Hematologic malignancy patients suffer higher mortality and morbidity from COVID-19 than solid tumor patients.

Abstract

e19048 Background: Cancer patients experience morbidity and mortality from infections, including COVID-19. Results are in conflict as to whether hematologic malignancy (heme) patients or solid tumor (solid) patients are at particular risk. We compared COVID-19 infection outcomes in these cohorts at our institution. Methods: Retrospective chart review of COVID-19 patients diagnosed from February 2020 to January 2021 who had a heme or solid diagnosis. We used Chi square analyses and t-tests for statistical correlations. Results: We analyzed outcomes of 146 COVID-19 episodes. Median age was 64 (25-97), 49% were female, and 24% were minorities. 49 (34%) had heme malignancy. Age, race, BMI, Karnofsky, and Charlson index (CCI) were similar for heme and solid cohorts. All-cause mortality (37 vs 15%) and COVID-induced mortality (33 vs 10%) were higher in heme vs solid, Table 1. Need for hospitalization (73 vs 46%, p = 0.009) and ICU care (31 vs 14%, p = 0.02) were worse for heme. Mean length of stay was 14.6 days and did not statistically differ for both groups. Overall risk of thrombosis was 11%. Severity of hypoxemia correlated with mortality. Compared with patients not requiring oxygen, relative risk (RR) of mortality was 3.06 (p = 0.06) for nasal cannula, 3.82 (p = 0.03) for high-flow oxygen, and 11.9 (p = 0.0008) for ventilation. Heme patients trended towards increased need of high-flow oxygen or ventilation vs solid patients, RR = 1.65 (p = 0.06). Additional mortality predictors were chemo within 6 months of COVID-19, age > 65, nursing home residency, CCI > 5, ICU care, max CRP > 100, ferritin > 1000, abs lymph count < 500, platelet count < 100, and bacteremia or C. difficile coinfection (Table). Race, gender, Appalachian status, and BMI did not predict mortality. Among the above factors, recent chemo (p=0.0001) and bacteremia (p = 0.03) were more common in heme vs solid patients. A higher proportion of heme patients received treatment for COVID-19 (51 vs 32%, p = 0.03). The most frequent therapies across the whole group were steroids (n=43), remdesivir (n=29), and azithromycin (n = 17). Conclusions: COVID-19 poses elevated mortality and morbidity risk for hematologic malignancy vs other cancer patients despite similar pre-infection demographics. Not surprisingly, need for ICU care, age, recent chemotherapy, and comorbidities elevate mortality risk in both heme and solid patients. Elevated inflammatory markers and severe lymphopenia appear to identify those patients who have immune dysregulation and greatest mortality risk. Mortality predictors.[Table: see text]