Organ preservation after neoadjuvant immunotherapy in mismatch repair-deficient colorectal cancer: response assessment, watch-and-wait, and salvage surgery Authors: Nan Yao, Guoshuai Xu, Ning Duan, Fuzhou Han, Wenqiang Li, Jun Qu Published: 2026 ABSTRACT Neoadjuvant immune checkpoint blockade has produced high pathological and clinical response rates in mismatch repair-deficient (dMMR) or microsatellite instability-high (MSI-H) colorectal cancer, raising the possibility of response-adapted organ preservation. However, the evidence is anatomically uneven and remains immature for routine surgery avoidance. In dMMR colon cancer, neoadjuvant studies primarily establish drug activity through resection-confirmed pathological and nodal response; reliable criteria for nonoperative assessment have not been validated. In dMMR rectal cancer, clinical complete response (cCR) after PD-1 blockade may support watch-and-wait in highly selected patients because local response can be reassessed using digital rectal examination, endoscopy, pelvic MRI, biopsy when appropriate, systemic imaging, and tumor markers. This narrative review critically appraises the biological rationale, prospective and observational evidence, response-assessment limitations, surgical decision points, surveillance requirements, toxicity, and unresolved controversies surrounding organ preservation after neoadjuvant immunotherapy. We use the term immune-response-adapted surgical deferral to emphasize that surgery remains available for incomplete response, unresolved uncertainty, or regrowth. Current implementation should be confined to experienced multidisciplinary programs with rigorous molecular confirmation, standardized multimodal assessment, reliable follow-up, informed patient participation, and timely salvage capacity. Particular attention is given to immunotherapy-specific uncertainty in defining durable cCR, the limitations of circulating tumor DNA (ctDNA) as a stand-alone local decision tool, hereditary-risk considerations in Lynch syndrome, and the need for location-specific prospective validation. Introduction: from surgical radicality to immune-mediated organ preservation Colorectal cancer remains a major global health burden (1, 2). Curative-intent treatment has traditionally relied on anatomical resection, including oncological colectomy for colon cancer and total mesorectal excision (TME) for rectal cancer (3, 4). Although these operations remain central to cure, their functional consequences create a strong rationale for safe organ preservation, particularly for low rectal tumors. The relevant question is therefore not whether surgery can be avoided at any cost, but whether response-adapted deferral can preserve function without compromising oncological rescue. Organ preservation was initially developed for selected rectal cancer patients after chemoradiotherapy or total neoadjuvant therapy. The emergence of mismatch repair-deficient (dMMR) or microsatellite instability-high (MSI-H) disease has changed the biological basis of this strategy because immune checkpoint blockade can produce deep regression by restoring pre-existing antitumor immunity. Contemporary rectal cancer guidelines recognize nonoperative management after a carefully established cCR in selected settings, but the evidence supporting immunotherapy-specific watch-and-wait remains substantially less mature than the broader chemoradiotherapy- or TNT-based literature (5–9). Molecular subtype may therefore inform treatment planning, but it does not replace anatomical staging, response verification, or surgical judgment. Historically, dMMR status was recognized as a molecular feature associated with distinctive clinicopathological characteristics, favorable prognosis in some localized colon cancers, and reduced benefit from fluoropyrimidine-only adjuvant chemotherapy (10, 11). Its therapeutic significance changed substantially after PD-1 blockade demonstrated marked activity in dMMR/MSI-H tumors (12, 13). In metastatic colorectal cancer, pembrolizumab improved progression-free survival compared with chemotherapy in KEYNOTE-177, while nivolumab-based regimens also showed durable activity in this molecular subgroup (14–16). These findings established dMMR/MSI-H status not only as a prognostic marker, but also as a predictive biomarker for immune checkpoint blockade. The key question has now moved into the curative-intent setting. Short-course checkpoint blockade produces frequent major pathological responses in dMMR colon cancer, whereas prospective rectal cancer cohorts show that selected patients may achieve a cCR and enter nonoperative surveillance (17–20). These observations are clinically compelling but arise from different anatomical sites, endpoints, and evidentiary designs. These findings create a location-specific surgical dilemma. In dMMR colon cancer, current evidence mainly supports neoadjuvant immunotherapy followed by planned resection, because pathological response and nodal clearance are generally established from the surgical specimen. In dMMR rectal cancer, cCR after PD-1 blockade raises the possibility of deferring TME, but the durability of response, immunotherapy-specific regrowth patterns, and salvage outcomes remain incompletely characterized. Organ preservation should therefore not be interpreted as passive omission of surgery or as a uniform strategy for all dMMR colorectal cancers. In this review, it is framed as immune-response-adapted surgical deferral: an evolving strategy in which surgery is repositioned according to tumor location, response depth, residual uncertainty, surveillance feasibility, patient preference, and salvage capacity. This conceptual framework is illustrated in Figure 1. Conceptual shift from surgery-first treatment to immune-response-adapted surgical deferral. The traditional pathway is contrasted with a biomarker-informed investigational framework in which dMMR/MSI-H confirmation leads to neoadjuvant immunotherapy, site-specific multimodal response assessment, and response-adapted management. Management options include