Patient characteristics
The cohort predominantly consisted of men, with 75 individuals accounting for 56.4% of the study population (Table 1). The median age of the patients was 61 years, with an IQR of 52 to 68 years. The majority of patients were diagnosed with B cell lymphoma/leukemia, representing 50.4% of the group (n = 67), followed by 21.0% (n = 28) with acute myeloid leukemia. In terms of comorbidities, 54.1% of the patients (n = 72) had at least one additional health condition. This included hypertension in 27 patients, pulmonary disease in 22, diabetes in 17, and obesity in 13 patients. At the time of contracting COVID-19, 53.4% of the patients (n = 71) had their malignant tumors in a controlled or stable condition, while 46.4% (n = 62) were experiencing onset/active disease states. Regarding treatment history prior to the COVID-19 infection, 77.4% of the patients (n = 103) had undergone some form of therapy. The administered treatments varied and included anti-CD20 combinations in 38 patients, bcl-2 combinations in 18, conventional chemotherapy in 12, and combinations of immunomodulating agents in 7 patients (Fig.1A). In this group of 133 patients, 39 cases (29.3%) were categorized as severe. Among these severe cases, a significant portion, 24 patients (61.5%), were either unvaccinated or had only received one dose of the vaccine. In contrast, the remaining 94 patients (70.7%) exhibited mild or moderate symptoms. Notably, within this latter group, a majority of 58 patients (61.7%) had received at least two doses of the vaccine (Fig.1B).
Univariate and Multivariate Logistic Regression Analysis for Severe/Critical COVID-19 in HM Patients
In the comparison between patients with severe/critical infection (n = 39) and non-severe infection(n = 94), significant differences were observed in median age, time from last treatment to COVID diagnosis, vaccine doses, median lymphocytes count, and C-reactive protein. 39 patients (29.3%) experienced severe/critical infection, with median age (IQR) of 65 (57-71) years. Among them, 24 individuals (61.5%) had not received or had only received a single dose of the vaccine, and 22 individuals (56.4%) had undergone tumor treatment in the last one month preceding the diagnosis of COVID-19. Concurrently, the C-response protein level (mg/L) in severe patients was 58 (14-149), and the lymphocytes count (×109/L) was 0.5 (0.1-1.0). A total of 94 patients (70.7%) were classified as non-severe, with median age (IQR) of 60 (49-67) years. Among these, 58 individuals (61.7%) had received at least two vaccine doses, and 36 individuals (38.3%) had undergone tumor treatment in the last one month preceding the diagnosis of COVID-19. Meanwhile, the C-response protein level (mg/L) in non-severe patients was 32 (10-66), and the lymphocytes count (×109/L) was 0.7 (0.4-1.2) (Table 2).
Univariate analysis revealed that severe/critical infection was associated with age at diagnosis (>50 years) (p = 0.046), vaccine dose (<2 doses) (p = 0.016), lymphocytes (≤0.2×109/L) (p = 0.017), and C-reactive protein ( ≥100mg/L) (p = 0.001). After adjusting for these factors, higher C-reactive protein levels (≥100 mg/L) remained significantly associated with severe/critical infection (odds ratio [OR]: 3.415, 95% confidence interval [CI]: 1.294-9.012; p = 0.013) (Table 3).
Univariate and Multivariate Logistic Regression Analysis for HM patients with persistent COVID-19 infection
Of the adult HM patients confirmed with COVID-19, 122 survived more than 30 days after diagnosis. In the comparison between patients with persistent infection (n = 47) and non-persistent infection (n = 75) (Table 4), significant differences were observed in disease classification, time from last treatment to COVID diagnosis, vaccine doses, median neutrophil count, serum immunoglobulin G (IgG) level, and Interleukin-6 level. 47 patients (38.5%) experienced persistent infection, with median age (IQR) of 64 (52-71) years. The majority of patients with persistent infection had B cell lymphoma/leukemia (n = 27, 57.5%). Among them, 26 individuals (55.3%) had not received or had only received a single dose of the vaccine, and 19 individuals (40.4%) had undergone tumor treatment in the last one month preceding the diagnosis of COVID-19. Concurrently, the Neutrophil count (×109/L) in patients with persistent infection was 2.8 (1.2-4.5), IgG level (mg/dl) was 833 (828-1070), and Interleukin-6 level (pg/ml) was 15.2 (7.4-47). A total of 75 patients (61.5%) were classified as non-persistent infection, with median age (IQR) of 60 (50-66) years. The majority of patients with B cell lymphoma/leukemia (n = 33, 44%) had non-persistent infection. Among these, 51 individuals (68%) had received at least two vaccine doses, and 34 individuals (45.3%) had undergone tumor treatment in the last one month preceding the diagnosis of COVID-19. Meanwhile, the neutrophil count (×109/L) in non-persistent infection patients was 1.7 (0.6-3.4), IgG level (mg/dl) was 1188 (637-1421), and Interleukin-6 level (pg/ml) was 8.2 (4.7-28.8) .
Univariate analysis linked higher persistent infection likelihood with factors such as age over 65 (p = 0.044), recent treatment (within 3 months, p = 0.011), less than two vaccine doses (p = 0.012), diminished IgG levels (<1000 mg/dl, p<0.001), and heightened Interleukin-6 (≥12 pg/ml, p = 0.019). After controlling for these variables, the correlation of persistent infection with less than two vaccine doses (OR: 0.202, 95% CI: 0.048-0.857; p = 0.030), lower IgG levels (<1000 mg/dl, OR: 0.129, 95% CI: 0.027-0.607; p = 0.010), and elevated interleukin-6 (≥12pg/ml, OR: 5.098, 95% CI: 1.118-23.243; p = 0.035) remained notably significant (Table 5).
Survival Analysis
For the entire study cohort, the median duration of follow-up was recorded at 303 days, with a 95% confidence interval (CI) ranging from 301 to 305 days. During this period, there were 5 patients who were lost to follow-up, with 2 belonging to the positive group and 3 to the negative group.The overall mortality rate in the study was found to be 30.8%, equivalent to 41 patients. Specifically, the 30-day mortality rate was noted to be 9.8%, accounting for 13 patients. The primary causes of death among these patients were infections, severe bleeding events, and multiple organ failure. A salient observation emerged regarding the substantial disparity in survival indices between the persistent and non-persistent cohorts (P<0.001). The dataset illuminated that subjects within the non-persistent faction manifested an augmented survival quotient relative to their counterparts in the persistent assemblage, underscoring a pronounced influence of the infection's temporal extent on patient prognoses (Figure 2).