The Lancet October 07, 2023

Blood-based tests for multicancer early detection (PATHFINDER): a prospective cohort study

Deb Schrag et al.

Bottom Line

A prospective cohort study demonstrating the clinical feasibility of using a blood-based multi-cancer early detection (MCED) test to screen for multiple cancers simultaneously and direct subsequent diagnostic workups in adults aged 50 and older.

Key Findings

1. Among 6,621 participants with analyzable blood test results, a cancer signal was detected in 92 individuals (1.4%).
2. Of the 92 participants with a positive signal, 35 were diagnosed with cancer (true positives, yielding a positive predictive value of 38%), while 57 had no cancer diagnosis (false positives, 62%).
3. The median time to diagnostic resolution was 79 days (IQR 37–219) overall, but was shorter for true positives (57 days; IQR 33–143) compared to false positives (162 days; IQR 44–248).
4. Diagnostic evaluations heavily utilized imaging (91% for true positives and 93% for false positives), but invasive procedures were largely avoided in those without cancer (only 30% of false positives underwent invasive procedures, and just 1 had surgery).

Study Design

Design
Prospective Cohort Study
Open-Label
Sample
6,662
Patients
Duration
12 mo
Median
Setting
7 US health networks
Population Adults aged 50 years or older without signs or symptoms of cancer, with or without additional known cancer risk factors.
Intervention Multi-cancer early detection (MCED) blood test using cell-free DNA methylation (GRAIL Galleri) with results and predicted cancer signal origin returned to physicians.
Comparator N/A (Single-arm study)
Outcome Time to, and extent of, diagnostic testing required to confirm the presence or absence of cancer following a detected cancer signal.

Study Limitations

Lack of a randomized control group prevents the assessment of the test's impact on cancer-specific mortality or overall survival.
A modest number of overall cancers detected (35) limits the statistical power for subgroup analyses by cancer type.
High proportion of false positives among those with a detected signal (62%), which can lead to psychological distress, care delays, and unnecessary healthcare expenditures.
The study enrolled a convenience sample with limited racial and ethnic diversity (91.7% of participants were White), restricting the generalizability to broader populations.
Without universal confirmatory screening (e.g., whole-body imaging) for all participants with a negative MCED result, the true clinical sensitivity and false negative rate of the test in this cohort cannot be definitively established.

Clinical Significance

The PATHFINDER study represents a pivotal milestone in cancer screening by proving the real-world operational feasibility of a blood-based MCED test (Galleri). It demonstrates that a single blood test can identify signals for multiple cancer types—including many that currently lack recommended screening pathways—and accurately predict the cancer signal origin to guide efficient clinical workups. Importantly, it proved that false-positive results did not trigger an unmanageable cascade of invasive surgical procedures, laying the foundation for implementing pan-cancer blood screening in broader clinical practice.

Historical Context

Historically, cancer screening has relied almost entirely on organ-specific modalities like mammography, colonoscopy, Pap smears, and low-dose CT scans. While highly effective, these isolated screenings leave the vast majority of lethal cancers undetected until advanced stages. The development of next-generation sequencing enabled the detection of tumor-derived cell-free DNA (cfDNA) methylation patterns circulating in the blood. Prior observational and case-control studies (such as the CCGA study) validated the analytical accuracy of the GRAIL MCED assay. PATHFINDER was the landmark first interventional study to actually return these novel test results to patients and clinicians in real-time, serving as the essential bridge between theoretical assay performance and massive ongoing randomized controlled mortality trials (like NHS-Galleri).

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

What is the biological mechanism by which the MCED test evaluated in PATHFINDER detects cancer, and why does this mechanism inherently lead to higher sensitivity for advanced-stage compared to early-stage tumors?

Key Response

The test identifies abnormal methylation patterns on cell-free DNA (cfDNA) shed into the bloodstream by tumors via apoptosis or necrosis. Because advanced-stage and larger tumors shed significantly more cfDNA than early-stage, localized tumors, the assay is intrinsically more likely to detect later-stage disease, representing a critical biological limitation for a tool intended for early detection.

Resident
Resident

A 65-year-old asymptomatic patient receives a positive MCED test indicating a gastrointestinal tract cancer signal. According to the PATHFINDER clinical workflow, what is the next step, and how should you counsel the patient regarding their actual risk of having cancer?

Key Response

The next step is targeted diagnostic imaging (e.g., endoscopy or CT of the abdomen/pelvis) directed by the predicted Cancer Signal Origin (CSO). The resident must counsel the patient that while the test is highly specific, its positive predictive value (PPV) is around 40 percent. This means there is roughly a 60 percent chance this is a false positive, which is essential to discuss upfront to mitigate severe anxiety during the diagnostic workup.

