Use of bisphosphonates was not associated with increased risk for esophageal or gastric cancer

Lim LS. — Tue, 21 Dec 2010 00:00:00 GMT

Question

Are bisphosphonates associated with increased risk for esophageal cancer?

Methods

Design

Cohort study with mean follow-up of approximately 4.5 years (maximum 12.9 y).

Setting

500 general practices in the UK that had up-to-standard reporting practices for the General Practice Research Database (GPRD).

Patients

41 826 patients (mean age 70 y, 81% women) who received a prescription for oral bisphosphonates between 1996 and 2006 (bisphosphonate cohort) and had ≥ 6 months of follow-up. Exclusion criteria were < 40 years of age at index date (date of first oral bisphosphonate prescription) or diagnosis of cancer (except nonmelanoma skin cancer) within 3 years of the index date. Each patient in the bisphosphonate cohort was matched, in sequential order of index date, to a single control patient of the same sex, birth year, and general practice, regardless of bisphosphonate use (to avoid excluding patients who received bisphosphonates for treatment of cancer-related osteoporosis or metastases) (control cohort, n = 41 826).

Risk factor

Prescription of oral bisphosphonate.

Outcomes

Incidence of esophageal and gastric cancer, starting 6 months after the index date.

Main results

Incidence of esophageal and gastric cancer combined was 0.34% (n = 287). Risk for combined gastric and esophageal cancer and for esophageal cancer only did not differ for the bisphosphonate and control cohorts (Table). Analyses by duration of use or type of bisphosphonate (nitrogen-containing, alendronate, or non–nitrogen-containing) showed no differences in risk between groups.

Conclusion

Use of oral bisphosphonates was not associated with increased risk for esophageal or gastric cancer.Association between prescription for oral bisphosphonates and esophageal and/or gastric cancer at a mean follow-up of 4.5 y*OutcomesIncidence/1000 person-years of riskHazard ratio (95% CI)†Bisphosphonate cohortControl cohortEsophageal and gastric cancer0.70.70.96 (0.74 to 1.25)Esophageal cancer only0.480.441.07 (0.77 to 1.49)*CI defined in Glossary.†Adjusted for body mass index, alcohol, and smoking; prescription for hormone therapy, nonsteroidal antiinflammatory drugs, H₂-receptor antagonists, and proton-pump inhibitors before index date; and diagnosis of Barrett esophagus or gastroesophageal reflux disease before index date.

Annals of Internal Medicine: Gastric Cancer Topic Collection

Use of bisphosphonates was not associated with increased risk for esophageal or gastric cancer