Chemoradiotherapy May Unlock Immunotherapy in Hard-to-Treat Rectal Cancer

For most patients with colorectal cancer, one of the most powerful weapons in modern oncology simply does not work. Immune checkpoint blockade, the class of drugs that has revolutionized treatment of many advanced cancers by releasing the brakes on T cells, produces spectacular and durable responses in the small subset of tumors that are mismatch repair-deficient. But the overwhelming majority of colorectal cancers, roughly 95 percent of metastatic disease, are mismatch repair-proficient, and in these tumors the drugs have shown no meaningful benefit in the advanced setting. Mismatch repair-proficient status has become shorthand for immunotherapy resistance. A new review published in Annals of Gastroenterological Surgery argues that this resistance may not be absolute, and that neoadjuvant chemoradiotherapy, the standard pre-surgical treatment for locally advanced rectal cancer, could be the key that unlocks immune checkpoint blockade responsiveness in tumors long considered immunologically cold.

The contrast between the two molecular subtypes could hardly be starker. In mismatch repair-deficient colorectal cancer, which carries high tumor mutational burden and abundant neoantigens, checkpoint blockade has redefined treatment at every stage of disease. In the advanced setting, median overall survival reaches 77.5 months, an extraordinary figure for metastatic disease. When given before surgery, neoadjuvant immunotherapy achieves pathological complete response rates of 60 to 68 percent, and in one landmark trial of mismatch repair-deficient rectal cancer, every one of 49 patients who completed treatment with the PD-1 antibody dostarlimab achieved a clinical complete response, allowing them to avoid surgery altogether. Mismatch repair-proficient tumors, by contrast, are microsatellite stable, carry low mutational burden, and tend to harbor non-inflamed, immunologically cold microenvironments with few neoantigens to target. In metastatic disease, checkpoint inhibitors have repeatedly failed in this population.

Yet a curious pattern has emerged in the neoadjuvant setting. In the NICHE study of early-stage colon cancer, neoadjuvant nivolumab plus ipilimumab produced responses even in mismatch repair-proficient tumors: an expanded cohort showed responses in 8 of 31 patients, including pathological complete responses in 3 of 30, and one responder maintained an ongoing clinical complete response without surgery. Comparable data using immunotherapy alone do not exist in rectal cancer, where neoadjuvant treatment is built around chemoradiotherapy rather than checkpoint blockade. But when the two modalities have been combined in mismatch repair-proficient rectal cancer, the results have caught researchers’ attention. The single-arm VOLTAGE and NECTAR trials reported pathological complete response rates of 30 to 40 percent with regimens combining standard fluoropyrimidine-based chemoradiotherapy with PD-1 blockade, well above what is typically seen with chemoradiotherapy alone.

The most informative evidence comes from randomized trials that tease apart how immunotherapy should be timed relative to radiation. In the three-arm POLARSTAR trial, conducted in a predominantly mismatch repair-proficient population, sequential PD-1 blockade delivered after completion of chemoradiotherapy significantly increased pathological complete response compared with chemoradiotherapy alone, reaching 32.7 percent versus 14.0 percent. The concurrent arm, in which the PD-1 antibody was given during radiotherapy, reached 27.1 percent, a numerically smaller increase that did not reach statistical significance. Meanwhile, the randomized CHINOREC trial found that adding dual checkpoint blockade with ipilimumab and nivolumab concurrently to chemoradiotherapy did not improve complete response at all, with 22 percent versus 30 percent for chemoradiotherapy alone. A separate randomized trial of sintilimab showed that single-agent PD-1 blockade added to intensified CAPOX-based chemoradiotherapy raised the composite complete response rate from 26.9 percent to 44.8 percent. Taken together, these trials suggest that the benefit depends critically on how and when checkpoint blockade is combined with radiation, with sequential delivery after chemoradiotherapy emerging as the most consistently successful strategy.

Why might immunotherapy succeed before surgery in some mismatch repair-proficient tumors yet fail once disease has advanced? The review proposes a two-level framework. At the level of disease setting, conditions that support checkpoint blockade responsiveness may simply be better preserved before surgery. Tumor-draining lymph nodes have emerged as crucial sites of the immune response to PD-1 and PD-L1 blockade: they harbor progenitor-exhausted CD8-positive T cells, stem-like cells marked by expression of the transcription factor TCF1, which expand after treatment and generate effector progeny that traffic to the tumor. Studies in head and neck cancer suggest that metastatic involvement of lymph nodes disrupts this progenitor-exhausted T cell compartment, and advanced disease additionally carries a larger tumor burden, deeper T cell exhaustion, and a more immunosuppressive microenvironment. Early-stage disease, treated before these defenses collapse, may offer checkpoint blockade a fighting chance that metastatic disease does not.

The second level concerns variation within the neoadjuvant setting itself. Checkpoint blockade amplifies antitumor immunity that already exists rather than creating it from nothing, so the immune state of a tumor before treatment constrains what the drugs can achieve. In the expanded mismatch repair-proficient NICHE cohort, responses occurred despite uniformly low tumor mutational burden and instead tracked with features of the pre-treatment immune state, including higher expression of TCF1, while a fibrotic, TGF-beta-rich stroma marked many non-responders. This dissociation between mutation burden and response is a recurring theme across gastrointestinal cancers. The implication is that immunogenicity in these tumors is not a binary property but a continuum, and shifting that continuum upward, through radiation or other means, could push more patients across the threshold of pathological complete response.

