Monoclonal Antibody Treatment of RSV Bronchiolitis in Young Infants: A Randomized Trial

Patients were recruited during 3 bronchiolitis seasons between May 2015 and January 2018 in the infirmary-observation unit of the Hamad Pediatric Emergency Center, the only pediatric emergency facility in the state of Qatar. There are ∼280 000 outpatient visits annually and 45 beds in the infirmary-observation unit where all inpatient services are provided except intensive care and surgery.

Infants aged ≤3 months presenting with a confirmed diagnosis of RSV bronchiolitis and requiring inpatient admission and treatment were eligible for the study. Inclusion required a prodromal history consistent with viral upper respiratory tract infection, including physical findings of bilateral chest crackles that could be associated with cough, rapid breathing, wheezing, and/or intercostal retractions in addition to a positive rapid RSV antigen test result on presentation. Patients were excluded for previous diagnosis of seizure disorder, suspected sepsis, previous history of renal or liver disease, known inborn error of metabolism, congenital heart disease or major congenital anomaly of the upper or lower respiratory tract, received steroid within 2 days before presentation, had received monoclonal antibody or intravenous immunoglobulin within the previous 3 months before random assignment, or known to have hypersensitivity to monoclonal antibodies or immunoglobulin products. Patients were assessed for study eligibility within 120 minutes of the initial physician assessment. Potentially eligible patients were examined on presentation in the emergency center, and those not medically ready for discharge from the emergency center after 120 minutes because of hypoxia (room air oxygen saturation <90%), inadequate feeding, prominent wheezing, or crackles or chest retractions were admitted as inpatients to the infirmary or observation unit. Written informed consent was obtained from 1 of the parents or legal guardians for consecutive eligible patients.

After enrollment, nasopharyngeal aspirate was obtained. Half the sample was processed within 30 minutes for an RSV antigen test (QuickVue RSV Test; Quidel Corporation, San Diego, CA), and half was sent to the main laboratory for a rapid respiratory virus panel capable of identifying 20 respiratory pathogens (multiplex real-time polymerase chain reaction [PCR] assay on ABI 7500 analyzer; Applied Biosystems, Foster City, CA). Patients with positive RSV antigen test results continued in the study, underwent chest radiography and intravenous line insertion, and were connected to a cardiorespiratory monitor. An unblinded study pharmacist used a computer-generated randomization list to prepare identical-appearing numbered syringes containing either 0.75 mL/kg (15 mg/kg) palivizumab or a similar volume of 0.9% sodium chloride for intravenous infusion by a syringe pump over 30 minutes. Then, patients were directed to the appropriate inpatient venue: the infirmary or observation unit, hospital, or PICU. Inhaled therapies, supplemental oxygen, respiratory support, hydration, and other interventions were given at the discretion of the treating physician. Corticosteroids were not permitted as inpatient or postdischarge treatment. Patients were determined to be ready for discharge when they were not requiring supplemental oxygen (room air oxygen saturation >90%); were feeding adequately; had minimal or absent wheezing, crackles, and chest retractions; and Wang bronchiolitis severity score¹⁴ <4. The Wang bronchiolitis severity score¹⁴ ranges from 0 to 12 and has 4 variables, each receiving a score from 0 to 3, with higher scores denoting worse status. On discharge, patients were sent home with salbutamol metered-dose inhalers with an appropriately sized aerochamber mask attachment (Forest Laboratories, New York City, NY). Discharge on salbutamol, although not evidence-based, was a common practice elsewhere¹⁵ and at our center. Daily telephone follow-up by a study nurse was mandatory for 3 weeks after discharge. Study participants returning for the same illness during the 3-week follow-up period were to undergo nasopharyngeal aspirates for RSV each time they revisited by using multiplex real-time PCR. Readmission during follow-up, if medically required, was to an inpatient service. For readmission during follow-up, a patient’s disposition to an infirmary-observation or inpatient hospital bed was based on perceived less versus more serious illness and bed availability, whereas admission to the PICU was required if there was a need for hemodynamic monitoring, assisted ventilation, or related advanced monitoring and treatment. Heart rate and blood pressure were monitored in all study patients from the start of infusion for 4 hours. The study was approved by the hospital institutional review board and funded by Sidra Medicine and Hamad Medical Corporation.

The primary efficacy outcome was readmission to either the infirmary-observation unit, hospital, or PICU during the 3 weeks after discharge. At study start, the primary outcome was the length of stay for the index episode, but this outcome and secondary outcomes were altered during a protocol review without data review in May 2016, a year after study start when additional funding became available for a larger study. Final secondary outcomes were time to medical readiness for discharge on the initial admission, revisit (but not requiring admission) to any medical facility for the same illness in the 3-week follow-up period, and need for transfer to the PICU during the initial admission. We further examined the subsets of patients presenting within 24 hours and within 48 hours of symptom onset and of patients with solely RSV-PCR positivity without other respiratory pathogen coinfection to optimize exploratory analysis of the specificity of the palivizumab intervention.

