Hyperglycemia and Insulin Utilization in the Setting of Multisystem Inflammatory Syndrome in Children

Objectives Multisystem Inflammatory Syndrome in Children (MIS-C), a hyperinflammatory illness that occurs 4-6 weeks after acute infection with SARS-CoV-2, has been associated with hyperglycemia. The aim of this study was to investigate the onset, duration and severity of hyperglycemia as well as its management in MIS-C. Methods A retrospective analysis of four previously healthy patients with no prior diagnosis of diabetes mellitus was conducted on patients who developed hyperglycemia requiring insulin administration during hospitalization with MIS-C. Results All four patients were admitted to the pediatric intensive care unit for multisystem organ dysfunction in the setting of MIS-C. Patient A was a 14-year-old Hispanic male admitted for pneumonia, cardiogenic shock and encephalopathy. Patient B was a 14-year old African American (AA) female admitted for acute renal failure and shock. Patient C was a 15-year-old AA male admitted for respiratory failure and cardiogenic shock. Patient D was a 16-year-old AA male admitted for cardiogenic shock. All patients tested positive for SARS-CoV-2 IgG confirming past infection. Two patients were normal weight and the other two were obese. They developed hyperglycemia within 1-3 days of staring IV glucocorticoids, requiring a maximum of 0.5-1.5 units/kg/day of subcutaneous basal bolus insulin. Hemoglobin A1c (HbA1c) prior to starting insulin had a median of 6.7% with a range of 5.3-6.9%. All severe symptoms of MIS-C and hyperglycemia in the patients reported here ultimately resolved. One patient stopped insulin prior to cessation of steroids and two patients stopped 2-5 months after cessation. One patient was lost to follow up but their insulin prescription for home treatment was never filled. Conclusions Children with MIS-C may experience transient hyperglycemia requiring insulin, which may be caused by a combination of factors. These include insulin resistance due to increased counterregulatory hormones in the setting of critical illness. There was likely pancreatic s-cell damage by SARS-CoV-2 infection preceding the development of MIS-C as evidenced by elevated HbA1c during hospitalization. Treatment for MIS-C includes glucocorticoids, which can further induce insulin resistance. Future studies are needed to determine long-term effects of SARSCoV- 2 infection and MIS-C on the glycemic outcomes in this population.

Association between Receipt of COVID-19, Influenza, and Pneumococcal Vaccination

Background. Whether receipt of COVID-19 vaccine associates with receipt of other routinely-recommended adult vaccines such as, influenza and pneumococcal vaccines is not well described. We evaluated this relationship in a population of adults who were hospitalized for acute respiratory infection (ARI). *Odds ratio describing odds of receiving at least one COVID-19 vaccine (vs not) by influenza vaccination status adjusted for race, employment status, chronic cardiac diseases, cancer, solid organ transplant, and chronic kidney disease. **Odds ratio describing odds of receiving at least one COVID-19 vaccine (vs not) by pneumococcal vaccination status adjusted for race and chronic kidney disease. Methods. We enrolled adults (>= 18 years of age) who were hospitalized at Emory University Hospital and Emory University Hospital Midtown with symptoms consistent with ARI. Participants were interviewed and medical records ed to gather demographic information, including social behaviors during the pandemic, medical history, and prior vaccination history (i.e., COVID-19, influenza, and pneumococcal). Using two separate logistic regression analyses, we determined the association between i) receipt of influenza vaccine in the prior year among adults >= 18 years and ii) receipt of any pneumococcal vaccine in the prior 5 years among adults >= 65 years on the receipt of at least one COVID-19 vaccine>= 14 days prior to admission. Adjusted models included demographic information (e.g., age, sex, race/ethnicity, employment status), social behaviors, and history of chronic medical conditions. Results. Overall, 1056 participants were enrolled and had vaccination records available. Of whom, 509/1056 (48.2%) had received at least one dose of COVID-19 vaccine. Adults >= 18 years who received influenza vaccine were more likely to have received >=1 dose of COVID-19 vaccine compared to those who did not (267/373 [71.6%] vs 242/683 [35.4%] P=< .0001;adjusted odds ratio [OR]: 3.3 [95%CI: 2.4, 4.4]). Similarly, adults >=65 years who received pneumococcal vaccine were more likely to have received >= 1 dose of COVID-19 vaccine compared to those who did not (195/257 [75.9%] vs 41/84 [48.8%] P=< .0001;adjusted odds ratio [OR]: 3.0 [95% CI: 1.8, 5.1]). Conclusion. In this study of adults hospitalized for ARI, receipt of influenza and pneumococcal vaccination strongly correlated with receipt of COVID-19 vaccination. Continued efforts are needed to reach adults who remain hesitant to not only receive COVID-19 vaccines, but also other vaccines that lessen the burden of respiratory illness.

