A randomized, placebo-controlled phase 3 trial of the PI3Kδ inhibitor leniolisib for activated PI3Kδ syndrome

Activated phosphoinositide 3-kinase delta (PI3Kδ) syndrome (APDS) is an inborn error of immunity resulting from pathogenic heterozygous variants in either of the genes encoding the PI3Kδ heterodimer. Gain-of-function variants in PIK3CD encoding the catalytic subunit p110δ cause APDS1, whereas loss-of-function variants in PIK3R1 encoding the regulatory subunit p85α cause APDS2.^1-3 Interaction of the 2 subunits is essential for heterodimer function and stability.⁴ Disruption of this interaction or loss of p85α-mediated inhibition because of pathologic variants in PIK3CD or PIK3R1 result in hyperactive PI3Kδ signaling.^1,2,4 This hyperactive signaling results in comparable immunologic consequences in both APDS1 and APDS2.^4-6 

APDS is a complex immune deficiency with striking immune dysregulation and diverse clinical presentation (supplemental Figure 1; available on the Blood website).^7,8 This heterogenous presentation is caused in part by dysregulated B and T cells.^9,10 Proper lymphocyte development and function depends on a tightly balanced PI3Kδ pathway: enzyme activity must be variably increased and dampened at different time points to allow for FOXO-dependent signaling.¹¹ Constitutively active PI3Kδ, as occurs in patients with APDS, results in disturbed immune cell development and function.^1,2 

Hallmarks of the clinical presentation are recurrent sinopulmonary infections and prolonged or intermittent herpesvirus viremia.^5-8 These infections are due to immunoglobulin dysfunction or related to increases in immature/dysfunctional B and T lymphocytes at the expense of mature/functional cells.^12-14 Patients with APDS often have an inverted CD4⁺/CD8⁺ T-cell ratio, which is primarily driven by an increase in terminal differentiation.^1,5,6 Naïve T cells decrease, and patients often have an increase in cytotoxic CD8⁺ effector memory and terminally differentiated effector memory cells at the expense of long-lived central memory cells.^1,2,5,6 These cells also tend to resemble highly inflammatory senescent or exhausted CD8⁺ T cells.^1,2,12 The shifts in T-cell subsets result in ineffective responses to infections, particularly chronic infections such as Epstein-Barr virus (EBV) and cytomegalovirus (CMV).^12-14 In the B-cell compartment, immature transitional B cells are frequently elevated whereas mature naïve B cells and memory B cells are often decreased, along with defects in class-switch recombination.^1,5,13 Hyperactive PI3Kδ signaling also drives terminal differentiation in B cells: immunoglobulin M⁺ (IgM⁺⁾ plasma cells are often increased in patients with APDS.^15,16 As a result of these disrupted lymphocyte subsets, many patients with APDS display elevated IgM and low IgG and IgA, with poor specific antibody responses, diminishing their ability to fight infections.^5,6,13 The percentage of naïve B cells out of the total B cells reflects altered B-cell subsets and can provide a sense of the impact of APDS on immunodeficiency.

Lymphoproliferation, including lymphadenopathy, splenomegaly, hepatomegaly, and nodular lymphoid hyperplasia of the airways and gut is typical.^5-8 Lymphadenopathy is driven by the proliferation of B cells and the infiltration of T follicular helper cells that are highly positive for programmed cell death protein 1 (PD-1).^5,6 T-cell subsets skewed towards the effector phenotype may also contribute to lymphadenopathy because these cells demonstrate an enhanced proliferative burst on encounter with antigen.¹ Depending on the location and size of the swollen lymph nodes, they may affect breathing or cause obstruction.^17,18 Lymphadenopathy and expansion of lymphoid tissues primarily reflects the immune dysregulation aspect of APDS and its reduction, especially in the neck, mediastinum, and gut, improves organ specific function.

