Bowel cancer immunotherapy pair keeps 3 in 4 alive at five years

Five years on, about three in four people with a particular type of advanced bowel cancer who were treated with two immunotherapy drugs together are still alive. The final overall survival results of the phase 3 CheckMate 8HW trial, presented by Sara Lonardi of the Veneto Institute of Oncology in Padua at the ESMO Congress 2026 and published online in Annals of Oncology on 24 October 2026, show 77% of people given nivolumab plus ipilimumab as their first treatment alive at five years, against 46% of those given chemotherapy.

That's an unusual figure for metastatic bowel cancer, and it applies only to a minority of patients: those whose tumours are MSI-high or mismatch repair deficient. The trial was funded by Bristol Myers Squibb, which makes both drugs.

What MSI-high means

Every time a cell divides, it copies its DNA, and a set of proofreading proteins, the mismatch repair system, fixes copying errors. In some tumours this system is broken, either through an inherited condition such as Lynch syndrome or through changes that happen in the tumour itself. These tumours are called mismatch repair deficient (dMMR). Because errors pile up, they show a distinctive pattern of instability in repetitive stretches of DNA known as microsatellites, so they're also called microsatellite instability-high (MSI-H). The two labels describe the same group.

About 5% of bowel cancers that have spread fall into this group. Their mass of mutations makes them look very foreign to the immune system, which is why immunotherapy works so well in them and so poorly in most other bowel cancers. Testing for MSI or mismatch repair status is now routine at diagnosis in the UK and many other countries.

What was tested

Nivolumab (brand name Opdivo) blocks PD-1, a brake on T cells. Ipilimumab (brand name Yervoy) blocks CTLA-4, a different brake that acts earlier in the immune response. Using both together releases more of the immune system, at the price of more side effects.

CheckMate 8HW randomised people with MSI-high or dMMR metastatic bowel cancer in a 2:2:1 ratio to nivolumab plus ipilimumab, nivolumab alone, or chemotherapy (with or without targeted drugs such as bevacizumab or cetuximab, chosen by the investigator). The chemotherapy arm was only open to people who hadn't had any treatment for metastatic disease. People who had already had treatment could be randomised between the two immunotherapy arms.

That design allowed three main comparisons:

1. Nivolumab plus ipilimumab against chemotherapy, as first treatment.
2. Nivolumab plus ipilimumab against nivolumab alone, across all lines of treatment.
3. Nivolumab plus ipilimumab against nivolumab alone, as first treatment.

Who took part

The trial randomised 839 people. Efficacy analyses were limited to those whose MSI-H or dMMR status was confirmed by a central laboratory, because local tests sometimes get it wrong. That gave 296 people on the combination and 286 on nivolumab alone across all lines, and, in the first-line comparisons, 171 on the combination, 170 on nivolumab alone and 84 on chemotherapy.

The trial is registered as NCT04008030 on ClinicalTrials.gov. Median follow-up for this final analysis was 68 months, more than five and a half years.

Results in plain numbers

Earlier analyses had already shown large gains in progression-free survival, which led to the combination being approved for first-line use in several regions. This final analysis was the planned test of overall survival.

Combination against chemotherapy, first treatment

Combination against nivolumab alone, all lines

Combination against nivolumab alone, first treatment

Sources: Lonardi et al., Annals of Oncology, 24 October 2026 (PubMed abstract); 4 and 5 year rates from Cancer Therapy Advisor's report of the ESMO presentation.

The headline is the comparison with chemotherapy. A hazard ratio of 0.37 means the risk of death was 63% lower with the combination. In absolute terms, about 31 more people in every 100 were alive at five years. Median survival on chemotherapy was a little over four years, which is better than many would expect, partly because people whose cancer progressed on chemotherapy could then receive immunotherapy.

The median survival with the combination hasn't been reached even after more than five years, meaning fewer than half of these patients have died.

Is ipilimumab worth adding?

This is the more practical question, since nivolumab alone (or the similar drug pembrolizumab alone) is also used first-line for MSI-high bowel cancer.

Across all lines of treatment, the combination clearly beat nivolumab alone on survival: a hazard ratio of 0.61 and 12 more people per 100 alive at five years.

In the first-line comparison, the direction was the same (HR 0.66, 77% against 66% at five years) but the result missed the strict statistical threshold the trial had set for this comparison, P below 0.006. The P value was 0.0432. The Annals of Oncology paper notes that this analysis was timed by protocol at three years or 243 events, whichever came first, and that fewer deaths than that had occurred, which limits its power. So for people starting first-line treatment, the extra survival benefit of ipilimumab over nivolumab alone is likely but not proven. The progression-free survival benefit (HR 0.68) supports it.

There's also an early dip to note. The survival curves for the two first-line immunotherapy arms crossed briefly around six months, because more people died in the first three months on the combination (10 against 2). The paper says only one of those deaths was due to drug toxicity, with the rest from disease progression or other causes.

Side effects

Adding ipilimumab increased side effects, as it does in other cancers. Chemotherapy was harder on patients than either immunotherapy option.

Sources: Annals of Oncology abstract; Cancer Therapy Advisor and OncoDaily reports of the ESMO presentation.

There were two treatment-related deaths with the combination (one from myocarditis, inflammation of the heart muscle, and one from pneumonitis, inflammation of the lungs) and one with nivolumab alone (pneumonitis). There were none with chemotherapy and no new treatment-related deaths during the extra follow-up.

Underactive thyroid is a common lasting effect. It usually means taking thyroid hormone tablets for life, which is manageable but permanent. Other immune-related effects, such as colitis, hepatitis or adrenal problems, are less common but can be serious, which is why people on these drugs are told to report new symptoms promptly.

Limitations

The trial was open-label, which can affect assessments of progression and side effects, though not survival itself.

The first-line comparison of the combination with nivolumab alone didn't reach statistical significance for survival, so the case for ipilimumab first-line rests partly on progression-free survival.

The chemotherapy group was small (84 people in the first-line analysis), so its survival estimates are less precise.

People in the chemotherapy group could later receive immunotherapy, which is right for patients but blurs the survival comparison. According to OncoDaily's report of the ESMO presentation, 67 of the 84 (80%) did.

Results apply only to MSI-H or dMMR tumours confirmed centrally. They say nothing about the 95% or so of metastatic bowel cancers that aren't MSI-high, where these drugs have little effect.

What it means for patients

For people with MSI-high or dMMR bowel cancer that has spread, CheckMate 8HW now gives long-term survival evidence for nivolumab plus ipilimumab as first treatment. Its advantage over chemotherapy is very large. Its advantage over nivolumab alone is smaller, more certain across all lines of treatment than first-line, and comes with more side effects.

In the UK, NICE has appraised nivolumab plus ipilimumab for previously treated MSI-high bowel cancer in the past. We couldn't confirm from the sources reviewed exactly what NICE currently recommends for first-line use, so check with your oncology team.

Questions worth asking:

• Has your tumour been tested for MSI or mismatch repair status, and was it confirmed?
• If it's MSI-high, whether to use one immunotherapy drug or two, and how the side effect trade-off looks for you.
• Whether Lynch syndrome testing is recommended for you and your relatives, since some MSI-high cancers are inherited.

This article explains published research and isn't medical advice. Talk to your own specialist team about treatment decisions.