Social Risk Factors for COVID-19-Related Hospitalizations in Adults

Background. During the COVID-19 pandemic, social interventions such as social distancing and mask wearing have been encouraged. Social risk factors for SARS-CoV-2 infection and subsequent hospitalization remain uncertain. Methods. Adult patients were eligible if admitted to Emory University Hospital or Emory University Hospital Midtown with acute respiratory infection (ARI) symptoms (<= 14 days) or an admitting ARI diagnosis from May 2021 - Feb 2022. After enrollment, an in-depth interview identified demographic and social factors (e.g., employment status, smoking history, alcohol use), household characteristics, and pandemic social behaviors. All patients were tested for SARS-CoV-2 using PCR. We evaluated whether these demographic and social factors were related to a positive SARS-CoV-2 test upon admission to hospital with ARI using a logistic regression model. Results. 1141 subjects were enrolled and had SARS-CoV-2 PCR results available (700 positive and 441 negative). The median age was greater in the SARS-CoV-2 negative cohort than in the positive cohort (60 and 53 years, respectively;P< .0001). Those who tested positive were more likely to have had at least some college education compared to those who tested negative (64.3% vs 52.3%, P< .0001;adjusted odds ratio [aOR]: 1.4 [95%CI: 1.1, 2.0]). Compared to those who tested negative, those who were SARS-CoV-2 positive were also more likely to be employed (48.9% vs 26.5%, P< .0001;aOR: 1.7 [95%CI: 1.1, 2.3]), have children 5-17 yo at home (27.6% vs 17.9%, P=.0002;aOR: 1.5 [95%CI: 1.1, 2.1]). Those with COVID-19 were less likely to receive home healthcare (6.2% vs 13.3%, P< .0001;aOR: 0.5 [95%CI: 0.4, 0.9]) and to be a current or previous smoker (7.6% vs 17.7%, P< .0001;aOR: 0.3 [95%CI: 0.2, 0.5]). Conclusion. Among adults admitted to the hospital for ARI, those who tested positive for SARS-CoV-2 were typically younger, more likely to care for school-aged children, more likely to work outside the home, but were less likely to receive home healthcare or smoke. Personal and public health strategies to mitigate COVID-19 should take into consideration modifiable social risk factors.

Prior SARS-CoV-2 infection and risk of subsequent COVID-19-related hospitalization: a test negative design

