Rectal dMMR: dostarlimab and complete clinical response

A 38-year-old physical education teacher has a rectal adenocarcinoma 6 centimeters from the anal verge, staged as cT3N1 on MRI, with no metastases. Immunohistochemistry showed loss of MLH1 and PMS2 expression, which indicates mismatch repair deficiency. She read on the internet that an immunotherapy drug made rectal cancer disappear in every patient in a study and asks whether she can use this drug instead of radiotherapy and surgery. She wants to become pregnant in the future. The scene is fictional and was written for teaching.

This text uses the case as a clinical reasoning scenario for students and health professionals. The question comes from Dr. Alexei's list for the oncology series: what did PD-1 blockade with dostarlimab change in rectal cancer with mismatch repair deficiency, and what has it not yet proven? The reference study is the phase 2 trial at Memorial Sloan Kettering, published in 2022, with its follow-ups and with series from other centers. The text does not replace your country's guideline or your institution's protocol. If you play the case of the day on NEXO, you will recognize the challenge: interpret an exceptional result without confusing complete response with proven cure.

Key points

• About 5 to 10 in every 100 rectal adenocarcinomas have mismatch repair deficiency.
• These tumors respond poorly to neoadjuvant chemotherapy, but very well to PD-1 blockade.
• In the phase 2 study with dostarlimab, all 12 initial patients had a complete clinical response and did not need radiotherapy or surgery.
• A complete clinical response is not the same as a proven cure, and follow-up is still short for long-term outcomes.
• Not every patient has a complete response, and rigorous follow-up is mandatory.
• Every rectal cancer should have mismatch repair assessed before starting treatment.

Why this is a reasoning scenario

The standard treatment for locally advanced rectal cancer combines chemotherapy, radiotherapy and surgery, with effects that can last a lifetime: bowel, urinary and sexual dysfunction, infertility and, in some patients, a permanent stoma. For a young woman who wants to become pregnant, pelvic radiotherapy and surgery carry special weight.

The teacher in the scenario has a tumor with mismatch repair deficiency, the subgroup in which PD-1 blockade produced unprecedented responses. At the same time, she arrives with an enthusiastic reading of a small study. The consultation needs to recognize the advance and explain honestly what is still uncertain.

The post about total neoadjuvant therapy in rectal cancer covers tumors with preserved repair. The post about why the multidisciplinary team differs in the colon and the rectum shows who takes part in the decision.

The mistake the scenario aims to prevent

The most serious mistake is starting standard total neoadjuvant therapy without knowing the mismatch repair status, losing the chance of a more effective and less mutilating treatment. The opposite mistake is presenting immunotherapy as a guaranteed cure. A third mistake is skipping rigorous follow-up after the complete response.

What mismatch repair deficiency is

Mismatch repair proteins, such as MLH1, MSH2, MSH6 and PMS2, correct errors that arise when DNA is copied. When one of them is missing, errors accumulate, especially in repetitive sequences, which produces microsatellite instability. The tumor then has many mutations and many abnormal proteins, which the immune system can recognize.

How frequent it is in the rectum

About 5 to 10 in every 100 rectal adenocarcinomas have mismatch repair deficiency, according to the study article with dostarlimab (Cercek et al., 2022). A 2025 review cites about 10 in every 100 (Loufopoulos et al., 2025). In a Chinese cohort of 854 patients with locally advanced rectal cancer, deficiency was found in 63, or 7.4 in every 100 (Wu et al., 2022). Lynch syndrome accounts for 3 to 5 in every 100 colorectal cancers with microsatellite instability, according to a 2022 commentary (Mi et al., 2022).

How it is assessed

Assessment can be done in two complementary ways:

• Immunohistochemistry for MLH1, MSH2, MSH6 and PMS2 on the biopsy, which shows loss of one or more proteins.
• Molecular microsatellite instability testing by polymerase chain reaction or sequencing.
• Testing for MLH1 promoter methylation or BRAF mutation when there is loss of MLH1, to separate sporadic cases from possible Lynch syndrome.
• Genetic counseling and germline testing when suspicion of Lynch syndrome remains.

Why chemotherapy works worse

The study article with dostarlimab notes that rectal tumors with repair deficiency respond poorly to standard chemotherapy regimens, including neoadjuvant chemotherapy (Cercek et al., 2022). In the Chinese cohort, the proportion of grade 0 to 1 tumor regression was 28.6 in every 100 with repair deficiency and 43.7 in every 100 with preserved repair (Wu et al., 2022). Among patients treated with chemotherapy alone, after propensity score matching, the difference was 9.1 versus 30.3 in every 100 (Wu et al., 2022).

