Long-Term Survival Outcomes Of A Placebo-Controlled Phase 3 Trial With Ngr-Htnf In Combination With Best Investigator Choice In Relapsed Malignant Pleural Mesothelioma (Mpm).

8567 Background: NGR-hTNF is a vascular-targeting agent able to increase intratumoral chemotherapy penetration and T-cell infiltration by modifying tumor microenvironment. MPM patients failing first-line therapy have limited treatment options. Methods: In the phase 3 trial NGR015 (ASCO 2015; abs 7501), 400 patients received single-agent gemcitabine, vinorelbine or doxorubicin ± NGR-hTNF as second-line therapy. By ITT analysis, overall survival (OS) did not differ between arms (median follow-up 18.7 months; data maturity 75%). In subgroup analyses, there was a significant interaction only between treatment and treatment-free interval (TFI) after first-line therapy. NGR-hTNF improved OS and PFS in the short TFI subset ( < median: 4.8 months; n = 198). We assessed long-term outcomes in patients with short TFI and prognostic TFI value after adjusting for baseline covariates (age, sex, PS, histology, EORTC score, NLR, response to prior therapy and selected chemotherapy). Results: At a median follow-up of 33.8 months (data maturity 86%), a treatment-by-TFI interaction for OS persisted in univariate (p = 0.004) and multivariable models (p = 0.002). In the short TFI subset, median OS was 9.2 months (95% CI 6.6-11.8) for NGR-hTNF vs 6.3 months (5.7-7.0) for placebo (HR 0.67; 0.48-0.93; p = 0.02; adjusted HR 0.64; 0.45-0.89; p = 0.008). Survival rates were 39% (29-50) vs 24% (16-32) at 1 year and 18% (9-26) vs 9% (3-15) at 2 years. The 25% survival rate occurred 7 months later with NGR-hTNF than with placebo (17.3 months vs 10.2 months). NGR-hTNF treatment was associated also with improved PFS (median 3.4 vs 1.9 months; HR 0.66; 0.48-0.91; p = 0.01; adjusted HR 0.65; 0.47-0.90; p = 0.009). In sensitivity analyses, consistent results were found using a 6-month TFI cutoff. The prognostic TFI value was assessed in the control group to avoid confounding effects of experimental treatment. A short TFI independently correlated with worse OS (HR 1.79; 1.29-2.50; p = 0.0006). Conclusions: Benefit with NGR-hTNF treatment in the short TFI subgroup was maintained after a 3-year follow-up, deserving a confirmatory randomized trial as these MPM patients have a poor prognosis. Clinical trial information: NCT01098266.