Other manifestations of APDS include autoimmunity (eg, cytopenias) and enteropathy.^5-8 Uncontrolled antigen- or EBV-driven PI3Kδ activity may partially drive malignant transformation of B cells.¹⁹ Lymphoma has been reported in up to 25% of patients with APDS, at rates much higher than those of the general population.^5-8,20 Less well-characterized manifestations include EBV positive posttransplant lymphoproliferative disease-like lymphoproliferation, sclerosing cholangitis, nodular regenerative hyperplasia of the liver with portal hypertension, allergy, asthma, eczema, neurodevelopmental delay, and seizures.^5,6,8,21,22 End-organ damage, including hearing loss, bronchiectasis, and liver disease reflects the complex and progressive nature of APDS with multisystem involvement.^5-8,23 APDS may lead to early mortality, although disease severity and survival remain variable.^8,13 

Current APDS management is mainly empirical. Treatment burden includes immunomodulatory therapies, prophylactic antimicrobials, immunoglobulin replacement therapy (IRT), and procedures such as splenectomies, repeated otosinopulmonary surgeries, and hematopoietic stem cell transplantation. Most strategies do not target the disease pathogenesis, hyperactive PI3Kδ signaling.^5-7 

Leniolisib, a novel, orally bioavailable small molecule inhibitor, was engineered to selectively target PI3Kδ signaling.²⁴ We previously reported leniolisib use in 6 patients with APDS in a 12-week dose-finding clinical trial. Leniolisib was well tolerated, reduced PI3Kδ pathway hyperactivity, partially reconstituted lymphocyte subsets, and decreased lymphoproliferation.²⁵ 

Here we report outcomes from a phase 3 trial (NCT02435173), a 12-week, randomized, triple-blinded, placebo-controlled, fixed-dose study of 31 patients.

Male and female patients aged 12 to 75 years and weighing ≥45 kg with pathogenic variants in PIK3CD or PIK3R1, clinical findings consistent with APDS, and ≥1 measurable lymph node on computed tomography or magnetic resonance imaging scan were eligible. Patients taking immunosuppressive agents, including rapamycin, underwent washout periods before entry. Glucocorticoid doses equivalent to ≤25 mg per day of prednisone were allowed within 2 weeks before the first dosing and throughout the study. supplemental Section 1 lists other eligibility criteria.

Patients from the United States, Europe, and Russia were enrolled in a 12-week, randomized, subject-, investigator-, and sponsor-blinded, placebo-controlled, fixed-dose study. On day (D) 1, patients were randomized 2:1 to receive 70 mg leniolisib or placebo orally every 12 hours twice daily. Efficacy and safety assessments were performed on D15, D29, D57, and D85, with pharmacokinetic (PK) assessments on D29, D57, and D85.

There were 2 coprimary endpoints. One was the negative change from baseline at D85 in the log₁₀-transformed sum of product diameters of the index lymph nodes. Index nodes comprised ≤6 of the largest lymph nodes as measured by magnetic resonance imaging or computed tomography scan and were selected per Cheson criteria.²⁶ The other was the positive change from baseline at D85 in the percentage of naïve B cells out of total B cells as assessed by flow cytometry. In order to score an improvement in the naïve B-cell percentages, patients were required to have <48% naïve B cells at baseline for analysis.²⁷ 

Secondary and exploratory endpoints included changes in bidimensional size and 3D volume of the spleen/liver, immunophenotyping of B- and T-cell subsets, and levels of serum immunoglobulins, cytokines, chemokines, and inflammatory markers. EBV and CMV loads were measured. Outcomes reported by patient and clinicians were evaluated via multiple instruments. PK assessments were performed in all patients receiving leniolisib. Safety was assessed in all patients; supplemental Section 1 includes assessment tools and further details.

Novartis AG, in collaboration with investigators, designed the study. Novartis AG oversaw its conduct, and Novartis AG and Pharming Group NV analyzed the data. The NCT02435173 trial was conducted and reported in accordance with the protocol (including the statistical analysis plan). Data were gathered locally, and protocol-defined laboratory samples and imaging were processed centrally. The trial was conducted in accordance with the Declaration of Helsinki and the International Council on Harmonization Good Clinical Practice guidelines. Independent ethics committees or institutional review boards at each center approved the protocol. Patients and/or their guardians provided written informed consent/assent. An independent data and safety monitoring committee monitored safety and protocol compliance.

Analysis of the primary lymphadenopathy endpoint determined sample size. Based on the standard deviation (SD) observed in the dose-finding trial (0.14), a sample size of 20 leniolisib and 10 placebo patients was estimated to provide 97% power to detect statistically significant differences using 5% type I error.