Background. Studies show that past SARS-CoV-2 infection provides a protective immune response against subsequent COVID-19, but the degree of protection from prior infection has not been determined. History of previous SARS-COV-2 Infection and Current SARS-COV-2 Infection Status at Admission. *Adjusted for chronic respiratory disease and prior COVID-19 vaccination Methods. From May 2021 through Feb 2022, adults (>= 18 years of age) hospitalized at Emory University Hospital and Emory University Hospital Midtown with acute respiratory infection (ARI) symptoms, who were PCR tested for SARS-CoV-2 were enrolled. A prior history of SARS-CoV-2 infection was obtained from patient interview and medical record review. Previous infection was defined as a self-reported prior SARS-CoV-2 infection or previous evidence of a positive SARS-CoV-2 PCR test >= 90 days before ARI hospital admission. We performed a test negative design to evaluate the protection provided by prior SARS-CoV-2 infection against subsequent COVID-19-related hospitalization. Effectiveness was determined using logistic regression analysis adjusted for patient sociodemographic and clinical characteristics and COVID-19 vaccination status. Results. Of 1152 adults hospitalized for ARI, 704/1152 (61%) were SARS-CoV-2 positive. 96/1152 (8%) had a prior SARS-CoV-2 infection before hospital admission. Patients with a previous history of SARS-CoV-2 infection were less likely to test positive for SARS-CoV-2 upon admission for ARI compared to those who did not have evidence of prior infection (31/96 [32%] vs 673/1056 [64%];adjustedOR: 0.25 [0.15, 0.41] (Table). Conclusion. Reinfections represented a small proportion (< 10%) of COVID-19-related hospitalizations. Prior SARS-CoV-2 infection provided meaningful protection against subsequent COVID-19-related hospitalization. The durability of this infection-induced immunity, variant-specific estimates, and the additive impact of vaccination are needed to further elucidate these findings.

Prospective Observational Cohort Study of Serological Responses to COVID-19 Vaccines in Pediatric Kidney Transplant Recipients at a Single Institution

Background. Pediatric kidney transplant recipients (PKTR) are at risk of poor outcomes from COVID-19. Data on serologic responses to COVID-19 vaccines in PKTR remain sparse. We characterized the magnitude, breadth, and longevity of SARS-CoV-2 spike protein binding antibody responses in PKTR. Methods. This single institution, prospective observational study enrolled PKTR presenting to a transplant clinic for routine care who had received or were eligible to receive a COVID-19 vaccine. Demographic data, history of prior COVID-19, and vaccination details were collected. Plasma samples obtained from standard-of-care residual specimens were analyzed for SARS-CoV-2 spike variant IgG using the MesoScale Discovery V-PLEX platform, which quantitatively measures antibodies to SARS-CoV-2 full-length spike wild-type (Wuhan-hu-1), Alpha (B.1.1.7), Beta (B.1.351), Delta (B.1.617.2), Gamma (P.1), and Omicron (B.1.1.529;BA.1) variants. Vaccine time points with > 5 samples available were analyzed. Geometric mean titers (GMTs) were calculated and log-transformed titers were compared using one-way ANOVA with Tukey's post-hoc comparisons test. Results. 61 PKTR enrolled (Table1);47 (77%) received at least 1 dose of COVID-19 vaccine in transplant clinic. 47 (77%) PKTR had at least one sample available for analysis, but serial specimens were lacking for many. By 6 months post-dose 2 of COVID-19 mRNA vaccination, spike (Wuhan-hu-1) IgG titers had waned to prevaccination levels (GMT 24 vs 47 binding antibody units (BAU)/mL, P=0.988). Administration of a 3rd dose of mRNA vaccine significantly boosted IgG antibodies (GMT 492 BAU/mL, P=0.007), and titers were maintained at 3 months (GMT 656 BAU/mL, P=0.001) but gradually waned by 6 months (GMT 223 BAU/mL, P=0.070). Administration of a 4th dose elicited a non-significant increase in titers (GMT 905 BAU/mL, P=0.870). Binding IgG antibodies to SARS-CoV-2 variant spike proteins post-vaccination were not significantly different from Wuhan spike. Conclusion. In this cohort of PKTR, a 3rd dose of COVID-19 mRNA vaccine significantly boosted broadly cross-reactive binding IgG antibodies to SARS-CoV-2 spike variants, including Omicron. Decreasing titers at 6 months post-dose 3 raise concern for waning protective immunity and support 4th dose vaccination.

