BACKGROUND AND OBJECTIVES: Cancer may increase the risk of developing Guillain-Barré syndrome (GBS) due to molecular mimicry or immunosuppression, but the exact relationship is unclear. We aimed to determine the association between incident cancer and the following risk of GBS development.
METHODS: We conducted a nationwide population-based case-control study of all patients with first-time hospital-diagnosed GBS in Denmark between 1987 and 2016 and 10 age-, sex-, and index date-matched population controls per case. We identified incident cancer diagnoses between 6 months before and 2 months after the GBS index date. We used conditional logistic regression to compute odds ratios (ORs) as a measure of relative risk and performed stratified analyses to assess the impact of cancer on GBS risk in strata of calendar periods, sex, and age. In sensitivity analyses, to assess any potential risk of survival bias induced by including cancer diagnoses potentially made after GBS diagnosis, we examined incident cancers in both a broader exposure window (1 year before to 3 months after GBS index date) and a narrower window (6 months to 1 month before the GBS index date).
RESULTS: Of the 2,414 patients with GBS and 23,909 controls included, 49 cases (2.0%) and 138 controls (0.6%) had a recent cancer diagnosis, yielding a matched OR of 3.6 (95% CI 2.6-5.1) for GBS associated with cancer. Stratification by calendar time, sex, and age showed robust results for the association between cancer and GBS, with no major variations. Broadening and narrowing the exposure window produced slightly weakened associations of OR of 2.4 (95% CI 1.8-3.3) and 2.5 (95% CI 1.5-4.1), respectively. The GBS ORs were highest for cancers of the lymphatic and hematopoietic tissue (OR 7.2, 95% CI 2.9-18.0), respiratory tract (OR 5.6, 95% CI 2.7-11.9), prostate and other male genital organ (OR 5.0, 95% CI 2.1-11.6), and breast (OR 5.0, 95% CI 1.7-14.5) cancer.
DISCUSSION: In this large nationwide epidemiologic study, incident cancer was associated with a markedly increased risk of subsequent GBS development. The results suggest that as-yet unidentified factors present in several types of cancer drive this association.