The launch of Exact Sciences’ Cancerguard (available August 22, 2025) is the latest milestone in the growing wave of multi-cancer early detection (MCED) innovation. But as Robert Green aptly put it in a chat we had a few years ago, “NGS doesn’t need another test, it needs a system.” That point cuts to the heart of the challenge (and opportunity) facing disruptive diagnostics, if we want this category to be perceived (and paid for) in a way that is commensurate to the value it provides.

The reason MCED matters is simple: prognosis for a cancer patient ties directly to stage of diagnosis. Catch a cancer at stage I, and five-year survival can be above 90%; catch it at stage IV, and survival rates can plummet into the single digits in some indications. For most cancers, we don’t screen at all. MCED promises to change that. But technology or a breakthrough assay alone won’t deliver the revolution; what matters is whether companies can build the system around it to ensure doctors are on board and patients actually benefit.

In a past piece, I argued that “Dx deserves better”: diagnostics are responsible for ~70% of clinical decisions, yet are underappreciated, under-reimbursed, and poorly integrated into care. MCED and its kin are a perfect case study for doing diagnostics differently: designing not just for analytical performance, but for adoption and impact!

‍

Why System-Building Wins

Strategically advanced players are showing that winning in MCED and adjacent markets means attacking the problem on multiple fronts:

• Integrating into clinical workflows. This integration touches on 4 parts: 1) Grail’s embedding of Galleri ordering into EHRs and its health system partnerships make adoption almost frictionless for physicians. Of note, while EHR ordering is the gold standard, alternatives (i.e., portals, test requisitions forms) exist. 2) So perhaps even more importantly, the same principle is critical for Cancerguard, not just listing it for sale, but embedding it into annual exam protocols. The test needs to be “on the agenda” for a patient appointment (e.g., similar to CBC, blood sugar). 3) Internally, within the diagnostics company, be ready to alleviate issues when they inevitably occur (e.g., tracking down the right sample or information when clinicians fill out the TRF incorrectly, to even fixing when the wrong test). And 4) Integrate the test within provider decision-support systems, which tend to follow guidelines and guidelines follow evidence.
• In parallel, consider a consumer channel. In June 2025, Grail announced a collaboration with Everlywell. Now, Galleri can be prescribed and ordered directly via everlywell.com, with kits delivered to patients’ homes, bypassing the need for an in-person HCP visit. That move signals a dual-channel strategy: reinforce your foothold in clinical workflows while simultaneously opening a consumer-facing avenue for proactive health seekers, a clever amplification of system thinking (e.g., network of channels). In the same vein, consumer-initiated ordering for Cologuard has seen meaningful growth (and positive reception). Given the broad availability of telemedicine and mobile phlebotomy / sample access, this approach is more doable than ever, especially in a post-COVID world that has normalized self-initiated, at-home testing for many consumers. Of note, this DTC trend is also increasingly seen on the pharma side.
• Securing payer and regulatory alignment early. Guardant Health’s experience with Shield, its blood-based colorectal cancer screening test, shows the power of the (already established Medicare) coverage pathway. Shield is on track for $55-60M in 2025 revenue (nearly 2x its February guidance) because FDA approval and CMS coverage turned it from “another” LDT into a reimbursed option that physicians can confidently recommend. This approach turns early launch ASP from “horrible” to merely “bad” (direct quote from a client!), with upside potential. In the therapy selection space, Caris’ FDA tissue test Medicare CDx pricing is $8,455 (for some of its volumes),~2.5x its former $3,500 LDT price. Clearly, Caris’ strategy to aim for comprehensiveness isn’t academic, but drives P&L, as it enables higher diagnosis rates and reimbursements.
• Generating and leveraging real-world evidence (RWE). Without compelling utility data, adoption stalls. Grail generated $34M in screening revenue in 2Q25 (+21% y/y) on >45,000 tests, supported by ongoing clinical trials and large-scale partnerships. Exact’s 50,000-patient FALCON study for Cancerguard will be essential to not only validate performance but also prove impact on downstream care and costs. These studies may be less important for guideline bodies than for broad sentiment change (key to drive usage growth).
• Owning a differentiated data asset. GeneDx’s dominance in pediatric rare disease genomics rests on >850,000 exome/genome sequences and >7M phenotypic datapoints. The differentiation of the actual molecular test (in this case, WGS) is actually minimal. This depth fuels interpretation accuracy, AI model training, and long-term defensibility (just as relevant as in oncology), where data flywheels improve test performance over time. The competitive moat is driven by truly unique sample cohorts or datasets.
• Designing for the patient journey. Exact’s Cancerguard playbook includes care navigators and reimbursement programs for follow-up imaging, turning a potentially anxiety-inducing “cancer signal detected” result into a guided pathway to definitive diagnosis. Delfi is applying similar thinking in lung cancer screening by aligning its FirstLook Lung test with routine check-ups for smokers to boost adherence where traditional CT screening has failed. Creating new test opportunities means addressing some of the complexity around what happens next, following a positive or negative result.
• Partnering with care delivery. Health system partnerships (e.g., Grail and Providence Health, Caris’ Precision Oncology Alliance, NeoGenomics’ decision-support tools in community oncology) create distribution, credibility, and integration that standalone marketing can’t match. Diagnostic companies don’t need to do it alone; they can tap into the mission of healthcare systems to not only treat patients but also advance healthcare.

