# Validation of an Integrated Clinical Biomarker Diagnostic Model for Acute Pancreatitis: Incorporating Trypsinogen-Activating Peptide and Trypsin-2 in a Romanian Population Study

## TL;DR

This study validates a new diagnostic model for acute pancreatitis using biomarkers to classify patients into risk groups for better clinical decisions.

## Contribution

A novel six-parameter risk assessment model integrating pancreatic biomarkers and systemic inflammation for early detection of severe acute pancreatitis.

## Key findings

- The model showed good internal consistency (Cronbach’s alpha = 0.72) and explained 65% of variability with two factors.
- TAP and trypsin-2 were significantly correlated with severe outcomes, with ROC analysis showing 82.4% sensitivity and 76.8% specificity.
- Temporal biomarker changes indicated TAP resolution and increasing trypsin-2, reflecting persistent pancreatic damage.

## Abstract

Introduction: Severe acute pancreatitis (SAP) is a critical condition that affects 20–30% of people with acute pancreatitis (AP). Prompt detection and accurate classification are crucial to direct prompt interventions, increase resource allocation, and improve patient outcomes. Current scoring systems, while beneficial, frequently face challenges related to speed, complexity, and early predictive accuracy. Method: We developed and validated an effective six-parameter risk assessment scale for AP, incorporating pancreatic-specific biomarkers (trypsinogen-activating peptide [TAP], trypsin-2), systemic inflammation markers (C-reactive protein), pancreatic enzyme concentrations, blood glucose, and patient age. The study cohort included 104 patient samples. Reliability was assessed using Cronbach’s alpha and Spearman–Brown coefficients, factorial validity was determined by principal component analysis, and predictive validity was analyzed using logistic regression and receiver operating characteristic (ROC) analysis. Biotemporal changes at 24 and 48 h were assessed to classify risk scoring. Results: The scale demonstrated satisfactory internal consistency (Cronbach’s alpha = 0.72) and a distinct structure with two factors representing local pancreatic damage and systemic inflammation, explaining 65% of the variability. Logistic regression established predictive validity for serious outcomes, with TAP and trypsin-2 showing significant correlations. ROC analysis demonstrated remarkable discriminative capacity (AUC = 0.85), showing a sensitivity of 82.4% and a specificity of 76.8%. Assessment of temporal biomarkers showed a reduction in TAP, signifying resolution of the initial enzymatic activation, while trypsin-2 levels continued to increase, indicating persistent damage to the pancreatic tissue. Patients were classified into low-, moderate- and high-risk groups, facilitating practical clinical decision-making. Discussion and Conclusions: This six-parameter risk score provides a rapid, biologically based, and clinically useful method for early detection of patients at risk for SAP. Combining indicators of local pancreatic involvement with systemic inflammation allows for prompt triage, improves the allocation of intensive therapy, and supports informed prognostic conversations.