Screening for Multiple Cancers, Including Urothelial Cancer, Using a Cell-Free DNA-Based Blood Test

A major recent focus in cancer research has been the development of a multi-cancer detection test that can facilitate early diagnosis and enhance survival across different cancer types. Accordingly, the Circulating Cell-free Genome Atlas (CCGA) is a large study established to develop and validate blood-based cell-free DNA (cfDNA) tests. In a recent study, Bryce et al. reported the performance of the test in a large cohort of patients with cancer-related symptoms.

The cohort consisted of 2,036 participants with cancer and 1,472 participants without cancer. Among the former group, 12.1% exhibited clinical signs and symptoms of cancer (CPC) and were labeled as the ‘high suspicion of cancer’ (HSC) subgroup. Those who did not have cancer were divided into three subcategories: those who were enrolled as HSC but were then confirmed to be cancer-free, those who were being seen for any medical condition except cancer (nonmalignant subgroup), and those who were being seen for hematology-related concerns (heme subgroup). The specificity of the test was high across the noncancer groups: 98.1% (95% CI, 90.2-99.9) for HSC individuals, 99.6% (95% CI, 97.5-100.0) for the nonmalignant subgroup, and 99.4% (95% CI, 97.7-99.8) for the heme subgroup. The aggregated specificity rate for all noncancer participants was 99.5% (95% CI, 98.4-99.8).

The sensitivity of cancer detection across cancer types and stages was 64.3% (95% CI, 62.2-66.4) in the CPC subgroup and 47.4% (95% CI, 41.2-53.6) in the CPC and HSC subgroup. In the CPC group, ten out of 24 cancer classes had a point estimate for sensitivity of ≥75% (lung, ovary, anus, pancreas, esophagus, head and neck, colon/rectum, urothelial tract, liver/bile duct, and cervix), sixteen had a point estimate for sensitivity of ≥ 50% (previous list plus breast, sarcoma, stomach, lymphoma, gallbladder, and plasma cell neoplasm). Four had a point estimate for sensitivity of < 25% (thyroid, kidney, myeloid neoplasm, and uterus). The sensitivity rate for detection was highest for gastrointestinal (GI) cancers (84.1%; 95% CI, 80.6-87.1) in the CPC group. The sensitivity was 65.5% (95% CI, 63.2-67.7) for solid tumors and 55.1% (95% CI, 49.3-60.8) for hematologic malignancies in the CPC group. The investigators subsequently tested performance using the more in-depth AJCC cancer classification and histologic subtypes. Here, sensitivity was reported in the test positive rate (TPR) due to the small number of samples per type. The TPR across types and stages was 63.9% (95% CI, 61.8-66.0) in the CPC group and 46.5% (95% CI, 46.5-52.8) in the CPC HSC group. The accuracy for cancer signal origin (CSO) prediction was 90.3% (95% CI, 88.6-91.9) in the CPC group and 78.7% (95% CI, 70.1-85.4) in the CPC HSC subgroup.

The test showed robust performance across age groups, cancer types, and stages. Importantly, accurate CSO prediction can help guide subsequent diagnostic decisions, thereby reducing the time to diagnosis and minimizing unnecessary testing. Further studies can characterize the specific real-world effects of integrating the test into standard clinical workflows. Understanding the biological differences in the shedding of ctDNA between different cancers and how that affects the sensitivity of these screening tests is an important area of research.

Written by: Bishoy M. Faltas, MD, Director of Bladder Cancer Research, Englander Institute for Precision Medicine, Weill Cornell Medicine

References:

Bryce AH, Thiel, DD, Seiden, MV, et al. Performance of a cell-free DNA-based multi-cancer detection test in individuals presenting with symptoms suspicious for cancers. JCO Precis Oncol. 2023; 7:e2200679.

Read the Abstract

Login