The prognosis remains good

The same cohort showed that, despite the worse response, repair deficiency was associated with longer disease-free survival, with a hazard ratio of 0.38 (Wu et al., 2022). Another Chinese cohort, with 269 patients treated with neoadjuvant chemoradiotherapy, showed better overall survival in the group with repair deficiency, even after matching (Chen et al., 2022). The combination of poor response to chemotherapy and good response to immunotherapy explains why this subgroup came to be treated differently.

The study with dostarlimab

The phase 2 study by Cercek et al. included patients with locally advanced rectal cancer and mismatch repair deficiency, without prior treatment (Cercek et al., 2022). Dostarlimab, an antibody against PD-1, was given at a dose of 500 mg intravenously every 3 weeks for 6 months, or 9 cycles (Cercek et al., 2022). The protocol then planned chemoradiotherapy with 5,040 cGy in 28 fractions and capecitabine, followed by total mesorectal excision, but those with a complete clinical response would continue without these treatments (Cercek et al., 2022).

What the study showed

All 12 patients who completed treatment and had at least 6 months of follow-up had a complete clinical response, with a confidence interval of 74 to 100 in every 100 (Cercek et al., 2022). Response was defined by the absence of tumor on magnetic resonance imaging, positron emission tomography, endoscopy, digital rectal examination and biopsy (Cercek et al., 2022). No patient had received chemoradiotherapy or surgery, none had progression or recurrence during follow-up of 6 to 25 months, and there were no grade 3 or higher adverse events (Cercek et al., 2022). The authors concluded that longer follow-up is needed to assess the duration of response (Cercek et al., 2022).

The update

A 2025 review describes the update by the same authors, with 41 patients who completed the 6 months of dostarlimab, all with a complete clinical response, without additional treatments and without recurrences or metastases (Loufopoulos et al., 2025). Of these, 20 maintained a sustained complete response at a median follow-up of 28.9 months from the start (Loufopoulos et al., 2025).

Series	Patients	Treatment	Main result
Cercek et al., 2022	12 with locally advanced rectal cancer	Dostarlimab for 6 months	12 of 12 with complete clinical response
STAR(t)-IT-RE, 2024	17 with locally advanced rectal cancer	Checkpoint inhibitor for 6 months	16 of 17 with complete clinical response
Chinese series, 2025	26 with locally advanced colorectal cancer	Immunotherapy followed by surgery	24 of 26 with complete pathological response

Source of the table numbers: Cercek and colleagues, 2022, Tosi and colleagues, 2024, and Luo and colleagues, 2025 (Cercek and colleagues, 2022) (Tosi and colleagues, 2024) (Luo and colleagues, 2025).

What other series show

In the Italian STAR(t)-IT-RE study, 16 of 17 evaluable patients had a complete clinical response, 94.1 in 100 in the intention-to-treat analysis (Tosi and colleagues, 2024). Three patients stopped before 6 months, due to immune pneumonitis, social reasons or non-oncological bowel obstruction, and there were two grade 3 adverse events (Tosi and colleagues, 2024). Median follow-up was only 9.5 months (Tosi and colleagues, 2024).

When surgery is performed

In a Chinese series of 26 patients operated after neoadjuvant immunotherapy, complete pathological response was 92.3 in 100, and immune adverse events occurred in 30.8 in 100 (Luo and colleagues, 2025). In another series, from Sichuan, 15 of 19 operated patients had complete pathological response (Deng and colleagues, 2024). In the NICHE study, with ipilimumab and nivolumab before surgery in colon cancer, all 20 mismatch repair deficient tumors had pathological response (Zhang and colleagues, 2022).

When the response is not complete

In STAR(t)-IT-RE, one patient with a near-complete response was operated, and the specimen showed minimal intramucosal residual tumor (Tosi and colleagues, 2024). Another patient had possible lung metastases, not yet confirmed (Tosi and colleagues, 2024). These findings show that the response is not universal outside a small, selected study.

Short regimens and surgery alone

A 2021 report described a 33-year-old man with Lynch syndrome and locally advanced rectal cancer who refused chemoradiotherapy due to the risk of sexual dysfunction (Trojan and colleagues, 2021). He received ipilimumab 1 mg/kg on day 1 and nivolumab 3 mg/kg on days 1 and 15, had a complete clinical response on day 21 and complete pathological response in the total mesorectal excision performed on day 30 (Trojan and colleagues, 2021). A single report does not define practice, but it shows how the topic is evolving.

For those without access to immunotherapy, a 2025 Chinese multicenter analysis of 119 patients with mismatch repair deficiency compared surgery alone, surgery with chemotherapy and surgery with chemoradiotherapy (Wang and colleagues, 2025). Surgery alone had better overall survival, progression-free survival and local control than surgery with chemoradiotherapy, and the authors suggest it may be an option when immunotherapy cannot be used (Wang and colleagues, 2025).

