Ultrasensitive ctDNA Turned Positive Before Every Esophageal Recurrence in a 36-Patient Cohort

MAESTRO enriches plasma for a patient's own mutations

MAESTRO uses mutation-specific oligonucleotide probes to enrich a cell-free DNA library for a patient's own tumor mutations, so those mutations can be called accurately with far less sequencing than brute-force depth would require. Timothy Blewett, Rhoades and colleagues at the Broad's Gerstner Center for Cancer Diagnostics described the pooled version in Clinical Chemistry in February 2024: probes for every patient in a cohort are synthesized together and applied to every sample, which makes each patient's plasma a control for everyone else's. In that paper the method screened 22,333 tumor mutations from nine melanoma patients across 98 plasma samples, detected residual disease down to 0.78 parts per million, and returned a single positive across 784 patient-unmatched tests. Sequencing ran on an Illumina NovaSeq S4 at a target of 50 million read pairs per sample.

Oncodetect V2 is the commercial expression of the same chemistry. Exact Sciences holds exclusive rights to MAESTRO and said on May 28, 2025 that the next-generation whole-genome version would track up to 5,000 patient-specific variants at a limit of detection below one part per million, with availability in 2026. Viktor Adalsteinsson, director of the Gerstner Center and senior author on the esophageal paper, was quoted in that release:

The precision and sensitivity seen in the next generation test reflect deep scientific collaboration and a shared commitment to advancing MRD technology. This approach to innovation will continue to raise the bar for recurrence monitoring, treatment response assessment, and, ultimately, patient outcomes.

Hazard ratios and specificity never reach the abstract

Sixty-four percent of the recurrence group was positive at the first postoperative draw, and the authors report longer lead times to recurrence with Oncodetect V2 than with MAESTRO-Pool without attaching a figure in months. Their conclusion states that ctDNA detection after curative-intent therapy carried strong positive and negative predictive value, and that esophageal testing may benefit from assays reaching part-per-million sensitivity. Repeated requests to aacrjournals.org came back as 403 refusals, and AACR sells the full article. That left the published abstract, the method paper behind MAESTRO-Pool and the company's own product disclosures. Hazard ratios, specificity figures, plasma input volume, the per-patient panel design and the draw schedule are not in the abstract.

Nine of the authors work for the company selling the test

Blewett, who led the 2024 method paper from the Broad, lists Exact Sciences on this one, and the company byline runs on through Limbo, Ratty, Fu, Mazloom, Ramesh, Costa, Hu and Culver. Garces lists Abbott. The academic side sits at Dana-Farber, where Enzinger and Mamon anchor the clinical cohort, at the Broad, and at Brigham and Women's Hospital, which contributed two authors. Nothing in that arrangement is unusual for an assay validation study, and none of it is concealed. It does mean the comparison that favors the commercial test was run by a team that includes its makers.

Why This Matters to the APO|APE Reader

Exact Sciences' published clinical validation for Oncodetect, the Beta-CORRECT study of more than 400 patients, covers stage II through IV colorectal cancer, which leaves esophageal ordering outside the indication the company has taken to a peer-reviewed endpoint. The paper establishes how low the two assays can read, and stops there. Nothing in the abstract tells a tumor board what to do about a patient sitting at 20 parts per million four weeks after esophagectomy, and the authors point to no study that randomized treatment on that result. Until one exists, an ultrasensitive positive in this disease buys prognosis and a surveillance schedule, and the burden of proving that acting on it changes anything falls to the next trial.