protocol-based watch-and-wait with structured surveillance and salvage surgery when feasible if local regrowth or progression occurs, delayed surgery, or oncologic resection. The lower pathway is conceptual and should not be interpreted as an established standard for colon cancer, in which planned resection remains the default outside prospective trials. CT, computed tomography; ctDNA, circulating tumor DNA; dMMR, mismatch repair-deficient; DRE, digital rectal examination; MRI, magnetic resonance imaging; MSI-H, microsatellite instability-high; TME, total mesorectal excision; W&W, watch-and-wait. Literature search, study selection, and evidence appraisal This article was designed as a narrative review rather than a systematic review or meta-analysis. PubMed/MEDLINE, Web of Science, Embase, and the Cochrane Library were searched from database inception through July 2026, and ClinicalTrials.gov was examined for relevant ongoing or recently completed studies. Search concepts combined terms for mismatch repair deficiency or microsatellite instability (“dMMR”, “MSI-H”, and “mismatch repair”) with terms for colorectal site and treatment (“colorectal cancer”, “rectal cancer”, “colon cancer”, “neoadjuvant immunotherapy”, “PD-1 blockade”, and “immune checkpoint inhibitor”) and organ-preservation outcomes (“clinical complete response”, “pathological complete response”, “watch-and-wait”, “nonoperative management”, “organ preservation”, “salvage surgery”, “ctDNA”, and “Lynch syndrome”). Reference lists of pivotal trials, guidelines, consensus documents, and major reviews were screened for additional studies. Eligible evidence addressed nonmetastatic or curative-intent dMMR/MSI-H colorectal cancer, neoadjuvant immune checkpoint blockade, response assessment, watch-and-wait, organ preservation, surveillance, salvage surgery, toxicity, quality of life, or hereditary-risk management. Prospective interventional studies were prioritized over retrospective series and case reports; peer-reviewed full-text publications were prioritized over conference abstracts; and studies reporting long-term oncological, organ-preservation, or salvage outcomes were given greater interpretive weight than studies reporting response rates alone. Conference abstracts were retained only when they provided clinically important information not yet available in full text. Colon and rectal cancer evidence was evaluated separately, and pCR, cCR, organ preservation, and survival were not treated as interchangeable endpoints. No formal risk-of-bias instrument or quantitative certainty grading was applied; accordingly, statements about clinical readiness represent a qualitative narrative appraisal based on study design, sample size, follow-up maturity, endpoint validity, response-assessment feasibility, and availability of salvage data. Preliminary or single-arm findings were interpreted as hypothesis-generating unless supported by convergent prospective evidence. For the registered-trial landscape, ClinicalTrials.gov was searched through July 21, 2026 using combinations of dMMR/MSI-H, colon or rectal cancer, PD-1/PD-L1 blockade, watch-and-wait, nonoperative management, clinical complete response, treatment duration, and deferred surgery. The trial table includes interventional studies of localized dMMR/MSI-H colorectal cancer that prespecified either an organ-preservation pathway after clinical response or a response-adapted treatment-duration/stopping rule. Mandatory-resection studies without an organ-preservation or response-adapted duration component were excluded from this focused table. Registry status, planned enrollment, regimen, assessment timing, surveillance, and surgery or salvage triggers were extracted from the current record and linked protocol when available; unreported details are identified as not specified rather than inferred. Trial status was rechecked on July 21, 2026. Terminology and interpretive scope For this review, organ preservation is an umbrella objective describing avoidance of radical resection while maintaining curative intent. Watch-and-wait refers specifically to protocol-based active surveillance after a multimodally established cCR; it is not synonymous with absence of treatment or follow-up. Nonoperative management is a broader management category that may include watch-and-wait and other trial-defined strategies that do not involve immediate radical surgery. Local excision is an operative organ-sparing intervention and should not be grouped with watch-and-wait. The phrase surgery omission is used only when describing a study endpoint or a completed treatment course, because at the time of decision-making surgery is more accurately considered deferred and remains available if response is incomplete, assessment is discordant, or regrowth occurs. These definitions are applied separately to rectal and colon cancer because the evidence, functional trade-offs, and ability to assess residual disease differ substantially by anatomical site. Biological rationale: why dMMR colorectal cancer is uniquely sensitive to immune checkpoint blockade Mismatch repair failure involving MLH1, PMS2, MSH2, or MSH6 permits accumulation of insertion-deletion errors and generates the MSI-H phenotype (21, 22). The resulting high mutational and neoantigen burden is commonly accompanied by dense immune infiltration, interferon signaling, cytotoxic T-cell activity, and adaptive expression of inhibitory checkpoints such as PD-1/PD-L1 (23–28). This immunogenic but restrained microenvironment provides the biological basis for checkpoint blockade. The neoadjuvant setting may enhance this effect because an intact primary tumor and draining lymphatic network provide abundant antigen during treatment. Preclinical and translational data suggest that checkpoint blockade before resection can expand tumor-reactive T-cell clones and generate systemic immune memory (29). This mechanism is consistent with the rapid pathological regression observed in colon cancer and the deep clinical regression reported in rectal cancer, but response is not uniform across patients