C. A. Hester1, M. R. Porembka1, M. A. Choti1, P. M. Polanco1,2, J. C. Mansour1, R. M. Minter1, S. C. Wang1, A. C. Yopp1 1University Of Texas Southwestern Medical Center,Surgical Oncology,Dallas, TX, USA 2Department Of Veterans Affairs North Texas Health Care System,Surgical Oncology,Dallas, TX, USA
Introduction:
Pancreatic adenosquamous carcinoma (PASC) is a histopathologic diagnosis distinct from pancreatic adenocarcinoma (PAC), characterized by ≥ 30% malignant keratinized squamous cell histology admixed with ductal adenocarcinoma. A paucity of data regarding the natural history of PASC and clinicopathological variables associated with outcome has limited value in individual patient counseling and therapeutic decision-making, especially in comparison to the more prevalent PAC histology. The aims of this study are to characterize the clinicopathological variables of PASC associated with outcome and compare these variables with PAC in surgically resected patients.
Methods:
We conducted a retrospective analysis of the prospectively collected National Cancer Database participant user file between 2004 and 2012. All patients with ICD-O-3 morphological codes corresponding to PASC and PAC were included for analysis. Patients with missing vital status data or in situ disease were excluded. Demographics, tumor characteristics, treatment regimens, and outcomes were abstracted. Differences between the groups were determined with Fisher’s exact and Chi-squared tests. Survival was estimated and compared using Kaplan-Meier and log-rank. Multivariate analysis was performed to determine variables associated with overall survival.
Results:
Of the 207,073 patients meeting the inclusion/exclusion criteria, 205,328 PAC and 1,745 PASC histologies were identified. There was no significant difference in age, race/ethnicity and insurance status between patients with PAC and PASC. PASC patients have tumors that are significantly larger (56% vs 33% with tumors ≥ 4 cm, p<0.001), more likely to originate in the pancreatic body and tail (36% vs 24%, p<0.001), undifferentiated histology (41% vs 17%, p<0.001), higher rate of positive lymph nodes (22% vs 15%, p<0.001), and presention at AJCC stage I/II (39% vs 32%, p<0.001). Patients with stage I and II PASC are more likely to undergo curative-intent surgery compared to PAC (75% vs 54%, p<0.001) and have lower rates of lymph node involvement within surgical specimens (29% vs 39%, p<0.001). There is no significant difference in overall survival when comparing all patients with PAC and PASC (6.2 months and 5.7 months, p=0.601). However, patients with PASC undergoing curative-intent surgery have significantly worse outcome (median survival 12.9 months vs 19.1 months, p<0.001). In patients with PASC, increased Charlson comorbidity score, positive lymph node status, AJCC stage III/IV, and lack of receipt of treatment (surgical, chemotherapy, and/or radiotherapy) were associated with worse overall survival.
Conclusion:
Although curative-intent surgery is more often performed in PASC patients, PAC histology is more favorable with regards to overall survival. Despite relative infavorable biology, receipt of therapy is associated with improved survival in patients with early and late stage PASC.