Exact Sciences Corp, the Madison, Wisconsin-based diagnostics company listed on NASDAQ as EXAS, has announced four presentations at the 2026 American Association for Cancer Research Annual Meeting, scheduled for April 17 to 22 in San Diego, California. The presentations cover new clinical data from the NSABP B-59/GBG-96-GeparDouze trial evaluating its Oncodetect tumor-informed circulating tumour DNA test in early triple-negative breast cancer, alongside updated performance data for its Cancerguard multi-cancer early detection test. Together the four abstracts span molecular residual disease testing in breast and colorectal cancer and two distinct aspects of the Cancerguard multi-biomarker classifier programme.
Why post-NAT ctDNA status in TNBC shifts the risk stratification conversation
The most consequential data point in the AACR disclosure concerns the Oncodetect test’s performance in the neoadjuvant setting for triple-negative breast cancer. Pre-surgery ctDNA status following neoadjuvant therapy was reportedly associated with both pathologic complete response and distant recurrence-free interval in the NSABP B-59 substudy. That pairing matters because it positions a blood-based molecular signal as a potential early indicator of treatment resistance before the surgeon opens the operative field, compressing the feedback loop between treatment and response assessment.
Triple-negative breast cancer is the subtype where clinicians most acutely feel the absence of targeted treatment options and the pressure to identify non-responders quickly. The current standard pathway, neoadjuvant chemotherapy followed by surgical assessment of pathologic complete response, delivers its verdict at a single time point. A pre-surgical ctDNA read that correlates with both pCR and long-term recurrence outcomes introduces the possibility of dynamic, biology-driven stratification during the neoadjuvant window rather than after it closes. The clinical utility question that follows, whether a positive ctDNA signal is actionable in time to change the treatment course, remains open and will likely drive the next wave of trial design in this indication.
Whole-exome sequencing underlies the Oncodetect test’s tumour-informed architecture, meaning the assay is built from a patient’s individual tumour mutation landscape rather than from a fixed panel of hotspot mutations. That design choice carries real sensitivity advantages in a subtype like TNBC where the mutational profile is heterogeneous and shared oncogenic drivers are limited. It also carries a cost and turnaround complexity that will shape commercial adoption pathways, particularly in community oncology settings where sequencing infrastructure is less uniformly available. Industry observers tracking the MRD space note that the tumour-informed versus tumour-naive design debate has become a defining axis of competitive differentiation in the liquid biopsy market, with each approach carrying distinct trade-offs between sensitivity, scalability, and clinical workflow integration.
How the colorectal and breast genomic profiling data extends the Oncodetect evidence base
A separate AACR presentation from Exact Sciences examines the value of testing for actionable alterations and circulating tumour DNA across breast and colorectal cancers. The key reported finding, that comprehensive tumour profiling identified actionable genomic alterations in nearly all patients tested and that approximately half had at least one alteration associated with FDA-approved therapies, does not break new scientific ground on its own. What it adds is clinical infrastructure evidence: the data support a framework in which ctDNA testing is layered onto, rather than replacing, existing molecular profiling workflows.
That framing is strategically important for Exact Sciences given the competitive environment. The liquid biopsy market is populated by platforms that have established earlier footholds in comprehensive genomic profiling, and the question of whether MRD testing sits alongside or competes with those platforms has material implications for hospital contracting and oncology workflow decisions. Presenting evidence that ctDNA monitoring and actionable alteration profiling are complementary rather than substitutable reinforces a portfolio argument that Exact Sciences has been constructing through successive conference cycles. Whether the commercial infrastructure to capture both revenue streams from a single patient encounter has been built out remains a separate question from the scientific one.
What the MP V2 classifier upgrade tells us about where Cancerguard is headed
The two Cancerguard-related presentations at AACR 2026 focus on the optimised methylation-protein Version 2 classifier and on the complementary contributions of methylation and protein biomarker classes to cancer signal detection. The MP V2 classifier reportedly demonstrates improved early-stage sensitivity compared with the prior version while maintaining high specificity. That combination, better sensitivity without a corresponding specificity trade-off, is the central technical challenge in multi-cancer early detection and has historically proven elusive across competing platforms.
