Dear Editor, An association between Borrelia burgdorferi, the spirochete carried by ticks that causes Lyme disease, and lymphoma due to chronic immune stimulation has been long suspected. No large population-based study has, to our knowledge, been published on the risk of solid tumors following Borrelia. Case series have reported elevated anti-Borrelia antibody titers and chronic Borrelia infection among patients with primary cutaneous B-cell lymphomas. A large Scandinavian populationbased case–control study of non-Hodgkin lymphoma (NHL) patients and controls observed a positive association between anti-Borrelia seropositivity and mantle cell lymphoma (OR 1⁄4 3.6, 95% CI 1⁄4 1.8–7.4, 12 exposed cases). Other hematological malignancies have not been studied. We, therefore, conducted a large cohort study based on a linked Swedish registry data to study the association between anti-Borrelia seropositivity and the risk of solid tumors and hematologic malignancies. The Karolinska University Hospital Clinical Laboratory database contained Borrelia testing laboratory results from 1992 to 2008. This analysis included all patients with blood drawn and tested for Borrelia antibodies by enzyme-linked immunoassay (IDEIA Borrelia burgdorferi IgM/IgG; Oxoid Ltd., Cambridgeshire, UK). Borrelia-positives were positive for anti-Borrelia IgG or IgM antibodies. The laboratory cohort as well as inpatient, outpatient and death registry files were linked to the national Swedish Cancer Registry to identify all cancer cases. We excluded patients diagnosed with cancer before Borrelia testing or within 1 year after testing or those having a history of organ transplant or HIV infection. We compared the risk of incident cancer among Borreliapositive versus Borrelia-negative patients to Cox proportional hazards models adjusted for age and sex (as appropriate). Of 107,177 total patients, 15,249 (14%) tested positive for Borrelia. The mean follow-up time was 6.6 years for both groups, and the mean age at testing was 45–46 years. In total, there were 5,380 incident cancers and 6,826 deaths during follow-up, of which 808 and 910, respectively, occurred in Borrelia-positive patients. Overall, there was no association between Borrelia seropositivity and risk of any cancer (Table 1). Cervical, lung and prostate cancer were significantly associated with Borrelia. Borrelia seropositivity was not significantly associated with NHL or any NHL subtypes. For all cancer sites, further analysis revealed that there was no dose response with antibody levels divided in tertiles and that the associations stratified by IgG and IgM positivity did not substantially differ. We did not observe an overall increased risk of cancer in people with antibodies to Borrelia. The associations for cervical, lung and prostate cancer were not significant at the Bonferroni-corrected p-value (0.002). We further speculate that the associations with cervical and prostate cancer may reflect a screening bias. Our findings did not support the previously reported association between Borrelia and mantle cell lymphoma. However, our study included few mantle cell lymphoma cases (N 1⁄4 8). Strengths of our study include the large cohort of patients tested for Borrelia and the use of high-quality laboratory and cancer registry data. Limitations include the relatively short mean follow-up time (6.6 years) and the inability to identify the subset of Borrelia-positives who developed a persistent Borrelia infection. In summary, our results indicate that ever having had a Borrelia infection does not increase the risk of solid tumors and hematologic malignancies.