Long-Term Use of Acetaminophen, Aspirin, and Other Non-Steroidal Anti-Inflammatory Drugs and Risk of Hematologic Malignancies: Results From the Large, Prospective VITamins and Lifestyle (VITAL) Cohort Study

Several epidemiological studies have examined the association of aspirin use and other over-the-counter analgesics or anti-inflammatory drugs and the incidence of hematologic malignancies. Although previous results have been inconsistent, some studies have suggested a reduced risk of leukemia or lymphoma with regular use of aspirin or non-aspirin non-steroidal anti-inflammatory drugs (NSAIDs). On the other hand, some but not all studies have reported an increased risk of leukemia or lymphoma with regular use of acetaminophen.

We evaluated the association of analgesic use to hematologic malignancies in a prospective cohort of 64,839 men and women aged 50 to 76 years from Washington State recruited in 2000 to 2002 to the VITamins And Lifestyle (VITAL) study. Eligible participants completed a 24-page baseline questionnaire, including detailed questions about medication use during the previous 10 years. Incident cases of hematologic malignancies (n=577, including MDS [n=54], AML [n=36], myeloproliferative disorders [n=46], CLL/SLL [n=88] and other non-Hodgkin lymphomas [n=235], Hodgkin lymphomas [n=22], plasma cell disorders [n=66], mature NK/T cell neoplasms [n=17], and other entities [n=13]) were identified through December 2008 by linkage to the Seattle-Puget Sound Surveillance, Epidemiology, and End Results (SEER) cancer registry. The censored date was the date of withdrawal from the study, death, move out of the SEER catchment area, or last date of linkage to SEER for diagnosis of hematologic malignancy. In addition, participants were excluded if they had any cancer prior to baseline other than non-melanoma skin cancer and were censored at the time of diagnosis of a non-hematologic malignancy during follow-up to remove treatment for a prior cancer as a cause of any subsequent hematologic cancer. Medication use was categorized as “no use”, “low use” (use for either less than 4 days/week or less than 4 years), and “high use” (use for at least 4 days/week and at least 4 years). Hazards ratios (HRs) and 95% confidence intervals (95% CI) associated with use of acetaminophen, aspirin, and non-aspirin NSAIDs for total incident hematologic malignancies and cancer subcategories were estimated by Cox proportional hazards models. Multivariable-adjusted models were fit by adjusting for age, sex, race/ethnicity, education, smoking, self-rated health, history of fatigue/lack of energy, and family history of leukemia or lymphoma. All models except low-dose aspirin were further adjusted for history of rheumatoid arthritis, history or non-rheumatoid arthritis or chronic neck/back/joint pain, and history of migraines or frequent headaches. The model for low-dose aspirin was further adjusted for history of coronary artery disease, stroke, diabetes, or use of antihypertensive or lipid-lowering medications.

After adjustment, there was an increased risk of incident hematologic malignancies associated with increasing use of acetaminophen (HR=1.81 [95% CI: 1.33–2.46] for high use; p=0.009 for trend). The association with high use of acetaminophen was seen for MDS/AML (HR=2.23 [1.09-4.56]), non-Hodgkin lymphomas (HR=1.82 [1.14-2.92]), and plasma cell disorders (HR=2.32 [0.98-5.50]) but not CLL/SLL (HR=0.83 [0.30-2.35]). By comparison, there was no association with risk of incident hematologic malignancies for increasing use of low-dose aspirin (HR=1.04 [0.81-1.33] for high use; p=0.856 for trend), regular-dose aspirin (HR=0.86 [0.67-1.11] for high use; p=0.329 for trend), non-aspirin NSAIDs (HR=1.04 [0.75-1.43] for high use; p=0.820 for trend), or ibuprofen (HR=0.98 [0.67-1.44] for high use; p=0.956 for trend).

Use of acetaminophen increased the risk of incident hematologic malignancies other than CLL/SLL in a usage-dependent manner, with an almost 100% increased risk for use least 4 days/week for of at least 4 years. Neither aspirin nor non-aspirin NSAIDs decreased risk and are unlikely to be useful for chemoprevention.

No relevant conflicts of interest to declare.