BACKGROUND

Despite a wealth of preclinical and observational data, prospective data regarding the use of metformin in lung cancer is extremely limited.

METHODS

We pooled individualized data from two prospective trials evaluating metformin plus platinum-based chemotherapy, with or without bevacizumab, in non-diabetic patients with untreated advanced NSCLC. In addition to reporting on clinical efficacy and safety endpoints, we also explored metformin's activity in key molecular cohorts.

RESULTS

33 patients were included in the pooled analysis, of whom 70% were current or previous smokers. 82% had standard tissue molecular testing results available. KRAS, EGFR, and LKB1 mutation prevalence was 48%, 26%, and 8.3%, respectively. Composite median PFS was 6 months for all patients (95% CI: [1.36, 7.96]), 7.2 months for KRAS mutants (95% CI: [1.18, 9.21]), and 6.6 months for EGFR mutants (95% CI: [1.18, 15.29]). Composite median OS was 14.8 months for all patients (95% CI: [8.25, 19.99]), 17.5 months for KRAS mutants (95% CI: [8.86, 26.96]), and 13.3 months for EGFR mutants (95% CI: [2.60, 25.86]). Lymphopenia was the most common grade 3 AE (12%), followed by leukopenia, nausea, vomiting, and hypertension (9% each). There were 2 grade 4 AEs, neutropenia (21%) and sepsis (3%), and 1 grade 5 AE (colonic perforation) attributed to bevacizumab.

CONCLUSION

Our results confirm the previously shown efficacy and tolerability of metformin in combination with chemotherapy and highlight encouraging activity in key molecular cohorts. Future efforts should build on this work by prospectively studying metformin in these molecular subgroups.