Early disease detection is a key pillar of strategies that aim to reduce the global burden of cancer. Effective screening technologies, such as mammography for breast cancer or colonoscopy and fecal tests for colorectal cancer, have had a profound impact on patient treatment and survival. They have now been implemented in cancer-prevention public health programs across the globe, with proven impact in reducing mortality rates. A few blood-based cancer biomarkers are now in clinical use also, but either their use is limited to people at high risk or they have not translated into reduced mortality1. Successful screening strategies are still the exception. Detecting cancer early, at a population-level scale, remains a major unmet clinical need. Liquid biopsies could fill that gap.

Next-generation liquid biopsy technologies, in particular those that target the detection of cell-free DNA — which includes fragments of circulating tumor DNA (ctDNA) — in the blood plasma of patients, show the most promise. In the future, they might be able to offer a window into the detection of multiple types of cancer at once, circumventing a critical limitation of other screening approaches that are tumor type specific. One potential roadblock is that during early-stage disease, a very small amount of ctDNA is present in the bloodstream. This means that to work as an early-detection tool, ctDNA tests must be highly sensitive. Researchers are trying to address this problem by improving the analytical performance of ctDNA tests2 or slowing the clearance of ctDNA to facilitate its detection in the blood3.

Despite such advances, it is too early to say if ctDNA technologies will facilitate the detection of early-stage primary cancers. It is also unclear if highly specific testing can be done on a population-wide basis and cover multiple cancer types at once, and whether multi-cancer early detection (MCED) screening is an achievable goal at all. In 2023, researchers reported the results of the prospective observational cohort study PATHFINDER, designed to test if a targeted methylation-based MCED test could be used for cancer screening in people 50 years of age or older, excluding young adults4. The MCED test showed a promising 99.1% specificity for cancer detection, although it also led to 57 people with ‘false positive’ results being subjected to downstream diagnostic testing, including invasive procedures. A separate study investigating the same MCED test in a large symptomatic population yielded an overall specificity of 98.4% and sensitivity of 66.3% that was lower for early-stage disease than for later-stage disease5.

Such findings raise questions not only about the current performance of ctDNA tests that are in clinical development for the detection of early-stage cancers but also about their cost-effectiveness and their potential impact on overdiagnosis and overtreatment. Prospective interventional trials, some of which are ongoing, are needed to determine the extent of benefit and harm to patients that would be achieved by such testing. Choosing the most appropriate endpoints in such trials is also a point of contention, as the obvious choice — mortality specifically due to cancer — can take years to determine6. In November 2023, an advisory panel to the US Food and Drug Administration agreed that stage shift (meaning more people are diagnosed at an earlier stage) or disease state (meaning whether or not curative treatment is available) could be used as potential surrogate endpoints for cancer-specific mortality7. That is a huge gamble, given the fact that surrogate endpoints such as event-free survival do not necessarily correlate with overall survival in each cancer type, and could instead increase the risk of overtreatment, which can be detrimental to the patient and have downstream effects on the sustainability of health systems as a whole.

The US Food and Drug Administration panel also recommended that after a positive test result, quick and comprehensive cancer-specific workups and educational tools and resources should be provided as guidance for patients and clinicians, in acknowledgment that time to diagnosis, as well as the long-term implications thereof, are health equity issues7. After all, a history of cancer affects people at the personal, health, economic and professional levels for the rest of their lives. These issues are becoming even more pressing, given the increase in cancer diagnosis earlier in life among young adults8.

Whatever their potential, MCED liquid biopsies, even if technically feasible, might not be the best fit for every scenario. For example, in the context of early-onset cancer, screening may need to be focused on testing for the most prevalent cancer types to avoid overdiagnosis and harm in young adult populations. Risk-based stratification will continue to be key, with screening strategies tailored to age brackets and overall disease risk level due to biological and environmental factors. Blood-based testing also may not be the most effective way to detect ctDNA in every tumor type. Indeed, studies of liquid biopsies of urine (for bladder cancer) or breast milk (for breast cancer) have shown promising results1. Multimodal approaches that include protein- or metabolite-based analyses in addition to ctDNA may further increase the level of confidence in the test results.

The ultimate goal of limiting the number of life years lost to premature mortality and suboptimal health due to cancer may still remain elusive. What is certain is that without better screening platforms, that goal cannot be met. Liquid biopsies may finally help turn the tide of cancer prevention.