831 Background: Diagnosis and first-course treatment of gastrointestinal malignancies pose multifactorial and evolving stressors, causing cancer-related distress with implications for treatment tolerance, adherence, and oncologic outcomes. Comprehensive longitudinal distress screening data in this population have not previously been reported; the natural history of distress and its association with overall survival (OS) is poorly characterized. Methods: Patients undergoing first-course treatment for gastrointestinal malignancies from 2019-2024 at our institution were longitudinally screened with a hybrid online/in-clinic approach. Fifteen unique domains of cancer-related distress were evaluated (no, mild, moderate, severe). The presence of multi-domain distress (moderate/severe distress in ≥5 domains) was evaluated with logistic regression; OS was assessed with Simon-Makuch Kaplan-Meier analysis and discrete-time survival models. Results: 2253 patients (n=10,926 screens) were identified ); screening response rate was 99.2%. The most common primary sites were pancreatic (27.7%), colonic (18.4%), rectal (11.4%), gastric (10.1%), small intestinal (9.9%) and hepatobiliary (8.1%). 819 (36.3%) patients developed multi-domain (≥5 domain) distress, most commonly at first screening (n=537, 65.6%); fatigue (76.6%), pain/numbness (73.1%), spouse/family concerns (69.0%), work/home life disruptions (67.4%), and anxiety (63.2%) were most common. On multivariable logistic regression, male sex (OR 0.61, P <0.01) was associated with reduced odds of multi-domain distress at diagnosis, while pancreatic primaries (OR 1.98, P <0.01), Black race (OR 2.85, P <0.001) and distant metastases (OR 1.74, P= 0.01) were associated with increased. Development of multi-domain distress was associated with worse subsequent OS at 1- (69.3% vs. 86.1% for no multi-domain distress) and 3-years (49.7% vs. 69.3% for no multi-domain distress, P <0.001). On survival modeling, multi-domain distress development was independently associated with higher hazard of subsequent death (HR 5.97, P <0.001). Conclusions: In the first longitudinal study of comprehensive distress outcomes in gastrointestinal malignancies, we demonstrate that multi-domain cancer-related distress is highly prevalent and independently associated with worse OS. Furthermore, one-third of multi-domain distress is identifiable only with longitudinal screening, highlighting its importance in helping to provide timely and tailored supportive care interventions. Characteristic N (%) or Median IQR Age, years 64 54–72 Female 1050 (46.6%) Metastatic at Diagnosis 697 (30.9%) Surgery 1367 (60.7%) Chemotherapy 1400 (62.1%) Radiation 400 (17.8%) Distress Screens 2 1–5 Inter-Screening Interval, Months 2.1 1.2–5.7 Oncologic Follow-Up, Months 30.2 13.9–66.0
Sutton et al. (2026) studied this question.