Clinical to pathological staging differences for upfront surgery vs neoadjuvant therapy for pancreatic adenocarcinoma

Abstract

Presenter: Omid Salehi MD | Saint Elizabeth's Medical Center Background: Inaccurate clinical staging of patients with pancreatic ductal adenocarcinoma (PDAC) may impact not only on prognostication but also on treatment, resulting in chemotherapeutic and surgical over- and undertreatment. This study assesses factors associated with stage migration stratified by treatment sequence (i.e. upfront surgery (UFS) vs neoadjuvant therapy (NAT)). Methods: The National Cancer Database was searched for patients diagnosed with clinical stage I-III PDAC undergoing curative-intent surgery between 2004-2016. Patients were stratified by stage migration and treatment sequence. Logistic regression was used for factors associated with downstaging, upstaging, and accuracy of clinical staging. Cox multivariable regression models and Kaplan-Meier survival analyses were performed to analyze the impact of treatment sequence on survival, stratified by stage migration. Results: 2466 patients met inclusion criteria; 2174 (88.2%) had UFS and 292 (11.8%) had NAT. 1523 (61.8%) had no stage migration between clinical to pathological stage, 850 (34.5%) were upstaged, and 93 (3.8%) were downstaged. Overall accuracy of clinical staging was 62.9%. When stratified by treatment sequence, UFS was associated with upstaging, whereas NAT was associated with downstaging and improved survival (HR 0.839, P=0.027). NAT was the only factor predicting downstaging of disease (OR 17.155, P < .001), whereas factors predicting upstaging included advanced tumor grade, more recent year of diagnosis, and treatment at an integrated cancer network. Factors associated with decreased clinical staging accuracy included Asian/other race, tumor size, government insurance, extent of resection, more recent year of diagnosis, and treatment at an integrated cancer network. Additionally, subgroup analysis of NAT vs. UFS patients demonstrated stage migration to be associated with improved survival [P=0.019]. Conclusion: Although image quality and staging accuracy is improving, clinical staging continues to underestimate stage II-III PDAC. This clinical staging inaccuracy is affected by not only image quality, but also modifiable factors such as insurance status and treatment facility. As NAT vs UFS is strongly associated with not only downstaging but also a survival benefits (even if the rare downstaging doesn’t occur), all resectable PDAC patients should be considered for NAT.