No Clear Benefit Found in Multi-Cancer Detection Tests

1. In this systematic review, no studies were found that assessed the benefits of multi-cancer detection tests for cancer screening.

2. Evidence on accuracy and harms of multi-cancer detection tests was highly variable and of insufficient strength.

Evidence Rating Level: 1

Study Rundown: Considering that few effective screening tests currently exist for many cancers, especially those typically found at advanced stages, blood-based multi-cancer detection (MCD) tests may offer an innovative approach to cancer screening. However, no MCD tests have been approved by the US Food and Drug Administration. This study aimed to evaluate evidence describing the benefits, accuracy, and harms of MCD tests for screening. A systematic review of twenty studies reporting on nineteen MCD tests found that none of the studies reported the benefits of MCD tests on cancer detection, mortality, or quality of life. Accuracy of MCD tests differed significantly based on analysis approach used, the number and types of cancers that each test reported, participant subgroups, and study designs. Pre-diagnostic performance studies largely reported lower sensitivity and area under the curve (AUC) compared with diagnostic performance studies. Harms of screening were assessed in one cohort study, but the results had a wide confidence interval that included the null. Overall, the strength of evidence for accuracy, harms of screening, and harms of diagnostic evaluation was considered insufficient. Risk of bias (ROB) was deemed to be either high or unclear for all studies. The generalizability of this review is limited by an absence of studies evaluating the direct benefits of screening, few studies evaluating harms of screening, and the heterogeneity of analytical methods. These findings suggests that further studies are required to evaluate the benefits, accuracy, and harms of MCD tests for cancer screening.

Click to read this study in AIM

Relevant Reading: Predictive Performance of Cell-Free Nucleic Acid-Based Multi-Cancer Early Detection Tests: A Systematic Review

In-Depth [systematic review and meta-analysis]: This systematic review aimed to synthesize evidence on the benefits, accuracy, and harms of MCD tests for cancer screening. Benefit and harm outcomes included cancer-specific and all-cause mortality, cancer detection and stage at diagnosis, quality of life, and functional status. Screening harm outcomes included psychosocial distress, adverse events, radiation exposure, reduced standard-of-care cancer screening, and out-of-pocket patient costs. Studies on test accuracy were included if they were case-control or cohort studies that assessed sensitivity, specificity, or AUC in a population that had not been involved in test development. This review included 20 unique studies, with a combined population of 109,177 participants (range, 102 to 41,516 per study), that reported on 19 unique MCD tests. Among these studies, 7 assessed 7 unique MCD tests using a prospective cohort design, a retrospective cohort design, or a nested case-control design; ROB was considered unclear in 5 of those studies and high in the other 2 studies. The remaining 13 studies used high ROB case-control designs to report 12 unique MCD tests, with bias largely originating from the use of a case-control design. No studies reported the benefit of MCD tests on cancer detection, mortality, or quality of life. The 2 tests commercially available in the United States reported sensitivities of 0.31 (95% confidence interval [CI], 0.21 to 0.43) and 0.57 (95% CI, 0.51 to 0.63) and positive predictive values of 43% and 4%, respectively. Strength of evidence for accuracy outcomes was graded as insufficient. Harms of screening were assessed in 1 controlled cohort study, which reported changes in adherence to standard-of-care screening. That study showed that standard-of-care screening increased in both MCD test and control groups and was higher in the MCD test group, although the CI was wide and crossed the null (odds ratio, 1.58 [95% CI, 0.47 to 5.31]). Strength of evidence for harms of screening was graded as insufficient. That study also reported 2 outcomes relating to harms of diagnostic evaluation, but strength of evidence for harms of diagnostic evaluation was graded as insufficient. Overall, this study suggests that no evidence currently exists for benefits of MCD tests, while evidence on their accuracy and harms is also of insufficient strength.