Optimizing colorectal cancer NGS pathways: Quality gaps, workflow barriers, and digital/AI readiness.

Abstract

181 Background: Next-generation sequencing (NGS) is integral to colorectal cancer care, as expanding targeted therapies depend on timely molecular profiling. However, timely processing and access to results depend on a multistep pathway, including biopsy, pathologist review, test initiation, laboratory processing, and report generation. Institutional variability across this pathway highlights opportunities to improve molecular testing timeliness and reliability. This study aimed to identify quality gaps in colorectal cancer NGS pathways, evaluate readiness for AI-enabled integration, and define targeted quality improvement (QI) opportunities. Methods: We conducted semi-structured interviews with 23 medical oncologists and pathologists from 15 academic and community hospitals across two Canadian provinces. Interviews explored colorectal cancer NGS workflow steps, including test initiation, laboratory processing, result tracking, reporting, and communication. A dedicated section examined digital infrastructure and AI integration readiness. Transcripts were analyzed thematically in MAXQDA. Results: Five themes emerged. (1) Clinical indications and test ordering: NGS initiation was inconsistent, reflecting uncertainty about clinician-initiated versus reflex testing, reliance on upstream clinicians, staging uncertainty at time of biopsy, evolving reflex testing indications beyond metastatic disease, and funding considerations. (2) Tissue, pathology, and laboratory workflow: Send-out versus in-house testing, laboratory capacity, staffing, and infrastructure contributed to bottlenecks and variable turnaround times. (3) Interdisciplinary communication and tracking: Communication relied on informal pathways, unclear ownership, and manual tracking, highlighting the need for closed-loop confirmation and alerts. (4) Reporting and clinical integration: Reports varied across sites in content, clarity, and clinical utility, with variable need for interpretive support and concerns about timing relative to treatment decisions. (5) EHR integration and AI readiness: Persistent paper/fax processes, fragmented EHR visibility, and limited interoperability contributed to delays; participants identified potential roles for AI in case triage and tracking, while emphasizing human oversight. Conclusions: Colorectal cancer NGS pathways are affected by modifiable system-level barriers across testing, tissue handling, laboratory capacity, reporting, communication, and digital infrastructure. Priority QI targets include standardizing reflex-testing criteria, improving community hospital NGS capacity, clarifying pathway ownership, supporting result processing and tracking through additional personnel and/or AI, and standardizing report formats. Addressing these gaps may improve timely, reliable, and equitable precision oncology access.