• None of the 32 patients had a cancer recurrence after nearly three years of monitoring.
• Protection held even for patients who still had small traces of disease remaining after treatment.
• The treatment course was short — up to nine weeks — compared with the longer schedules typical of conventional chemotherapy.

The durability matters as much as the response itself. A cancer that "melts away" on scans means little if it returns within months. Here, the follow-up window of nearly three years offers early but meaningful evidence that the responses are lasting.

Why would immunotherapy work here?

The genetic profile of these tumours is the key. Mismatch repair-deficient cancers accumulate large numbers of mutations as they grow. Each mutation can produce an abnormal protein that the immune system recognises as foreign — like a burglar leaving fingerprints at every window.

Pembrolizumab works by blocking a molecular brake called PD-1 on immune cells. Once that brake is released, T cells (the immune system's killing forces) can find and destroy tumour cells carrying those fingerprints. The drug does not attack the cancer directly; it persuades the patient's own immune system to do the work, which helps explain why responses can persist long after treatment stops.

How confident should patients be?

Caution is warranted, and the trial's own design signals that.

Thirty-two patients is a small sample. A single-arm study of this size cannot rule out that some recurrences would appear with longer follow-up, nor can it directly compare outcomes against standard chemotherapy in a matched control group. The findings also apply only to the subset of bowel cancer patients whose tumours carry mismatch repair deficiency — a minority of cases overall.

Whether the same approach benefits patients with the more common, genetically stable form of bowel cancer remains an open question that this trial was not designed to answer.

What happens next?

The results will need confirmation in larger trials before pre-surgery pembrolizumab becomes standard care for this group. If those studies reproduce these findings, the practical change could be substantial: a nine-week infusion course before surgery, instead of months of post-operative chemotherapy, for patients whose tumours carry the relevant genetic marker.

For now, the trial offers something rare in oncology — a clean result, a plausible mechanism, and patients who remain disease-free three years on. Zero recurrences in 32 patients is a small number written in large letters.