Features of hospitalized pediatric patients with seizures and respiratory viruses: a comparison of SARS-CoV2, other corona viruses, and influenza

Objective: To compare features of hospitalized pediatric patients with seizures and respiratory viruses. Background: Neurological symptoms are reported in children with COVID-19, caused by SARS-CoV2. Limited data is available on neurological symptoms in other respiratory viruses, including influenza and other coronaviruses. Design/Methods: Retrospective data was extracted from EPIC on all children between 0-21 years old admitted to Children's Hospital of Atlanta from January 1, 2014 to June 1, 2021 for seizures and had positive PCR for SARS-CoV-2, other coronaviruses (Coronavirus NL63 and Coronavirus OC34) and influenza (A and B). Patient characteristics including age, race, sex, ethnicity, hospital length of stay, intensive care unit admission, intubation, chest x ray, MRI, and disposition were included. Comparison of patient characteristics between children with COVID-19 and the two other groups were conducted respectively. Results: A total of 487 pediatric patients were included: 68 COVID-19, 232 influenza, and 187 with other coronaviruses (OC), with median age of 6.2 (interquartile range (IQR): [3.1,11.2]). COVID-19 patients had higher ICU admission rates (50% versus 31%, p = 0.008) but lower intubation rates than OC (19% versus 35%, p = 0.021). Stroke rates were higher in COVID-19 versus OC (6% versus 0.5%, p= 0.019). Initial white blood cell counts were lower in the COVID-19 (median 7.32[6.05, 11.07]) than OC (median 10.66[ 7.61, 14.26];p = 0.001). The rate of abnormal MRI was lower among COVID-19 patients, compared to OC patients (57% versus 82%) with a medium to large effect size (standard mean difference) of 0.55. However, no differences were observed between COVID-19 and influenza cohorts. Conclusions: Differences in patients with COVID-19 versus other coronaviruses were observed, whereas no differences were observed in COVID-19 versus influenza patients. Vigilance should be undertaken in treatment of children presenting with all respiratory illnesses.

Burden of Respiratory Syncytial Virus (RSV) Infection among Hospitalized Older Adults and Those with Underlying Chronic Obstructive Pulmonary Disease (COPD) or Congestive Heart Failure (CHF)

Background. The burden of Respiratory Syncytial Virus (RSV)-associated hospitalization in adults is incompletely understood. The COVID-19 pandemic has resulted in multiple public health measures (e.g., social distancing, handwashing, masking) to decrease SARS-CoV-2 transmission, which could impact RSV-associated hospitalizations. We sought to compare RSV-associated hospitalizations from 2 pre- and one mid-COVID-19 winter viral respiratory seasons. Methods. We conducted an IRB-approved prospective surveillance at two Atlanta-area hospitals during the winter respiratory viral seasons from Oct 2018-Apr 2021 for adults ≥ 50 years of age admitted with acute respiratory infections (ARI) and adults of any age with COPD or CHF-related admissions. Adults were eligible if they were residents of an 8 county region surrounding Atlanta, Georgia. Those with symptoms > 14 days were excluded. Standard of care test results were included. Asymptomatic adults ≥ 50 years of age were enrolled as controls in Seasons 1 and 2. Nasopharyngeal swabs from cases and controls were tested for RSV using BioFireR FilmArrayR Respiratory Viral Panel (RVP). We compared the demographic features and outcomes of RSV+ cases and controls. Results. RSV was detected in 71/2,728 (2.6%) hospitalized adults with ARI, CHF, or COPD and 4/466 (0.9%) controls. In Season 1, RSV occurred in 5.9% (35/596 patients), in Season 2 3.6% (35/970 patients), but in only 0.09% (1/1,162 patients) in Season 3 (P < 0.001 for both seasons). RSV detection in Season 3 was similar to RSV detection among controls during Seasons 1 and 2 (P=0.6). Median age of cases and controls was 67 years (Table 1). Of cases with RSV 11% were admitted to the ICU and two required mechanical ventilation. The majority of hospitalized patients were discharged home (95.8%) with a median length of hospitalization of three days (IQR 2-7). Conclusion. Over 3 seasons, RSV was detected in 2.6% of adults admitted to the hospital with ARI, CHF or COPD. The rate of RSV dramatically declined during the 2020-21 winter respiratory viral season, likely due to public health measures implemented in response to COVID-19.