In response to editorial and reviewer requests, we also post hoc analyzed time to medical readiness for discharge in the subgroup of patients with the highest bronchiolitis severity scores.

Study data were analyzed by the authors. For the primary and secondary outcomes, we planned a proof-of-concept per-protocol analysis, excluding patients lost to follow-up. χ² test or Fisher’s exact test as appropriate were used for qualitative variable and outcome comparisons. Time to medical readiness for hospital discharge was plotted by univariate Kaplan-Meier survival analysis, and the geometric mean time to readiness for discharge for each treatment group was determined by accelerated failure time analysis. We estimated a 20% readmission rate⁴ and hypothesized a reduction to 10%, consequent to investigational palivizumab treatment. For 80% power to find such a difference at a 2-sided P value of .05, 199 patients per group were required. We increased each group by 10 patients to account for dropouts. This expanded sample size was determined during protocol review without data review in May 2016, a year after study start. The initial sample size planned (65 patients per treatment group) had been based on convenience and limited funding that was subsequently increased.

Categorical and continuous data values were expressed as frequency (percentage) and mean (SD). Descriptive statistics were used to summarize baseline demographic, laboratory, and clinical characteristics of the patients. Quantitative variable means between the 2 independent groups were analyzed by using unpaired t and Wilcoxon rank sum tests. Associations between 2 or more qualitative variables were assessed by using the χ² test. Significant values and differences were reported with their corresponding 95% confidence intervals (CIs). Statistical analyses were performed by using a statistical software package (SPSS version 22.0; SPSS Inc, Chicago, IL). Data were transferred from SPSS package to Stata SE 14.0 (Stata Corp, College Station, TX) for geometric mean times to readiness for discharge for each treatment group.

From May 2015 to January 2018, until the enrollment goal was reached, 420 previously healthy infants diagnosed with RSV viral bronchiolitis (median age 49.0 days [range 4–97 days]) were enrolled in the study (patient flow is summarized in Fig 1). Data for 7 of the enrolled infants were not included in the analysis; 3 parents refused the intervention after preparation of study medication, and 4 infants were lost to follow-up. Of the 413 (98.3%) infant episodes remaining, 208 had been randomly assigned to receive intravenous palivizumab, and 205 received the placebo. Subjects’ baseline characteristics were similar in the 2 treatment arms at enrollment (Table 1). Among palivizumab recipients, 193 remained in infirmary-observation, 5 were cared for on the ward, and 10 were cared for in the PICU. Among placebo recipients, there were 189 in the infirmary-observation unit, 3 on the ward, and 13 in the PICU. Treatment with nebulized albuterol, nebulized epinephrine, and systemic antimicrobial agents occurred with similar frequencies in both patient groups (Table 1).

For the primary outcome, 23 of 208 (11.1%) infants treated with palivizumab required readmission to the infirmary or observation unit, hospital, or PICU during the follow-up period compared with 19 of 205 (9.3%) placebo recipients (difference 1.8%; 95% CI −4.4% to 7.7%; P = .51).

The geometric mean time to readiness for discharge was 29.5 hours (95% CI 25.7 to 33.9) for the palivizumab group and 30.2 hours (95% CI 26.3 to 34.7) for the placebo group, resulting in a ratio of 0.98 (95% CI 0.81 to 1.20; P = .83; Fig 2). Revisits for the same illness not requiring readmission during the 3 weeks after discharge occurred in 41 palivizumab recipients (19.7%) and 45 placebo recipients (22.0%). Altogether, 64 palivizumab and 64 placebo recipients made 85 and 94 visits, respectively (difference −0.4%; 95% CI −8.9% to 9.8%; Table 2). Transfer to the PICU due to worsening illness severity during the initial admission was required for 10 patients (4.8%) in the palivizumab group and 13 (6.3%) in the placebo group (difference −1.5%; 95% CI −5.8% to 3.2%).

We considered that evidence favoring palivizumab in patients with short symptom duration before presentation and in patients without a pathogen other than RSV would support a positive efficacy hypothesis. Among palivizumab recipients with symptoms ≤1 day before presentation, 3 of 18 (16.7%) required readmission during the 3 weeks after discharge compared with 1 of 17 (5.9%) placebo recipients (difference 10.8%; 95% CI −13.0% to 21.6%). Among patients with symptoms ≤2 days before presentation, 8 of 78 (10.3%) palivizumab recipients and 7 of 71 (9.9%) placebo recipients were readmitted within 3 weeks (difference 0.4%; 95% CI −9.7% to 10.0%). For the subgroup of patients without PCR evidence of coinfection with a respiratory pathogen other than RSV, readmission within 3 weeks occurred in 15 of 142 (10.6%) palivizumab recipients and 12 of 138 (8.7%) placebo recipients (difference 1.9%; 95% CI −5.5% to 8.8%). The time to medical readiness for discharge and frequency of other secondary outcomes for each subgroup comparison were also similar for the palivizumab and placebo treatment groups (data not shown) as were subjects’ baseline characteristics for each comparison (data not shown).

