Background: Chronic lymphocytic leukemia (CLL), diffuse large B-cell lymphoma (DLBCL), follicular lymphoma (FL) and multiple myeloma (MM) are B-cell lineage malignancies all sharing an immune dysfunction leading to an increased risk of infection. CLL, FL and MM are preceded by indolent precursor states; more specifically monoclonal B-cell lymphocytosis (MBL), in situ follicular neoplasia (ISFN), and monoclonal gammopathy of uncertain significance (MGUS), respectively. MBL, ISFN and MGUS may be detected decades prior to overt malignancy. Although CLL and FL may transform into DLBCL, evidence for a premalignant DLBCL precursor is scarce. We have previously demonstrated an increased use of antibiotics up to 20 years prior to CLL diagnosis (Andersen et al. Leukemia. 2020). Whether patients with DLBCL, FL or MM share the same trait as patients with CLL remains unknown. Aims: To assess use of antimicrobials for patients with FL, DLBCL and MM prior to diagnosis Methods: We included all patients diagnosed with CLL since 2008, small lymphocytic lymphoma (SLL), DLBCL, FL and symptomatic MM since 2005 through Danish national patient registries (DCLLR, LYFO and DaMyDa; https://www.rkkp.dk/). We analyzed patients with CLL and SLL jointly, as they are recognized as the same disease entity and treated similarly. Likewise, patients with FL grade 3b are clinically similar to DLBCL, thus patients with FL treated with R-CHOP-like regimens within 3 months of diagnosis were considered grade 3b and analyzed together with DLBCL. Data on antimicrobial (AM) prescriptions (Rx) were retrieved from the Danish national database for reimbursed prescriptions (DNDRP) using WHO ATC codes. To characterize the use of prediagnostic AM use, we included AM Rx from January 2004 until date of diagnosis. Patients with multiple malignancies were classified according to the first occurring diagnosis. Statistical analyses were performed using R software. Results: We followed 21,822 patients for a mean of 9.8 person-years prior to diagnosis (SD+/- 4.8), from which a total of 138,805 AM Rx were identified for 18,840 (86.3%) distinct patients. Within the cohort, 6377 (29.2%) were diagnosed with CLL accounting for 42,107 (30.33%) AM Rx, 5803 (26.6%) were diagnosed with MM accounting for 38,417 (27.7%) AM Rx, 6735 (30.9%) were diagnosed with DLBCL accounting for 41,623 (30.0%) AM Rx, and 2907 (13.3%) were diagnosed with FL accounting for 16,658 (12.0%) AM Rx. Thus, 2982 (13.7%) patients had no Rx prior to diagnosis of whom 989 (33.2%) had DLBCL, 775 (26.0%) had CLL, 722 (24.2%) had MM, and 496 (16.6%) had FL. Stratified by disease, patients later diagnosed with CLL and MM accounted for more AM Rx than patients later diagnosed with DLBCL and FL (45.0% vs 41.6% vs 37.8% vs 35.4% with >3 AM Rx, respectively; Table 1). In particular, more penicillin and macrolide were prescribed prior to diagnosis of CLL and MM compared to patients later diagnosed with DLBCL and FL. Image:Summary/Conclusion: These preliminary results indicate a higher rate of infections, as assessed by number of AM Rx, in individuals later developing CLL and MM compared to individuals later developing DLBCL and FL. Thus, MM and CLL seems to be prone to immune dysfunction also prior to diagnosis. Whether immune dysfunction plays a role in development of the disease or is due to the premalignant MBL and MGUS warrants further investigations.