Association of lymphoma and COVID-19 among hospitalized cancer patients: A population-based cohort study.

Abstract

e19090 Background: Recent data has suggested there might be a link between COVID-19 and Lymphoma more complex than only an immunocompromised state. Methods: We used publicly available, deidentified, and state-wide data to conduct a population-base cohort study of hospitalizations aged ≥ 18 years admitted in Texas during 2020 Q2 through 2022 Q4. Target population was identified using ICD-10-CM codes for Mantel Cell Lymphoma, Small B-cell lymphoma, Chronic lymphocytic leukemia, Follicular lymphoma, Marginal zone lymphoma and diffuse large B-cell lymphoma selected from CCSR category NEO058: Non-Hodgkin lymphoma and C18x and C20 selected from CCSR category NEO015: Gastrointestinal cancers - colorectal . Exposure variable was diagnosis lymphoma with hospitalizations diagnosed with only colon cancer used as the reference group. Response variable was diagnosis of the COVID-19. COVID-19 was identified using ICD-10-CM code U071. Three methods were used to measure the association of lymphoma and COVID-19. Propensity score matching was the primary analysis approach with propensity adjusted multilevel logistic regression, and propensity score augmented overlap weighting estimators using multilevel logistic regression as propensity and outcome model were applied as alternative analysis procedures. Subgroup analyses included sex, age, and race/ethnicity, insurance, obesity, smoking, and mental disorders. Results are reported as adjusted odds rations and 95% confidence intervals (aOR [95% CI]). Results: 88,641 hospitalizations were included in the study of which 34,238 (38.6%) had a diagnosis of lymphoma and 136,795 (61.4%) had a diagnosis of colon cancer and no diagnosis of lymphoma. Hospitalizations with lymphoma were older (64.3 % vs 50.5% aged ≥ 65 years), had lower (mean [SD]) Deyo comorbidity index (3.62[2.12] vs 5.55[3.22]), more often had substance use disorders, (27.4% vs 11.3%), more often autoimmune disorders (3.3% vs 1.9%); p < 0.0001 for all comparisons. The rate of COVID-19 was over ten-fold higher in lymphoma hospitalizations (9.6% vs 0.9%). Lymphoma remained associated with COVID-19 on adjusted analysis using the primary model (aOR 11.1340 [95% CI 10.0281 – 12.3618]). The association between COVID-19 and lymphoma was consistent in every subgroup considered. Propensity score matching resulted in 25,454 pairs of hospitalizations. In the propensity matched cohort, lymphoma hospitalizations had a nearly tenfold higher rate of COVID-19 diagnosis (9.9% vs 1.0%) corresponding to (aOR 10.7780 [95% CI 9.4727 – 12.2631]). Augmented propensity score overlaps weighting estimation similarly reported (aOR 10.3287 [95% CI 9.1916 – 11.6064]). Conclusions: There is an association between COVID-19 and Lymphoma. The link between the two diseases may root beyond the immunocompromised state. Furter research is needed to determine factors that may link these diseases.