Specificity is not an academic metric in this context. The Cancerguard test launched commercially in September 2025 at a list price of 689 US dollars as a laboratory-developed test, distributed through Quest Diagnostics’ patient access network. At that price point and without broad insurance coverage, the downstream cost of a false positive, including the imaging workup, specialist referral, and patient anxiety that follows, falls disproportionately on patients and healthcare systems already absorbing the cost of a novel screening intervention. Industry observers tracking multi-cancer screening adoption suggest that the specificity threshold for asymptomatic screening acceptance by major US health systems and payers sits considerably higher than for symptomatic or surveillance populations, which is why the maintained high specificity claim in the MP V2 data is as commercially significant as the sensitivity improvement.
The fourth AACR presentation, examining how methylation and protein biomarker classes contribute independently and complementarily to cancer signal detection, provides scientific grounding for the multi-biomarker architecture that differentiates Cancerguard from earlier single-class approaches. The argument that each class captures a distinct biological signal, methylation reflecting epigenetic cancer-associated changes and protein biomarkers tracking secreted or shed tumour-associated proteins, has been part of the Cancerguard scientific narrative for several years. The AACR data appears to quantify that complementarity specifically in early-stage disease, which is the population where the clinical and commercial case for MCED testing is most fragile and most important to strengthen.
Regulatory pathway and payer coverage remain the unresolved variables
The transition from laboratory-developed test status to a formally cleared or approved assay under US Food and Drug Administration oversight remains the pivotal outstanding variable for the Cancerguard programme. Laboratory-developed tests operate under Clinical Laboratory Improvement Amendments oversight and do not require premarket approval, but this regulatory status limits the ability to claim clinical validity in promotional materials to the same degree as an FDA-cleared device and complicates payer coverage determinations. No FDA-cleared multi-cancer early detection test existed in the US market as of early 2026, meaning that any company achieving clearance first would gain a significant coverage negotiation advantage.
Reimbursement is not merely a downstream commercial detail for a test priced at nearly 700 US dollars aimed at an asymptomatic screening population. The Falcon registry, Exact Sciences’ ongoing prospective real-world study targeting 25,000 participants, is in part an evidence-generation vehicle designed to support eventual coverage applications. Clinicians tracking the MCED field note that the Centers for Medicare and Medicaid Services has not signalled a coverage determination timeline for any multi-cancer early detection test, and that without Medicare coverage, the addressable population in the US is largely limited to commercially insured, higher-income individuals, a demographic that biases utilisation data and limits the public health impact projections that make MCED testing compelling in policy discussions.
What competitors and clinicians are watching after this AACR disclosure
Exact Sciences is not alone in bringing MCED data to major oncology conferences in 2026. The broader competitive landscape includes Illumina-backed Grail, whose Galleri test has been subject to its own regulatory and commercial scrutiny, along with emerging platforms from Guardant Health and others advancing in the ctDNA monitoring and early detection space. The specific advantage that Exact Sciences is attempting to establish through its multi-biomarker class architecture, combining methylation, protein, and potential mutation reflex signals, has been presented as a differentiated approach to the sensitivity-specificity problem. Whether that differentiation is sufficient to establish durable commercial separation from competing platforms will depend as much on coverage decisions and sales force penetration as on scientific performance.
From a clinical perspective, the most actionable near-term question is how the Oncodetect TNBC data from the NSABP B-59 substudy will influence trial design in the neoadjuvant setting. If the pre-surgical ctDNA association with distant recurrence-free interval holds in larger prospective cohorts, the case for incorporating tumour-informed MRD testing into standard neoadjuvant TNBC protocols becomes substantially stronger. Regulatory bodies would then face pressure to define what an MRD-informed treatment decision looks like in terms of labelling requirements and clinical practice guideline incorporation. That process, from conference data to guideline inclusion, typically takes three to five years in oncology, meaning the AACR 2026 data represents a relatively early signal in what will be a protracted adoption curve.
For Exact Sciences, the strategic significance of the four AACR presentations lies less in any single data point and more in the consistent accumulation of evidence across multiple indication types and test platforms. The company is presenting simultaneously in breast cancer MRD, colorectal genomic profiling, and multi-cancer early detection, all under a unified diagnostic portfolio argument. Whether that breadth becomes a strength in commercial execution or stretches the company’s evidence-generation and commercial resources across too many simultaneous development pathways is a question that the next twelve months of sales data, coverage decisions, and conference disclosures will begin to answer.
How Exact Sciences is building a multi-platform cancer diagnostics case at AACR 2026 added by Pallavi Madhiraju on
View all posts by Pallavi Madhiraju →