This is where chemoradiotherapy enters the picture with a plausible biological mechanism. Radiation can trigger immunogenic cell death, releasing tumor antigens and danger signals that enhance antigen presentation and prime antitumor T cells, providing a rationale for greater responsiveness to subsequent checkpoint blockade. The POLARSTAR timing signal fits this biology: sequential PD-1 blockade after chemoradiotherapy was the strategy that reached significance, consistent with the idea that radiation first remodels the tumor immune microenvironment and immunotherapy then capitalizes on the newly inflamed terrain. The review argues that these findings justify prospectively testing whether chemoradiotherapy can induce, and thereby expand, checkpoint blockade responsiveness, rather than assuming that higher pathological complete response rates merely reflect the additive cytotoxic effects of two independent treatments. Because pathological complete response cannot distinguish induced responsiveness from simple additivity, the authors call for larger trials that compare biomarkers of immunotherapy responsiveness before and after chemoradiotherapy within the same patients.

Pathological complete response itself carries important caveats. Patients who achieve it do have better long-term outcomes, but treatment-induced increases in pathological complete response have not been shown to translate into improved five-year survival, and response at surgery is not equivalent to the sustained clinical complete response required for organ-preserving watch-and-wait strategies. The safety of omitting surgery depends on local regrowth and salvage success: in a meta-analysis of watch-and-wait series, ultimate local failure was 5 percent overall but rose to 24 percent among patients who experienced regrowth. Long-term follow-up will therefore be essential to determine whether higher response rates genuinely improve disease control and whether organ preservation after a sustained clinical complete response is safe in mismatch repair-proficient disease.

Because treatment intensification exposes all patients to additional therapy, the review emphasizes biomarker-based selection as the path forward. The leading candidate is a pre-treatment T-cell-inflamed phenotype, assessed by Immunoscore or by measuring CD3-positive and CD8-positive T cell density. In the single-arm SILAR trial, enrolling only patients with intermediate or high pre-treatment Immunoscore and treating them with intensified mFOLFOX6-based chemoradiotherapy plus sintilimab produced a striking pathological complete response rate of 65.2 percent, although without a low-Immunoscore comparator the result reflects patient selection and cannot validate Immunoscore as a predictive biomarker. T cell density has independently been associated with response to chemoradiotherapy in earlier rectal cancer studies. Other signals include higher pre-treatment TCF1 expression in responders and TGF-beta-rich stroma in non-responders, along with an exploratory association between PD-L1 combined positive score and sintilimab benefit. Interferon-gamma and antigen-presentation signatures remain plausible but untested candidates. None is yet a clinically validated selection assay, and the authors argue that developing and prospectively validating such an assay, ideally assessed at the post-chemoradiotherapy, pre-immunotherapy time point when treatment selection actually occurs, is the central task ahead.

The message of the review is ultimately one of cautious optimism. Resistance to checkpoint blockade in mismatch repair-proficient rectal cancer is not the absolute wall it once appeared to be: selected neoadjuvant regimens combining chemoradiotherapy with immunotherapy have produced pathological complete response rates substantially higher than expected from chemoradiotherapy alone, and randomized evidence now pinpoints sequential PD-1 blockade after radiation as the most promising timing strategy. Whether these gains reflect radiation-induced reprogramming of the tumor immune microenvironment or simply additive effects remains an open and testable question, one that within-patient biomarker comparisons before and after chemoradiotherapy can directly address. If the induction hypothesis holds, the implications would extend far beyond rectal cancer, offering a strategy to convert immunologically cold, microsatellite-stable tumors, the majority of colorectal cancer, into tumors that respond to some of the most effective drugs in the oncology arsenal. For the vast majority of colorectal cancer patients who have so far been excluded from the immunotherapy revolution, that possibility represents a genuinely transformative horizon.

Subject of Research: Whether neoadjuvant chemoradiotherapy can induce immune checkpoint blockade responsiveness in mismatch repair-proficient rectal cancer

Article Title: Can Chemoradiotherapy Expand Immune Checkpoint Blockade Responsiveness in Mismatch Repair‐Proficient Rectal Cancer?

Article References: Yamashita, K., Matsuda, T., Mukohyama, J., Sasaki, R., & Kakeji, Y. (2026). Can Chemoradiotherapy Expand Immune Checkpoint Blockade Responsiveness in Mismatch Repair‐Proficient Rectal Cancer?. Annals of Gastroenterological Surgery, Article ags3.70281. https://doi.org/10.1002/ags3.70281

Image Credits: AI Generated

DOI: 10.1002/ags3.70281

Keywords: rectal cancer, colorectal cancer, immune checkpoint blockade, chemoradiotherapy, mismatch repair-proficient, neoadjuvant therapy, pathological complete response, PD-1 blockade, Immunoscore, organ preservation, tumor microenvironment, biomarkers