Clinical Summary:

• Design/Population: The ongoing AGEO-NEO-MSI study evaluated real-world neoadjuvant immunotherapy in patients with non-metastatic mismatch repair-deficient or microsatellite instability–high colorectal cancer.
• Key Outcomes: Neoadjuvant immunotherapy produced high complete response rates, with pathological complete responses consistent with prior trials. Doublet immune checkpoint inhibition appeared more active but had higher toxicity.
• Clinical Relevance: Neoadjuvant immunotherapy is a promising approach in this setting, but treatment selection should be individualized to balance response potential with toxicity risk.

Annalice Gandini, IRCCS, Université Paris Cité, Paris, France, discusses results from the AGEO-NEO-MSI study evaluating real-world use of neoadjuvant immunotherapy in patients with non-metastatic mismatch repair-deficient or microsatellite instability (dMMR/MSI) colorectal cancer. The study examined the effectiveness and safety of single- and doublet-immune checkpoint inhibitor strategies outside the setting of prospective clinical trials.

The findings demonstrated complete response rates comparable to those reported in prospective studies while highlighting important differences in toxicity between treatment approaches. Ongoing translational analyses aim to identify biomarkers that can better guide patient selection and optimize the use of neoadjuvant immunotherapy in clinical practice. 

Dr Gandini presented these results at the European Society for Medical Oncology (ESMO) Gastrointestinal Cancers Congress in Munich, Germany. 

Transcript:

My name is Annalice Gandini, I'm a medical oncologist in Paris, France, and I'm also a PhD candidate at University Paris Cité. During the last ESMO GI Congress 2026, I had the opportunity to present as a rapid oral presentation the results of our study, AGEO-NEO-MSI, which evaluates real-world neoadjuvant immunotherapy in dMMR/MSI colorectal cancer.

In the last few years, we have observed an increasing number of trials evaluating neoadjuvant immunotherapy in dMMR/MSI colorectal cancer, but lots of questions are still open in particular regarding the applicability on large scale of this regimen and how to choose between the different treatment regimen on how to manage the risk of overtreatment of toxicity and also of potentially early progression with the risk of compromising cure.

We have designed the AGEO-NEO-MSI study, which is still ongoing. It's an ambispective, observational study that was conducted in more than 30 international centers and enrolled patients with dMMR/MSI colorectal cancer with no distant metastasis who were treated with neoadjuvant immunotherapy. The primary end point was complete response rate defined as pathological complete response or clinical complete response if patient underwent non-operative management. The secondary end points include event-free survival, safety, and pathological and surgical outcomes.

Between April 2019 and February 2026, we have enrolled 316 patients. Median age was 68 years old and 27% of the patients had Lynch syndrome. Regarding baseline clinical stage, most of the patients had stage 3 disease (75%). Patients were mostly treated with single-ICI (77%) and 23% of the patient received doublet-ICI. Non-operative management was more frequent when single-ICI was administered. At the end, complete response, both pathological and clinical, depending on the surgical management was achieved by 67% of the patients.

If we focus on patients who underwent surgery, pathological complete response was 66%, and we were actually not able to precisely predict pathological complete response. Indeed, lots of patients who were at the end in pathological complete response still presented disease at the imaging evaluation before surgery, both partial response or stable disease, or even in rare cases also progressive disease.

Regarding toxicity, we registered 35% of immune-related adverse events including 6% of grade ≥3 and 9% of tox-related discontinuation. Severe adverse events were more frequent when patients received doublet-ICI (12%) versus single-ICI (4%). Also, 1 toxic death occurred in the group of patients receiving doublet-ICI.

We conducted few exploratory analyses in trying to identify potential predictors of complete response and we saw that a doublet short strategy, meaning <1 month, was associated with higher rate of complete response when compared with a long single-ICI strategy. Immune-related adverse events seem to be associated with higher rates of complete response only in patients receiving mono-ICI, possibly related to the fact that if patients develop an adverse event during a doublet-ICI treatment, which is usually shorter, they do not receive the whole treatment and they undergo surgery earlier. This can potentially compromise the efficacy of this strategy.

In conclusion, we can say that our results, in terms of complete response, especially pathological complete response of 66% is absolutely in line with the available literature where range of pathological responses are around 65 to 70% depending on the studies. We also evaluated event-free survival that was still immature so we cannot conclude on this, but we have a 2-year event-free survival rate of 86%, which is lower than the reported event-free survival rate in literature. This can also be related to the fact that we have a real-world population. We had 1/4 of the patients who were not resectable at baseline. We have 6 (7%) patients with ECOG PS2 at the diagnosis. We have a real-world population which is less selective than patient-population trials. However, event-free survival, it's absolutely encouraging.

Doublet-ICI seem to increase complete response rate, but also toxicity. Probably, when possible, a doublet-ICI combination can be more efficient when compared with a long singlet ICI therapy, but we need to balance choice based on patient characteristics and comorbidities and general conditions of the patient. We must decide it case-by-case, we cannot generalize on this point. 

We have ongoing some translational analysis which can potentially help us in patient selection that would be helpful in our daily practice because for the moment we do not have still short answers. We hope that our work can help clinicians in managing patients with dMMR/MSI colorectal cancer with no distant metastasis. Thank you very much.

Source:

Gandini A, Overman MJ, Rasschaert GAC, et al. Treatment strategy and immune-related toxicity shape response to neoadjuvant immunotherapy in dMMR/MSI colorectal cancer: Results from the AGEO-NEO-MSI study. Presented at ESMO Gastrointestinal Cancers Congress. July 1-4, 2026. Munich, Germany. Abstract 3RO.