White River Junction VA Medical Center

Abstract Background Computer-assisted diagnosis (CADx) systems employ artificial intelligence algorithms to generate probabilistic outputs, assisting in polyp characterization. The clinical utility of these AI-generated predictions ultimately depends on the endoscopist’s ability to interpret and appropriately act upon them. CADx demands expertise from endoscopists, who must employ knowledge of polyp histology based on surface features and clinical judgement to accept or reject the AI-based polyp classifications. This requires a higher level of endoscopist knowledge to discern whether an AI-generated classification is accurate and should be incorporated into the final diagnosis or whether it requires correction. Aims Our study aimed to evaluate endoscopists’ decision-making when interpreting CADx predictions for each polyp category, compared to histopathology diagnosis as ground truth. Specifically, we examined the accuracy of accepted and rejected CADx outputs, the frequency and direction of overrides, and the patterns of reclassification following rejection. Methods We conducted a secondary analysis of a prospective cohort between 2022-2025 (NCT06822816) in which CADx was used during colonoscopies to classify polyps in real-time as either neoplastic or hyperplastic. Endoscopists could either accept the diagnosis made by CADx or reject it by providing an alternate prediction. They could chose to predict polyp histology as adenoma, hyperplastic or sessile serrated lesion. This histology prediction was made prior to resection and independent of pathology. All small (<10mm) polyps with documented CADx output, endoscopist histology prediction, and adenoma, hyperplastic or SSL pathology diagnosis were included. Results Our cohort included 754 polyps from 367 patients. Accepted CADx diagnoses were correct for 89.1% of neoplastic and 68.7% of hyperplastic CADx predictions. Endoscopists accepted incorrect classifications more frequently for hyperplastic than neoplastic polyps (31.3% vs.10.9%). Upon rejecting correct CADx neoplastic predictions, endoscopists misclassified 23.4% as sessile serrated lesions. Similarly, 29.5% of rejected correct CADx hyperplastic predictions were incorrectly reassigned as SSLs. Conclusions Endoscopists demonstrate a pronounced diagnostic bias towards neoplastic lesions, frequently misclassifying hyperplastic polyps as SSLs. While clinically safe, this phenomenon might have implications increasing downstream endoscopy procedures and cost-effectiveness of CADx, highlighting the necessity of incorporating comprehensive cost analyses including downstream colonoscopy utilization. Developing CADx systems with robust SSL identification capabilities will likely be critical in overcoming these diagnostic challenges and achieving effective integration into routine clinical practice. Funding Agencies None

ABSTRACT Development of novel CRISPR/Cas systems enhances opportunities for gene editing to treat infectious diseases, cancer, and genetic disorders. CasX2 (PlmCas12e) belongs to the class II CRISPR system derived from Planctomycetes, a non-pathogenic bacterium present in aquatic and terrestrial soils and offers several advantages as a potential therapeutic CRISPR system over Streptococcus pyogenes Cas9 (SpCas9) and Staphylococcus aureus Cas9 (SaCas9). These advantages include its smaller size, distinct protospacer adjacent motif (PAM) requirements, staggered cleavage cuts that promote homology-directed repair, and the absence of pre-existing immunity in humans. We compared the cleavage efficiency and double-stranded break repair characteristics between CasX2 and CasX2Max, a recently generated CasX2 variant with three amino acid substitutions, for targeting CCR5, a gene that encodes the CCR5 receptor important for HIV-1 infection. Two single guide RNAs (sgRNAs) were designed that flank the 32 bases deleted in the natural CCR5 ∆32 mutation. Nanopore sequencing demonstrated that CasX2 using sgRNAs with spacers of 17 nucleotides (nt), 20 nt or 23 nt in length were ineffective at cleaving genomic CCR5. In contrast, CasX2Max using sgRNAs with 20 nt and 23 nt spacer lengths, enabled cleavage of genomic CCR5. Structural modelling indicated that two of the CasX2Max amino acid substitutions enhanced sgRNA-DNA duplex stability, while the third improved DNA strand alignment within the catalytic site. These structural changes likely underlie the increased activity of CasX2Max in cellular gene excision. In sum, CasX2Max consistently outperformed native CasX2 across all assays and represents a superior gene-editing platform for therapeutic applications.

Background Colorectal cancer (CRC) is the second leading cause of U.S. cancer mortality. This study evaluated the utilization of CRC screening modalities from 2017 to 2023. Methods This retrospective, cross-sectional study analyzed data from 2017 to 2023 using a national multi-payer claims database, supplemented with laboratory data related to CRC screening. Patients aged 45–75 years, at average risk for CRC, with no prior CRC diagnosis, and who had continuous health insurance enrollment for 24 months (from January 1 of the baseline year to December 31 of the study year) were included. Annual proportions for colonoscopy, multi-target stool DNA (mt-sDNA) test, fecal immunochemical test/fecal occult blood test (FIT/FOBT), and other modalities were assessed, along with sociodemographic factors. Descriptive statistics and chi-square tests were used to assess utilization trends across the years. Results Colonoscopy remained the most commonly used screening modality, with its share increasing slightly from 53.0% in 2017 to 58.7% in 2023. The mt-sDNA test proportion increased significantly from 2.4% in 2017 to 20.4% in 2023, while the proportion of FIT/FOBT declined significantly, from 44% to 20.4%. Similarly, significant age-related shifts in screening utilization were observed, with colonoscopy proportion increasing from 48.1% to 61.6%, mt-sDNA rising from 0.0% to 24.0%, and FIT/FOBT declining from 50.8% to 14.1% in the 45–49 age group from 2017 to 2023. Conclusion CRC screening utilization shifted significantly from 2017 to 2023, with increased colonoscopy and mt-sDNA use and a marked decline in FIT/FOBT. Continued monitoring is critical to ensure equitable access to effective modalities.

<i>Author affiliations are listed at the end of this article.</i>

Background and Study AimsRecent research has identified an association between proximal sessile serrated lesions (SSLs) and an increased risk of advanced metachronous neoplasia (TMAN), with no significant impact from distal SSL. This study aimed to assess the risk of TMAN at follow-up colonoscopy after detecting proximal hyperplastic polyps (HP), adenomas, or their combination at the initial colonoscopy.MethodsMedical records from patients who underwent colonoscopies in 2014 and 2015 were reviewed. The primary outcome was the presence of TMAN (advanced adenomas or high-risk SSL) at follow-up, based on the presence of proximal HP, adenomas, or their combination during the index colonoscopy.ResultsOut of 2014 patients screened, 764 were included in the final analysis (44.1% male; mean age 63 years; median follow-up of 3.46 years). Patients with both proximal HPs and adenomas during the initial colonoscopy had a significantly higher risk of developing TMAN compared with patients with adenomas and distal HP or adenomas alone (30.5% vs 19%; HR = 1.87; 95% CI, 1.3-2.7). Additionally, a combination of proximal HPs and adenomas posed a higher risk of TMAN than proximal HP alone (30.5% vs 13.9%; HR = 3.6; 95% CI, 1.4-9.5). No significant difference in TMAN risk was observed between patients with adenomas alone versus proximal HP (19.1% vs 13.9%; HR = 1.8; 95% CI, 0.73-4.4).ConclusionThe presence of both proximal HPs and adenomas significantly increases the risk of TMAN compared with adenomas or HPs alone, highlighting the need for further studies to evaluate the effect of these variables on postcolonoscopy CRC.