Survival Benefit Associated With Metformin Use In Lung Cancer Patients With Type 2 Diabetes

Background: Although preclinical studies have suggested that metformin may improve the survival outcome of lung cancer,observational clinical studies of metformin for treatment of lung cancer have reported mixed findings.Due to methodological biases(such as immortal time bias)and lack of control of confounding variables in several studies,systematic reviews and meta-analysis have suggested cautious interpretations.In addition,there were no prospective cohort studies to explore the association between metformin use and the survival of lung cancer patients.Objective: The objective of this study was to assess the difference between progressionfree survival(PFS)and overall survival(OS)of patients with lung cancer who were treated with metformin and those treated with other hypoglycemic agents after lung diagnosis.Methods: This study was a prospective cohort study.The hospital information system from Renmin Hospital and Zhongnan Hospital of Wuhan University,face-to-face survey and 3 regular follow-up were used to obtain the information of patients,e.g.,demographic information,hospital inpatient information,drug prescription information and survival information.The subjects were patients diagnosed and/or treated between April 2017 and October 2019,who were diagnosed with lung cancer with type 2 diabetes by cytology or pathology.Kaplan Meier method and univariate and multivariate Cox proportional hazards regression models were used to evaluate the association between metformin use and PFS and OS during follow-up.We conducted a primary analysis in patients which met inclusion criteria.Then,sensitivity analyses were performed according to exclusion criteria by considering the influence of immortal time bias.Follow-ups were up to October 20,2019.Results: 1.Primary analysis1.1 In the primary analysis,a total of 237 lung cancer patients with type 2 diabetes were included.106 patients were exposed to metformin after lung cancer diagnosis,and 88 patients died during follow-up(mean follow-up time: 80.88 weeks).The log-rank test of Kaplan Meier showed that there was no significant difference of OS(P = 0.122)between the two groups,but the PFS(P = 0.012)was statistically significant.Univariate Cox analysis found that metformin could not improve the OS(HR = 0.71,95% CI = 0.45-1.10),but could improve PFS(HR = 0.52,95% CI = 0.31-0.87).After adjusting for covariates with P <0.15 in univariate analysis,metformin was not associated with OS(HR = 1.09,95% CI = 0.61-1.93)and PFS(HR = 0.90,95% CI = 0.48-1.68).2.Sensitivity analysis2.1 In the sensitivity analysis,according to the exclusion criteria,after excluding 93 patients,a total of 143 patients were included.During the follow-up,66 patients used metformin and 77 patients did not use metformin.Log rank test of Kaplan-Meier showed that there was no statistical difference in OS(P = 0.178)and PFS(P = 0.074)between the two groups.In the univariate Cox regression analysis,in terms of OS(HR = 0.64,95% CI = 0.34-1.20)and PFS(HR = 0.59,95% CI = 0.33-1.06),the survival outcomes of patients taking metformin were better than non-metformin,but did not reach statistical significance.2.2 In the sensitivity analysis,in all multivariate analyses,no improvement in PFS and OS was found in the metformin group.2.3 Metformin group and non-metformin group was no difference in PFS and OS in different clinicopathological types.2.4 Stratified analyses were conducted based on whether the patients received chemotherapy,insulin therapy at baseline,and were in stage I–II or III–IV.In the stratification,metformin did not show the improvement of OS and PFS.2.5 In all histological types and non-small cell lung cancer,there is no significant relationship between the start time of metformin use and the clinical outcomes.2.6 Further analysis of the use time after diagnosis of lung cancer showed that,compared to the non-metformin group,metformin group which the cumulative use time was between 0 and 30 weeks,the hazard risk was high,then with the time increase,the hazard risk decreased,the smallest of hazard risk occurred when the metformin has been taken for more than 60 weeks in OS and PFS.Conclusion: In this hospital‐based study,we found no survival benefit with duration of metformin use among lung patients with type 2 diabetic patients in most analyses.However,we observed protective associations between metformin exposure and OS and PFS in the over 60‐week post‐diagnosis period.