Background: MetMAb is a monovalent monoclonal antibody that binds specifically to the Met receptor, blocking HGF-mediated activation. Met expression is associated with worse prognosis in NSCLC, and
Met activation has been implicated in resistance to EGFR inhibition in EGFR-mutated NSCLC. Dual inhibition of Met/EGFR may result in promising activity in NSCLC.
Materials and methods: OAM4558g is a global randomized, double-blind phase II study comparing MetMAb (15mg/kg IV q3wks) plus erlotinib (ME) to placebo plus erlotinib (PE) in 2nd/3rd line NSCLC.
Eligible patients on PE arm were allowed to crossover to ME following progression. Patients were stratified by histology, ECOG PS and smoking status. Archival tissue was mandatory for determination
of Met expression by IHC. Co-primary endpoints were PFS in the Met+ and ITT populations. Safety and OS were additional endpoints.
Results: 128 patients were randomized from 3/2009 to 3/2010 to ME (n=64) or PE (n=64). Tissue was evaluable for Met in n=121 and for EGFR and KRAS mutations in n=112. 23 patients crossed over to
ME, following progression from PE. Baseline characteristics were well-balanced in the ITT population including Met+ (51%/57%; PE/ME) KRAS mutation (23%/23%) and EGFR mutation (11%/13%). As of the
June 8 data cut there were 85 PFS and 49 OS events. Both a PFS benefit (HR 0.56; 95% CI 0.31, 1.02; p=0.05,) and an OS benefit (HR 0.55; 95% CI 0.25, 1.16; p=0.11) were observed in the Met+
patients treated with ME. However, in the Met- population, PFS (HR 2.01; 95% CI 1.04, 3.91; p=0.04) and OS (HR 3.26; 95% CI 1.20, 8.80; p=0.01) were worse in the ME cohort. HR for PFS and OS in the
ITT population were 1.09 (95% CI 0.71, 1.67; p=0.70) and 1.13 (95% CI 0.64, 1.97; p=0.68). Selective benefit of ME was not observed in other subgroups, including: nonsquamous, EGFR or KRAS mutants.
Rash, diarrhea, and fatigue were comparable between treatment arms in both Met+ and Met- subpopulations. The incidence of Gr ≥3 AEs was similar in ME v. PE in the Met+ group (54% v. 53%);
however, was higher in the ME arm in the Met- group (52%vs 35%).
Conclusions: The addition of MetMAb to erlotinib in patients with Met+ NSCLC improved both PFS and OS and with no unexpected safety signal. Conversely, Met- NSCLC patients had worse PFS and OS when
treated with ME, and a higher incidence of Gr ≥3 AEs. A prospective study of ME v. PE in patients selected for Met+ tumors is warranted.
We acknowledge that the presentation of this data would not be possible without the significant contributions of Drs. Jiping Zha and Robert Yauch towards the development and interpretation of the
diagnostic assay.