Active disease in leukaemia patients produces more severe COVID-19 and higher mortality if neutropenic and with a poor pre-COVID prognosis
Active disease in leukaemia patients gives rise to more severe disease after infection with COVID-19 and a higher incidence of mortality, but only where individuals are neutropenic or of their pre-COVID-19 prognosis was under 6 months. This was the finding of a retrospective analysis by researchers based at the Division of Haematology and Oncology, Weill Cornell Medical College, New York, US, presented at ASH 2021.
The most important predictors of both a severe infection and related outcomes after infection with COVID-19 in patients acute leukemias including with acute myeloid leukaemia (AML), acute lymphocytic leukaemia (ALL) and myelodysplastic syndromes (MDS) are uncertain. However, it is possible that active disease in leukaemia might result in more severe disease due to the presence of cytopenias. Data held in the American Society of Haematology ASH RC COVID-19 Registry, which was established on 1 April, 2020 for data collection and which is regularly updated, provides researchers with a rich source of data on the impact of COVID-19 among those with leukaemia. The registry includes patient characteristics such as age, comorbidities, type of haematological malignancy (AML, MDS, ALL), neutrophil and lymphocyte count and active treatments prescribed at the time of COVID-19 diagnosis as well as the outcomes associated with infection.
Using this registry data, the US team documented patient outcomes after infection with COVID-19 stratified by disease status (active initial diagnosis and relapsed/refractory vs. remission) and type of haematologic malignancy. COVID-19 severity was defined as mild (i.e., no hospitalisation required), moderate (hospitalisation required), or severe (ICU admission required). Categorical patient characteristics for each response group and associations between response groups and characteristics (i.e., alive vs. dead, severity vs. non-severity) were summarised by frequency and multivariable analyses used to identify independent predictors of outcomes.
Findings
Data were analysed for 257 patients with AML (n=135), MDS (n=40), and ALL (n=82) with 47% of patients aged less than 60years of age. At the time of infection, active disease in leukaemia was present in 44% of individuals and the overall mortality from COVID-19 infection was 21%. Those with active disease were significantly more likely to present with moderate and severe COVID-19 compared to those in remission when the data were dichotomised as severe vs. non severe and this difference was significant (p = 0.002).
Multivariable analyses revealed that for the whole cohort, increased COVID-19 related mortality was significantly associated only with neutropenia at diagnosis (odds ratio, OR = 3.15, 95% CI 1.31-8.08, p=0.01) and where the estimated pre-COVID-19 prognosis was under 6 months (OR = 8.58, 95% CI 3.24-24.46, p<0.001). However, a further factor appeared to be avoidance of intensive care (OR = 6.66, 95% CI 2.56-18.23, p<0.001).
Restricting the analysis to only hospitalised patients revealed how an increased COVID-19 mortality was again associated with estimated pre-COVID-19 prognosis of < 6 months (OR = 6.77, 95% CI 2.34-22.24, p<0.001) and in those forgoing ICU care (OR = 3.98, 95% CI 1.45-11.66, p = 0.007). Furthermore, mortality was not affected by the type of leukaemia.
The authors concluded that their data suggest that patients with active disease experience significantly higher COVID-19 severity but not increased mortality from COVID-19, apart from those with neutropenia and a pre-COVID-19 prognosis of < 6 months, for which mortality was higher, adding that if desired by patients, aggressive support for those hospitalised patients with COVID-19 is appropriate regardless of remission status.
Desai P et al. Clinical Predictors of Outcome in Adult Patients with Acute Leukemias and Myelodysplastic Syndrome and COVID-19 Infection: Report from the American Society of Hematology Research Collaborative (ASH RC) Data Hub. ASH Conference 2021
