PPS58. Distinct tumor biology between IPMN-derived and PanIN-derived early-stage pancreatic cancer

Intraductal papillary mucinous neoplasm (IPMN) and pancreatic intraepithelial neoplasia (PanIN) are well-established precursor pathways to pancreatic ductal adenocarcinoma (PDAC). Neoadjuvant therapy can alter tumor cellularity, potentially reducing the sensitivity of molecular analyses. This study examines treatment-naïve early-stage PDAC patients, focusing on the clinicopathologic and genomic differences between these two subtypes.


Methods:

Patients from 1999 to 2024 with resected, early-stage (T1N0M0) PDAC who did not receive neoadjuvant therapy were identified from our institutional database. Tumors associated with IPMN (IPMN-derived) on pathology were compared to those that did not have this distinction (PanIN-derived). In addition, we conducted an exploratory molecular analysis on a subset of tumors with next-generation DNA targeted sequencing data. Clinicopathological features and genomic variables were compared using Fisher’s Exact test for categorical variables and Mann-Whitney test for continuous variables. Kaplan-Meier method was used for time-dependent endpoints. 


Results:

185 patients met inclusion criteria with 112 in the IPMN-derived group and 73 in the PanIN-derived group. No patients received neoadjuvant therapy. PanIN-derived PDAC patients had larger tumors (P < 0.001) that were more likely to be poorly-differentiated (P < 0.001) or have perineural invasion (P < 0.001). Median overall survival was significantly longer in the IPMN-derived group (124 months vs. 59 months, P < 0.001). 11 patients with IPMN-derived PDAC and 22 patients with PanIN-derived PDAC had genomic data. Mutations in KRAS (96% vs. 64%, P = 0.033), with a trend towards KRAS G12D (41% vs. 9%, P = 0.11), were enriched in the PanIN-derived group compared to KRAS G12V (45% vs 9%, P = 0.027) in the IPMN-derived group. Mutations in APC were also enriched in the IPMN-derived cohort (27% vs 0%, P = 0.030). Alterations in TP53, SMAD4, and RNF43 were not statistically significant.   


Conclusions:

Our findings not only reinforce that PanIN-derived PDAC demonstrates more aggressive tumor biology but also highlight the distinct patterns of genomic enrichment. KRAS G12V mutations were more frequently observed in IPMN-derived tumors, compared to KRAS G12D mutations in PanIN-derived tumors. Validation in larger cohorts will be essential to confirm these findings. Greater emphasis should be placed on integrating comprehensive genomic profiling to guide personalized treatment decisions, especially with emerging mutation-specific therapies.