Can early cancer detection through liquid biopsy save lives?

How accurate are liquid biopsies for catching cancer early?

Liquid biopsies can detect early-stage cancers with high accuracy, often outperforming traditional blood markers. A large study of 2,092 patients using a spectroscopic blood test (Dxcover®) detected 64% of Stage I cancers when set to 99% specificity, meaning only 1% of healthy people would be falsely flagged [2]. When tuned for higher sensitivity, it caught 99% of Stage I cancers, though with more false positives (59% specificity) [2]. Another study using methylation patterns in cell-free DNA found sensitivities of 76–83% for Stage I–IV colorectal cancer and 44–79% for lung cancer, all at 99% specificity [4]. For gastric cancer, a 3-miRNA signature achieved an AUC of 0.85 for Stage I disease in a prospective cohort of 349 patients, meaning it correctly distinguished cancer from non-cancer 85% of the time [5].

These numbers are impressive, but they also show that no single test is perfect. The same methylation study missed about 22% of colorectal cancers and 34% of lung cancers at the highest specificity setting [4]. A 2025 review of multiple commercial liquid biopsy tests concluded that current data do not support their use for population screening due to false negatives and false positives [7]. So while liquid biopsies are powerful tools, they are best used in combination with other methods or in high-risk populations.

Can liquid biopsies find cancer before symptoms appear?

Yes, several studies show liquid biopsies can detect cancer at Stage I, often before symptoms arise. For gastric cancer, a long non-coding RNA (GClnc1) signature identified early-stage (Stage I/II) disease with an AUC of 0.94 in a discovery cohort of 2,141 patients, and it distinguished early gastric cancer from precancerous lesions [1]. Another gastric cancer study using a 5-miRNA signature achieved an AUC of 0.89 for Stage I disease in retrospective cohorts [5]. For early-onset colorectal cancer (EOCRC), a 4-miRNA panel identified patients with an AUC of 0.92 in training and 0.88 in validation, and it robustly detected early-stage EOCRC [3].

The key advantage is that these tests can detect cancer when it is still localized and potentially curable. For example, the gastric cancer GClnc1 test dropped to low levels after surgery, confirming its tumor specificity [1]. The spectroscopic test detected 64% of Stage I cancers at high specificity, meaning it could catch cancers that would otherwise be missed until later stages [2]. However, the 2025 review warns that small tumors shed very little DNA into the blood, so some early cancers will be missed [7]. This means liquid biopsies are promising for early detection but not yet a standalone screening tool.

What are the limitations and caveats of liquid biopsy for early detection?

Despite their promise, liquid biopsies have significant limitations. The most critical is that early-stage tumors often release very little DNA into the bloodstream, leading to false negatives. A 2025 analysis estimated that only a tiny fraction of the 3 liters of plasma is tested (3–4 mL), so the amount of tumor DNA retrieved may be insufficient for diagnosis, especially in small tumors [7]. This is why even the best tests miss 20–30% of early cancers [4].

Another limitation is false positives, which can cause unnecessary anxiety and invasive follow-up procedures. The spectroscopic test, when tuned for high sensitivity, produced false positives in 41% of non-cancer patients [2]. Additionally, many liquid biopsy tests are still expensive and not standardized across labs. A 2022 review noted that point-of-care biosensors are being developed to make testing cheaper and faster, but these are not yet widely available [6]. A 2025 review of breakthroughs and challenges also highlighted cost barriers and regulatory complexities as hurdles to widespread clinical use [8].

Finally, most studies here are on specific cancer types (gastric, colorectal, lung) and may not generalize to all cancers. The spectroscopic test covered eight cancers, but others focused on one or two types [2][4]. So while the evidence is strong for certain cancers, more research is needed for others.

About These Sources

This answer is built on 8 peer-reviewed studies — published from 2021 to 2025, 2 from 2024 or later, 7 in Q1 journals, collectively cited 342 times — selected as the most relevant from 8 studies that passed quality screening, drawn from 61 papers retrieved from a database of over 500 million.

Sources used in this answer

Related research questions