Remdesivir shortens time to recovery in adults with severe coronavirus disease 2019 (COVID-19), but its efficacy and safety in children are unknown. We describe outcomes in children with severe COVID-19 treated with remdesivir.
Seventy-seven hospitalized patients <18 years old with confirmed severe acute respiratory syndrome coronavirus 2 infection received remdesivir through a compassionate-use program between March 21 and April 22, 2020. The intended remdesivir treatment course was 10 days (200 mg on day 1 and 100 mg daily subsequently for children ≥40 kg and 5 mg/kg on day 1 and 2.5 mg/kg daily subsequently for children <40 kg, given intravenously). Clinical data through 28 days of follow-up were collected.
Median age was 14 years (interquartile range 7–16, range <2 months to 17 years). Seventy-nine percent of patients had ≥1 comorbid condition. At baseline, 90% of children required supplemental oxygen and 51% required invasive ventilation. By day 28 of follow-up, 88% of patients had a decreased oxygen-support requirement, 83% recovered, and 73% were discharged. Among children requiring invasive ventilation at baseline, 90% were extubated, 80% recovered, and 67% were discharged. There were 4 deaths, of which 3 were attributed to COVID-19. Remdesivir was well tolerated, with a low incidence of serious adverse events (16%). Most adverse events were related to COVID-19 or comorbid conditions. Laboratory abnormalities, including elevations in transaminase levels, were common; 61% were grades 1 or 2.
Among 77 children treated with remdesivir for severe COVID-19, most recovered and the rate of serious adverse events was low.
What’s Known on This Subject:
Currently, the only evidence concerning the safety and efficacy of remdesivir in hospitalized children with coronavirus disease 2019 (COVID-19) comes from individual case reports and small series.
What This Study Adds:
We report the largest cohort to date of children with COVID-19 receiving remdesivir. Although the lack of a control in this compassionate-use program precludes efficacy assessment, these data on remdesivir safety and outcomes in children with COVID-19 are important.
Coronavirus disease 2019 (COVID-19) is a respiratory illness caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) with clinical manifestations ranging from asymptomatic disease to hypoxemic respiratory failure and death. As of March 11, 2021, >3 million cases of COVID-19 in children 0 to 17 years of age have been reported in the United States, representing ∼13% of all cases.1 According to the American Academy of Pediatrics and the Children’s Hospital Association, up to 3.0% of all COVID-19 cases in children resulted in hospitalization, but COVID-19-related mortality among children is low (<0.2%).1 Centers for Disease Control and Prevention surveillance data through July 25, 2020, indicate that the pediatric hospitalization rate for COVID-19 was 8 per 100 000 population compared with 164.5 per 100 000 in adults; however, 1 in 3 hospitalized children required ICU care, a proportion similar to that observed in adults. Children hospitalized for COVID-19 were more likely to be Black or Hispanic and had a high prevalence of underlying medical conditions, including obesity, chronic lung disease, and prematurity.2 In addition to pulmonary disease, SARS-CoV-2 infection in children has been associated with multisystem inflammatory syndrome in children and cardiac dysfunction.3–5
Remdesivir is a nucleotide analogue that selectively inhibits the RNA-dependent RNA polymerase of several viruses, including SARS-CoV-2.6,7 Part 1 of the Adaptive COVID-19 Treatment Trial demonstrated that a 10-day course of remdesivir was superior to placebo in reducing time to recovery in hospitalized adults with severe COVID-19.8 The SIMPLE Severe trial showed that outcomes with 5 days of remdesivir treatment of severe COVID-19 were similar to 10 days of treatment.9 The SIMPLE Moderate trial revealed clinical benefit of 5 days of remdesivir for moderate COVID-19 relative to standard of care.10 These studies did not demonstrate a mortality benefit of remdesivir. In contrast, the World Health Organization SOLIDARITY trial did not show an effect on duration of hospitalization of remdesivir, hydroxychloroquine, lopinavir, or interferon.11
Optimal therapy for children with severe COVID-19 is unknown; although supportive care is likely to be adequate for most cases, initial expert guidance suggests that antiviral therapy with remdesivir should be considered in some cases.12
Here, we report clinical outcomes in 77 hospitalized children who received remdesivir as part of the compassionate-use program. Although this study cannot evaluate the efficacy of remdesivir, it provides valuable data on safety and the clinical features of severe COVID-19 in children.