or compartments. dMMR/MSI-H status strongly enriches for benefit but does not guarantee eradication. Primary or acquired resistance may involve impaired antigen presentation, HLA or beta-2-microglobulin alterations, JAK-STAT pathway disruption, immune exclusion, or intratumoral heterogeneity. Moreover, regression of the mucosal primary does not prove clearance of deeper wall or nodal disease. Biology therefore supports neoadjuvant immunotherapy, whereas the safety of organ preservation must be established through location-specific response assessment and long-term clinical outcomes. Clinical evidence for neoadjuvant immunotherapy in dMMR colorectal cancer Colon cancer: compelling surgery-confirmed activity, unproven nonoperative safety The strongest surgery-confirmed evidence comes from localized dMMR colon cancer. NICHE and NICHE-2 demonstrated rapid major pathological responses, including frequent pCR, after short-course nivolumab plus ipilimumab before planned colectomy (17, 18). These studies establish biological activity and perioperative feasibility; they do not validate colectomy omission because residual mural and nodal disease was determined from the surgical specimen. Evidence outside selected prospective cohorts is more heterogeneous. A multicenter real-world study reported lower pathological response and greater serious toxicity than landmark trials, while a systematic review identified substantial variation in regimens, response definitions, and follow-up (30, 31). The colon-inclusive NEOCAP phase II study provides direct but still preliminary nonoperative evidence: among 52 evaluable patients treated with camrelizumab plus apatinib, 28 (54%) achieved cCR and 24 entered W&W, including 20 patients with colon or multiple colorectal primaries; median follow-up was 16.4 months and no regrowth was reported during that short interval (32). These findings establish feasibility in selected patients, not the diagnostic accuracy or long-term oncological safety required for routine colectomy omission. Rectal cancer: emerging organ-preservation evidence with limited long-term validation Rectal cancer presents a different clinical opportunity because the primary site is repeatedly accessible and avoidance of TME may yield substantial functional benefit. In the dostarlimab program, selected patients with dMMR locally advanced rectal cancer achieved cCR after PD-1 blockade and entered nonoperative surveillance; subsequent prospective reporting broadened this experience (19, 20). The cohorts remain nonrandomized, concentrated in expert programs, and limited in size and follow-up relative to the time required to establish durable organ preservation. Other prospective studies, including PICC and NEOPRISM-CRC, confirm sensitivity across checkpoint inhibitors and treatment schedules, but most used resection-confirmed pathological endpoints rather than durable nonoperative control (33, 34). Comparisons of response percentages are therefore confounded by site, stage, regimen, treatment duration, mandatory surgery, and endpoint definition. The evidence supports high neoadjuvant activity in colon cancer and emerging protocol-based W&W in selected rectal cancer—not a uniform nonoperative strategy across colorectal sites. The evidence base for routine organ preservation remains immature: most studies are phase II, single-arm, and selected; immunotherapy-specific cCR definitions vary; randomized comparisons are lacking; and long-term regrowth, distant relapse, and salvage outcomes are incompletely characterized. Clinical readiness should therefore be judged by durable cCR, organ-preservation survival, salvageability, functional outcomes, and late toxicity rather than initial response alone. Tables 1, 2 summarize this location- and endpoint-specific interpretation, while the principal clinical studies of neoadjuvant immunotherapy are summarized in Supplementary Table 1. Table 1Differences between rectal and colon cancer in organ preservation after immunotherapy.DimensionRectal cancerColon cancerClinical implicationOrgan preservation valueHigh functional value, especially for low rectal tumors, because avoidance of TME may prevent permanent stoma, low anterior resection syndrome, and urinary or sexual dysfunction.Usually lower functional gain than rectal preservation, because colectomy is often better tolerated and does not usually cause the same pelvic functional burden.The clinical rationale for organ preservation is generally stronger in rectal cancer, but implementation still requires selected patients, expert assessment, and prospective validation.Assessment methodMultimodal local assessment is feasible using digital rectal examination, endoscopy, biopsy when appropriate, pelvic MRI, diffusion-weighted imaging, and systemic restaging.Local assessment is less mature. Colonoscopy can assess mucosal healing, but it cannot reliably exclude transmural or nodal residual disease; CT/MRI are less validated for cCR determination.cCR-based decision-making is more clinically actionable in rectal cancer; colon cancer generally still requires resection to confirm pathological response.W&W experienceExtensive experience exists after chemoradiotherapy/TNT; early dMMR immunotherapy studies provide emerging, not definitive, support for selected patients with cCR.Very limited experience. Most neoadjuvant immunotherapy studies in dMMR colon cancer still use surgery to confirm pCR or major pathological response.Immunotherapy-based W&W may be considered selectively for rectal cancer; colon cancer W&W remains investigational.Local regrowth detectionEarly local regrowth can often be detected by symptoms, DRE, endoscopy, and pelvic MRI during intensive surveillance.Detection is more difficult because subtle mural or nodal residual disease is harder to monitor repeatedly and reliably outside surgical pathology.Surveillance infrastructure is a key reason why organ preservation is more mature in rectal cancer.Salvage feasibilitySalvage TME or local surgery may preserve curative intent when regrowth is detected early, although immunotherapy-specific