Citations

Background/Objectives: We performed a systematic review of preclinical literature on the use of high-LET particle therapy, DDRi, and/or immunotherapy specifically in pMMR colorectal cancer. Methods: A systematic review of the literature published between 2014 and 2025 was conducted across major databases. Studies were included if they examined particle radiotherapy (e.g., proton, alpha, and carbon) or X-ray radiation either alone or in combination with DDRi and/or immune checkpoint inhibitors (ICIs) in pMMR colorectal cancer models. Results: In total, 131 studies met the inclusion criteria, including 70 preclinical studies. These studies consistently demonstrate that high-LET radiation amplifies immunogenic cell death, increases cGAS-STING pathway activation, and enhances tumor antigen presentation, thereby fostering greater immune infiltration and systemic antitumor responses. Concurrent irradiation with DDRi enhances persistent DNA damage and cytosolic DNA accumulation. In murine models, high-LET therapies show excellent local control, with manageable toxicity profiles. Combination regimens with ICIs exhibit improved local control and elicit systemic antitumor immune responses. Conclusions: High-LET particle radiation and/or the use of concurrent DDRi with ICI have significant preclinical evidence of immunostimulatory effects in pMMR rectal adenocarcinoma and increased response rates to immunotherapy. The clinical evidence will be reviewed in the companion manuscript.

Background/Objectives: We performed a systematic review of the current clinical and preclinical literature on the use of particle therapy, DDRi, and/or immunotherapy, specifically for pMMR colorectal cancer. Methods: A systematic review of the literature published between 2014 and 2025 was conducted across major databases. Studies were included if they examined particle radiotherapy (e.g., proton, alpha, carbon) either alone or in combination with DDRi and/or immune checkpoint inhibitors (ICI), or X-ray radiotherapy (XRT) in combination with DDRi/ICI. Results: In total, 133 studies met the inclusion criteria, including 62 clinical studies. Clinically, particle therapies show excellent local control and normal tissue sparing, with manageable toxicity profiles. Trials with any form of radiation and DDRi are few, potentially owing to toxicity concerns. ICI combinations showed promising efficacy with XRT, with no randomized trials comparing them to particle radiation. Conclusions: Particle radiation and/or DDRi have significant preclinical evidence of immunostimulatory effects in pMMR rectal adenocarcinoma and increase response rates to immunotherapy (presented in the companion manuscript). Despite strong preclinical evidence and rationale, clinical trials including all three modalities are scarce. Existing evidence suggests a potential benefit based on extrapolation from photon-based studies and supports prospective evaluation with careful attention to treatment-related toxicity, which remains a concern.

Abstract Generative artificial intelligence (AI) is rapidly populating medical records with synthetic or partially AI-generated content, creating a feedback loop where future models are increasingly at risk of training on uncurated AI-generated data. However, the clinical consequences of this AI-generated data contamination remain unexplored. Here, we show that in the absence of mandatory human verification, this self-referential cycle drives a rapid erosion of pathological variability and diagnostic reliability of medical data at population scale. By analysing more than 800,000 synthetic data points across clinical text generation, vision–language reporting, and medical image synthesis, we find that models progressively converge toward generic phenotypes regardless of the model architectures. Specifically, rare but critical findings, including pneumothorax and effusions, vanish from the synthetic content generated by AI models, while demographic representations skew heavily toward middle-aged male phenotypes. Crucially, this degradation is masked by false diagnostic confidence. Models continue to issue reassuring reports while failing to detect life-threatening pathology, with false reassurance rates tripling to 40%. Blinded physician evaluation confirms that this decoupling of confidence and accuracy renders AI-generated documentation clinically useless after just two generations. We systematically evaluate three mitigation strategies that can be easily integrated into existing clinical workflows, finding that while synthetic volume scaling fails to prevent collapse, mixing real data with quality-aware filtering effectively preserves diversity. Ultimately, our results suggest that without policy-mandated human oversight, the deployment of generative AI threatens to degrade the very healthcare data ecosystems it relies upon.

