Impact of inaccurate clinical staging for gastric cancer on patient survival outcomes.

Abstract

Listen

Translate article

345 Background: Accurate clinical staging (CS) in gastric cancer is critical for appropriate treatment selection and prognostication, but CS remains highly ineffective. Our study aims to evaluate the factors associated with inaccurate CS, the impact of inaccurate CS on patient outcomes, and effect of adjuvant therapy in patients with inaccurate CS. Methods: We conducted a retrospective review of the NCDB of patients diagnosed with clinical early-stage gastric adenocarcinoma based on AJCC 8th edition (cT1-2, N0, M0) between 2004-2016. Those who did not undergo upfront surgery or had missing pathologic staging data were excluded. Patients were classified into 3 groups: accurately staged (AS) if pathologic staging confirmed early-stage cancer, inaccurately staged with receipt of adjuvant therapy (IS+), and inaccurately staged with no receipt of adjuvant therapy (IS-). Logistic regression using stepwise selection was utilized to assess the impact of factors on CS accuracy and receipt of adjuvant therapies. Kaplan-Meier and Cox Proportional Hazard methods were used to compare survival outcomes. Results: Approximately 39% of patients (2841/7199) were understaged. T2 tumors, non-well differentiated tumors, and diffuse type histology were associated with increased likelihood of inaccurate CS. Age >60, female sex, Asian/Black race, and non-cardia tumor location were associated with decreased likelihood of inaccurate CS. Only 44% of patients who were inaccurately staged received adjuvant chemotherapy/radiation. Age >75 and fundus/body tumor location were associated with decreased likelihood of receiving adjuvant therapies, while more advanced pT and pN stage were associated with increased likelihood. Treatment facility type (community vs. academic) had no impact on likelihood of accurate CS or receipt of adjuvant treatment after inaccurate CS. 5-year overall survival was significantly different between groups (71.7% AS, 48.3% IS+, 51.1% AS-; p<0.001). Conclusions: CS is inadequate, and understaging has detrimental effects on patient survival outcomes. Novel strategies for improved CS are needed to improve patient care.