Jul 27, 2026
What colorectal cancer screening teaches us about the coming era of multi-cancer early detection
A recent essay in The Cancer Letter asks whether it is time to reassess the practice and practicality of screening colonoscopy. Its answer, grounded in the American Cancer Society's updated 2026 colorectal cancer guideline and a body of microsimulation modeling, is a confident yes: with one in three eligible Americans still unscreened and colorectal cancer now the leading cancer killer of adults under 50, the burden is simply too large to solve with colonoscopy alone. Redistribute scarce colonoscopy capacity toward diagnostic and therapeutic procedures, the argument goes, and lean on high-performing noninvasive tests for initial screening — and detection rises, cancers are prevented, and costs fall.
We agree with the premise. The math of access is undeniable, and a strategy that meets patients where they are is the right one. Noninvasive, at-home testing is how we close the gap between screening eligibility and screening completion. That is not in dispute.
But there is a quieter assumption buried inside the optimism, and it deserves scrutiny before we carry it into the next, larger wave of cancer screening. The assumption is that a positive noninvasive test is a good outcome. It is not. A positive test is not a diagnosis, a treatment, or a life saved. It is a question — one that only a follow-up procedure can answer. And in the real world, that question goes unanswered far more often than the models assume.
The loop that keeps breaking
The ACS guideline is explicit on this point: for screening to work, a positive result on any non-colonoscopy test must be followed by a timely colonoscopy, preferably within six months, or the screening process is not complete. Robert Smith, who led the guideline update, put it plainly — many patients with a positive stool test never get that follow-up, and without it the screen is meaningless.
How often does the loop break? The honest answer is: much more often than the headlines about test performance would suggest.
Manufacturer-reported figures are encouraging. Exact Sciences reports that roughly 79% of patients with a positive Cologuard result complete a follow-up colonoscopy, and a company-funded analysis of patients aged 45–49 found follow-up adherence of 85% for multi-target stool DNA testing — dramatically better than the 35% observed for older fecal tests. Those are real gains, and they are worth defending.
But even at that best-case number, roughly one in five positive results never reaches resolution. And those numbers reflect programs wrapped in patient-navigation support that most patients, in most systems, never encounter. When researchers step outside those supported cohorts, the picture darkens considerably. An independent 2025 real-world study published in The Journal of Gastrointestinal Cancer found that only 44% of patients completed a follow-up colonoscopy within a year of a positive Cologuard result — meaning more than half did not. Broader systematic reviews put real-world follow-up somewhere between 18% and 56% within one year, against the 80–90% completion seen in the landmark trials that established screening's benefit in the first place.
This is the uncomfortable truth the modeling can obscure. Microsimulation studies frequently assume follow-up adherence at or near 100%. Reality operates at half that. And the consequence is not neutral: a large retrospective study found that a delay of 13 months or more before follow-up colonoscopy raised the odds of a colorectal cancer diagnosis, and a delay of 19 months or more raised the odds of dying from it. A positive test that leads nowhere is not a benign non-event. It is a missed cancer, a false sense of security, and a cost incurred for no clinical return.
So the reassessment The Cancer Letter calls for is right, but it is only half a reassessment. Shifting the front door of screening from colonoscopy to noninvasive testing does not reduce the need for diagnostic completion. It multiplies it. Every additional patient we successfully screen with a stool or blood test is another patient who, if positive, now needs a colonoscopy they may never schedule. Expanding access without engineering resolution simply moves the failure point downstream, where it is harder to see and easier to ignore.
Why this matters more for MCED, not less
We raise this now, in the context of colorectal cancer, because CRC is the rehearsal. The main event is multi-cancer early detection.
MCED blood tests promise to screen for dozens of cancers from a single draw, many of them cancers with no screening pathway at all. The appeal is obvious and the potential is genuine. But everything that makes the follow-up problem hard in CRC is harder in MCED — and several things are entirely new.
In colorectal cancer, a positive noninvasive test has a single, well-defined next step: a colonoscopy. The pathway is standardized, the specialist is identifiable, the procedure is covered, and the system has decades of muscle memory for delivering it. Even with all of that, we complete the loop maybe half the time.
An MCED signal offers none of that clarity. A positive result points to a predicted tissue of origin, not a destination. Resolution may require CT or PET imaging, referral to one of several specialties, sequential workups, and in some cases an extended search for a cancer that diagnostic tools cannot yet localize. There is no single "colonoscopy" to schedule — there is a diagnostic odyssey to navigate, often across departments that do not talk to one another. If a positive stool test loses half its patients on the way to one known procedure, what happens to a positive MCED result that hands a patient and a primary care physician an ambiguous signal and a map with no roads?
The lesson from CRC is therefore not "MCED will be fine because the tests are getting better." Test performance was never the bottleneck. The bottleneck is the space between the result and the resolution — the referral that is never placed, the appointment that is never booked, the patient who is reassured by a normal-sounding delay, the finding that falls into the gap between primary care and specialty care. Better tests do not close that gap. In some ways they widen it, by generating more signals than the diagnostic system is built to resolve.
Detection is a means; resolution is the outcome
If there is one principle worth carrying from the colorectal experience into the MCED era, it is this: the metric that matters is not how many cancers a test “detects”, but how many patients reach “diagnostic resolution”. A screening program should be judged, funded, and designed around completed pathways — positive result to definitive answer — not around test volume or positivity rates.
That reframing has practical consequences. It means building navigation into the screening pathway from day one rather than bolting it on after adherence disappoints. It means measuring and reporting loop-closure as a core quality metric, the way HEDIS and the US Multi-Society Task Force already treat follow-up colonoscopy. It means designing the handoff between a positive result and the diagnostic workup as carefully as we design the test itself. And for MCED specifically, it means solving the resolution pathway — who owns the workup, how it is coordinated, how the patient is guided and covered — “before” we scale the screening, not after.
The Cancer Letter is right that the moment calls for an adaptive strategy. We would only add: an adaptive strategy that stops at detection is not a strategy at all. Screening does not save lives. Completed screening does. As we accelerate toward a future of one-draw, many-cancer testing, the hardest and most important work is not the signal — it is everything that has to happen after it.
Looking for the executive summary? Read the five-minute companion essay.
*Sources referenced: The Cancer Letter, "Is it time to re-assess the practice and practicality of screening colonoscopy?" (July 2026); American Cancer Society 2026 Colorectal Cancer Screening Guideline update (CA: A Cancer Journal for Clinicians); Exact Sciences Cologuard/Cologuard Plus adherence materials and company-funded adherence studies (Int J Colorectal Dis 2025; ScienceDirect 2025); "From Detection to Delay: Real-World Gaps in Post-Cologuard Colonoscopy Adherence," J Gastrointest Cancer (2025); systematic review of follow-up after abnormal stool-based tests, Gastroenterology (2024); Mohl et al., JAMA Network Open (2023).