salvage outcomes require longer follow-up.Salvage colectomy may be feasible if localized disease is detected, but delayed recognition of residual or nodal disease could compromise oncologic safety.Any organ-preservation program requires realistic access to timely salvage, although immunotherapy-specific salvage outcomes remain uncertain.Current evidence maturityEmerging and clinically important, but based mainly on selected single-arm cohorts with limited long-term follow-up.Biologically promising but clinically immature; high pCR rates after neoadjuvant immunotherapy do not yet validate routine omission of colectomy.Future trials should evaluate rectal and colon cancer separately, with different endpoints and surveillance rules.cCR, clinical complete response; DRE, digital rectal examination; dMMR, mismatch repair-deficient; MRI, magnetic resonance imaging; pCR, pathological complete response; TME, total mesorectal excision; TNT, total neoadjuvant therapy; W&W, watch-and-wait.The table is intended to support conceptual comparison in a narrative review rather than to define formal treatment guidelines. Differences between rectal and colon cancer in organ preservation after immunotherapy. cCR, clinical complete response; DRE, digital rectal examination; dMMR, mismatch repair-deficient; MRI, magnetic resonance imaging; pCR, pathological complete response; TME, total mesorectal excision; TNT, total neoadjuvant therapy; W&W, watch-and-wait. The table is intended to support conceptual comparison in a narrative review rather than to define formal treatment guidelines. Table 2Evidence strength and clinical readiness of organ preservation after neoadjuvant immunotherapy.Clinical settingEvidence maturityCurrent implicationKey uncertaintyClinical readinessdMMR colon cancer after neoadjuvant immunotherapyHigh surgery-confirmed pathological response evidence; limited nonoperative evidenceSupports neoadjuvant immunotherapy before planned colectomy in selected settingsWhether colectomy can be safely omitted and how nodal clearance can be verified without resectionLow for routine W&W; investigational outside trialsdMMR rectal cancer with robust cCREmerging prospective organ-preservation evidence; selected cohorts, nonrandomized designs, and limited long-term follow-upMay support protocol-based W&W in selected patients within expert MDT programsLong-term durability, local regrowth biology, and immunotherapy-specific salvage outcomesConditional and limited; specialist programs or prospective protocolsdMMR rectal cancer with near-cCRIndirect and evolving evidence; usually extrapolated from rectal W&W and early immunotherapy cohortsSupports time-limited reassessment or protocol-defined management rather than open-ended observationOptimal reassessment interval, number of cycles, and role of local excision or additional immunotherapyLimited; requires predefined MDT protocolResidual or progressive disease after immunotherapyConventional oncologic evidence remains strongest for resection when response is incompleteRadical surgery generally remains the reference oncologic pathwayWhether selected patients benefit from additional immunotherapy before surgeryEstablished role for surgery when residual disease is convincingColon cancer watch-and-wait after apparent complete responseVery limited evidence; validated local cCR criteria are lackingNot supported as routine careInability to reliably exclude transmural or nodal residual disease without colectomyInvestigational; prospective trials or exceptional individualized circumstancesLynch syndrome-associated dMMR colorectal cancerEstablished hereditary-risk evidence; very limited organ-preservation-specific evidenceRequires integration of tumor response with hereditary-risk counseling and long-term surveillance planningOptimal balance between segmental preservation, metachronous cancer risk, function, and patient preferenceIndividualized; requires hereditary-risk planningcCR, clinical complete response; dMMR, mismatch repair-deficient; MDT, multidisciplinary team; near-cCR, near-complete response; W&W, watch-and-wait.This table provides a narrative synthesis of current evidence strength and clinical readiness across different organ-preservation scenarios after neoadjuvant immunotherapy. The categories reflect the authors’ interpretation based on available clinical evidence, response-assessment feasibility, surveillance reliability, and salvage-surgery considerations, and should not be interpreted as formal treatment guidelines. Evidence strength and clinical readiness of organ preservation after neoadjuvant immunotherapy. cCR, clinical complete response; dMMR, mismatch repair-deficient; MDT, multidisciplinary team; near-cCR, near-complete response; W&W, watch-and-wait. This table provides a narrative synthesis of current evidence strength and clinical readiness across different organ-preservation scenarios after neoadjuvant immunotherapy. The categories reflect the authors’ interpretation based on available clinical evidence, response-assessment feasibility, surveillance reliability, and salvage-surgery considerations, and should not be interpreted as formal treatment guidelines. Response assessment after neoadjuvant immunotherapy: cCR, pCR, and the problem of discordance Response assessment is the central bottleneck between neoadjuvant immunotherapy and organ preservation. pCR is established after resection by demonstrating no viable tumor in the treated primary site or sampled regional nodes. cCR is an inferential, time-dependent clinical state based on the absence of detectable disease across examination, endoscopy, imaging, and other restaging tests. It is therefore a management threshold rather than a pathological diagnosis, and its false-negative consequences are greatest when surgery is deferred. Rectal cancer: multimodal cCR assessment and immunotherapy-specific uncertainty Rectal cCR criteria were developed mainly after chemoradiotherapy or total neoadjuvant therapy. They combine no palpable tumor on digital rectal examination, a flat white scar or mucosal normalization