Severity of COVID-19 in Individuals with Chronic Lymphocytic Leukemia throughout the Pandemic in Sweden: A Nationwide Multiple Register Cohort Study Conducted from 2020 to 2023

Introduction

Due to disease- and treatment-related immune defects, patients with chronic lymphocytic leukemia (CLL) have an increased risk of severe disease and death from COVID-19, as well as impaired responses to SARS-CoV-2 vaccination. Thus, we performed a retrospective nationwide analysis on the risk of severe disease and death in individuals with CLL compared to matched controls without CLL in Sweden during the first 3 years of the COVID-19 pandemic in Sweden, using multiple national and population-based registers. The impact of vaccination status and specific CLL directed therapies on outcome, during different phases of the pandemic was also studied.

Methods
We conducted a nationwide cohort study considering all SARS-CoV-2 infection episodes (>90 days between positive PCR tests) from individuals born 1930-2003, residing in Sweden from 1 February 2020 to 31 March 2023. Data from multiple nationwide registers with high coverage (including, but not limited to, the Total Population Register, the National Cause of Death Register, the National Vaccination Register and the INCA CLL register) were used. An infection episode was classified as exposed to CLL when a CLL diagnosis was registered any time before and up until 90 days after the positive PCR test. Each exposed episode was matched to an unexposed episode using a combination of exact and propensity score-based matching without replacement. Exact matching included age category, sex, born in Sweden, residential region, infection episode number, calendar month of positive PCR test, and SARS-CoV-2 variant. Propensity score matching included calendar week, age, and more detailed information on region of birth and residential region. Primary outcome was 90-day all-cause mortality, and secondary outcomes were 90-day COVID-19 mortality, COVID-19 hospital admission and ICU admission. Standardized mean difference (SMD) was used to assess balance before and after matching. Risk ratios (RRs) adjusted for matching factors as well as income quartile, education level, COVID-19 vaccination status, and comorbidities (based on prescription drug use) were calculated using modified Poisson regression with confidence intervals (CIs) derived from a sandwich variance estimator.

Results
From a population of 8,275,839 individuals (6,653 with CLL, 8,269,186 without CLL), 2,088,163 first infection episodes (1,289 CLL, 2,086,874 no CLL) and 140,041 subsequent infection episodes (83 CLL, 139,958 no CLL) were identified. From these, 1,369 episodes from individuals with CLL were matched to the same number of controls. The matching removed substantial differences observed in age, sex, region of birth, and SARS-CoV-2 variant. After matching, SMD values >0.1 were only observed for education level, immunosuppressive drug use, and COVID-19 vaccination status. The 90-day all-cause mortality was 15% (n=199) in individuals with CLL and 8% (n=113) in controls, of which 63% (n=126) and 47% (n=53) had a COVID-19 diagnosis as main cause of death. The adjusted RR (95% CI) was 1.71 (1.38-2.11) for CLL compared with controls. The 90-day all-cause mortality rate in CLL was 25% (64/258) for Wild-type, 12% (23/194) for Alpha/Delta, and 12% (112/916) for Omicron. The adjusted RRs (95% CIs) for Wild-type, Alpha/Delta, and Omicron were 1.79 (1.26-2.54), 2.17 (1.08-4.33), and 1.59 (1.19-2.12). The adjusted RR (95% CI) was 1.64 (1.21-2.24) for unvaccinated, 1.57 (1.17-2.10) for >2 doses, 1.59 (1.16-2.18) for >3 doses, and 1.97 (1.21-3.20) for >4 doses. The adjusted RR (95% CI) was 2.41 (1.79-3.24) when restricting analyses to 90-day COVID-19 mortality. Risks were also significantly increased for COVID-19 hospital admission and ICU admission. When including all 2,228,204 episodes, the adjusted RR (95% CI) was 1.68 (1.48-1.90), similar to the matched cohort analysis.

Conclusions
SARS-CoV-2 infected individuals with CLL had a 2.4 times higher 90-day all-cause mortality, COVID-19 hospital and ICU admission compared with matched controls. This increased risk remained throughout different SARS-CoV-2 variant periods, reinforcing the importance of sustained efforts to protect this frail patient population from infection also during the endemic phase of COVID-19. However, the consistent RR of death compared to controls through different vaccination statuses indicates a benefit in protection from death in individuals with CLL similar to that seen in controls.