Sucharu Prakash, Director of Quality Services at Texas Oncology, shared a post on LinkedIn:
“What do the GRAIL trials results mean? The recent update from GRAIL on its MCED blood test has sparked strong reactions across oncology and the markets. In the 142,000-participant NHS Galleri trial, the study did not meet its primary endpoint of significantly reducing late-stage cancer diagnoses across the entire cohort. This does not mean that it’s a bad test… just that the trial’s scientific endpoint was not met.
Let me explain:
The trial showed that across 12 (hard to treat) cancers, screening reduced stage IV diagnoses and increased early-stage detection which is significant! Even more important is that these included cancers without any existing screening programs–ovarian, pancreatic, liver, esophageal etc.
In any other context, these results would be met with optimism, but we live in an age of headline and social media driven processes which makes it difficult to have a fundamental analysis or long-term planning.
- The Challenge of MCED Screening – Cancer screening trials measure mortality reduction or stage shift, not simply detection accuracy. Even a highly sensitive ctDNA assay can miss its endpoint if follow-up is too short or if there aren’t good treatments available for that cancer. In screening science, detection is only step one. Impact requires time + intervention + system readiness.
- Optimism with Biology – While the trial missed its primary endpoint, data showed reduction in stage 4 diagnoses across 12 high-mortality cancers, increased early-stage (I–II) detection and stronger effects with sequential screening rounds. This aligns with prior findings from GRAIL’s SYMPLIFY study, where performance improved significantly in higher-risk populations.
- Extended Follow up Needed – MCED takes a long time to show impact. Detection → Diagnostic workup → Treatment → Survival impact. Extending follow-up 6–12 months, as GRAIL plans while pursuing approval from the U.S. FDA may provide critical data.
- ctDNA Technology Still Needs to Evolve – ctDNA biology is complex. Signal fractions in early-stage disease are extremely low. Fragments, methylation signatures, tumor-informed modeling, and AI-based classification must continue to mature. Sensitivity for stage I cancers — particularly in low-shedding tumors like pancreatic or ovarian — remains a frontier.
- MCED is not a finished product – It’s a platform in development. I expect improved assay sensitivity, better tumor-of-origin prediction, integration with imaging and clinical risk models, and AI-driven longitudinal signal tracking.
- Increased Competition – The MCED space has many labs/companies with huge investments. We will likely see improved technology, stratification and system protocols – ctDNA continues to have the potential to redefine population oncology.”
Other articles featuring Sucharu Prakash on OncoDaily.