Matteo Delli Carria,b, Giacomo Pepea, Eduardo Maria Sommellaa, Manuela Giovanna Basilicatac, Pietro Campigliaa

aDepartment of Pharmacy, University of Salerno, Via G. Paolo II, Fisciano, 84084 Salerno, Italy.
bPhD Program in Drug Discovery and Development, University of Salerno, Fisciano, 84084 Salerno, Italy
cDepartment of Advanced Medical and Surgical Sciences, University of Campania “Luigi Vanvitelli”, Naples, Italy

Intraductal Papillary Mucinous Neoplasms (IPMNs) are pancreatic cystic lesions that can progress to invasive cancer, making early and accurate differentiation between Indolent and High-Grade IPMN critical for patient management. While Indolent IPMNs exhibit slow progression and a lower risk of malignancy, High-Grade IPMNs are more aggressive, with a significantly higher potential for cancer development [1]. Metabolomics, the study of small molecules (metabolites) in biological samples, has emerged as a powerful tool for unraveling disease mechanisms and identifying potential biomarkers. By analyzing metabolic profiles, it is possible to gain insights into disease progression and the underlying biochemical alterations in conditions like IPMN. In this study, we applied an untargeted metabolomics approach using Gas Chromatography-Mass Spectrometry (GC-MS) to analyze 126 plasma samples from patients diagnosed with either Indolent or High-Grade IPMN. 

The resulting metabolomic data underwent comprehensive pre-processing, including probabilistic quotient normalization (PQN), log10 transformation and autoscaling. Both univariate and multivariate techniques were employed using an in-house MATLAB script: univariate analysis was conducted using volcano plots to identify metabolites that were significantly different between the two IPMN groups, while multivariate analysis, including Principal Component Analysis (PCA), was used to visualize patterns and clusters within the data, aiding in the distinction between Indolent and High-Grade IPMNs. Glycolic acid notably emerged as a key metabolite, significantly distinguishing between Indolent and High-Grade IPMNs, consistent with previous studies and reinforcing its potential role in IPMN diagnostics [2]. In addition, our study uncovered several novel metabolites not previously linked to IPMN pathology. Moreover, through data preprocessing and feature selection of identified metabolites, we developed a classification model using a Linear Support Vector Machine (SVM). This model achieved an accuracy of 84.2% in distinguishing between Indolent and High-Grade IPMNs during testing, demonstrating the potential of GC-MS-based metabolomic profiling to enhance diagnostic precision and inform clinical decision-making for IPMN management.

Keywords: Intraductal Papillary Mucinous Neoplasms, IPMN, metabolomics, GC-MS

References

1. Machado N.O., al Qadhi H., et al., Am J Med Sci., 7 (2015) 160. DOI: 10.4103/1947-2714.157477 

2. Luo X., Liu J., et al, Pharmacological Research, 156, (2020) 104805. DOI: 10.1016/j.phrs.2020.104805