TWO MULTI-CANCER EARLY DETECTION BLOOD TESTS ARE NOW AVAILABLE
Cancerguard and Galleri detect cancer signals in cell-free DNA
Straight Healthcare
October 2026
October 2026
Multi-cancer early detection (MCED) tests are blood tests that screen for many cancer types at once, including cancers that have no recommended screening test, such as pancreatic, ovarian, and liver cancer. Two MCED tests are commercially available in the U.S.: Galleri (GRAIL) and Cancerguard (Exact Sciences). Both analyze cell-free DNA (cfDNA) that tumors shed into the bloodstream. Galleri looks for cancer-associated methylation patterns, while Cancerguard combines cfDNA methylation with tumor-associated protein levels. The tests differ notably in how they report results. Cancerguard reports only a positive or negative result, with no indication of where a cancer might be, so a positive result is followed by a standardized imaging workup: CT of the neck, chest, abdomen, and pelvis, and whole-body PET-CT if the CT is unrevealing. In contrast, a positive Galleri result includes a predicted cancer signal origin (CSO), such as the pancreas or colon/rectum, which directs the diagnostic workup toward a specific organ. In PATHFINDER 2, the predicted CSO was correct in 91.3% of participants with cancer. Both are prescription-only laboratory-developed tests that have not been approved by the FDA, and neither is intended to replace guideline-recommended screening. GRAIL has submitted a premarket approval application for Galleri, and in September, an FDA advisory panel voted 7 to 2 that its benefits outweigh its risks.
Galleri's performance was evaluated in PATHFINDER 2, a prospective, single-arm study of 32,007 adults aged 50 years or older without suspected cancer. Over 12 months, 0.9% of participants tested positive, and 60.3% of those were diagnosed with cancer. Episode sensitivity was 39.3% for all cancers, 54.9% when breast and prostate cancers were excluded, and 69.8% for 12 cancers responsible for two-thirds of U.S. cancer deaths. Cancerguard was evaluated in a case-control validation study (N=1,124) of people already known to have cancer or to be cancer-free. Its sensitivity was 55.6% (excluding breast and prostate), and its specificity was 97.4%. Because the study designs differ, the numbers are not directly comparable. However, Cancerguard's lower specificity means more false positives, and its PPV has to be estimated from the expected prevalence of undiagnosed cancer.
No study has shown that MCED screening reduces cancer mortality. In a study evaluating the effects of widespread screening with Galleri, 142,250 adults aged 50 to 77 years in England were randomized to three annual rounds of Galleri plus usual care or to usual care alone. The primary outcome, the incidence of stage III or IV cancer among 12 prespecified cancers, did not differ between groups (IRR 1.03; 95% CI 0.92 to 1.14; P=0.63). Stage IV cancers were less common in the screening group (IRR 0.86; 95% CI 0.74 to 1.00), but overall, more cancers were diagnosed, raising the possibility of overdiagnosis. In both groups, only about one-third of cancers (32.3%) were found through screening; the remaining two-thirds were diagnosed clinically (e.g., after symptoms developed). Other limitations include low sensitivity for early-stage disease, invasive workups after false-positive results, and limited insurance coverage. NHS-Galleri cancer-specific mortality results are expected 3 and 6 years after the last study visit. Until then, MCED testing is best viewed as an optional supplement to standard screening. Our MCED blood tests page provides extensive information on both tests, including ordering and specimen requirements, accuracy by cancer type and stage, positive-test follow-up, cost, and detailed summaries of the NHS-Galleri and PATHFINDER 2 studies.
Galleri's performance was evaluated in PATHFINDER 2, a prospective, single-arm study of 32,007 adults aged 50 years or older without suspected cancer. Over 12 months, 0.9% of participants tested positive, and 60.3% of those were diagnosed with cancer. Episode sensitivity was 39.3% for all cancers, 54.9% when breast and prostate cancers were excluded, and 69.8% for 12 cancers responsible for two-thirds of U.S. cancer deaths. Cancerguard was evaluated in a case-control validation study (N=1,124) of people already known to have cancer or to be cancer-free. Its sensitivity was 55.6% (excluding breast and prostate), and its specificity was 97.4%. Because the study designs differ, the numbers are not directly comparable. However, Cancerguard's lower specificity means more false positives, and its PPV has to be estimated from the expected prevalence of undiagnosed cancer.
| Measure | Cancerguard | Galleri |
|---|---|---|
| Sensitivity | 55.6% (excluding breast and prostate) |
39.3% (all cancers) 54.9% (excluding breast and prostate) |
| Specificity | 97.4% | 99.64% |
| PPV | 9.7% - 30.4% (estimated) |
60.3% (observed) |
| NPV | 99.1% - 99.8% (estimated) |
99.2% (observed) |
| Self-pay price | $689 | $799 |
- NHS-Galleri - Effect of Screening with Multicancer Early-Detection Test on Late-Stage Cancer Diagnosis, NEJM (2026) [PubMed abstract]
- PATHFINDER 2 - Performance and Safety of a Multi-Cancer Early Detection Test, Nat Med (2026) [PubMed abstract]
- Multi-cancer early detection (MCED) blood tests