Complete response is not proven cure

Complete clinical response means that the available tests find no tumor. It does not prove that no tumor cells remain, and it is not the same as complete pathological response, which can only be seen in the surgical specimen. A 2022 commentary recalls that the original study had no pathological response data, that imaging response does not appear to predict pathological response accurately and that follow-up was short (Mi and colleagues, 2022).

The limitations that matter

The same commentary listed the limitations of the original study (Mi and colleagues, 2022):

• Small sample, from a single center.
• No separate analysis between Lynch syndrome and sporadic cases.
• No pathological response data.
• Short median follow-up.
• Response as the primary endpoint, with no 3-year disease-free survival data.

Does it need a randomized trial?

A 2022 methodological commentary discussed whether such a large effect would make randomization unnecessary, using the parachute analogy (Powell et al., 2022). The authors note that randomized studies reduce bias and are essential for modest effects, but may be less important when the effect is enormous, and they proposed an alternative design to evaluate dostarlimab (Powell et al., 2022). The discussion shows that the strength of the evidence depends not only on the size of the effect, but also on the length of follow-up.

For the professor in the scenario, the honest message is that the chance of a complete response is very high and that many patients avoided radiotherapy and surgery, but that long-term data are still being gathered. The word cure should be avoided until follow-up is long enough. It is also fair to say that, if the response is not complete, radiotherapy and surgery remain available, as the original study protocol foresaw, which planned chemoradiotherapy and surgery for those who did not have a complete response (Cercek et al., 2022).

How to explain it to the patient

The conversation with the professor has three tasks: show why her tumor is different, present the numbers at their true size and agree on a plan for each possible result. The first task is the simplest. The biopsy test showed that the tumor has a DNA repair defect, which leaves it full of mutations and more visible to the immune system.

The numbers at their true size

The original study had 12 patients, and the confidence interval of the response rate ranged from 74 to 100 per 100 (Cercek et al., 2022). This means that, even with 12 responses out of 12, the true rate may be lower than the observed one. The methodological commentary notes that, in an analysis of 822 articles that cited the parachute analogy, 35 compared a medical practice to a parachute and 22 of those practices had already been tested in randomized studies (Powell et al., 2022). The lesson is that effects that seem obvious are not always confirmed.

Other studies with PD-1 blockade in the colon and rectum show high responses, but not identical ones. In the Danish study with pembrolizumab before surgery in 42 patients with colon cancer and repair deficiency, the pathological complete response was 46 per 100, according to the 2025 review (Loufopoulos et al., 2025). Differences in organ, drug, duration and response criteria explain part of this variation.

A plan for each result

In the original study, response was assessed with endoscopy and digital rectal examination at baseline, at 6 weeks, 3 months and 6 months, and then every 4 months, in addition to MRI, positron emission tomography and CT of the chest, abdomen and pelvis (Cercek et al., 2022). Agreeing on this schedule before starting keeps the patient from seeing each exam as a surprise. If the response is complete, observation follows. If it is incomplete, the chemoradiotherapy and surgery foreseen in the protocol come in (Cercek et al., 2022). If there is progression, the team decides without delay.

It is also worth explaining that immunotherapy has its own adverse effects, different from those of chemotherapy, and that they can appear months after the start. In a Chinese series, immune adverse events occurred in 30.8 per 100 patients (Luo et al., 2025). The patient should know which symptoms to report right away, such as diarrhea, shortness of breath, intense fatigue or skin changes.

How to follow up with the patient

Follow-up after a complete clinical response must be as rigorous as that of the watch-and-wait strategy after total neoadjuvant therapy. A 2025 review described a series of 24 patients under observation after response to immunotherapy, with no local regrowth or metastasis over a median follow-up of 29.1 months (Loufopoulos et al., 2025). The review also highlights that organ preservation maintains bowel, urinary and sexual function, which is important for young patients (Loufopoulos et al., 2025).

The practical steps

A follow-up roadmap usually includes:

• Reassessment during treatment, to detect progression early.
• Digital rectal exam, endoscopy and pelvic MRI at short intervals after the response.
• CT of the chest and abdomen to look for metastases.
• Surveillance for immune adverse events, such as thyroiditis, colitis and pneumonitis.
• A clear plan for radiotherapy and surgery if there is an incomplete response or regrowth.

For the teacher, who wants to become pregnant, the discussion with the assisted reproduction team should start early, because the plan may change if the response is incomplete. The post about diagnostic uncertainty helps with conversations about decisions with short follow-up.

How NEXO helps train this reasoning

In NEXO, the case of the day is fictional, with a real clinical presentation and up to six progressive clues, and you propose the diagnosis by looking for the ICD-10 code. Rectal cancer with repair deficiency is a good exercise because the decisive clue is in a biopsy test that must be requested before any treatment. The post about differential diagnosis with clues shows how to train with progressive clues.

Three exercises with cases

Change the repair. Say what changes if immunohistochemistry shows preserved expression of all four proteins.