Outcomes among Influenza and SARS-CoV-2 Infection in Hospitalized Adults Age ≥ 50 Years and with Underlying Chronic Obstructive Pulmonary Disease (COPD) or Congestive Heart Failure (CHF)

Background. A significant burden of disease exists for adults infected with influenza (flu) and SARS-CoV-2, which causes COVID-19. However, data are limited comparing outcomes between hospitalized adults infected with these viruses. Methods. Over the course of 3 consecutive winter respiratory viral seasons, adults ≥ 50 years of age admitted with acute respiratory tract infections (ARI) and adults of any age with COPD or CHF-related admissions were enrolled from 2 Atlanta area hospitals. For the 2018-19 and 2019-20 seasons, participants were approached in the hospital. If the participant enrolled, nasopharyngeal (NP) and oropharyngeal (OP) swabs were collected and tested using BioFire® FilmArray® respiratory panel. Due to the COVID-19 pandemic in 2020-21 and limitations involving participant contact, only NP standard of care (SOC) swabs were collected. A comprehensive medical chart review was completed for each subject which encompassed data on their hospitalization, past medical history, and vaccination history. Co-infected patients were excluded from the analyses. Results. Of the eligible participants, 118 were flu positive (three RSV-influenza co-infections were excluded) and 527 were COVID-19 positive. Median age was lower for the flu cohort at 62 (IQR 56-71) than those with COVID-19 (67, IQR 59-77) (p < 0.0001). Length of stay (LOS) was shorter in flu-infected patients (median 3 d, IQR 2-6), but was longer for COVID-19 patients (median 5 d, IQR 3-10). ICU admission occurred in 20% of those with flu, and among those admitted to the ICU mechanical ventilation (MV) occurred in 12.5%. ICU admission and MV was significantly higher for those with COVID-19, with 28% of patients admitted to the ICU and 47% of those requiring MV. Among patients with COVID-19, 8.9% died. This was significantly higher than that of flu (3.4%) (p=0.008). Hospital discharge occurred more frequently to a nursing home or LTCF with COVID-19 (10.3%) than with flu (0%) (p< 0.0001). Table 1. Breakdown of age, hospitalization course, and discharge disposition for participants diagnosed with influenza or COVID-19 during hospitalization. Conclusion. COVID-19 resulted in a longer hospital admission, a greater chance of ICU admission and MV as compared to flu. Additionally, COVID-19 participants had a high rate of discharge to a nursing home/LTCF and a significantly higher risk of death. While the clinical course was not as severe as COVID-19, influenza contributed a significant burden.

The Burden of Influenza and Rhinovirus among Hospitalized Adults Post the COVID-19 Pandemic

