Pancreatic adenosquamous carcinoma (PASC): A comparative genomic landscape study.

Abstract

4153 Background: PASC accounts for 1-4% of primary exocrine pancreatic malignancies and is associated with more aggressive disease biology and worse clinical outcomes compared to conventional pancreatic ductal adenocarcinoma (PDAC). Despite its aggressive nature, PASC lacks targeted treatment choices and is frequently excluded from clinical trials, underscoring a key unmet need. Here, we performed a comparative analysis of the genomic landscapes of PASC versus PDAC using the FoundationOne database to identify distinct molecular drivers and potential therapeutic targets. Methods: Comprehensive genomic profiling using hybrid capture-based next-generation sequencing (NGS) was performed on 244 PASC and 29,021 PDAC tumors to identify genomic alterations (GAs). All patients (pts) had clinically advanced disease, predominantly stage IV, at the time of profiling. Genomic ancestry, MSI status, tumor mutational burden (TMB), homologous recombination deficiency signature (HRDsig) and cosmic trinucleotide signature were assessed. PD-L1 expression was quantified using the tumor proportion score (TPS) via the Dako 22C3 immunohistochemistry assay. Statistical analysis was performed using Fisher’s exact test, with false discovery rate (FDR) correction applied through the Benjamini-Hochberg method. Results: PASC harbored a median of 6 GAs per tumor (range: 1-23) with a similar genomic ancestry profile compared to PDAC. Of note, PASC featured a higher frequency of cases with MSI-high status (2.1% vs 0.5%; p = .027) and TMB > 10 mutations/Mb (3.7% vs 1.3%; p = .013). Additionally, PD-L1 expression (TPS > 1%) was significantly more common in PASC compared to PDAC (66.7% vs 37.0%; p < .0001). The frequency of KRAS mutations and HRDsig positivity was similar between the two subtypes. Disease-associated GAs more frequent in PASC than PDAC included mutations in CDKN2A (77.0% vs 56.6%; p < .0001), KMT2D (8.2% vs 3.2%; p < .0001), TP53 (89.3% vs 78.0%; p < .0001), and MTAP loss (33.3% vs 23.8%; p = .002). Conclusions: PASC exhibits a genomic profile with molecular features that are both shared and distinct compared to PDAC. Given the similar frequency of KRAS mutations, PASC pts should be included in clinical trials of emerging RAS-targeted therapies. Furthermore, immunotherapy-based strategies (↑MSI-high, TMB, and PD-L1) and PRMT5/MAT2A inhibitors (↑MTAP loss) warrant consideration in this rare and understudied disease subtype. PDAC (n=29,021) PASC (n=244) P-value Median GAs/tumor (range) (IQR) 5 (0-61) (3-6) 6 (1-23) (4-8) <.0001 MSI-high 0.5% 2.1% .027 TMB > 10 muts/Mb 1.3% 3.7% .013 PD-L1 TPS > 1% 37.0% 66.7% <.0001 HRDsig+ 4.6% 3.1% NS CDKN2A 56.6% 77.0% <.0001 KRAS 92.8% 95.5% NS MTAP loss 23.8% 33.2% .002 TP53 78.0% 89.3% <.0001