Kelong Han, Pascal Chanu, Fredrik Jonsson, Helen Winter, Rene Bruno, Jin Jin, Mark Stroh
Clinical Pharmacology, Genentech Inc., South San Francisco, CA; Pharsight Consulting Services, Lyon, France; Pharsight Consulting Services, Stockholm, Sweden; Pharsight Consulting Services, Marseille, France
Objectives: The phase III trial OAM4971g comparing onartuzumab (ONT) plus erlotinib vs erlotinib in MET-positive patients with 2nd and 3rd line NSCLC did not meet the primary endpoint of overall survival (OS). An exposure-response analysis was performed to assess whether a higher ONT dose may yield better efficacy without significantly changing the safety profile.
Methods: Data of 636 patients from the phase II study OAM4558g and phase III study OAM4971g were pooled. Tumor growth inhibition (TGI) models were fit to longitudinal tumor size data to estimate individual TGI metrics including time to tumor re-growth (TTG), growth rate constant and tumor size ratio to baseline at week 8. Cox regression models were developed for time-to-event endpoints (progression-free survival [PFS], OS and TTG) and investigated relationships with baseline prognostic (Px) factors and ONT exposure. Incidence of adverse events was modeled by logistic regression.
Results: An exposure-response analysis of MET-positive patients in OAM4558g and OAM4971g suggested longer PFS and OS for patients whose ONT steady-state trough concentration is above the upper quartile. However, exposure-response relationship with these clinical endpoints may be confounded by unobserved Px factors that may impact both OS and exposure. One way to address the confounding issue is to incorporate an explanatory variable, e.g. TGI metrics. After TGI metrics were included while building the OS model, ONT exposure was no longer significant. TTG is the only TGI metric remaining in the final OS model. The final OS model indicated that longer OS is associated with higher albumin, longer TTG, fewer metastatic sites, female gender, ECOG performance score equal to 0, and higher age. On the other hand, ONT exposure was not significantly associated with TTG after adjusting for Px factors. Longer TTG is associated with the presence of EGFR mutation and higher albumin. Finally, higher ONT exposure is associated with increased incidence of infusion reactions and peripheral edema. However, the clinical significance of this exposure-safety relationship has not been determined.
Conclusions: Higher ONT exposure is not significantly associated with improved OS after adjusting for Px factors and TTG, but there was a trend of unknown clinical significance toward increased incidence of infusion reactions and peripheral edema. These results do not support testing a higher ONT dose in this population.
Reference: PAGE 24 (2015) Abstr 3642 [www.page-meeting.org/?abstract=3642]
Poster: Drug/Disease modeling - Oncology