ABSTRACT Background Several colorectal cancer (CRC) screening modalities are guideline-recommended in the United States, yet they vary considerably in screening interval and real-world adherence. As a result, single-round test performance may not reflect cumulative effectiveness over time. This study compared the 10-year longitudinal outcomes of two CRC screening strategies— triennial next-generation multitarget stool DNA testing (ng mt-sDNA) and decennial screening colonoscopy. Methods This study used the validated, microsimulation-based Colorectal Cancer and Adenoma Incidence and Mortality (CRC-AIM) model to estimate 10-year cumulative outcomes for two guideline-recommended screening strategies: triennial ng mt-sDNA and decennial colonoscopy. Model inputs included test-specific performance and real-world adherence. Outcomes included CRC and precancerous lesions detected, CRC mortality reductions, and life-years gained (LYG). Sensitivity analyses examined the effects of varying both screening adherence and follow-up colonoscopy adherence. Results Over 10 years per 1,000 individuals offered screening, the ng mt-sDNA screening test detected 13% more precancerous lesions and 11% more CRC cases than colonoscopy, with a greater proportion of CRCs identified through screening rather than symptomatic detection. ng mt-sDNA resulted in greater CRC mortality reduction (33% vs 20%) and 62% more life-years gained, with consistent findings across sensitivity analyses. Conclusions With real-world adherence, screening with triennial ng mt-sDNA demonstrates superior cumulative effectiveness compared with decennial colonoscopy, driven by higher adherence and favorable longitudinal performance. These findings support the expanded use of noninvasive stool-based screening to reduce CRC mortality and alleviate capacity constraints associated with colonoscopy-based screening. Broader adoption of ng mt-sDNA may enhance population-level CRC prevention by increasing participation and improving early detection across the screening eligible population. Plain language summary Colorectal cancer screening tests are recommended at different time intervals and completed at different adherence rates in clinical practice. The analysis used a validated simulation model to compare 10-year outcomes of triennial next-generation multi-target stool DNA test (ng mt-sDNA) with a single colonoscopy, accounting for real-world screening and follow-up colonoscopy adherence. Our findings indicate that repeated ng mt-sDNA provides greater cumulative screening effectiveness than colonoscopy over a 10-year period.

Abstract In adult-type diffuse gliomas CDKN2A and/or CDKN2B (CDKN2A/B) deletions often co-occur with deletion of MTAP, suggesting that MTAP immunohistochemistry (IHC) may be a surrogate marker of CDKN2A/B status. However, the association between CDKN2A/B and MTAP deletion at the genomic level remains unknown. We assessed CDKN2A/B and MTAP deletions by chromosomal microarray in 333 adult-type diffuse gliomas and performed MTAP IHC on a subset (n = 63). CDKN2A/B and MTAP deletions were detected in 216 and 215 cases, respectively, and were concurrent in 99.5% (215/216). While most tumors with CDKN2A/B homozygous deletion (n = 148) showed concurrent MTAP homozygous deletion (108/148; 73.0%), a subset harbored MTAP heterozygous deletion (39/148; 26.4%). By analyzing the size of the chromosomal alterations, we demonstrate that initial large chromosomal 9p losses result in concurrent heterozygous deletion of CDKN2A/B and MTAP whereas smaller “second hit” deletions leading to homozygous CDKN2A/B deletion do not always encompass the MTAP locus. Discordant CDKN2A/B and MTAP tumors affect the association between MTAP IHC and copy number status of MTAP and CDKN2A/B. These findings suggest that adult-type diffuse gliomas, regardless of IDH status, follow a stereotypic pathway involving concurrent CDKN2A/B and MTAP heterozygous deletion but may diverge for CDKN2A/B and MTAP homozygous deletion.

Abstract Generative artificial intelligence (AI) is rapidly populating medical records with synthetic or partially AI-generated content, creating a feedback loop where future models are increasingly at risk of training on uncurated AI-generated data. However, the clinical consequences of this AI-generated data contamination remain unexplored. Here, we show that in the absence of mandatory human verification, this self-referential cycle drives a rapid erosion of pathological variability and diagnostic reliability of medical data at population scale. By analysing more than 800,000 synthetic data points across clinical text generation, vision–language reporting, and medical image synthesis, we find that models progressively converge toward generic phenotypes regardless of the model architectures. Specifically, rare but critical findings, including pneumothorax and effusions, vanish from the synthetic content generated by AI models, while demographic representations skew heavily toward middle-aged male phenotypes. Crucially, this degradation is masked by false diagnostic confidence. Models continue to issue reassuring reports while failing to detect life-threatening pathology, with false reassurance rates tripling to 40%. Blinded physician evaluation confirms that this decoupling of confidence and accuracy renders AI-generated documentation clinically useless after just two generations. We systematically evaluate three mitigation strategies that can be easily integrated into existing clinical workflows, finding that while synthetic volume scaling fails to prevent collapse, mixing real data with quality-aware filtering effectively preserves diversity. Ultimately, our results suggest that without policy-mandated human oversight, the deployment of generative AI threatens to degrade the very healthcare data ecosystems it relies upon.

