775PTrastuzumab and pertuzumab added to neoadjuvant chemoradiotherapy in resectable HER2+ esophageal adenocarcinoma patients: An update on survival and predictive biomarkers in the TRAP study

Abstract Background Preliminary data from the TRAP study demonstrated feasibility of neoadjuvant chemoradiation (nCRT) with trastuzumab and pertuzumab. Here, we present updated survival results with propensity score matching, including potential predictive biomarkers for response to this treatment. Methods Patients (pts) with HER2+ resectable esophageal adenocarcinoma (EAC) received nCRT with trastuzumab and pertuzumab (Schokker et al, JCO 2018 suppl. 4057). Survival data was compared with patients not receiving study medication in the Dutch Cancer Registry, using propensity score matching (1:4) based on a logistical regression model, matching for baseline demographic and clinical characteristics. HER2 status was not included as a matching variable, since it was unknown in the majority of pts in the matched cohort. Associations between baseline F-FDG uptake on PET scans, measured by the maximum standardized uptake value (SUVmax), or ΔSUVmax and pathological complete response (pCR) were assessed. Furthermore, expression of HER2 and growth factor receptor-bound protein (Grb)7 was assessed by immunohistochemistry on baseline biopsies. Results were correlated with pCR and survival. Results 40 pts were enrolled and pCR was seen in 34%. Progression-free survival (PFS) rates at 1 and 3 years were 83% and 72%, respectively, with 1- and 3-year overall survival (OS) rates of 90% and 71% (median follow-up 32.1 months). A statistically significant difference in OS was observed for nCRT with trastuzumab and pertuzumab compared to nCRT (3-year OS rate 71% vs 54%, p = 0.039). Baseline and post-treatment PET scans were available for 40 and 34 pts, respectively. Baseline SUVmax was not predictive for recurrence or death, neither did ΔSUVmax correlate with pCR (p > 0.05). HER2 3+ pts and pts with Grb7+ tumours at baseline demonstrated a significantly better response to treatment (p = 0.016 respectively p = 0.007). Conclusions Compared to a propensity score matched cohort receiving nCRT, OS of pts with resectable HER2+ EAC receiving nCRT with trastuzumab and pertuzumab is statistically better. Grb7 and HER2 3+ are potential predictive biomarkers for response to this treatment. Clinical trial identification NCT02120911 April 23, 2014. Legal entity responsible for the study The authors. Funding Roche. Disclosure R.H.A. Verhoeven: Research grant / Funding (self): Roche; Research grant / Funding (self): Bristol-Myers Squibb. M. van Oijen: Research grant / Funding (institution): Roche; Research grant / Funding (institution): Servier; Research grant / Funding (institution): Amgen; Research grant / Funding (institution): Merck; Research grant / Funding (institution): Nordic. G.A.P. Hospers: Advisory / Consultancy, Payment to institution: Amgen; Advisory / Consultancy, Payment to institution: Roche; Advisory / Consultancy, Payment to institution: MSD; Advisory / Consultancy, Research grant / Funding (institution), Payment to institution: BMS; Advisory / Consultancy, Payment to institution: Pfizer; Advisory / Consultancy, Payment to institution: Novartis; Research grant / Funding (institution), Payment to institution: Seerave. M. Bijlsma: Research grant / Funding (institution): Celgene; Advisory / Consultancy: Servier. M.I. van Berge Henegouwen: Honoraria (institution), Research grant / Funding (institution): Stryker; Honoraria (institution), Research grant / Funding (institution): Olympus; Honoraria (institution), Advisory / Consultancy: Medtronic; Honoraria (institution), Research grant / Funding (institution): Mylan. H.W.M. van Laarhoven: Honoraria (self), Advisory / Consultancy: Lilly/ImClone; Advisory / Consultancy, Research grant / Funding (institution): Nordic Group; Advisory / Consultancy, Research grant / Funding (institution): Bristol-Myers Squibb; Travel / Accommodation / Expenses: AstraZeneca; Research grant / Funding (institution): Bayer Schering Pharma; Research grant / Funding (institution): Celgene; Research grant / Funding (institution): Janssen-Cilag; Research grant / Funding (institution): Lilly; Research grant / Funding (institution): Philips Healthcare; Research grant / Funding (institution): Roche; Research grant / Funding (institution): Merck Sharp & Dohme. All other authors have declared no conflicts of interest.