without ulceration or nodularity on endoscopy, and fibrotic low-signal change without suspicious intermediate T2 signal, restricted diffusion, or residual nodes on pelvic MRI (5–8, 35–38). Concordance across modalities is more informative than any single test, but these criteria have not been fully validated after immunotherapy. Checkpoint blockade may produce immune infiltration, edema, ulcer healing, fibrosis, mucinous change, delayed regression, and heterogeneous clearance that complicate interpretation. Endoscopy and superficial biopsy cannot exclude deeper mural or nodal disease; MRI may be confounded by treatment effect and reader variability; CT is primarily a systemic restaging tool; and CEA lacks sufficient sensitivity or specificity for local decisions (35–38). A negative biopsy should not overrule suspicious examination, endoscopic, or MRI findings. Timing and discordance require prespecified management. When the patient is stable, short-interval reassessment by the same expert teams may distinguish delayed regression from persistent disease; progressive or unresolved discordance should lower the threshold for tissue clarification or surgery. Durable cCR should denote serially concordant absence of detectable disease, with separate reporting of treatment completion, first cCR, confirmation, and sustained response. No immunotherapy-validated interval or minimum duration currently defines durability. Colon cancer: tentative investigational response frameworks No validated colon-specific cCR framework exists. Published and registered studies use heterogeneous investigator-derived constructs. NEOCAP assessed response after four and eight cycles using site-adapted radiology and endoscopy and offered W&W after cCR; NAIO defines major clinical response after 18 weeks by an asymptomatic primary tumor, no metastatic disease, and stable or declining ctDNA; and PREMICES requires two successive response assessments and uses scheduled CT plus colonoscopy with biopsy in its nonoperative arm (32, 39–42). WINDOW uses at least four cycles of tislelizumab and requires two consecutive tumor-informed MRD-negative samples before W&W, whereas its linked protocol also incorporates RECIST imaging, colonoscopy with biopsy, tumor markers, and site-specific examination (43). In a 2025 conference abstract, 24 patients were reported, 20 entered W&W, and no recurrence was observed at a median follow-up of 20.3 months; these preliminary, non-peer-reviewed abstract data do not establish long-term safety or validate ctDNA as a stand-alone decision rule (44). These are protocol eligibility or management rules—not validated diagnostic criteria—and they differ in whether local mucosal, transmural, nodal, and molecular clearance must be demonstrated. Across these investigational frameworks, recurring domains include resolution of obstruction, bleeding, or other tumor-related symptoms; no radiological progression or distant disease; marked endoscopic regression or disappearance of visible tumor; biopsy without viable tumor when required; stable or declining CEA; and ctDNA/MRD decline or clearance. However, the evidence supporting each domain is uneven. Superficial biopsy cannot exclude transmural or nodal tumor, CT has not been validated to exclude microscopic bowel-wall disease, radiological CR under RECIST is not synonymous with sterilization of the primary and regional nodes, and ctDNA negativity may reflect low shedding. Serial confirmation and multidisciplinary review reduce but do not eliminate this uncertainty; no current combination has a validated negative predictive value sufficient to replace colectomy outside a prospective trial. Accordingly, response assessment must be location-specific. A concordant rectal cCR may serve as an entry criterion for protocol-based W&W, whereas an apparent colon response remains investigational and should not be assumed to represent pCR. Prospective studies should correlate each proposed criterion with resection pathology or sufficiently long nonoperative outcomes and should report false-negative local and nodal disease explicitly. Watch-and-wait after immunotherapy: who may be suitable? Immunotherapy-based W&W is an evolving rectal cancer strategy rather than an established colorectal standard. It may be considered for selected patients with molecularly confirmed dMMR/MSI-H rectal cancer, no distant disease, concordant and confirmed multimodal cCR, informed acceptance of residual uncertainty, reliable adherence, and immediate access to expert surveillance and salvage surgery (9, 19, 20, 35–38, 45–48). When these conditions are not met, surgery may provide the safer curative pathway. Conventional W&W cohorts show that many rectal regrowths occur early and can often be salvaged when detected promptly (6, 45–48). These data inform surveillance design but are not direct proof for immunotherapy-treated disease, in which the timing of regrowth and outcomes of salvage remain less mature. W&W is inappropriate for progressive or convincing residual disease, uncontrolled symptoms, threatened obstruction or perforation, unreliable follow-up, or unavailable salvage. Near-cCR requires a predefined reassessment window rather than open-ended observation. For colon cancer, nonoperative management should remain confined to prospective trials or exceptional circumstances in which standard surgery is infeasible and uncertainty is explicitly documented. Surgical decision-making after response: immediate surgery, delayed surgery, local excision, or nonoperative management? After immunotherapy, the choice among planned radical surgery, time-limited reassessment, selected local excision, and W&W should integrate anatomical site, baseline nodal and margin risk, response category, concordance, patient priorities, and salvageability. These pathways differ in both purpose and evidentiary support. Radical surgery remains appropriate for residual or progressive disease, high-risk anatomy, persistent symptoms, or unresolved assessment uncertainty. It also remains the default after neoadjuvant immunotherapy for operable