Methods
Compassionate-Use Program Description and Population
Gilead Sciences began accepting requests from clinicians for compassionate use of remdesivir in children with COVID-19 on March 21, 2020. Instructions for the use of remdesivir were outlined in a single patient protocol (SPP) (see Supplemental Information), including guidance on dosing, administration, clinical procedures, laboratory tests, and concomitant medication administration. Access was limited to hospitalized patients with polymerase chain reaction–confirmed SARS-CoV-2 infections and severe manifestations of COVID-19. Patients were generally required to have oxygen saturation of ≤94% while breathing ambient air or a need for oxygen support, although decisions were based on an individualized risk/benefit assessment, and, in some cases, patients were accepted because of concern for extrapulmonary manifestations of SARS-CoV-2 infection, serious comorbidities, or radiographic findings.13,14 Remdesivir was not provided for patients with creatinine clearance (by an age-appropriate estimated glomerular filtration rate formula) of <30 mL per minute, serum levels of alanine aminotransferase (ALT) >5 times the upper limit of normal, or evidence of multiorgan failure. Concomitant administration of other investigational agents for COVID-19 was not permitted in the SPP; however, treating physicians did administer these in some cases. Children weighing ≥40 kg received a loading dose of 200 mg intravenously on day 1, plus 100 mg intravenously on subsequent days. Children weighing <40 kg received a loading dose of 5 mg/kg intravenously on day 1, plus 2.5 mg/kg intravenously on subsequent days. A 10-day course of therapy was recommended, but shorter courses were sometimes given at the clinician’s discretion. Supportive therapy and study drug discontinuation were also at clinicians’ discretion. Patients were followed through 28 days postinitiation or until discharge or death, if either occurred before 28 days had passed. Some patients included in this analysis have been reported previously in case reports.
Program Oversight
For each patient, we obtained regulatory and institutional review board or independent ethics committee approval, with consent secured for all patients on the basis of local regulations. The sponsor (Gilead Sciences) designed and conducted the program according to the SPP, collected the data, monitored program conduct, and performed all statistical analyses. All authors were given access to the reported data and took responsibility for their integrity and completeness. A professional writer employed by the sponsor assisted in the writing of this article.
Procedures
This program had no prespecified end points. An electronic case report form was used by clinicians to report patient medical history, changes in clinical status, remdesivir administration, adverse events, and laboratory results on each day of treatment, day 15, and day 28. Medical history and adverse events were coded by using the Medical Dictionary for Regulatory Activities version 22.1. Clinicians ascertained and documented in the electronic case report form the date of COVID-19 symptom onset. The timing of onset for medical history was not collected, and, therefore, the chronological relationship to COVID-19 diagnosis cannot be ascertained. Clinicians were asked to perform daily laboratory tests during remdesivir therapy (see SPP in the Supplemental Information). No further follow-up under the compassionate-use program protocol was required after hospital discharge or day 28, whichever came first.
Outcomes
We assessed clinical outcomes, including safety, respiratory support status, hospital discharge, and recovery. Baseline was defined as the day of first remdesivir dose. Duration of symptoms was defined as the time elapsed between date of COVID-19 symptom onset and baseline. Recovery was defined as hospital discharge for children on room air at baseline (N = 8) and improvement to room air or discharge for all others. For patients who were on invasive mechanical ventilation at baseline, we report extubation and time to extubation. Changes in clinical status are reported by using the modified ordinal scale (Fig 1).9,15,16
Modified ordinal scale. IMV, invasive mechanical ventilation (by endotracheal tube or tracheostomy); NIPPV noninvasive positive pressure ventilation.
