Prognostic impact of pSTAT3 on overall survival (OS) in gastrointestinal (GI) cancers: Data from 3 phase 3, randomized controlled trials (RCTs).
Emily Brooks,Claudia Lebedinsky,Tanios S. Bekaii-Saab,Manish A. Shah,Derek J. Jonker,Kamesh Kuchimanchi,Ewa Wybieralska,Marilyn Fontaine,Cindy Oh,Suha Sari,Emma Foos,Bo Xu,Matthew Hitron
DOI: https://doi.org/10.1200/JCO.2021.39.15_suppl.4147
IF: 45.3
2021-01-01
Journal of Clinical Oncology
Abstract:4147 Background: Retrospective studies of phosphorylated STAT3 (pSTAT3, a regulator of gene expression) in cancer-lineage cells have suggested it is a biomarker of poor prognosis. We analyzed data from 3 phase 3 RCTs (BRIGHTER [T1; NCT02178956], CO.23 [T2; NCT01830621], CanStem111P [T3; NCT02993731]) to evaluate the prognostic impact of pSTAT3 on OS in GI cancer patients (pts). Methods: pSTAT3 was evaluated in archival tumor tissue from pts with gastric or gastroesophageal junction adenocarcinoma (T1), colorectal cancer (T2), or pancreatic cancer (T3) who received standard of care in the control arms (CTL) of their respective trials. pSTAT3 positive (+) or negative (−) status was determined by detecting pSTAT3 in both tumor and tumor microenvironment (TME) cells in a laboratory-developed immunohistochemistry (IHC) assay (T1; T2), later developed into the investigational pSTAT3 IHC D3A7 assay (Agilent Technologies, Inc.) (T3). The positivity cutoff requires a sample have both a TME score of 2 and ≥5% of tumor cells staining positively for pSTAT3. Pre-analytical variables (eg, specimen collection, handling, and processing), tumor indication, and slight differences in the assays used for each study may contribute to relative differences in evaluable rates between studies. In an exploratory analysis, OS was compared in pSTAT3+ vs pSTAT3− pts via an unstratified log-rank test. Results: In each study, intent-to-treat pts with evaluable pSTAT3 results were considered pSTAT3-evaluable. This abstract includes pSTAT3-evaluable pts randomized to the CTL arms (T1: n/N=248/357, 69.5%; T2: n/N=110/144, 76.4%; T3: n/N=243/569, 42.7%). Baseline characteristics are presented in the table. Median (m) OS was shorter in pts with pSTAT3+ tumors across all studies (T1 hazard ratio [HR]=1.32 [95% confidence interval (CI): 1.00, 1.74]; T2 HR=2.40 [1.50, 3.83]; T3 HR=1.08 [0.77, 1.50]). Conclusions: In pSTAT3-evaluable populations from 3 RCTs across GI cancers, CTL pts with pSTAT3+ (vs pSTAT3−) tumors had shorter mOS. These results suggest that pSTAT3 positivity may be a biomarker of poor clinical outcome. Clinical trial information: NCT02178956; NCT01830621; NCT02993731. [Table: see text]