colon cancer outside a trial. Deferral should not allow obstruction, perforation, systemic progression, or loss of a technically curative window. Near-complete response: the key grey zone between organ preservation and surgery Near-cCR is a time-limited state rather than treatment success or failure. Selected stable rectal cancer patients may undergo short-interval reassessment or protocol-defined additional therapy, with the interval and surgical triggers documented in advance. Persistent or progressive abnormality should prompt resection rather than indefinite extension. Local excision has a narrower investigational role for small superficial rectal abnormalities when nodal and mesorectal assessment is otherwise reassuring. It may clarify pathology while preserving the rectum, but cannot reliably address occult nodal disease. A confirmed cCR may instead lead to W&W, with prompt restaging and consideration of salvage resection if regrowth develops. Immunotherapy-specific salvage resectability, morbidity, and long-term control require prospective reporting. Practical MDT workflow for implementation A response-adapted pathway begins with validated dMMR/MSI-H confirmation, appropriate germline evaluation, high-quality baseline anatomical and systemic staging, and documentation of symptoms and surgical options. Before treatment, the MDT should define site-specific restaging, response categories, reassessment intervals, and triggers for surgery or salvage. Figure 2 summarizes the proposed rectal pathway; its cCR branches should not be extrapolated to colon cancer, for which planned colectomy remains the default outside trials. Response assessment and response-adapted management algorithm for dMMR/MSI-H rectal cancer after neoadjuvant immunotherapy. After multimodal assessment in a specialist multidisciplinary program, patients are categorized as having cCR, near-cCR, or residual/incomplete response. A concordant and sustained cCR may support protocol-based watch-and-wait. Near-cCR should prompt a predefined short-interval reassessment strategy, after which selected patients may enter watch-and-wait following sustained cCR, undergo delayed surgery or highly selected local excision, or proceed to radical surgery. Residual/incomplete response generally warrants standard oncologic resection. Only local regrowth or progression during watch-and-wait proceeds to restaging and consideration of salvage surgery when feasible. This algorithm has not been validated for colon cancer, for which planned colectomy remains standard outside a clinical trial. cCR, clinical complete response; CEA, carcinoembryonic antigen; CT, computed tomography; ctDNA, circulating tumor DNA; dMMR, mismatch repair-deficient; DRE, digital rectal examination; MRI, magnetic resonance imaging; MSI-H, microsatellite instability-high; W&W, watch-and-wait. Surveillance strategy during organ preservation Surveillance is the active treatment component of organ preservation. During the first two years after rectal cCR, frequent digital rectal examination, endoscopy, and pelvic MRI should evaluate local regrowth, with periodic CT of the chest, abdomen, and pelvis and longitudinal CEA for systemic relapse. Exact intervals vary among programs and should be protocolized; the key requirements are consistent modalities, experienced interpretation, rapid review of discordant findings, and timely salvage access (6, 7, 35–38, 45–48). An illustrative surveillance framework is provided in Supplementary Table 2. Intervals may lengthen after the highest-risk early period, but structured follow-up should continue for at least five years and be individualized by baseline stage, initial nodal risk, treatment duration, response kinetics, hereditary risk, and adherence. New bleeding, pain, altered bowel habit, weight loss, or obstructive symptoms require assessment outside the routine schedule. ctDNA is an investigational adjunct whose sensitivity varies with burden, site, biological shedding, assay design, and sampling time (49–51). A negative result cannot exclude small-volume disease confined to the rectal wall or mesorectal nodes; a positive result does not localize disease. No threshold or ctDNA-guided rule is validated for omitting rectal or colon surgery after immunotherapy. It should therefore complement, not replace, anatomical and clinical assessment, with prospective trials reporting assay, time points, false-negative local failures, and management consequences. Toxicity, quality of life, and patient-centered outcomes The aim of organ preservation is improved patient-centered outcome without loss of oncological safety. Checkpoint blockade can cause acute or delayed immune-related toxicity, including colitis, hepatitis, pneumonitis, endocrinopathies, dermatological events, arthritis, and rare neurological or cardiac complications (52, 53). Trials should report cumulative exposure, treatment discontinuation, hospitalization, long-term hormone replacement, and late toxicity rather than assuming immunotherapy is functionally neutral. The functional argument is strongest for rectal cancer. TME, especially for low tumors, may result in permanent or temporary stoma, low anterior resection syndrome with urgency, clustering, fragmentation or incontinence, and urinary or sexual dysfunction (54, 55). Avoiding TME may improve bowel function and quality of life in selected patients, although W&W itself can generate anxiety and repeated-procedure burden; prospective cohorts should measure both benefits and harms (56). Shared decision-making should elicit how each patient values functional preservation, definitive pathology, recurrence uncertainty, immune toxicity, surveillance burden, and possible salvage. Success should be evaluated with disease-free and overall survival, local regrowth-free and organ-preservation survival, stoma-free survival, validated bowel/urinary/sexual function, quality of life, treatment burden, and patient-reported outcomes—not cCR or surgery avoidance alone (57, 58). Discussion: current controversies and unanswered questions The central uncertainty