Integrative Transcriptomic and Proteomic Profiling Defines Secretory and Immune Related Subtypes in Pancreatic Ductal Adenocarcinoma

Background: Pancreatic ductal adenocarcinoma (PDAC) is
characterized
by extreme molecular heterogeneity and poor prognosis. Understanding
its multiomics molecular heterogeneity may facilitate precise classification
and individualized treatment. Methods: Transcriptomic and proteomic
profiles from 140 PDAC patients were integrated using Similarity Network
Fusion (SNF) to identify robust molecular subtypes. Sparse partial
least-squares discriminant analysis (sPLS-DA), differential expression
analysis, and network-based functional enrichment were employed to
characterize subtype-specific features. Associations between molecular
subtypes, clinicopathological variables, and patient survival were
further assessed. Results: Three PDAC subtypes were identified with
distinct molecular and clinical profiles. The basic subtype (SNF-1)
showed no significant transcriptomic or proteomic distinction. The
pancreatic secretory subtype (SNF-2) exhibited highly active endocrine
and exocrine secretion pathways, suggesting strong dependence on pancreatic
secretory function. The glycolytic–immune suppressive subtype
(SNF-3) displayed prominent immune-evasion features and activation
of multiple drug-resistance pathways, consistent with its poorest
prognosis. These findings reveal distinct molecular programs underlying
PDAC heterogeneity and its clinical manifestations. Conclusion: This
integrative multiomics analysis delineates three biologically and
clinically distinct PDAC subtypes. The results highlight metabolic
and functional diversity underlying tumor heterogeneity, offering
new insights for precision stratification and the development of targeted
therapeutic strategies.