Cancer risk in patients with migraine: A population-based cohort study in Denmark.

Abstract

The purpose of this study was to examine overall and site-specific cancer risk among individuals diagnosed with migraine compared with the general population. Current evidence regarding migraine and risk of cancer is sparse and inconclusive. We conducted a nationwide population-based cohort study with data collected routinely and prospectively from Danish population-based registries from 1995 to 2017. We computed the age- and sex-standardized incidence ratio (SIR) as the ratio of observed to expected cancers among patients diagnosed with migraine in the study population overall, and by encounter type of first diagnosis (inpatient, outpatient specialty clinic, and emergency department). Site-specific cancers were grouped according to etiology. We identified 72,826 patients with a first-time hospital migraine diagnosis. There were 3090 observed overall cancer cases among individuals diagnosed with migraine as compared with 3108 expected cases (SIR 0.99, 95% confidence interval [CI]: 0.96-1.03). The cumulative incidence of all cancers combined from 1995 to 2017 among those with a first-time migraine diagnosis was 9.47% (95% CI: 9.08-9.87). The SIRs for most cancers were consistent with absence of an association: 1.00 (95% CI: 0.94-1.06) for hormone-related cancers, 0.96 (95% CI: 0.88-1.03) for smoking-related cancers, 1.10 (95% CI: 0.98-1.24) for hematologic cancers, and 0.95 (95% CI: 0.85-1.06) for immune-related cancers. Exceptions were SIRs for gastrointestinal cancers (0.78, 95% CI: 0.70-0.87) and for cancers of neurological origin (1.57, 95% CI: 1.40-1.76). For most cancer groups, our results did not support an association with migraine. The exceptions were an increased risk for cancers of neurological origin and a decreased risk for gastrointestinal cancers. These findings may reflect a true difference in risk among individuals with migraine, or more plausibly they reflect other forces, such as differences in medication use, detection bias and reverse causation, or shared risk factors.