A post hoc secondary analysis of efficacy in patients with Wang bronchiolitis severity scores¹⁴ well above the mean and median (≥7; Figs 3 and 4) suggested a (not statistically significant) reduced time to medical readiness for discharge for patients treated with palivizumab: 33.1 hours (95% CI 24.0 to 45.7) for the palivizumab (n = 37) patients versus 52.5 hours (95% CI 35.5 to 75.9) for placebo (n = 33) recipients (ratio 0.66; 95% CI 0.41 to 1.06; P = .08). Baseline characteristics and other outcomes were similar for the 2 treatment subgroups.

During the 3 weeks of follow-up, 64 infants in the palivizumab group and 64 in the placebo group made at least 1 revisit for the same illness (Table 2). However, protocol nasopharyngeal aspirates for RSV were repeated for only 19 infants in the palivizumab group and 16 infants in the placebo group. PCR was RSV-positive for 17 palivizumab and 14 placebo patients. When compared with the initial PCR results, 7 children in the palivizumab group and 6 in the placebo group now tested positive for a respiratory pathogen not previously present.

No patient was withdrawn from the study because of apnea, cyanosis, or hemodynamic instability. Blood pressure and heart rate profiles were similar between the palivizumab and placebo groups throughout the period of monitoring.

For young infants with RSV-associated bronchiolitis, treatment with intravenous recombinant monoclonal anti-RSV antibody was not superior to normal saline for preventing the need for readmission, reducing time to medical readiness for discharge, and the remainder of the secondary and exploratory outcomes.

A systematic review¹³ of the available evidence for monoclonal RSV antibody treatment of acute RSV bronchiolitis presented studies conducted on patients with advanced disease who mostly required ICU and ventilatory support. Two small placebo-controlled randomized trials of human polyclonal anti-RSV antibody in children <2 years of age were conducted 2 decades ago, with 1 regarding previously healthy infants¹⁶ and the second regarding infants at high risk¹⁷ for severe RSV infections. Both revealed safety but were unconvincing regarding efficacy. Palivizumab was well tolerated, found to be safe when given intravenously,^13,18,–25 and produced adequate serum antibody levels for at least 3 weeks when given at 15 mg/kg.²⁶ Yet, there was no survival benefit despite a decrease in RSV quantified from tracheal secretions.¹³ A randomized trial using motivizumab²⁷ in hospitalized infants with RSV infection did not show reduced viral load or severity of illness. Motavizumab, which is derived from palivizumab, had considerably more potent RSV neutralizing activity in tissue culture.

Our findings do not support the routine use of RSV monoclonal antibody to treat acute RSV bronchiolitis in young infants, particularly considering the high cost of treatment. We think these results are robust because we strictly applied enrollment criteria in a population at high risk for severe disease in a randomized double-blind trial. Our study outcomes allowed ample time for the treatment to reveal a difference on clinical grounds should one exist, because our considerably enlarged patient sample was managed for 3 weeks after discharge, with <2% dropouts. We chose need for readmission during 3-week follow-up as our primary outcome because previously published work revealed that 25% of such children remained symptomatic at 21 days and 37% had further unscheduled medical visits after discharge.² A 33% shortened time to medical readiness for discharge after palivizumab treatment in the 17% of patients with the most severe bronchiolitis presentation is a tantalizing finding but requires confirmation in robust and prospective investigation.

A complex host inflammatory response to RSV infection has been described, with massive peribronchial inflammatory cell infiltration with activation of cytokines and chemokines, leading to bronchial structural damage, cell necrosis, and intraluminal obstruction.^28,–30 From our study, we speculate that host inflammatory response appears to be the more important factor affecting disease severity and resolution in the overall patient group we studied.

Our study has limitations. We did not attempt to quantify viral load in upper and lower specimens to validate palivizumab’s mechanism of action. Our study follow-up was not long enough to examine possible time-remote effects, such as the incidence of future wheezing or the possibility of the type of disease enhancement that followed the use of an inactivated RSV vaccine. Our protocol for nasopharyngeal aspirate reexamination upon revisit after discharge had poor adherence. Finally, despite randomizing >400 patients, we may have failed to identify a small group of children who might have benefitted from palivizumab therapy, such as those presenting with the highest bronchiolitis severity scores or a group with a lesser clinical benefit than our sample size allowed us to detect.

We conclude that intravenous palivizumab, which is known to suppress replication of RSV, did not help or harm young infants with acute RSV-positive bronchiolitis and did not improve any of multiple relevant clinical outcomes.

We thank Dr Rajvir Singh for his thoughtful guidance in biostatistical analysis.

CI

confidence interval

PCR

polymerase chain reaction

RSV

respiratory syncytial virus