Background. Acute respiratory tract infections (ARIs) are a significant cause of morbidity in adults. Influenza is associated with about 490,600 hospitalizations and 34,200 deaths in the US in the 2018-2019 season. The burden of rhinovirus among adults hospitalized with ARI is less well known. We compared the burden of influenza and rhinovirus from 2 consecutive winter respiratory viral seasons in hospitalized adults and healthy controls pre-COVID-19 and one season mid-COVID-19 to determine the impact of rhinovirus as a pathogen. Methods. From Oct 2018 to Apr 2021, prospective surveillance of adults ≥50 years old admitted with ARI or COPD/CHF exacerbations at any age was conducted at two Atlanta hospitals. Adults were eligible if they lived within an eightcounty region around Atlanta and if their symptom duration was < 14 days. In the seasons from Oct 2018 to Mar 2020, asymptomatic adults ≥50 years old were enrolled as controls. Standard of care test results were included and those enrolled contributed nasopharyngeal swabs that were tested for respiratory pathogens using BioFire® FilmArray® Respiratory Viral Panel (RVP). Results. During the first two seasons, 1566 hospitalized adults were enrolled. Rhinovirus was detected in 7.5% (118) and influenza was detected in 7.7% (121). Rhinovirus was also detected in 2.2% of 466 healthy adult controls while influenza was detected in 0%. During Season 3, the peak of the COVID-19 pandemic, influenza declined to 0% of ARI hospitalizations. Rhinovirus also declined (p=0.01) but still accounted for 5.1% of all ARIs screened (Figure 1). Rhinovirus was detected at a greater rate in Season 3 than in asymptomatic controls in the first 2 seasons (p=0.008). In the first two seasons, Influenza was detected in 8.6% (24/276) of those admitted to the ICU. Rhinovirus was detected in 6.1% (17/276) of those admitted to the ICU but declined to 3.1% (8/258) in Season 3. Conclusion. Dramatic declines occurred in influenza in adults hospitalized with ARI, CHF, or COPD in Atlanta during the COVID-19 pandemic and with enhanced public health measures. Although rhinovirus declined during the COVID-19 pandemic, it continued to be identified at a rate higher than in historical controls. Additional data are needed to understand the role of rhinovirus in adult ARI, CHF, and COPD exacerbations.

Clinical features and management of pediatric patients presenting with new onset acute leukemia and concomitant COVID-19

Background: Infections represent a significant cause of morbidity and mortality in pediatric patients undergoing treatment for hematologic malignancies. Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection has led to a worldwide pandemic of coronavirus disease 2019 (COVID-19) and pediatric patients with cancer appear to be at higher risk of severe disease than reported in the general pediatric population. Data are limited on the optimal management of children infected with SARS-CoV-2 and a new diagnosis of leukemia. The objective of this study was to describe our experience of six children who presented with a new diagnosis of acute leukemia and concurrent COVID-19.

Elucidating sars-cov-2 t-cell responses in pediatric covid-19 and mis-c

Background: Multisystem Inflammatory Syndrome in Children (MIS-C) can develop 1-2 mo post SARS-CoV-2 infection. MIS-C is characterized by fever, multiorgan dysfunction requiring hospitalization, and systemic inflammation. To evaluate a potential role for aberrant T-cell responses as a potential mechanism for MIS-C pathogenesis, we quantified SARS-CoV-2 -reactive T cells in children with COVID-19, MIS-C, and healthy children (HC). Methods: Hospitalized children ages 0-20 yrs with COVID-19 (n=13) or MIS-C (n=18) were enrolled from May-Sep 2020. Peripheral blood mononucle ar cells (PBMC) were obtained from convalescent phase of infection (28-54 d from illness onset) for COVID-19 or at hospitalization for MIS-C to approximate similar time since infection. Plasma SARS-CoV-2 receptor binding domain (RBD) antibody titers were determined by ELISA. PBMC from HC (n=20) with undetectable RBD antibodies served as controls. T-cell responses were quantified using activation-induced marker (AIM) assay after stimulation with SARS-CoV-2 peptide “megapools” (MP): CD4 MP-S with 253 spike-spanning peptides, CD4 MP-R with 221 remaining non-spike;spike-containing CD8 MP-A and non-spike CD8 MP-B with 314 each. Frequency of AIM+ T-cells and stimulation index (SI) were compared across donor groups. Results: Among COVID-19, majority had SARS-CoV-2 specific CD4+ (100% spike, 83% non-spike) and CD8+ (85% spike-containing, 83% non-spike) T-cells. There was a trend for lower frequencies of AIM+ T-cells to all peptide MP in MIS-C, with significantly lower responses to non-spike antigens in CD4+ (p<0.05) and CD8+ (p<0.05) T-cells compared to those in COVID-19. In addition, COVID-19 had higher reactivity to stimulation, with significantly greater SI for spike CD4+ T-cell responses compared with HC (4.62 vs 1.93, p<0.05) and non-spike compared to both MIS-C (3.27 vs 1.44, p<0.05) and HC (3.27 vs 1.60, p<0.01). Interestingly, most HC also had detectable CD4+ (70% spike, 50% non-spike) and CD8+ T-cells (90% spike, 75% non-spike) against SARS-CoV-2 antigens, possibly attributable to prior infection by endemic coronaviruses. RBD IgG levels were similar between MIS-C and convalescent COVID-19. Conclusion: We find more robust CD4+ and CD8+ T-cell responses against non-spike SARS-CoV-2 peptides in convalescent COVID-19 compared to MIS-C. Equivalent humoral responses against spike RBD among MIS-C and COVID-19 suggest that impaired SARS-CoV-2-specific T-cell response to non-spike antigens may contribute to the immunopathogenesis of MIS-C.