Modified ordinal scale. IMV, invasive mechanical ventilation (by endotracheal tube or tracheostomy); NIPPV noninvasive positive pressure ventilation.
Statistical Methods
The analysis population included all children who initiated compassionate use of remdesivir on or before April 22, 2020, and for whom clinical data for baseline and at least 1 subsequent day were available as of May 9, 2020. The population was stratified by age category (≤12 and >12 years) and requirement for invasive mechanical ventilation at baseline. No sample size calculation was performed.
Time to recovery and time to discharge were analyzed with stratification by baseline respiratory support status (invasive versus noninvasive). The proportions of patients achieving the outcomes of recovery or discharge were assessed through day 28 by using the cumulative incidence function method, with death as the competing risk. P values for these analyses were from a Cox regression model.
Multivariate analysis was performed by using a Cox regression model. Variables in the final model were selected by using a stepwise procedure (entry variables included baseline invasive status [yes or no], sex [male or female], age group [≤12 or >12 years], and duration of symptoms [<8 or ≥8 days]). Hazard ratios and P values were calculated with death as a competing risk.
The Division of AIDS Table for Grading the Severity of Adult and Pediatric Adverse Event, Version 2.1 (July 2017) was used for assigning toxicity grades (0–4) to laboratory results.17 Treatment-emergent laboratory abnormalities were defined as results that increase at least 1 toxicity grade from baseline at any postbaseline time point.
Results
Baseline Demographics and Clinical Characteristics
The analysis population included 77 children: 58 (75%) in the United States, 7 (9%) in Spain, 6 (8%) in the United Kingdom, 4 (5%) in Italy, and 1 each (1%) in France and Germany (Table 1). The median age of patients was 14 years (interquartile range [IQR] 7–16; range <2 months to 17 years), and 46 (60%) of patients were male. At baseline, 39 patients (51%) were on invasive respiratory support (38 on invasive mechanical ventilation and 1 on venovenous extracorporeal membrane oxygenation [ECMO]). Of the remaining 38 patients, 6 were receiving noninvasive positive pressure ventilation, 14 were receiving high-flow oxygen, 10 were receiving low-flow oxygen, and 8 were breathing room air. The median duration of COVID-19 symptoms before baseline was shorter in the group receiving invasive respiratory support at baseline (7 days, IQR 5–8) than in the noninvasive group (9 days, IQR 7–12) (P = .005).
Baseline Demographic and Clinical Characteristics
| Invasive Oxygen, n = 39 | Noninvasive Oxygen, n = 38 | Total, N = 77 | |
|---|---|---|---|
| Median age (range), y | 11 (0–17) | 15 (0–17) | 14 (0–17) |
| Age, n (%) | |||
| <2 mo | 4 (10) | 0 | 4 (5) |
| 2 mo to <1 y | 5 (13) | 3 (8) | 8 (10) |
| 1–<5 y | 3 (8) | 1 (3) | 4 (5) |
| 5–12 y | 11 (28) | 9 (24) | 20 (26) |
| >12 y | 16 (41) | 25 (66) | 41 (53) |
| Sex, n (%) | |||
| Male | 23 (59) | 23 (61) | 46 (60) |
| Female | 16 (41) | 15 (39) | 31 (40) |
| Median duration of symptoms (quartile 1, quartile 3), d | 7 (5, 8) | 9 (7, 12) | 8 (6, 10) |
| Median duration of hospitalization (quartile 1, quartile 3), d | 4 (3, 5) | 4 (2, 7) | 4 (3, 5) |
| Median duration of invasive oxygen support (quartile 1, quartile 3), d | 2 (2, 3) | 0 | 2 (2, 3) |
| ALT level ≤50 U/L, n (%) | 25 (66) | 31 (84) | 56 (75) |
| Median ALT level (quartile 1, quartile 3), U/L | 33 (21, 69) | 31 (20, 44) | 32 (20, 51) |
| Invasive Oxygen, n = 39 | Noninvasive Oxygen, n = 38 | Total, N = 77 | |