Three data sets were analyzed. The safety analysis set included all patients who received any study drug. The pharmacodynamics (PD) analysis set included all patients who received any study drug with no protocol deviations and with relevant impact on endpoints. The PK analysis set included all patients with ≥1 valid PK concentration measurement who received any study drug with no protocol deviations and with relevant impact on PK data. For all data sets, there was no imputation of missing data.

Analysis of covariance was performed on the PD analysis set for each coprimary endpoint. Baseline intake of glucocorticoids and IRT were included as categorical (Yes/No) covariates. Comparison of the 2 treatment groups was 2-sided, with 5% type I error. Treatment was a fixed effect and baseline was a covariate. For the lymphadenopathy endpoint, patients with 0 nodes at D85 were excluded. For the B-cell endpoint, only patients with <48% naïve B cells at baseline were included in the primary analysis (B-PD data set). A supportive analysis using the full PD analysis set was also performed. As a sensitivity analysis, the change from baseline in naïve B-cell levels across other study dates was analyzed in the B-PD data set using a longitudinal mixed model, with treatment, time, treatment by time interaction, baseline, and baseline by time interaction as fixed effects. Multiplicity was controlled across the coprimary endpoints implicitly by requiring both to be statistically significant for a positive study.

Analyses of covariance with the same parameters as the primary endpoint analyses were also used to assess changes from baseline to D85 in the PD analysis set for patient-reported outcomes, non-index lymph nodes, and spleen/liver sizes. The same repeated measures analysis used for B-cell frequencies was used for patient-reported outcomes. For all secondary and exploratory endpoints, summary statistics were calculated; 95% confidence intervals (CIs) have not been adjusted for multiplicity and inferences drawn may not be reproducible. See supplemental Section 1 for details.

From December 2017 to August 2021, 32 patients were screened and 31 entered the trial (Figure 1). Baseline demographics and clinical characteristics of the patients were similar between groups and representative of the broader APDS population (Table 1, supplemental Section 2 and supplemental Table 1).^7,8 All randomized patients completed treatment and were included in the safety analysis set. The PD analysis set (N = 27) was used to assess efficacy outcomes; 2 patients from each treatment arm were excluded because of protocol deviations. The B-PD analysis set for the primary naïve B-cell endpoint excluded an additional 14 patients; reasons included >48% naïve B cells at baseline (n = 8), no assessment performed at D85 (n = 4), no baseline measurement of total B-cell levels (n = 1), and no naïve B cells at baseline (n = 1). Baseline demographics and clinical characteristics of the B-PD analysis set were comparable between treatment arms (data not shown).

Leniolisib significantly reduced lymphadenopathy, meeting a coprimary endpoint (Figure 2A). The difference in the adjusted mean change (95% CI) between leniolisib (n = 18) and placebo (n = 8) was −0.25 (−0.38, −0.12; P = .0006). One patient receiving leniolisib was excluded from the PD analysis set because the baseline index node fully resolved by D85. Figure 3A shows index lymph node size for patients in the safety analysis set. In the safety analysis set, 26% of patients in the leniolisib group (n = 19) achieved complete absence of index lymphadenopathy, whereas the other 74% achieved partial response. In the placebo group (n = 9), 45% achieved partial response, 44% had stable disease, and 11% had an unknown response. The single patient with the unknown response moved during the scan, which may explain the data loss. Though no measurements were available at D85 for 2 of 6 index nodes, the rest demonstrated continued lymphadenopathy. Leniolisib decreased spleen size compared with placebo: the adjusted mean difference (95% CI) in bidimensional size between the groups was −13.5 cm² (−24.1, −2.91; P = .0148) and in 3D volume was −186 cm³ (−297, −76.2; P = .0020). Of the patients in the safety analysis set with baseline splenomegaly, in the leniolisib group (n = 13), 38% achieved complete response, 54% achieved partial response, and 8% had stable disease at 12 weeks. In the placebo group (n = 5), 20% achieved complete response with the remaining 80% experiencing worsening disease. Figure 3B shows 3D spleen volumes of each patient with measurements in the safety analysis set. Figure 3C-J shows radiographic renderings of nodes and spleens. supplemental Section 2 reports supportive analyses and changes in liver size.

