BACKGROUND: Serum cholesterol levels are known to be low in patients newly diagnosed with a lymphoma. Only one study has examined the trajectory of cholesterol levels in the years before a lymphoma diagnosis.
OBJECTIVES: To describe the trajectories of lipid serum levels (total, low and high-density lipoprotein cholesterol and triglycerides) in the years before a diagnosis of lymphoma in a European population.
METHODS: We conducted a case-control study on patients with a lymphoma diagnosed between 2000 and 2017. Up to 5 noncancer controls per case were matched on practice, sex, calendar year, and age at lymphoma diagnosis date or an equivalent date among the controls (index date). Study participants were aged ≥ 40 years at index date; had no evidence of another cancer, HIV infection, or organ transplantation before or on the index date; and had ≥ 1 cholesterol measurement before the index date and before the first prescription of a hypolipidemic drug. General practitioner and hospital data from the Clinical Practice Research Datalink in the United Kingdom were used. Multilevel, multivariable linear longitudinal models were fit to examine the trajectories of serum lipid levels in the years before the index date with time since index date, case status, and confounders in the model.
RESULTS: During the study period, 3,161,492 patients had a valid cholesterol measurement. Among these, 11,969 cases of non-Hodgkin lymphoma and 473 of Hodgkin lymphoma met the study’s inclusion and exclusion criteria. Controls were selected for non-Hodgkin lymphoma and Hodgkin lymphoma cases (59,537 and 2,357, respectively). The most common subtypes of lymphoma were plasma cell neoplasm, chronic lymphocytic leukemia, and diffuse large B cell lymphoma. For all lymphoma types, crude mean cholesterol levels in the years before index date was lower in cases than in controls. For most lymphoma subtypes, the adjusted cholesterol levels showed a more pronounced decrease in the 4 years before diagnosis. On the other hand, triglycerides levels were not clearly related to case status. Results remained similar when restricting to patients who had never used any hypolipidemic drug and had ≥ 1 measure of cholesterol within 1 year before index date.
CONCLUSIONS: This research is the first to replicate a similar study conducted in the United States, while more thoroughly adjusting for potential confounders. The different behavior of cholesterol and triglycerides, described here for the first time, offers insight into the role of lipids in lymphomagenesis. The study suggests the potential role of cholesterol as a biomarker for lymphoma in the years prior to its diagnosis.