|---|---|---|---|
| Median age (range), y | 11 (0–17) | 15 (0–17) | 14 (0–17) |
| Age, n (%) | |||
| <2 mo | 4 (10) | 0 | 4 (5) |
| 2 mo to <1 y | 5 (13) | 3 (8) | 8 (10) |
| 1–<5 y | 3 (8) | 1 (3) | 4 (5) |
| 5–12 y | 11 (28) | 9 (24) | 20 (26) |
| >12 y | 16 (41) | 25 (66) | 41 (53) |
| Sex, n (%) | |||
| Male | 23 (59) | 23 (61) | 46 (60) |
| Female | 16 (41) | 15 (39) | 31 (40) |
| Median duration of symptoms (quartile 1, quartile 3), d | 7 (5, 8) | 9 (7, 12) | 8 (6, 10) |
| Median duration of hospitalization (quartile 1, quartile 3), d | 4 (3, 5) | 4 (2, 7) | 4 (3, 5) |
| Median duration of invasive oxygen support (quartile 1, quartile 3), d | 2 (2, 3) | 0 | 2 (2, 3) |
| ALT level ≤50 U/L, n (%) | 25 (66) | 31 (84) | 56 (75) |
| Median ALT level (quartile 1, quartile 3), U/L | 33 (21, 69) | 31 (20, 44) | 32 (20, 51) |
Data are n (%) or median (IQR), except age, which is median (range).
Of the 77 children treated, 61 (79%) had at least 1 comorbid medical condition (Supplemental Table 4). Twenty-two (29%) had preexisting pulmonary or thoracic conditions (including 7 [9%] with asthma), 17 (22%) had metabolism and nutrition disorders (including 10 [13%] with obesity), 22 (29%) had congenital disorders, 4 [5%] had cerebral palsy, 13 (17%) had other neurologic disorders (including 7 [9%] with seizure disorders), and 5 (6%) had history of premature birth. Among those with history of premature birth, 2 were >1 year old at baseline. The remaining 3 were an ex–33-week infant 5 weeks old at baseline, an ex–33-week infant 9 weeks old at baseline, and an ex–32-week infant 5 weeks old at baseline (Supplemental Table 4).
Remdesivir Exposure, Disposition, and Concomitant Medications
Forty-eight patients (62%) received all 10 doses of remdesivir, 23 (30%) received 5 to 9 doses, and 6 (8%) received fewer than 5 doses. Almost all patients (75 of 77 [97%]) had the full 28 days of follow-up or had died or been discharged by day 28. The 2 patients who did not complete 28 days of follow-up both completed 10 days of remdesivir and were on room air at the last available follow-up. Concomitant medications with potential effects on COVID-19 included hydroxychloroquine (in 31% of patients in the invasive group and 21% in the noninvasive group), methylprednisolone (in 23% and 16%, respectively), anakinra (in 8% and 5%, respectively), tocilizumab (in 8% and 3%, respectively), hydrocortisone (in 5% of patients in the invasive group), and dexamethasone (in 5% of patients in the invasive group).
Clinical Outcomes
Recovery
By day 28 after the initiation of remdesivir dosing, 83% of patients (64 of 77) had recovered: 79% (31 of 39) of those on invasive respiratory support at baseline were extubated, and 87% (33 of 38) of those not on invasive respiratory support had improved to room air or were discharged. A smaller proportion of patients aged ≤12 years recovered by day 28 than those aged >12 years (75% vs 90%, P = .019). The median time to recovery for patients receiving invasive respiratory support at baseline was 16 days (IQR 11–28), which was significantly longer than the median of 7 days (IQR 5–15) to recovery for those not receiving invasive respiratory support at baseline (P = .005) (Fig 2A).
Time to recovery and discharge in children who received remdesivir. A and B, Time to recovery (A) and time to discharge (B) stratified by baseline respiratory support status. P values were derived from a Cox regression model.
Time to recovery and discharge in children who received remdesivir. A and B, Time to recovery (A) and time to discharge (B) stratified by baseline respiratory support status. P values were derived from a Cox regression model.
