Prospective observational study of prevalence, assessment and treatment of pancreatic exocrine insufficiency (PEI) in patients with advanced pancreatic cancer (aPC): PanDA

Abstract

Background: PEI in patients with advanced pancreatic cancer is well documented, but there is a lack of consensus regarding optimal screening. Methods: Eligible patients for this observational study (NCT03616431) were those diagnosed with aPC referred for consideration of palliative therapy who consented to evaluation by a research dietitian. In addition to symptom and full dietetic assessment (including Mid-Upper Arm Circumference (MUAC), handgrip and stair climb test), full nutritional blood panel, faecal elastase (FE) and 13C mixed triglyceride breath test (for diagnostic cohort (DiC)) were performed. Primary objectives: prospective assessment of PEI prevalence (dietitian-assessed; demographic cohort (DeC)), and to design (using breath test as gold standard; DiC) and validate (follow-up cohort (FuC)) the most suitable screening tool for PEI in patients with aPC. Logistic and Cox regression were used for statistical analysis (Stat v.12). Results: Between 1st July 2018 and 30th October 2020, 112 eligible patients [50 (DeC), 25 (DiC), 37 (FuC)]. Prevalence of PEI in the DeC was 64.0% (PEI-related symptoms were flatus (84.0%), weight loss (84.0%), abdominal discomfort (50.0%) and steatorrhea (48.0%)); 70.0% of patients required pancreatic enzyme replacement therapy and 74.0% had anorexia (low appetite); 44.0% and 18.0% had low vitamin D and vitamin A levels, respectively. Designed PEI screening panel (DiC; 19 patients with breath test completed) included FE [normal/missing (0 points); low (1 point)] and MUAC [normal/missing ( > percentile 25 for age/gender) (0 points); low (2 points)] and identified patients at high-risk (2-3 total points) of PEI [vs. low-medium risk (0-1 total points)]. When patients from DeC and DiC) were analysed together, those classified as “high-risk of PEI” according to the screening panel had shorter overall survival (multivariable Hazard Ratio (mHR) 1.86 (95% CI 1.03-3.36); p-value 0.040) when adjusted for other prognostic factors, including presence of PEI symptoms (mHR 2.28 (95% CI 1.19-4.35); p-value 0.013). The screening panel was tested in the FuC; 78.38% were classified as patients at “high-risk of PEI”; of these, 89.6% were confirmed to have PEI by the dietitian. The panel was feasible for use in clinical practice, (64.8% of patients completed fully the assessments required) and acceptability was high (87.5% of patients would do it again). The majority of patients (91.3%) recommended that all future patients with aPC should have dietitian input. Conclusions: PEI is present in the majority of patients with aPC, and early dietetic input is important to provide a holistic nutritional overview, including, but not limited to, PEI. This proposed screening panel could be used to prioritise patients at higher risk of PEI requiring urgent dietitian input. Its prognostic role needs further validation.