancer,” Olszanski said. “Conversely, about 60% of patients with a positive test result likely suffered from a considerable amount of anxiety that may persist even after further testing did not reveal a malignancy.”

Another important issue is that such testing may incur substantial healthcare cost. “Less than two participants per 100 had a positive test result, and those patients underwent further testing to interrogate the result,” he added. “It also remains unclear if detecting cancer early will lead to better outcomes.”

Whether or not the test will be cost-effective remains unknown, as Schrag emphasized they do not have a formal cost analysis at this time. “This technology is not ready for population-wide screening, but as the technology improves, costs will go down,” she said.

Schrag also added that this is a new concept and the trial shows it is feasible to detect cancer using a blood test. “It was not designed to determine if the test can decrease cancer mortality, which is obviously the purpose of screening, but it’s premature for that,” she said.

Details of the Results

The Galleri test uses cell-free DNA (cfDNA) and machine learning to detect a common cancer signal across more than 50 cancer types as well as to predict cancer signal origin (CSO).

Overall, the test detected a cancer signal in 1.4% (n = 92) of participants with analyzable samples.

A total of 90 participants underwent diagnostic testing (33 true positives and 57 false positives). Of the true positives, 81.8% underwent more than one invasive diagnostic test, as did 29.8% of false positives.

Specificity was 99.1%, positive predictive value (PPV) was approximately 40%, and 73% of those who were true positives had diagnostic resolution in less than 3 months.

Of the cancers that were diagnosed, 19 were solid tumors and 17 were hematologic cancers; seven were diagnosed in a person with a history of cancer, 26 were cancer types without standard screening, and 14 were diagnosed at an early stage.

“What is exciting about this new paradigm is that many of these were cancers for which we don’t have standard screening,” said Schrag.

Schrag noted that given the immense interest in this study, the manufacturer is working towards refining the assay and improving the test. A reanalysis was conducted on all specimens using a refined version of the test.

“Importantly, the new analysis identified fewer patients with having positive signals, from 1.4% to 0.9%,” she said. “Specificity improved to 99.5% as did PPV — from 38% to 43.1% — and more people need to be screened to find a cancer — up to 263 from 189.”

False Positives Concerning

Previous, and very similar, results from the PATHFINDER trial were presented last year at the annual meeting of the American Society of Clinical Oncology and reported by Medscape Medical News at the time.

Max Diehn, MD, PhD, associate professor of radiation oncology at Stanford University, California, was an invited discussant for the study.

He pointed out that there were more false positives than true positives and noted that “there were a significant number of invasive procedures in false positives, which could cause harm to these patients who don’t have cancer.”

Diehn also explained that most true positives were for lymphoid malignancies, not solid tumors, and it is not known whether early detection of lymphoid malignancy has clinical utility. 

The Galleri test is already available in the United States and is being offered by a number of US health networks. However, it is not approved by the US Food and Drug Administration and is not covered by medical insurance, so individuals have to pay around $950 for it out of pocket. 

Although some experts are excited by its potential, describing it as a “game-changer,” others are concerned that there are no clinical pathways in place yet to deal with the results of such a blood test, and say it is not ready for primetime, as previously reported by Medscape Medical News

The study was funded by Grail, a subsidiary of Illumina. Shrag has reported relationships with Grail, the Journal of the American Medical Association, and Pfizer. Several co-authors also have disclosed relationships with industry. Olszanski has reported participating in advisory boards for BMS, Merck, and Instil Bio, and running trials for them.

ESMO 2022. Abstract 9030. Presented September 11, 2022.

Roxanne Nelson is a registered nurse and an award-winning medical writer who has written for many major news outlets and is a regular contributor to Medscape.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube.

Source: Read Full Article