Phase II study of AZD4547 in FGFR amplified tumours: Gastroesophageal cancer (GC) cohort pharmacodynamic and biomarker results.

Cancers of the Esophagus and Stomach
Session Type and Session Title: 
Poster Session A: Cancers of the Esophagus and Stomach
Abstract Number: 


Poster Board Number: 
Poster Session A Board #N2
J Clin Oncol 34, 2016 (suppl 4S; abstr 154)
Elizabeth Catherine Smyth, Nicholas C. Turner, Alex Pearson, Clare Peckitt, Ian Chau, David J. Watkins, Naureen Starling, Sheela Rao, Angela Gillbanks, Elaine Kilgour, Katherine Anne Sumpter, Neil R. Smith, Ros Cutts, Claire Rooney, Anne L. Thomas, Mazhar A. Ajaz, Sue Chua, Gina Brown, Sanjay Popat, David Cunningham; Royal Marsden, London, United Kingdom; Institute of Cancer Research, London, United Kingdom; Royal Marsden Hospital, London, United Kingdom; AstraZeneca Oncology Innovative Medicines, Macclesfield, United Kingdom; Northern Center for Cancer Treatment, Newcastle-upon-Tyne, United Kingdom; Leicester Royal Infirmary, Leicester, United Kingdom; Royal Surrey County Hospital, Guildford, United Kingdom; The Royal Marsden and The Institute of Cancer Research, London, United Kingdom

Abstract Disclosures


Background: FGFR amplified tumors demonstrate oncogene addiction in vitro and in vivo. We examined the efficacy of AZD4547, an orally available inhibitor of FGFR 1, 2 & 3, in FGFR1/2 amplified cancers, and herein present results from a gastroesophageal cancer (GC) cohort. Methods: Phase II Simon 2 stage design (3 independent cohorts) for previously treated patients (pts) with advanced FGFR1 (HER2 negative breast/NSCLC) or FGFR2 amplified tumors. FGFR1/2 amplification was centrally assessed using FISH. Pts received AZD4547 80mg twice daily. Primary endpoint is centrally confirmed response rate (RR), with the study concluding efficacy if ≥ 3/17 patients in a cohort had a confirmed response. PET-CT was performed at baseline, D14 and 8 wks, biopsy at baseline and D14 and optionally on progression. Biomarkers included FGFR copy number variation in tumor and plasma, NanoString gene expression, MIRAX digital FISH imaging and phospho-immunohistochemistry. Results: Of 288 pts with screening results, 9% (12/138) gastroesophageal (GC) pts had FGFR2 amplification with 5% (7/138) highly amplified (FGFR2 ratio > 5.0). Confirmed RR was 33% (3/9) in FGFR2 amplified GC. Mean duration of response in responders was 5.7 months. All GC responders had a PET response on D14 PET. Common toxicities (all pts n = 18) included fatigue (72%), constipation (50%) mucositis, skin and eye changes (44% each). Phospho-immunohistochemistry on pre-treatment and D14 biopsies demonstrated an increase in S6 and decrease in p-ERK following treatment. Using digital droplet PCR (ddPCR) elevated FGFR2 copy number was detected in free plasma DNA of all responding GC pts, and no non-responding pt. NanoString and MIRAX imaging analysis revealed the presence of truncated C3 FGFR2 isoform and high level homogenous FGFR2 amplification in all GC responders respectively. Conclusions: AZD4547 demonstrated promising activity in FGFR2 amplified GC. On treatment biopsies demonstrate evidence of target inhibition. Clonal, high level amplification appears to be required for response to AZD4547. As all responding patients had FGFR2 copy number gain detected in plasma using ddPCR this technique will be developed as a screening tool. Clinical trial information: NCT01795768