# Morphological and functional alterations in type 2 diabetes pancreata assessed with MRI-based metrics and [18F]FP-(+)-DTBZ PET

## TL;DR

This study combines PET and MRI imaging with machine learning to better predict beta-cell function in type 2 diabetes.

## Contribution

The novel contribution is combining PET-derived beta-cell mass with MRI morphology metrics to predict insulin responses in T2D.

## Key findings

- T2D individuals had significantly reduced acute and maximum insulin responses compared to healthy obese volunteers.
- A model combining PET-derived SUVR-1 and clinical covariates best predicted acute beta-cell function.
- Predicting maximum functional reserve required adding MRI-based morphology metrics to PET and clinical data.

## Abstract

To determine if combining PET-derived beta-cell mass (BCM) estimates with MRI-based morphology metrics improves the prediction of beta-cell functional mass in type 2 diabetes (T2D).

We performed a retrospective analysis of 40 participants—19 T2D individuals, 16 healthy obese volunteers (HOVs), and five prediabetes individuals—who underwent [18F]FP-(+)-DTBZ PET to quantify vesicular monoamine transporter type 2 (VMAT2) density [standardized uptake value ratio (SUVR-1)], T1-weighted MRI for 3D morphology metric analysis, and an arginine stimulation test to measure acute (AIRarg) and maximum (AIRargMAX) insulin responses. Least Absolute Shrinkage and Selection Operator (LASSO) regression models identified the optimal combination of positron emission tomography (PET), MRI, and clinical variables to predict beta-cell function for the whole pancreas and its subregions.

Compared to HOVs, individuals with T2D exhibited significantly reduced AIRarg and AIRargMAX. Only the pancreas body volume was significantly smaller in the T2D cohort. For the whole pancreas, a model including PET-derived SUVR-1 and a subset of clinical covariates best predicted acute beta-cell function (AIRarg). However, predicting maximum functional reserve (AIRargMAX) required the addition of MRI-based morphology metrics in combination with SUVR-1 and a subset of clinical covariates.

We combined PET imaging of BCM and MRI morphology metrics with a robust machine learning-based variable selection method to extract useful PET- and MRI-based metrics for predicting acute and maximum insulin responses. This synergistic approach offers a novel combination of biomarkers for staging disease and evaluating therapeutic interventions.