Purpose
There is a growing body of evidence suggesting the decisive involvement of the human microbiome in cancer development. The consumption of antibiotics may fundamentally change the microbiome and thereby create a precancerous environment promoting cancer development and growth. However, clinical data on the association between the consumption of antibiotics and cancer incidence have remained inconclusive. In this study, we quantified the association between the intake of different antibiotics and various cancer entities among outpatients from Germany.
Methods
This retrospective case–control study based on the IQVIA Disease Analyzer database included 111,828 cancer patients and 111,828 non-cancer controls who were matched to cancer cases using propensity scores. Patients were categorized as non-users, low-consumption (up to 50th percentile), and high-consumption (above 50th percentile) users of antibiotics overall and for each antibiotic class. Multivariable logistic conditional regression models were used to study the association between antibiotic intake within 5 years prior to the index date (first cancer diagnosis for cases or randomly selected date for controls) and cancer incidence.
Results
The probability of cancer was significantly higher among patients with a history of antibiotic intake than in matched controls. Patients using penicillin or cephalosporins displayed a higher incidence of cancer, while the intake of tetracyclines and macrolides actually reduced the risk of cancer development slightly. A complex picture was observed in our cancer site-stratified analyses. Most notably, the consumption of penicillin was significantly and positively associated with cancer development in the respiratory organs only (low consumption OR: 1.33, 95% CI 1.20–1.47; high consumption OR 1.42, 95% CI 1.22–1.64) and cephalosporin consumption was significantly associated with respiratory organ cancer (low consumption OR: 1.32, 95% CI 1.17–1.48, high consumption OR: 1.47, 95% CI 1.29–1.66), breast cancer (high consumption OR: 1.40, 95% CI 1.25–1.56), and lymphoid and hematopoietic tissue cancer (high consumption OR: 1.50, 95% CI 1.35–1.66).
Conclusion
Our data strongly support the hypothesis that the intake of antibiotics is positively associated with the risk of cancer development.