Change the response. Say what to do if, in the third month of dostarlimab, the MRI shows the tumor has grown.

Change the genetics. Say what to investigate if the loss is of MSH2 and MSH6 and the patient has a mother with endometrial cancer.

The post about clinical decision-making covers tests that change management. The limits of the game's use are on the NEXO economy page. If you research teaching of clinical reasoning, the page for researchers explains how to get in touch.

Frequently asked questions

What is the dostarlimab dose in the study?

500 mg intravenously every 3 weeks for 6 months, or 9 cycles (Cercek et al., 2022). Check the package insert and the approval in your country.

Do all patients respond?

In the original study, all 12 had a complete clinical response, but in the Italian study 16 of 17 responded completely (Cercek et al., 2022) (Tosi et al., 2024).

Does a complete clinical response mean cure?

No. It means absence of tumor on the tests, and long-term data are still limited (Mi et al., 2022).

Are the adverse effects serious?

In the original study, there were no grade 3 or higher events, but in another series there was interruption due to immune pneumonitis (Cercek et al., 2022) (Tosi et al., 2024).

Does chemotherapy still have a role?

Tumors with repair deficiency respond worse to chemotherapy, and radiotherapy and surgery remain an option if the response to immunotherapy is incomplete (Wu et al., 2022) (Cercek et al., 2022).

Final checklist

• I confirmed mismatch repair on the biopsy before starting treatment.
• I investigated Lynch syndrome when indicated.
• I discussed immunotherapy with the multidisciplinary team.
• I explained the difference between complete response and proven cure.
• I planned reassessments during and after treatment.
• I arranged surveillance for immune adverse events.
• I defined the plan if there is an incomplete response or progression.
• I referred for fertility preservation when the patient wishes.

Want to practice reasoning with progressive clues? Play the case of the day.

References

1. Cercek A, Lumish M, Sinopoli J, et al. PD-1 Blockade in Mismatch Repair-Deficient, Locally Advanced Rectal Cancer. N Engl J Med. 2022. PMID 35660797.
2. Chen L, Yang X, Zhang Y, et al. Survival outcomes analysis according to mismatch repair status in locally advanced rectal cancer patients treated with neoadjuvant chemoradiotherapy. Front Oncol. 2022. PMID 36003789.
3. Deng Z, Luo Y, Chen X, et al. Pathological response following neoadjuvant immunotherapy and imaging characteristics in dMMR/MSI-H locally advanced colorectal cancer. Front Immunol. 2024. PMID 39399495.
4. Loufopoulos P, Perivoliotis K, Chatziathanasiou D, et al. Immunotherapy-Induced Complete Response in dMMR Rectal Cancer-A Surgical Dilemma?. Cancers (Basel). 2025. PMID 41097680.
5. Luo Z, Wang D, Wang Z, et al. Is neoadjuvant immunotherapy feasible for patients with dMMR/MSI-H locally advanced colorectal cancer? a retrospective study. Front Immunol. 2025. PMID 40873558.
6. Mi M, Ye C, Yuan Y. Neoadjuvant PD-1 blockade: a promising nonoperative strategy for mismatch repair-deficient, locally advanced rectal cancer. Signal Transduct Target Ther. 2022. PMID 36216819.
7. Powell K, Olivier T, Prasad V. A novel trial methodology to test interventions with very large effect sizes: the case of dostarlimab in mismatch repair-deficient, locally advanced rectal cancer. Trials. 2022. PMID 36564825.
8. Tosi F, Salvatore L, Tamburini E, et al. Curative immune checkpoint inhibitors therapy in patients with mismatch repair-deficient locally advanced rectal cancer: a real-world observational study. ESMO Open. 2024. PMID 39357124.
9. Trojan J, Stintzing S, Haase O, et al. Complete Pathological Response After Neoadjuvant Short-Course Immunotherapy with Ipilimumab and Nivolumab in Locally Advanced MSI-H/dMMR Rectal Cancer. Oncologist. 2021. PMID 34431576.
10. Wang HH, Yan YY, Zeng HY, et al. The efficacy and safety of neoadjuvant and adjuvant chemo(radio)therapy combined with surgery in patients with locally advanced rectal cancer harboring defective mismatch repair system: a large-scale multicenter propensity score analysis. Front Immunol. 2025. PMID 40692776.
11. Wu Z, Hu H, Wang C, et al. The prognostic and predictive value of mismatch repair status in patients with locally advanced rectal cancer following neoadjuvant therapy. Ann Transl Med. 2022. PMID 35571401.
12. Zhang X, Wu T, Cai X, et al. Neoadjuvant Immunotherapy for MSI-H/dMMR Locally Advanced Colorectal Cancer: New Strategies and Unveiled Opportunities. Front Immunol. 2022. PMID 35371084.