is whether clinically measured regression can support durable organ preservation without compromising cure. Colon studies largely validate drug activity through resection pathology, whereas rectal nonoperative studies depend on serial cCR and salvageable follow-up (17–20, 30, 31, 33–38, 59). These endpoints are not interchangeable. High colon pCR rates should motivate validation of response prediction, not immediate adoption of W&W. Three evidence gaps are particularly consequential: no validated colon cCR framework; no immunotherapy-specific definition or confirmation interval for durable rectal cCR; and limited long-term data on regrowth, distant relapse, and salvage. Optimal checkpoint inhibitor, combination, duration, and stopping rule also remain undefined. Trial protocols now range from brief preoperative regimens to six months or longer of therapy, underscoring that schedule choice is still an empirical research question rather than a clinical standard. Lynch syndrome adds a lifelong hereditary-risk horizon. Response of the index tumor does not remove gene-dependent risks of metachronous colorectal or extracolonic cancer. When indicated, germline clarification, genetic counseling, cascade testing, and gene-, age-, sex-, and family-history-informed surveillance should therefore precede or accompany a definitive organ-preservation plan (60–63). The trade-off is site-specific. In Lynch-associated colon cancer, preserving the index segment must be weighed against metachronous risk and the conventional choice between segmental and extended colectomy. In rectal cancer, avoiding TME may offer greater functional benefit, but surveillance of the remaining colon and extracolonic organs remains necessary. Current cohorts cannot establish whether cCR durability differs between germline and sporadic dMMR disease; decisions should therefore integrate hereditary risk rather than generalize from tumor response alone. Future directions: toward risk-adapted organ preservation trials Future trials should be location-specific and pathway-focused. Rectal studies need prespecified cCR and near-cCR definitions, confirmation time points, surveillance schedules, and salvage triggers. Colon studies must validate whether any multimodal construct can exclude residual mural and nodal disease with sufficient accuracy for nonoperative management. Comparative cohorts, independent response review, and long follow-up are essential. Endpoints should include local regrowth, distant relapse, organ-preservation survival, salvage R0 resection, postoperative morbidity, stoma-free survival, immune toxicity, function, quality of life, and patient-reported burden. Biomarker work should prespecify assays and sampling times and should evaluate incremental value beyond anatomy rather than substituting molecular results for local assessment. Any multivariable response model should be transparently developed, validated, and reported (64). Registered trials addressing treatment duration and W&W protocols The predefined registry search identified eight organ-preservation or response-adapted duration studies (Table 3), including the multicenter AZUR-1 registrational phase II study of dostarlimab in untreated dMMR/MSI-H locally advanced rectal cancer (65). In July 2026, an official sponsor announcement reported that AZUR-1 had met its primary objective of sustained cCR at 12 months; 154 participants received treatment. However, numerical response data and full peer-reviewed results were not yet publicly available, and the finding should therefore be interpreted cautiously (66). Treatment exposure ranges from four 3-weekly sintilimab cycles in a completed rectal cohort to at least four and up to eight response-guided tislelizumab cycles in WINDOW, six dostarlimab doses over 18 weeks in NAIO, eight pembrolizumab cycles with possible extension to 17 in PREMICES, and nine dostarlimab doses over six months in the rectal dostarlimab programs (39–43, 65, 67, 68). NEOCAP assessed camrelizumab plus apatinib after four and eight cycles. This heterogeneity precludes cross-trial inference that any cycle number is optimal. Table 3Registered interventional trials meeting predefined criteria for treatment-duration or organ-preservation pathways after neoadjuvant PD-1 blockade.Trial/site; status*Design/sample sizeRegimen/durationResponse assessment/timingW&W eligibilitySurgery/salvage triggerSurveillance informationNCT04165772/rectum; active, published cohortPhase II, single arm/expansion ongoingDostarlimab 500 mg q3w ×9/6 monthsDRE, endoscopy/biopsy, pelvic MRI and CT/PET at protocol time points; cCR assessed during and after therapyConcordant cCR after PD-1 blockadeIncomplete response or later regrowth: protocol-directed additional treatment and/or surgeryIntensive serial endoscopy, MRI, examination and systemic imaging; exact schedule per protocolNCT05723562 AZUR-1/rectum; active, not recruitingGlobal multicenter phase II, single arm/154 enrolledDostarlimab 500 mg q3w ×9/6 monthsDRE, endoscopy, pelvic MRI and systemic imaging; cCR assessed after therapy, with 12-month cCR by independent central review as a primary endpointcCR after dostarlimab permits nonoperative managementIncomplete response or recurrence: protocol-directed oncologic treatment, including surgery when indicatedClose nonoperative surveillance for recurrence; complete public schedule not specifiedNCT04304209/rectum; closed, publishedPhase II dMMR cohort/17 treatedSintilimab q3w ×4/12 weeksDRE, endoscopy, rectal MRI, CT and biopsy; reassessment after four cyclescCR patients could choose W&WNon-cCR: surgery; suspected regrowth: restaging and salvageProspective W&W follow-up reported; long-term immunotherapy-specific salvage data limitedNCT04715633 NEOCAP/colon and rectum; completed, publishedPhase II, single arm/53Camrelizumab plus apatinib; response checks after cycles 4 and 8CT for colon, pelvic MRI for rectum, endoscopy and biopsy; site-adapted cCRcCR after treatment; patient choiceNon-cCR/inadequate response: site-specific therapy or surgeryProtocol follow-up; published median follow-up 16.4 