Discharge
Overall, the rate of hospital discharge by day 28 was 73% (56 of 77). As with recovery, the rate of discharge was lower among patients on invasive respiratory support at baseline than among those not on invasive ventilation (67% [26 of 39] vs 79% [30 of 38]). Similar to recovery, the median time to discharge for patients receiving invasive respiratory support at baseline was significantly longer at 20 days (IQR 17 to not available) than the median of 13 days (IQR 7–22) for those not receiving invasive respiratory support at baseline (P = .005) (Fig 2B).
Shift in Distribution of Status on Ordinal Scale
At day 28, the majority of patients experienced an improvement in the distribution of clinical support status from baseline. Sixty-eight (88%) improved by at least 1 category in clinical support, 5 (6%) did not change status, and 4 (5%) had worsened status (Fig 3). One patient who was on ECMO at baseline required invasive mechanical ventilation (but not ECMO) at day 28. Three additional patients started ECMO after baseline; of these, 2 were discharged and 1 required low-flow oxygen at day 28. Improvement was less consistent among younger patients: of the 36 patients who were aged ≤12 years, 4 (11%) had worsened clinical status, 3 (8%) did not change, and 29 (81%) had improved clinical status. Of the 41 patients who were aged >12 years, none had worsened clinical status, 2 (5%) did not change, and 39 (95%) had improved clinical status. The proportion of younger children with invasive respiratory support at baseline was higher than that in older patients (59% of patients aged ≤12 years versus 41% of patients aged >12 years).
Clinical outcomes in children treated with remdesivir at day 28. For each oxygen-support category, percentages were calculated with the number of patients at baseline as the denominator. Improvement (blue cells), no change (beige), and worsening (pink) in oxygen-support status are shown. Invasive ventilation includes invasive mechanical ventilation (IMV), ECMO, or both. Noninvasive ventilation includes nasal high-flow oxygen therapy, noninvasive positive pressure ventilation (NIPPV), or both. a A fourth death (reported in Table 1) occurred after day 28. BL, baseline.
Clinical outcomes in children treated with remdesivir at day 28. For each oxygen-support category, percentages were calculated with the number of patients at baseline as the denominator. Improvement (blue cells), no change (beige), and worsening (pink) in oxygen-support status are shown. Invasive ventilation includes invasive mechanical ventilation (IMV), ECMO, or both. Noninvasive ventilation includes nasal high-flow oxygen therapy, noninvasive positive pressure ventilation (NIPPV), or both. a A fourth death (reported in Table 1) occurred after day 28. BL, baseline.
Multivariate Analysis
In a multivariate regression analysis conducted to determine which baseline factors may have been associated with time to recovery, patients on invasive ventilation at baseline had significantly longer time to recovery than those not on invasive ventilation at baseline (hazard ratio of 0.47 [95% confidence interval 0.28–0.78; P = .0035]) and patients aged ≤12 years had significantly longer time to recovery than those >12 years (hazard ratio of 0.54 [95% confidence interval 0.33–0.89; P = .016]).
Safety
Overall, 25 patients (32%) experienced at least 1 adverse event, with a higher proportion among patients on invasive respiratory support at baseline than among those not on invasive respiratory support (38% [15 of 39] vs 26% [10 of 38]) (Table 2, Supplemental Table 5). Twenty-six serious adverse events were reported in 12 patients, with a rate of 21% (8 of 39) in the invasive respiratory support group and 11% (4 of 38) in those not on invasive respiratory support (Supplemental Table 6).