Clinical features and antibody responseof two pediatric patients presenting with new-onset acutemyeloid leukemia and concomitant severe COVID-19

Objective: SARS-CoV-2 infection has led to a worldwide pandemic of COVID-19 (coronavirus disease 2019), placing individuals with pre-existing medical conditions at a higher risk for morbidity and mortality. Limited data inpediatric patients with malignancies suggest that severe COVID-19 illness is rare. The objective of this study was todescribe our experience of two adolescents who presented with new diagnoses of acute myeloid leukemia (AML)and concurrent COVID-19. Methods: The clinical presentation, treatment, and serology of two patients who presented with AML andconcurrent SARS-CoV-2 infection were abstracted. Residual blood was tested for serial quantitative IgG by ELISA tothe SARS-CoV-2 spike protein receptor binding domain, which has high sensitivity and specificity to SARS-CoV-2.The study was approved by Children's Healthcare of Atlanta and Emory University IRBs. Results: Patient 1 was a 16-year-old Caucasian male with previously treated classical Hodgkin's lymphoma whopresented with fever, cough, hyperleukocytosis, and pulmonary infiltrates and was diagnosed with therapy-relatedAML (TR-AML). SARS-CoV-2 was detected by nasopharyngeal (NP) RT-PCR testing on admission. He receivedremdesivir for treatment of COVID-19 and modified induction therapy with cytarabine alone starting on hospital day(HD) 3. He demonstrated high SARS-CoV-2 IgG titer (1:1327.3) on HD 4 and cleared SARS-CoV-2 with a negativeNP RT-PCR on HD 14. He went on to receive additional myelosuppressive AML therapy on HD 26 with azacitidineand gemtuzamab ozogamicin. On HD 34, his IgG titer remains elevated (1:5621.4) and he is currently awaitingcount recovery. Patient 2 was a 19-year-old Hispanic, previously healthy male who presented with fever, cough, dyspnea, and hyperleukocytosis and was diagnosed with de novo AML (D-AML). He also tested positive for SARS-CoV-2 via NP RT-PCR on admission. He began standard induction therapy with cytarabine, etoposide, anddaunorubicin on HD 2. He developed hypoxemic respiratory failure on HD 4 and received COVID-19 directedtherapies of convalescent plasma, remdesivir, and tocilizumab. His serologic testing showed low SARS-CoV-2 IgGtiter (1:619.3) on HD 4 despite administration of convalescent plasma. His titers waned over the subsequent twoweeks and he continued to test positive for SARS-CoV-2 via NP RT-PCR on HD 21. He remains critically ill inmultiorgan failure with signs of neutrophil recovery on HD 25. Conclusion: COVID-19 can be severe in children with AML and make treatment decisions challenging. Clinicalpresentation, curative modalities (hematopoietic stem cell transplantation for TR-AML versus potentiallychemotherapy alone for D-AML), and concurrent COVID-19 were considered in determining induction therapy. Whiledifficult to draw definite conclusions from two patients, the differential serologic response in these patients seems tocorrelate with the intensity of therapy they received and may have contributed to the overall severity of their COVID-19.