What must happen after a cancer signal?

A blood test that looks for signals associated with multiple cancers invites a powerful intuition: earlier information should create better outcomes. GRAIL's multicancer detection work made that possibility tangible. Yet a screening service is more than an assay. It also includes the population tested, the interpretation of a result, the diagnostic work that follows, and the balance between benefit and harm. This case focuses on the PATHFINDER evidence published in 2023, rather than offering a current recommendation about whether an individual should be screened.

PATHFINDER took an important step beyond testing stored samples from people with known diagnoses. Its prospective design returned results to participants and their clinicians, allowing researchers to observe the workup that followed. The Lancet report described 6,621 participants with analyzable results, aged fifty or older. Ninety-two received a cancer signal. Thirty-five were ultimately classified as true positives and fifty-seven as false positives. Those denominators matter: a positive signal was not itself a diagnosis, and most people with a signal did not receive a cancer diagnosis in that evaluation.

GRAIL's contemporary announcement framed the study as evidence that multicancer screening could be incorporated into clinical care. That is the company's interpretation of feasibility. The study also made downstream work visible. Investigations took time and could involve imaging, procedures, and additional consultations. The narrative should not turn either the detected cancers or the false positives into a complete verdict. Each is part of the evidence a screening program has to account for, alongside cancers the test does not detect.

The National Cancer Institute had already explained in 2022 why the larger question remained difficult. Detecting more cancers, or detecting them earlier, does not by itself establish that a screening strategy reduces deaths or produces a favorable overall balance of outcomes. Overdiagnosis, false reassurance, unnecessary investigations, and displacement of established screening are relevant concerns. Those are general screening principles, not proof that this particular technology cannot succeed. They define the evidence that would make the promise more or less convincing.

The encounter is therefore between two legitimate forms of expertise. Assay developers can characterize a biological signal. Clinicians and health systems have to decide how to act on it. Patients must live through the uncertainty between a result and its resolution. Payers and policymakers need evidence relevant to a population, not only a technically impressive demonstration. A product that crosses these boundaries needs a pathway that connects them, including responsibility for follow-up and an honest account of what remains unknown.

The teaching interpretation is that the first successful measurement can reveal a second innovation problem. Once a test can identify a signal, the next bottleneck may be diagnostic capacity, the clarity of referral decisions, or the ability to evaluate meaningful outcomes. More testing can increase work elsewhere in the system. That is not an argument against testing; it is a reason to include downstream consequences in the design and evaluation rather than treat them as someone else's implementation detail.

For a clinical innovator, a bounded first move is to specify the decision that will change after each possible result and then study the entire pathway. What happens after a positive result that cannot be resolved quickly? What should a negative result never be allowed to imply? Who pays for the investigation? PATHFINDER is useful because it made part of that pathway observable. The broader lesson is to preserve scientific ambition while remaining precise about the difference between detecting a signal, diagnosing disease, and demonstrating that a screening program improves lives.

For a discussion tailored to your team, explore a workshop with Julien.