Aspirin did not prevent recurrence after colorectal liver metastasis surgery

Daily aspirin did not reduce cancer recurrence in people who had undergone surgery or other curative-intent treatment for colorectal cancer that had spread to the liver, according to the Scandinavian phase 3 ASAC trial [s1]. If anything, the signal ran the wrong way: survival at three years was lower in the aspirin group than in the placebo group — a sobering result for a cheap, widely available drug that many had hoped could hold cancer at bay [s1].

Aspirin has a long and genuinely encouraging track record in colorectal cancer, where observational data and some randomised evidence suggest it can lower the risk of developing the disease and, in certain tumours, of recurrence. ASAC asked a narrower, tougher question: once the liver metastases have been treated with intent to cure, does adding aspirin keep the cancer from coming back? The answer, in this trial, was no.

What the trial did

ASAC (NCT03326791) was a phase 3, randomised, double-blind, placebo-controlled trial run at 14 centres in Norway, Sweden, and Denmark [s1][s2]. It was funded by public and charitable bodies — the South-Eastern Norway Regional Health Authority, the Research Council of Norway, and the Norwegian Cancer Society — and sponsored by Oslo University Hospital, an independent academic trial with no pharmaceutical company standing to gain from a positive result [s1][s2].

Patients who had completed curative-intent treatment for colorectal cancer liver metastases were randomly assigned 1:1 to aspirin 160 mg per day or matching placebo, continued for up to three years or until recurrence, death, or stopping treatment [s1]. The primary outcome was disease-free survival. Between 15 December 2017 and 27 January 2022, 466 patients were randomised, of whom 428 started treatment and formed the full analysis set — 217 on aspirin and 211 on placebo [s1]. Their mean age was 62.2 years; 272 (64%) were male and 156 (36%) female [s1].

What it found

Aspirin did not improve disease-free survival. There were 269 disease-free survival events — 138 in the aspirin group and 131 on placebo — giving a hazard ratio of 1.08 (95.44% confidence interval 0.85 to 1.38, P=0.58) [s1]. Median disease-free survival was 1.13 years with aspirin (95.44% CI 0.92 to 1.44) and 1.40 years with placebo (1.11 to 1.69) — numerically shorter on the drug, though the difference was not statistically distinguishable [s1]. Time to recurrence was likewise no different, with an adjusted cause-specific hazard ratio of 1.06 (95% CI 0.83 to 1.35, P=0.64) [s1].

The overall-survival finding is the one that will draw attention. At 36 months, 76% of the aspirin group were alive (95% CI 69.4 to 82.7) versus 85% on placebo (79.5 to 90.8), a hazard ratio for death of 1.64 (95% CI 1.01 to 2.65, P=0.045) [s1]. Serious adverse events were also more common with aspirin, occurring in 17 patients (8%) of 217 versus four (2%) of 211 on placebo [s1].

How to read it

The disease-free survival result is clearly null: a confidence interval straddling 1 and a P value of 0.58 give no reason to think aspirin helped [s1]. The survival signal is harder to interpret and should be handled with care. A hazard ratio of 1.64 with a P value of 0.045 is nominally significant, but it sits on a secondary endpoint, the interval only just excludes 1, and a mortality difference not mirrored by any recurrence difference is biologically odd — the kind of result that can reflect play of chance or imbalances rather than true harm [s1]. The right summary is that aspirin did not help and may have carried net harm in this setting, not that it has been proven dangerous.

It is worth holding this result alongside aspirin's better news elsewhere. Coverage has described the ALASCCA trial, in which aspirin cut recurrence in colorectal cancers with specific PI3K-pathway mutations, and a trial of high-dose vitamin D in metastatic colorectal cancer. The contrast is the lesson: a drug that helps a biomarker-selected group early in the disease need not help an unselected group after metastases have already been treated. Readers weighing screening and prevention more broadly may also find our explainer on colorectal cancer screening and the move to start at 45 useful.

What to watch

ASAC does not close the book on aspirin in colorectal cancer; it narrows it. The open questions are whether tumour genetics identify patients for whom aspirin still earns its place, and whether its established role in prevention and in biomarker-selected early disease holds up as those trials mature [s1]. What this trial settles is the specific question it asked: routine daily aspirin after curative-intent treatment of colorectal liver metastases does not prevent the cancer from returning.

This article describes research and is not medical advice. Anyone taking or considering aspirin should discuss it with their doctor.

Sources

• Aspirin after curative-intent treatment of colorectal cancer liver metastases (ASAC) — The Lancet Gastroenterology & Hepatology, 23 September 2026
• ASAC trial registration (NCT03326791) — ClinicalTrials.gov