‍

The Strategic Takeaway

In MCED, SCED, and other high-impact diagnostics, the market leaders will be those who design for adoption (“the system”) as rigorously as they design for accuracy (“the test”). The playbook is becoming clear:

1. Embed in workflows so ordering and acting on results are easy and natural, and positive results yields clear next steps, limiting ambiguity.
2. Secure coverage early by aligning regulatory strategy with payer needs from day one.
3. Engineer the patient experience to remove friction and uncertainty.
4. Leverage partnerships to accelerate adoption and scale.
5. Invest in data as a compounding asset that improves performance and strengthens your moat. Data is the final part of this playbook that connects to further innovation and turning that flywheel faster.

The next decade in diagnostics will not be won by the cleverest assay alone (e.g., Cancerguard doesn’t provide tissue of origin, a clear imperfection). It will be won by the companies that build the systems (i.e., clinical, commercial, and operational) to make their tests clinically seamless from a patient journey perspective, and indispensable. MCED may be the latest poster child for this shift, but the same strategic principles apply across the diagnostic spectrum. Because the real innovation is detecting disease earlier, *and* creating the infrastructure to make early detection actionable in healthcare systems at scale. The best diagnostics companies in the future are going to have lots in common with the best logistics and operations companies.

From an investor perspective, this is not a “best assay wins” market; it’s best system wins. Shield’s revenue ramp tracks coverage and workflow; GRAIL’s enterprise growth tracks integration and RWE; Cancerguard’s differentiation is the codified downstream pathway; GeneDx’s defensibility is its data advantage. Capital should flow to teams that ship technology + operations + payer model (i.e., systems), because that’s where adoption, durability, and margin expansion compound. Execution risks are higher than ever before; strategy without operational discipline (e.g., IP litigation, cash burn) can bleed value.

_______________________

Case Study: Cancerguard as a System (Not a SKU)

Exact Sciences can deploy Cancerguard across its 300K+ provider base at near-zero marginal cost. But the strategic significance of Exact’s Cancerguard isn’t the test alone or channel access; it’s the downstream design:

• Imaging‑first confirmation: Rather than predicting a cancer signal of origin, Cancerguard directs positive results into a standardized, radiology-led diagnostic workup (e.g., PET/CT + targeted follow‑up as indicated). That choice accomplishes 3 goals: 1) reduces diagnostic ambiguity, 2) aligns with how health systems already resolve “incidentalomas,” and 3) creates a repeatable pathway payers can evaluate. It remains to be seen whether this approach will prove optimal, given the costs of a full-body PET scan, potential scan availability constraints, and uncertainty around the exact downstream workups.
• Navigation as product: Dedicated care navigators, templated orders, and pre-authorization workflows turn an anxiety-provoking screen into a coordinated, supported patient journey. Patient support (including follow‑up imaging assistance) isn’t a nice‑to‑have; it’s core to activation, adherence, and outcomes.
• Provider workflow: Ordering, result delivery, and action prompts live in the EHR; the product is the closed loop (i.e., order → result → next step), not the assay alone.
• RWE loop: Each positive/negative feeds a learning system (diagnostic resolution, time‑to‑diagnosis, stage shift, treatment initiation), strengthening the clinical and economic story with every cohort.

This article was prompted by Cancerguard’s recognition that the test is only half the product; the real differentiation lies in owning “what happens next.” Going forward, MCED/SCED and other disruptive diagnostics will need to meet that same standard to deliver real value.

If you’re only building tests, you’re already behind.

‍

_______________________

Note: Some of the companies listed in this article may be DeciBio Consulting clients.

Author: Stephane Budel, Partners at DeciBio Consulting, LLC

Connect with Stephane on LinkedIn