The narrative around liquid biopsy in early cancer detection has had a rough couple of years. The headline trials did not produce the sensitivity numbers that the field hoped for. Reimbursement coverage decisions were slow and, in some cases, discouraging. Public market valuations for the major liquid biopsy companies fell substantially from peak. Investor sentiment, which had been unusually buoyant, recalibrated.
We have not changed our fundamental view. We remain invested in liquid biopsy diagnostics and would invest again at the right entry point. Here is the reasoning.
What the Disappointments Actually Mean
The clinical trial results that drew the most skepticism showed multi-cancer early detection tests with sensitivities in the 50-70% range at high specificity. For Stage I cancers specifically, the numbers were lower. Critics correctly noted that a test missing roughly half of early cancers is a different value proposition than what was implied during the hype period.
That criticism is fair. It does not, however, mean the technology is not valuable. It means the technology has a specific performance profile that needs to be matched to the right clinical context. A screening test that detects 60% of early cancers with a false positive rate below 1% still saves lives if deployed in populations with elevated cancer risk — the question is whether that performance profile is good enough to justify the cost and the downstream workup of positives.
For some cancer types, the sensitivity numbers are much better than the aggregate suggests. Ovarian cancer, certain lymphomas, and colorectal cancer have shown consistently better early detection sensitivity with ctDNA-based approaches than the overall multi-cancer numbers imply. The tests are not equally good at finding all cancers early — and understanding which cancers drive the most clinical value is the real analytical work that the market sometimes elides.
The useful frame is not "does liquid biopsy work" — it clearly does for specific applications. The useful frame is "for which cancer type, in which population, at what clinical time point, does the performance profile justify the test."
Where We Think the Durable Value Is
We are more bullish on minimal residual disease monitoring and treatment response assessment than on general population multi-cancer screening, at least in the near term. MRD applications have cleaner clinical utility logic: you have a patient with a confirmed cancer diagnosis, you treat them, and you want to know whether circulating tumor DNA is detectable post-treatment as a signal of residual disease or early recurrence. That clinical question is well-defined, the patient population is identified, and there is strong correlation between ctDNA clearance and survival outcomes across multiple cancer types.
Several payers have already moved to cover ctDNA monitoring in specific oncology contexts. The reimbursement pathway there is clearer than it is for population screening, where the coverage question involves health economic modeling over decades and population-scale benefit estimates that are inherently uncertain.
Companion diagnostic applications also remain compelling. As targeted therapies require increasingly precise patient stratification — and as resistance mechanisms become more important to monitor during therapy — the ability to perform serial liquid biopsy at treatment initiation and during disease course has real clinical value that is already beginning to translate into coverage.
The Reimbursement Timeline
National coverage determinations for multi-cancer early detection are moving. The regulatory guidance that has come out of FDA review processes for the leading tests has been more constructive than the bear case anticipated. CMS is engaged in formal evaluation processes, and the evidence generation requirements — while substantial — are not unreasonable given the scope of what these tests claim to do.
Our estimate is that at least one multi-cancer early detection test achieves broad Medicare coverage within the next two to three years, most likely with initial indications in high-risk populations — smokers, patients with Lynch syndrome, patients with prior cancer history — rather than universal population screening. That is a meaningful market even before general population use is considered.
The companies that survive to that coverage decision will be the ones with the strongest health economic data, the largest prospective clinical datasets, and the most defensible specificity profiles. The field has thinned since the peak valuations, which is actually healthier for the companies still running rigorous trials.
Where We Are Looking
Our interest has shifted somewhat from pure-play early detection companies — where the public market leaders have established substantial data moats — toward companies that are building liquid biopsy into disease management workflows for specific cancer types with high unmet need. Pancreatic cancer, where early stage detection dramatically changes outcomes and current screening methods are inadequate, is one area we are tracking carefully. Rare cancers with limited tissue biopsy access, where liquid biopsy may become the primary diagnostic modality rather than a supplement, are another.
We also watch the multi-analyte space: companies combining ctDNA with protein biomarkers, methylation signatures, and other circulating analytes in integrated panels that exceed the sensitivity of any single modality alone. The biological rationale for multi-analyte approaches is strong, and some of the early validation data is encouraging.
The liquid biopsy story is not over. It has moved from a phase of exuberant expectations to a phase of evidence-driven clinical validation. That is how mature diagnostic technologies develop. We are comfortable in this phase and positioned accordingly.