Fellow
Fellow

The PATHFINDER study utilized the test's predicted Cancer Signal Origin (CSO) to direct initial workup. What is the clinical dilemma when the initial CSO-directed workup is completely negative, and how does this impact the risk-benefit ratio of the test?

Key Response

When the CSO-directed workup is negative, the clinician faces a 'diagnostic odyssey.' They must decide whether to pursue broader, untargeted whole-body imaging (like PET/CT) or invasive procedures to find a hidden malignancy, versus recommending watchful waiting. This risks exposing the patient to unnecessary radiation, complications from biopsies, and prolonged psychological distress, heavily complicating the test's risk-benefit ratio in standard practice.

Attending
Attending

While PATHFINDER demonstrates the feasibility of detecting signals for cancers without standard screening programs, why might widespread implementation of this test currently fail to achieve the ultimate goal of reducing all-cause or cancer-specific mortality?

Key Response

Feasibility of detection does not equate to clinical utility. Detecting cancers earlier (lead-time bias) or detecting indolent tumors that would never have caused symptoms in the patient's lifetime (length-time bias and overdiagnosis) can falsely inflate apparent survival times without actually changing the disease's natural history. True clinical utility requires proving that this early detection alters treatment pathways and ultimately prevents mortality.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

The PATHFINDER study is a single-arm prospective cohort. What specific methodological limitations regarding the ascertainment of the true negative rate exist in this design, and how should future trials be designed to definitively evaluate diagnostic performance?

Key Response

In a feasibility study without systematic pan-scanning of all participants who test negative, false negatives are only identified if a cancer clinically manifests during the limited 12-month follow-up window. This differential verification bias artificially inflates perceived test performance. Future research requires large randomized controlled trials comparing MCED screening to usual care, powered to detect true differences in late-stage cancer incidence and mortality.

Journal Editor
Journal Editor

If evaluating the PATHFINDER manuscript for publication, what major concerns should be raised regarding the representative nature of the study cohort, and how does this affect the external validity of the reported positive predictive value (PPV)?

Key Response

A seasoned reviewer would flag healthy volunteer bias; trial participants were likely more affluent, health-conscious, and potentially had different baseline cancer risks than the general population. Because PPV is mathematically dependent on disease prevalence, evaluating the test in a skewed demographic threatens the external validity and generalizability of the reported PPV to a broader, unselected primary care population.

Guideline Committee
Guideline Committee

The USPSTF currently recommends targeted screening for specific cancers (e.g., colonoscopy, mammography) but has no recommendations for MCED tests. Based on the PATHFINDER findings, what evidentiary thresholds remain unmet that prevent a positive recommendation for population-wide MCED screening?

Key Response

The USPSTF requires high-certainty evidence that a screening test improves patient-centered outcomes (like reducing cancer-specific mortality) and that the benefits outweigh the harms (e.g., overdiagnosis, false-positive cascades). PATHFINDER only establishes clinical feasibility and diagnostic resolution time frames. Without RCT data demonstrating a population-level stage shift or mortality benefit, MCED testing cannot yet be integrated into current evidence-based guidelines.

Clinical Landscape

Noteworthy Related Trials

2020

DETECT-A Trial

n = 10,006 · Science

Tested

CancerSEEK multi-cancer blood test followed by PET-CT

Population

Women aged 65-75 years with no history of cancer

Comparator

Standard of care screening

Endpoint

Safety and feasibility of MCED testing

Key result: The test successfully identified 26 cancers across 10 organs in asymptomatic women, enabling curative surgery without increasing unsafe diagnostic procedures.
2021

CCGA Study

n = 15,254 · Ann Oncol

Tested

Targeted methylation-based cfDNA assay (Galleri)

Population

Individuals with and without newly diagnosed cancer

Comparator

Non-cancer controls

Endpoint

Sensitivity, specificity, and tissue of origin accuracy

Key result: The assay showed 99.5% specificity and correctly predicted the tissue of origin in 88.7% of true positive cases across more than 50 cancer types.
2023

SYMPLIFY Trial

n = 6,238 · Lancet Oncol

Tested

Targeted methylation-based cfDNA assay (Galleri)

Population

Patients with non-specific symptoms referred for cancer investigation

Comparator

Standard diagnostic workup

Endpoint

Sensitivity, specificity, and positive predictive value for cancer diagnosis

Key result: The MCED test demonstrated high specificity and accurately predicted the primary cancer site in over 85% of cases, aiding diagnostic triage.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis