Phase III trial (NGR015) with NGR-hTNF plus best investigator choice (BIC) versus placebo plus BIC in previously treated patients with advanced malignant pleural mesothelioma (MPM).

Abstract

7501 Background: Currently, there are no standard options for MPM patients who failed a pemetrexed-based chemotherapy (CT). NGR-hTNF, a tumor-targeted antivascular agent, displays antitumor activity through a vessel normalization that improves intratumor CT uptake and T-cell infiltration. Methods: MPM patients who progressed on or after a front-line pemetrexed-based regimen, stratified for performance status (PS) and CT agent, were randomly assigned to receive weekly NGR-hTNF 0.8 μg/m2 (arm A; n = 200) or placebo (arm B; n = 200), both given with BIC (gemcitabine [G], vinorelbine [V], doxorubicin [D] or supportive care). Primary endpoint was overall survival (OS). Hypothesis testing: hazard ratio (HR) = 0.72, 1-β = 0.80, α = 0.05. Results: Baseline characteristics were balanced between arms (A vs B): median age (65 vs 67 years); men (76% vs 74%); PS ≥ 1 (72% vs 69%); nonepithelial histology (15% vs 19%); poor EORTC score (30% vs 23%); prior treatment-free interval (TFI) < median of 4.8 months (47% vs 53%). Investigator-selected CT (n = 381, 95%): G 55%, V 42%, D 3%. Patients completing six CT cycles: 41% vs 32% (p = 0.08). Most common grade 3/4 toxicity: neutropenia (17% vs 19%) and fatigue (5% vs 8%). After a median follow-up of 18.9 months, OS did not differ significantly between arms in ITT analysis (median 8.4 vs 7.9 months; HR = 0.94 p = 0.61). By predefined OS analyses, there was a significant interaction only between treatment group and TFI (p = 0.008). In 198 patients with TFI shorter than 4.8 months after first-line therapy, median OS for NGR-hTNF vs placebo was 9.0 vs 6.3 months and 1-year OS was 39% vs 23%, respectively (HR = 0.69 p = 0.02; stratified HR = 0.65 p = 0.01). By CT agent, median OS for NGR-hTNF plus G vs placebo plus G was 9.0 vs 6.2 months and for NGR-hTNF plus V vs placebo plus V was 9.7 vs 6.9 months. A significant treatment-by-TFI interaction was also observed for PFS (p = 0.009), with 6-month rates in the short TFI subset of 25% for NGR-hTNF and 12% for placebo (HR = 0.71 p = 0.03). Conclusions: Though the primary endpoint was not met, OS and PFS benefit reported with NGR-hTNF plus CT in patients with short TFI deserves a confirmatory first-line phase III trial. Clinical trial information: NCT01098266.