Effect of Exposure to Inhaled Corticosteroids on Lung Cancer Incidence - A Comparison of Earlier Study Methodology on the Financial Database of the Hungarian National Health Insurance Fund.

BACKGROUND

Lung cancer (LC) is one of the main causes of death among COPD patients. Similar risk factors and pathophysiology motivated multiple retrospective studies investigating a possible risk reduction effect of inhaled corticosteroids (ICS) use among COPD patients, yet there is still no consensus on the issue.

METHODS

Using the financial database of the National Health Insurance Fund, involving the overall population of Hungary, two retrospective analyses were performed between 2009 and 2019, based on the contradicting works of Suissa and Raymakers. Both analytical methods were used to recreate the study populations and compare the outcomes. Cox proportional hazard models were used to estimate the effect of ICS use on the risk of LC. The models were corrected for immortal time bias.

RESULTS

Altogether, 96,686 and 196,560 patients were enrolled in the Suissa and Raymakers models, respectively. Average age was 65.7 (standard deviation - SD: 9.14) and 65.3 (SD: 9.37) years, with 48.4% and 56% women correspondingly. Through the study period 26.2% and 60.8% used ICS for longer than 1 year. Overall, 4941 and 8969 new LC cases were identified. Based on the adjusted Cox models ICS use had a protective effect for LC in both populations with hazard ratios being 0.77 (0.63-0.94) and 0.61 (0.53-0.7), respectively.

CONCLUSION

Even though the original publications reported contradicting results, in our analysis of over 200,000 patients, both reproduced models showed a significant reduction of LC risk in COPD patients using ICS therapies. Main limitation: models were adjusted for age and sex; individual smoking data were not available in the administrative database. A higher proportion of patients with predominantly asthma-coded prescriptions was observed in the ICS-exposed group of Cohort 2.