Blood-based Test Shows Promising Accuracy in Detecting Cancer-Signal Across Multiple Tumor Types

Multicancer early detection (MCED) tests yielded high specificity and accuracy in predicting cancer signal origin (CSO) and identified cancer signals in multiple tumor types, according to data from The Circulating Cell-Free Genome Atlas Study (CCGA; NCT02889978) published in Annals of Oncology.¹ These results provide rationale to utilize these tests as a compliment to existing single-cancer screening tests.

Specifically, data from the study showed that cancer signals were detected across more than 50 tumor types and that the MCED tests demonstrated a specificity for cancer signal detection of 99.5% (95% CI, 99.0%-99.8%). The overall sensitivity for cancer signal detection was 51.5% (n = 1453/2823; 95% CI, 49.6%-53.3%) and the sensitivity of the tests increased with disease stage; sensitivity for stage I disease was 16.8% (95% CI, 14.5%-19.5%), 40.4% (95% CI, 36.8%-44.1%) for stage II, 77.0% (95% CI, 73.4%-80.3%) for stage III, and 90.1% (95% CI, 87.5%-92.2%) for stage IV.

Overall, the accuracy of CSO prediction in true positives was 88.7% (95% CI, 87.0%-90.2%).

Investigators also reported that the MCED test demonstrated a cancer signal detection sensitivity of 40.7% (95% CI, 38.7%-42.9%) in an analysis was conducted in a cohort of patients with 12 prespecified cancers which account for approximately 66% of annual cancer deaths annually in the United States. Specifically, this cohort included patients with either anal, esophageal, lung, pancreatic, bladder, head and neck, stomach, ovarian, liver/bile duct, and colorectal cancers, as well as lymphoma and plasma cell neoplasms.

Further, an analysis of patients with stage I-III disease showed that the sensitivity of cancer signal detection with the MCED tests was 67.6% (95% CI, 64.4%-70.6%).

“Finding cancer early, when treatment is more likely to be successful, is one of the most significant opportunities we have to reduce the burden of cancer,” Eric Klein, MD, the chairman of the Glickman Urological and Kidney Institute at Cleveland Clinic, said in a press release from ESMO.² “These data suggest that, if used alongside existing screening tests, the multicancer detection test could have a profound impact on how cancer is detected and, ultimately, on public health.”

Widespread detection of cancer prior to stage IV disease could reduce cancer-related deaths by more than 15% within 5 years. Five single-cancer screening tests available for use in the United States for detection of breast, colorectal, cervical, lung, and prostate cancers, which account for 42% of annual cancer incidence rates. The use of cancer screening tests have contributed to the reduction of cancer-specific mortality, however, they are associated with high rates of false positives, over-diagnosis, and over treatment.

MCED tests represent a potential new approach to detecting signals for multiple cancer types by utilizing cell-free DNA (cfDNA) or other circulating analytes in the blood shed by tumors. As such, CCGA study investigators sought to develop and utilize a MCED test to detect cancer signals across multiple cancer types, as well as predict CSO with just a single blood draw.1

From August 2016 and February 2019, this prospective, multi-center, case-control, observational study with longitudinal follow-up, enrolled 15,254 patients from 142 sites in North America. The study was then divided into 3 prespecified sub-studies that included a discovery cohort, a cohort that examined training and validation with the selected and updated assay and classifiers, and clinical validation cohort, reported here.

Patients eligible to enroll on the study were those 20 years and older, who received a diagnosis of cancer and/or were scheduled to undergo biopsy or surgical resection for a confirmed or highly suspected malignancy. Those who had previously undergone chemotherapy, radiotherapy, or definitive local therapy were not eligible to enroll.

The primary end point of the sub-study was to evaluate MCED test performance for cancer signal detection, CSO prediction, and the rates of both combined. Key secondary end points included test performance by age group, test performance for cancer signal detection by method of cancer diagnosis, and test performance for cancer signal detection in a prespecified group of 12 cancer classes.

In total, 5309 patients were enrolled on the third sub-study, 4077 of whom were included in the analysis. Of the evaluable patients, 2823 had a confirmed cancer diagnosis, and 1254 did not have a cancer diagnosis. The most common reasons for exclusion from the analysis were incomplete 1-year follow-up for noncancer participants (n = 324), presence of nonmalignant conditions at enrollment (n = 283), and unconfirmed cancer or treatment status at blood draw (n = 171).

Patients had a mean age of 60.6 years, and a majority were female (55.4%) and White (81.2%). Among patients with a confirmed cancer diagnosis, most presented with stage I or stage II disease (54.9%).

“These data add to a growing body of literature that supports the use of next-generation sequencing for the detection of cell-free DNA in blood samples as a tool for earlier detection of common cancers that account for a significant number of deaths and other health problems worldwide,” Klein said. “In addition, a screening test that requires only a simple blood draw could provide an option for communities that have poor access to medical facilities.”

References

1. Klein E, Richards D, Cohn A, et al. Clinical validation of a targeted methylation-based multi-cancer early detection test using an independent validation set. Ann Oncol. Published online ahead of print June 23, 2021. doi:10.1016/j.annonc.2021.05.806
2. Blood test for early detection of cancer: final study results support screening use. News release. European Society for Medical Oncology. June 25, 2021. Accessed June 25, 2021. https://bit.ly/35W4dNH