months, insufficient for durable safetyNCT05239546 NAIO/colon; recruitingPhase II, single arm/registry-planned enrollmentDostarlimab q3w ×6/18 weeks; continued treatment in eligible respondersSymptoms, metastatic staging and ctDNA after induction; colonoscopy/biopsy and imaging per protocolMajor clinical response: asymptomatic primary, no metastases, stable/declining ctDNANonresponse, progression or failure of protocol criteria: colectomyRegistry specifies structured nonoperative follow-up; detailed universal cCR scar criteria not establishedNCT06262581/colon and rectum; unknown status (last verified February 2024)Phase II, single arm/40Tislelizumab q3w; protocol-defined neoadjuvant courseImaging and endoscopic/clinical assessment; detailed confirmation interval incompletely specified publiclyPatients qualified as cCR may enter W&WNon-cCR or progression: planned surgery; recurrence: MDT-directed treatmentPublic record does not fully specify all W&W intervalsNCT06477991 WINDOW/colon and rectum; recruiting; preliminary conference reportSingle-center prospective phase II/registry-planned 22; abstract-reported 24Tislelizumab 200 mg q3w, ≥4 cycles; response-guided to 8Tumor-informed MRD after cycle 4 and sequentially; RECIST imaging, colonoscopy/biopsy, markers and site-specific examinationTwo consecutive MRD-negative resultsMRD still positive after cycle 8: surgery; conversion to positive or clinical recurrence: reassessmentAbstract: 20 entered W&W, with no recurrence at median 20.3 months; preliminary conference evidence. Registry: MRD at months 1, 4, 7, 13, 19, 25 and 37, with protocolized clinical and imaging follow-upNCT06646445 PREMICES/colon; not yet recruitingMulticenter randomized phase II/84Pembrolizumab 200 mg q3w ×8; responders may extend to 17 cyclesCT and two successive colonoscopies with biopsy; strategy success assessed at 6 monthsComplete response without tumor on serial assessment/biopsy in experimental strategyProgression, persistent tumor on biopsy or decision for primary-tumor surgery constitutes strategy failureScheduled CT and colonoscopy/biopsy; 24-month strategy success and salvage-related outcomes plannedcCR, clinical complete response; CT, computed tomography; ctDNA, circulating tumor DNA; DRE, digital rectal examination; MDT, multidisciplinary team; MRD, molecular residual disease; MRI, magnetic resonance imaging; PD-1, programmed cell death protein 1; PET, positron emission tomography; q3w, every 3 weeks; RECIST, Response Evaluation Criteria in Solid Tumors; W&W, watch-and-wait.*Trial status reflects the ClinicalTrials.gov records accessed on July 21, 2026. NCT06262581 was listed as having an unknown status because its record had not been verified within the preceding two years. “Not specified” or equivalent wording indicates that sufficient information was not available in the public registry record or linked protocol. Sample sizes represent planned enrollment, actual enrollment, or abstract-reported enrollment, as explicitly indicated. Registry protocols and preliminary conference abstracts do not constitute evidence of established efficacy, and the eligibility and assessment criteria summarized here should not be interpreted as validated cCR standards or formal clinical recommendations. Registered interventional trials meeting predefined criteria for treatment-duration or organ-preservation pathways after neoadjuvant PD-1 blockade. cCR, clinical complete response; CT, computed tomography; ctDNA, circulating tumor DNA; DRE, digital rectal examination; MDT, multidisciplinary team; MRD, molecular residual disease; MRI, magnetic resonance imaging; PD-1, programmed cell death protein 1; PET, positron emission tomography; q3w, every 3 weeks; RECIST, Response Evaluation Criteria in Solid Tumors; W&W, watch-and-wait. *Trial status reflects the ClinicalTrials.gov records accessed on July 21, 2026. NCT06262581 was listed as having an unknown status because its record had not been verified within the preceding two years. “Not specified” or equivalent wording indicates that sufficient information was not available in the public registry record or linked protocol. Sample sizes represent planned enrollment, actual enrollment, or abstract-reported enrollment, as explicitly indicated. Registry protocols and preliminary conference abstracts do not constitute evidence of established efficacy, and the eligibility and assessment criteria summarized here should not be interpreted as validated cCR standards or formal clinical recommendations. Response and rescue rules are similarly inconsistent. Rectal studies generally combine digital examination, endoscopy, MRI, and systemic imaging, but confirmation intervals and durable-cCR definitions differ or remain incompletely reported. Colon-inclusive protocols variously rely on symptoms, CT, colonoscopy/biopsy, RECIST response, or serial ctDNA/MRD; WINDOW specifies surgery when MRD remains positive after eight cycles, while PREMICES defines failure through progression, persistent tumor on biopsy, or primary-tumor surgery (39–43, 67, 68). Protocols provide much less mature evidence on regrowth localization, salvage R0 rates, and morbidity than on initial response. The absence of a common treatment duration, cCR definition, surveillance schedule, and salvage trigger is therefore a principal finding rather than a basis for choosing among regimens. Conclusions Neoadjuvant immunotherapy has demonstrated marked activity in dMMR/MSI-H colorectal cancer, but organ-preservation evidence is anatomically uneven. Colon cancer data primarily support surgery-confirmed pathological efficacy; proposed nonoperative response criteria remain investigational. In selected rectal cancer patients, a confirmed cCR may support W&W within expert multidisciplinary programs, provided that uncertainty, surveillance, and timely salvage are integral to the plan. The immediate research priority is not broader surgery omission, but prospective validation of site-specific response criteria, treatment duration, durable control, salvage outcomes, ctDNA utility, and hereditary-risk-informed care.