Multi-cancer early detection blood tests: a signal is not a cancer diagnosis

Multi-cancer detection (MCD or MCED) tests analyze blood for biological signals associated with several cancers. They are screening signals, not diagnoses. As of August 24, 2026, no MCD test is FDA authorized in the United States; some are sold as laboratory-developed tests. A positive result requires a diagnostic pathway, a negative result does not rule out cancer, and either result should not replace established breast, cervical, colorectal, lung, or other risk-based screening.¹²⁴

The central question is not whether a blood draw can find a molecular pattern. It is whether testing an asymptomatic population, following every signal, and managing missed or overdiagnosed disease ultimately produces more benefit than harm. NCI is building randomized evidence to answer that question; the answer is not yet established.³

Place the test in the screening chain

Skipping from the first row to “early cancer detected” changes a prediction into a diagnosis. A provider should explain the full chain before drawing blood, including who owns an unresolved positive result.

FDA authorization and laboratory status are separate

NCI states that no MCD tests are FDA authorized as of the review date.¹ Some tests are commercially available as LDTs designed, manufactured, and used within a single laboratory. A Breakthrough Device designation, if present, can facilitate development and review but is not FDA authorization and does not mean premarket review is complete.¹

Verify the legal laboratory, CLIA certificate, exact assay and version, ordering clinician, specimen logistics, report module, and current FDA status. The CLIA-versus-FDA guide explains why laboratory certification does not prove clinical benefit or convert an LDT into an authorized test.

Ask the seller to say “not FDA authorized” plainly if that is the status. “Physician ordered,” “CLIA certified,” “breakthrough,” “validated,” and “available nationwide” are not substitutes.

A positive result is a workup assignment

NCI says a positive MCD result means a person might have cancer and that further tests—potentially imaging, invasive procedures, or biopsy—are needed to determine whether cancer is present.¹ Some assays offer a predicted tissue of origin, but that estimate does not confirm a site.

Before testing, request a written positive-result pathway:

NCI notes that in published research, more than half of people with a positive MCD result had no cancer found during follow-up.¹ That does not prove those signals were all errors; it shows why an unresolved result can create repeated testing, invasive procedures, cost, time, and anxiety.

A negative result has a defined blind spot

No MCD test covers every cancer, every subtype, or every stage. Early tumors may release too little of the measured signal, a covered cancer may be missed, and an uncovered cancer is outside the assay by design. NCI says people have been diagnosed with cancer despite a negative MCD test.¹

Ask for sensitivity by cancer type and stage, not only one pooled detection rate. A test that detects more late-stage disease than stage I disease can post an attractive combined percentage while missing the exact early disease the buyer expects it to find.

Negative predictive value depends on population and follow-up. A “no signal detected” report should not be translated into “cancer free,” permission to ignore symptoms, or a reason to stop ordinary screening.

Standard screening continues on its own calendar

NCI says standard-of-care cancer screenings remain recommended regardless of an MCD result.¹⁴ Depending on age, anatomy, history, exposure, and risk, that can include established breast, cervical, colorectal, and lung screening pathways and other individual recommendations.

Create two calendars: one for recommended standard screening and one for the experimental or commercially offered MCD test and its follow-up. Never let the latter overwrite the former in a portal or membership plan.

The same principle applies to symptoms. Screening is for people without signs or symptoms. A new mass, bleeding, unexplained weight change, persistent pain, neurologic change, or another concerning symptom needs clinical evaluation, not a consumer screening test as a gatekeeper.

Read performance in the offered population

Sensitivity and specificity are not enough. Ask for the number of asymptomatic people in the intended-use population, cancer prevalence, distribution by type and stage, false positives, false negatives, positive predictive value, indeterminate results, follow-up completion, and length of outcome ascertainment.

One test may count people with a known diagnosis, symptoms, or later-stage disease in a development dataset; those results do not directly predict performance in an average-risk screening population. Case-control discrimination and prospective population screening answer different questions.

More detection is not automatically more benefit

Screening can find aggressive disease earlier, which is the hoped-for benefit. It can also detect slow-growing disease that would never affect health, create false alarms, expose people to procedures, or divert resources. NCI’s professional screening overview says no MCD assay has yet shown a mortality reduction in a randomized clinical trial.²

The actively recruiting Vanguard Study is a pilot enrolling up to 24,000 people to inform a much larger randomized trial that would evaluate whether benefits outweigh harms and whether MCD screening can reduce deaths.³ Participation in that evidence-building program and buying a commercial test are different propositions.

A citation to observational detection rates cannot replace outcome evidence. Ask what endpoint supports the marketing claim: analytic detection, cancer found after workup, stage distribution, reduced late-stage diagnosis, quality of life, or mortality.

Cost includes the diagnostic cascade

The blood draw and laboratory fee are only the entry cost. Build possible costs for clinician review, repeat testing, imaging, specialist consultation, travel, biopsy, anesthesia, pathology, time away from work, and longitudinal follow-up when no source is found. Ask which services may be covered and which are self-pay.

An ordering service should not promise that insurance “will cover the follow-up.” Coverage depends on plan, order, diagnosis code, network, authorization, and medical-necessity review. Get the likely route and billing entities in writing.

Decide whether the unanswered questions are acceptable

1. Protect established screening. Write every standard screening and risk-based appointment due regardless of the MCD result.
2. Verify test status. Record exact assay, version, laboratory, CLIA certificate, ordering clinician, and absence or presence of an FDA authorization.
3. Read population-level performance. Use prospective asymptomatic data with type- and stage-specific sensitivity, specificity, PPV, indeterminate rates, and follow-up.
4. Pre-author the positive pathway. Identify the clinician, diagnostic sequence, unresolved-result plan, insurance route, and likely costs before the draw.
5. Constrain the negative meaning. A negative does not exclude cancer, cover every type, replace screening, or adjudicate symptoms.
6. Separate detection from outcome. Ask whether evidence shows analytic signal detection, a confirmed diagnosis, stage change, or reduced mortality.

The most important pretest question is: “If this result is positive, who owns the diagnostic workup until cancer is confirmed or the signal remains unexplained—and if it is negative, which screenings and symptoms remain unchanged?”