Overall Safety Summary
| Invasive Oxygen (n = 39), n (%) | Noninvasive Oxygen (n = 38), n (%) | Total (N = 77), n (%) | |
|---|---|---|---|
| Any AE | 15 (38) | 10 (26) | 25 (32) |
| Any serious AE | 8 (21) | 4 (11) | 12 (16) |
| Death | 2 (5) | 2 (5) | 4 (5) |
| AE occurring in >1 patient | |||
| ALT level increased | 2 (5) | 3 (8) | 5 (6) |
| AST level increased | 3 (8) | 1 (3) | 4 (5) |
| Anemia | 2 (5) | 0 | 2 (3) |
| Laboratory abnormality, any grade | 30 (77) | 31 (82) | 61 (79) |
| ALT level increased | 14 (36) | 23 (61) | 37 (48) |
| AST level increased | 23 (64) | 18 (47) | 41 (55) |
| Creatinine level increased | 15 (38) | 15 (39) | 30 (39) |
| Laboratory abnormality, grade 3–4 | 15 (38) | 11 (29) | 26 (34) |
| ALT level increased (>5 × ULN) | 5 (13) | 5 (13) | 10 (13) |
| AST level increased (>5 × ULN) | 11 (28) | 4 (11) | 15 (19) |
| Creatinine level increased (>1.8 × ULN) | 8 (21) | 6 (16) | 14 (18) |
| Invasive Oxygen (n = 39), n (%) | Noninvasive Oxygen (n = 38), n (%) | Total (N = 77), n (%) | |
|---|---|---|---|
| Any AE | 15 (38) | 10 (26) | 25 (32) |
| Any serious AE | 8 (21) | 4 (11) | 12 (16) |
| Death | 2 (5) | 2 (5) | 4 (5) |
| AE occurring in >1 patient | |||
| ALT level increased | 2 (5) | 3 (8) | 5 (6) |
| AST level increased | 3 (8) | 1 (3) | 4 (5) |
| Anemia | 2 (5) | 0 | 2 (3) |
| Laboratory abnormality, any grade | 30 (77) | 31 (82) | 61 (79) |
| ALT level increased | 14 (36) | 23 (61) | 37 (48) |
| AST level increased | 23 (64) | 18 (47) | 41 (55) |
| Creatinine level increased | 15 (38) | 15 (39) | 30 (39) |
| Laboratory abnormality, grade 3–4 | 15 (38) | 11 (29) | 26 (34) |
| ALT level increased (>5 × ULN) | 5 (13) | 5 (13) | 10 (13) |
| AST level increased (>5 × ULN) | 11 (28) | 4 (11) | 15 (19) |
| Creatinine level increased (>1.8 × ULN) | 8 (21) | 6 (16) | 14 (18) |
AE, adverse event; ULN, upper limit of normal.
The only adverse events that occurred in >1 patient were elevated ALT levels (2 patients [5%] in the invasive group and 3 [8%] in the noninvasive group), elevated aspartate aminotransferase (AST) levels (3 patients [8%] in the invasive group and 1 [3%] in the noninvasive group), and anemia (2 patients [5%] in the invasive group). Likewise, the only serious adverse events that occurred in >1 patient were elevations of transaminase levels: both groups had 1 grade 3 and 1 grade 4 elevation of ALT and AST levels (Supplemental Table 6). There were 3 renal adverse events reported in 1 invasive group patient each: hematuria, toxic nephropathy, and renal impairment. None of the renal adverse events were attributed to remdesivir by the clinician or led to discontinuation. Laboratory safety findings included 5 cases of grade 3 or 4 elevations of ALT levels in each group, which resolved or improved in all cases in which postelevation follow-up was available (Table 2, Supplemental Fig 4). There were 8 cases of grade 3 to 4 elevations of creatinine levels in the invasive group and 6 in the noninvasive group (Table 2).
Five patients discontinued remdesivir because of adverse events: 2 patients in the group receiving invasive respiratory support (both for elevated liver enzyme levels) and 3 patients not receiving invasive respiratory support (1 because of rash, 1 because of a relapse of acute lymphoid leukemia, and 1 because of elevated liver enzyme levels).