Background/Objectives: Modeling analyses for colorectal cancer (CRC) screening focusing solely on the costs of screening do not fully capture the value of screening programs. We evaluated the clinical and economic effects of CRC stool-based screening tests, including impacts on cancer-related outcomes. Methods: A cohort-based decision-analytic cost-estimator model estimated outcomes for a single round of screening with next-generation multi-target stool DNA (ng mt-sDNA) test or fecal immunochemical test (FIT) from a US payer perspective. Undiagnosed cancers were assumed to become symptomatic (and detected) within 10 years. Clinical assumptions, advanced precancerous lesion and CRC prevalence, and test performance inputs were from clinical trial data. Adherence rates for initial screening and follow-up colonoscopy after a positive result were from real-world data. Input costs included the screening tests, follow-up colonoscopy (with and without polypectomy), and CRC treatment. Results: Compared with FIT, more individuals completed ng mt-sDNA (321,000 vs. 713,000, respectively), leading to the detection of more CRC cases (436 with FIT vs. 2235 with ng mt-sDNA), more advanced precancerous lesions, and more CRC at earlier stages. The cost of screening per patient screened was USD 801 for ng mt-sDNA and USD 124 for FIT. Follow-up colonoscopy cost was USD 149 million with ng mt-sDNA versus USD 22 million with FIT, whereas CRC treatment costs were lower for ng mt-sDNA (USD 1423 million versus USD 1474 million, respectively). When accounting for both direct and CRC averted costs, the total cost of screening and treatment was USD 1383 million with ng mt-sDNA versus USD 1427 million with FIT. Conclusions: Higher screening costs with ng mt-sDNA versus FIT are counterbalanced by savings realized from enhanced CRC prevention and earlier detection due to the superior test performance and better adherence with ng mt-sDNA.

INTRODUCTION: Colorectal cancer remains a leading cause of cancer associated death in the United States and colonoscopy the primary screening strategy for prevention. Rates of adenomatous and serrated neoplasia detection are inversely associated with postcolonoscopy colorectal cancer. This crucial quality metric depends on accurate ascertainment of colorectal neoplasia findings from both endoscopy and histopathology records. We aimed to assess the feasibility of a random forest machine learning model to rapidly and accurately categorize colorectal neoplasia from electronic health record data. METHODS: A retrospective cohort study compared neoplasia detection rates among individuals undergoing colonoscopy at a large academic institution to develop a rule-based algorithm to categorize colorectal neoplasia from endoscopy reports and pathology systematized nomenclature of medicine – clinical terms. This cohort provided a large training set to develop a natural language processing system using a random forest approach to automatically classify unstructured pathology findings into adenoma, serrated, or advanced neoplasms. This system was manually validated through an independent holdout set. RESULTS: The training set comprised 35,953 unstructured pathology reports with matched systematized nomenclature of medicine – clinical terms from 95,188 unstructured colonoscopy reports. The final model was assessed on an independent holdout set of 337 manually annotated procedures obtaining an area under the receiver operating characteristic curve of 0.997 (confidence interval [CI] 0.994–1), 0.99 (CI 0.98–1), and 0.99 (CI 0.98–0.99) for prediction of adenoma, serrated, and advanced lesions, respectively. DISCUSSION: The random forest-based hybrid natural language processing system for classification of colonoscopy results was both accurate and explainable. NLP combined with effective machine learning algorithms can provide a scalable strategy for colonoscopy quality monitoring.

Abstract Background Recommended noninvasive strategies for average-risk colorectal cancer (CRC) screening include multitarget stool DNA and fecal immunochemical test from ages 45 to 75 years. With new clinical trials, performance data for next-generation multitarget stool DNA, multitarget stool RNA, and blood-based screening tests are now available. This decision analytical model study evaluated the estimated benefit-to-burden ratio by means of efficient frontiers for noninvasive established and emerging CRC screening strategies. Methods Outcomes were estimated using the Colorectal Cancer and Adenoma Incidence and Mortality microsimulation model for average-risk individuals in the United States. Screening strategies were next-generation multitarget stool DNA (an updated marker panel), fecal immunochemical tests, multitarget stool RNA, or blood-based tests every 1-3 years, over various age ranges. Test performance inputs were derived from recent large clinical trials. A strategy was deemed efficient if no other strategy provided more life-years gained with equivalent or fewer lifetime colonoscopies and near-efficient if within 3 days of life-years gained of the efficient frontier. Results All modeled screening strategies resulted in life-years gained vs no screening. No strategy using blood-based tests was efficient or near-efficient. Overall, 10 strategies were efficient (6 next-generation multitarget stool DNA and 4 fecal immunochemical tests), including 2 strategies among those ages 45-75 years (biennial and triennial next-generation multitarget stool DNA). Overall, 22 strategies were near-efficient, including 4 strategies among those ages 45-75 years (annual, biennial, or triennial fecal immunochemical test; annual next-generation multitarget stool DNA). Conclusion Based on this modeling study, next-generation multitarget stool DNA was the only noninvasive screening test at guideline-endorsed interval and age-recommended ranges that was deemed efficient. Blood-based and multitarget stool RNA strategies were deemed not efficient for primary screening.