The authors compared metformin combined with radiotherapy and chemotherapy vs chemoradiotherapy alone in patients with unresected stage IIIA/B non–small-cell lung cancer.
CHICAGO-In patients with unresected stage IIIA/B non–small-cell lung cancer (NSCLC), treatment with metformin combined with radiotherapy and chemotherapy did not improve 1-year progression-free survival (PFS) or overall survival (OS) compared with chemoradiotherapy alone, according to research (abstract 8502) presented at the 2019 American Society of Clinical Oncology (ASCO) Annual Meeting, held May 31–June 4 in Chicago.
The unwelcome news follows preclinical models that previously suggested that metformin, which inhibits mitochondria complex I, could boost the efficacy of radiotherapy and chemotherapy.
Metformin blocks mitochondrial respiration, thus causing metabolic stress, which led researchers to believe that it might activate tumor suppression pathways and inhibit lung cancer proliferation, among other beneficial effects. It could also enhance NSCLC response to chemotherapy and radiotherapy.
The NRG-LU001 trial was designed to detect a 15% improvement in 12-month PFS. Researchers randomized 167 non-diabetic subjects to carboplatin/paclitaxel chemotherapy concurrent with 60 Gy chest radiotherapy for 6 weeks, followed by 6 weeks of consolidation carboplatin/paclitaxel chemotherapy (control) or the same regimen plus 2,000 mg oral metformin administered alongside 6 weeks of cytotoxic therapy, with 10 weeks of metformin.
A pre-planned analysis occurred in February 2019, after 102 PFS events. Metformin tolerability was generally good, with levels of gastrointestinal disorders, respiratory/thoracic/mediastinal disorders, vascular disorders, and metabolism/nutritional disorders generally similar between the two groups.
However, metformin compliance was worse than expected. The per-protocol analysis included 63.4% of patients in the metformin arm, and just 39% completed the study with no modification to metformin dose.
There was no significant difference between the treatment and control arm with respect to PFS in the intention-to-treat analysis (median metformin, 12.2 months vs 16.6 months placebo; hazard ratio [HR], 1.15; P = .2441). There was also no difference in 1-year PFS in the per-protocol analysis (60.4% control vs 51.3% treatment) or 2-year PFS (40.1% vs 34.5%). Median PFS in the metformin group was 15.4 months vs 16.6 months placebo.
Results were similarly disappointing with respect to OS. There was no significant difference in the metformin group on the intention-to-treat analysis (median metformin, 40.1 months vs 38.5 months placebo; HR, 1.03; P = .8910) or in a 2-year OS per-protocol analysis (65.4% vs 64.9%). A total of 91% of deaths in the control arm were due to disease, compared with 70% in the metformin arm.
More work needs to be done to understand the reduced compliance in the metformin treatment group, said Theodoros Tsakiridis, MD, PhD, associate professor at McMaster University, during his presentation of the results.