Overall, 4 of the 77 patients (5%) died, 2 in each oxygen-support group. In the group receiving invasive respiratory support, a 14-year-old boy with history of autism, seizure disorder, and gastrostomy tube dependency died 3 weeks after treatment from COVID-19 with suspected cytokine storm contributing, and an 11-year-old girl with history of dermatomyositis and interstitial lung disease died of multiorgan failure due to Gram-negative bacteremia and hemophagocytic lymphohistiocytosis the day after stopping treatment on day 9. In the noninvasive group, a 4-month-old with history of atrial septal defect and pulmonary hypertension died of COVID-19 9 days after completing a 10-day course of treatment. A 5-year-old patient in the noninvasive group died 6 days after completing a 10-day course of remdesivir. This patient presented with clinical and radiographic evidence of meningoencephalitis and subsequently died because of brain herniation; the cause of the meningoencephalitis is unknown.
Discussion
Although most children who become infected with SARS-CoV-2 experience mild or no symptoms and recover fully without medical care, a subset of pediatric patients develop life-threatening symptoms and require hospitalization. For those children requiring admission to the ICU, mortality is considerably lower than that reported in adults and many recover fully with only supportive care.18 A number of case reports and small case series have been published regarding the use of remdesivir in these patients,18–23 but the present report, describing outcomes in 77 pediatric patients with severe COVID-19 who received remdesivir on a compassionate-use basis, is by far the largest set of data on the use of remdesivir to treat pediatric COVID-19 to date. Despite the inherent limitations of data collected in a program of this type, we believe these results represent an important contribution to understanding COVID-19 in pediatric patients and the safety and possible benefit of remdesivir in this population.
In this cohort of children, the majority recovered regardless of the need for invasive mechanical ventilation at baseline. Without a control arm, we cannot know whether remdesivir contributed to recovery. Children who required invasive respiratory support at baseline were less likely to recover and recovered more slowly, but nevertheless, nearly 80% had recovered by day 28, demonstrating that good clinical outcomes can be achieved even in those with severe presentations.
At least 79% of the children in this compassionate-use program had comorbid conditions. The most common comorbidities reported included asthma, obesity, neurologic disorders (including seizure disorders), prematurity, and hematologic conditions (both malignant and nonmalignant). These findings are remarkably similar to those reported in Centers for Disease Control and Prevention surveillance data of children hospitalized with COVID-19, highlighting the potential role of these conditions in predisposing patients to severe illness.2
Overall, remdesivir was well tolerated. Only anemia and transaminase elevations were reported in >1 patient. The observed transaminase elevation may be attributable to COVID-19 illness, remdesivir, or a combination of the two. COVID-19 has a well-established association with liver injury,24–28 and remdesivir has been implicated in transient mild-to-moderate transaminitis in healthy volunteers and in patients infected with Ebola virus.29 Most of the treatment-emergent transaminase elevations we observed were mild, and the more severe (grade 3–4) elevations were reversible.
Interpretation of the results of this descriptive analysis is necessarily limited by the nature of the data. This was not a clinical study designed for the collection of data, but a compassionate-use program undertaken to provide remdesivir to patients who were seriously ill. Without comparative data from a randomly assigned control group, it is not possible to say if the high level of recovery observed in these patients was due to the effects of remdesivir, the natural course of the disease, or other therapeutic interventions. Another limitation is the relatively short duration of follow-up (28 days), at which point 4 patients remained on invasive respiratory support.
Conclusions
Children receiving remdesivir for severe COVID-19 in this compassionate-use program had a high rate of clinical recovery with a favorable tolerability profile; however, in the absence of a control arm, no conclusions can be made about the efficacy of remdesivir in children. The safety, tolerability, pharmacokinetics, and efficacy of remdesivir in children are currently being assessed in a phase 2 and 3 study (NCT04431453).
Acknowledgments
We thank the children who participated in this program and their partners and families and the principal investigators and their colleagues and staff, as well as the Gilead remdesivir compassionate-use eligibility review team (Supplemental Information). We express our solidarity with those who are or have been ill with COVID-19, their families, and the health care workers on the front lines of this pandemic. We also thank Sarah Tse, Gretchen Schmelz, and Deborah Ajayi (BioScience Communications) for preparing graphics and David McNeel (Gilead) for providing editorial assistance.
Drs Osinusi, DeZure, Cao, Chokkalingam, and Brainard conceptualized and designed the compassionate-use program and reviewed and revised the manuscript; Dr Carter coordinated and supervised data collection, drafted the initial manuscript, and reviewed and revised the manuscript; Dr Zhao designed the data collection methodology, conducted the initial analyses, and reviewed and revised the manuscript; Drs Goldman, Aldrich, Hagmann, Camacho-Gonzalez, Méndez-Echevarría, Lapadula, Lee, Bonfanti, and Bamford collected data and reviewed and revised the manuscript; Ms Telep and Drs Pikora and Das coordinated and supervised data collection and reviewed and revised the manuscript; Ms Naik, Mr Marshall, Dr Katsarolis, and Ms Desai coordinated data collection and critically reviewed the manuscript for important intellectual content; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
FUNDING: Funded by Gilead Sciences. The program was designed and conducted by the sponsor (Gilead Sciences). Gilead collected the data, monitored conduct of the program, and performed the statistical analyses.
COMPANION PAPER: A companion to this article can be found online at www.pediatrics.org/cgi/doi/10.1542/peds.2021-050212.
Comments
RE: Remdesivir use in premature neonates with SARS-CoV-2 infection
US FDA has not approved RDV use even for Emergency Use Authorisation in children less than 3.5 kgs of weight [1]. We would like to briefly describe our experience on compassionate use of RDV on three premature neonates who presented with profound apnoeas needing mechanical ventilation support between November 2020 and March 2021. We reported two of these cases recently [2].
All three neonates were positive for SARS-CoV-2 RNA on multiple specimens from naso-pharyngeal aspirate (NPA) as well as broncho-alveolar lavage (BAL). All neonates were discharged from neonatal unit and have spent at least a week or so at home before representing to our Emergency Department. Parents of all three neonates were positive for SARS-CoV-2 RNA - all adults had mild symptoms and were isolating at home. Case 1 was born at 31/40 weeks, presented at 6 weeks of age and the weight at presentation was 2.5 kgs. Case 2 was born at 33/40 weeks and presented at 2.5 weeks of age with a weight of 1.9 kgs. Case 3 was born at 33/40 weeks and presented at 5 weeks of age at a weight of 2.8 kgs. All of them needed oxygen and ventilator support. All three had a negative screen for other common respiratory viruses and also had negative blood cultures. C-Reactive protein, lactate dehydrogenase, ferritin, d-dimer and NT pro-BNP were elevated in all babies.
RDV was given at a dose of 2.5mg/kg on day 1 and 1.25mg/kg between day 2 to 5 [2]. This dosing schedule was agreed at a multi-professional team meeting with special consideration to the age and the weight and subsequently a local guideline was produced [2]. Case 1, 2 and 3 were extubated on day 3, 2 and 3 of RDV treatment respectively. There were no significant side effects noted in our cohort except for case 2 who showed a three fold elevation of aspartate transaminase, AST (highest 162 IU/L) that came back to normal after completion of five days of RDV therapy. All three patients were discharged home successfully and have remained well.
Our experience of RDV use in treatment of these ex-preterm babies has been a positive one and larger studies are needed before recommending RDV as standard treatment in this cohort.
References:
1. Fact sheet for healthcare providers emergency use authorization (EUA) of veklury® (remdesivir) for hospitalized pediatric patients weighing 3.5 kg to less than 40 kg or hospitalized pediatric patients less than 12 years of age weighing at least 3.5 kg. Available at: https://www.fda.gov/media/137566/download (Accessed on 15 November 2020).
2. Saikia B, Tang J, Robinson S, Nichani S, Lawman KB, Katre M, Bandi S. Neonates With SARS-CoV-2 Infection and Pulmonary Disease Safely Treated With Remdesivir, The Pediatric Infectious Disease Journal: May 2021 - Volume 40 - Issue 5 - p e194-e196. doi: 10.1097/INF.0000000000003081