As an indication of improvement in immunodeficiency, leniolisib significantly increased the percentage of naïve B cells, meeting the other coprimary endpoint (Figure 2B). The difference in the adjusted mean change (95% CI) between leniolisib (n = 8) and placebo (n = 5) from baseline to D85 was 37.30 (24.06, 50.54; P = .0002). The supportive analysis, which included all patients in the PD analysis set with measurements at baseline and D85, was consistent with the primary analysis: the adjusted mean difference (95% CI) between leniolisib (n = 13) and placebo (n = 8) was 27.94 (15.02, 40.85; P = .0003). A repeated measures sensitivity analysis confirmed these findings (supplemental Section 2, supplemental Figure 2). Figure 4A shows the mean percentage of naïve B cells over time for all patients in the safety analysis set with measurements.

Leniolisib improved key lymphocyte subsets. Changes in frequency in the PD analysis set are shown in supplemental Table 3, whereas Figures 4-5 show all patients in the safety analysis set with measurements. Each stage of lymphocyte trafficking and development is impacted in APDS, as evidenced by the subsets of B cells that accumulate. Elevated transitional B cells and CD38⁺ plasmablasts decreased in the leniolisib group (Figure 4B-C). Switched and nonswitched memory B-cell percentages decreased slightly (Figure 4E). Importantly, leniolisib greatly reduced serum IgM, which is often elevated in patients with APDS (Figure 4D). In the PD analysis set, the mean IgM level decreased 208.26 mg/dL from baseline to D85 in the leniolisib arm and 10.00 mg/dL in the placebo arm.

CD8⁺ senescent CD57⁺ T cells and PD-1⁺ T cells, often elevated in patients with APDS, were reduced with leniolisib (Figures 5A-B). The inverted CD4:CD8 T-cell ratio normalized with leniolisib, increasing from 0.73 to 1.05. Although leniolisib treatment decreased the overall proportion of total CD8⁺ T cells, including terminally differentiated effector memory (T_EMRA) CD8⁺ T cells, it increased naïve CD8⁺ T-cell percentages (Figure 5D-E). Leniolisib increased the percentage of total CD4⁺ T cells, with a decrease in the CD4⁺ T_EMRA subset (Figure 5D-E). We observed little effect of leniolisib on central memory T cells and effector memory T cells after 12 weeks (supplemental Figure 2).

CXCL13 is a chemokine facilitating recruitment of B and T cells to lymphoid follicles and a marker for T follicular helper cell activity and aberrant lymphocyte trafficking; mean levels decreased 286.77 pg/mL from baseline to D85 in the leniolisib group and increased 59.31 pg/mL in the placebo arm in the PD analysis set (Figure 5F). We observed decreases in tumor necrosis factor α and other systemic inflammatory markers (supplemental Section 2, supplemental Figure 3).

We used quantitative polymerase chain reaction to detect EBV and CMV. Although we detected CMV (n = 6), EBV (n = 5), or both (n = 5), no clinical progression occurred.^30,31 The highest value for either virus occurred in 2 different patients: EBV, 696 DNA copies per mL; CMV, 591 DNA copies per mL.

Baseline cytopenias varied among patients: in the leniolisib group, baseline cytopenias included anemia (n = 8), thrombocytopenia (n = 4), lymphopenia (n = 4), and neutropenia (n = 1), with 2 patients having multiple cytopenias. In the placebo group, baseline cytopenias included anemia (n = 4) and thrombocytopenia (n = 1). Among the safety analysis set, 82% of cytopenias improved in patients receiving leniolisib, compared with 60% in patients receiving placebo (supplemental Section 2; Figure 6).

We observed no statistically significant changes in patient- and clinician-reported outcomes in 12 weeks (supplemental Section 2 and supplemental Figures 4-6). However, patient and physician general assessment scores reflected clinically meaningful improvements for the leniolisib group, whereas the placebo group only showed meaningful improvement for the physician general assessment score. Elements of the Work Productivity and Activity Impairment plus Classroom Impairment Questionnaire also suggest clinically meaningful improvements for the leniolisib but not placebo group. Improvements in SF-36 scores were not clinically meaningful except for the physical functioning subscale for the placebo group. Corresponding investigator narratives described positive improvements including increased tolerance for physical activity and decreased fatigue in 70.0% of patients receiving leniolisib and 44.4% receiving placebo.

After a single dose of 70 mg leniolisib, the absorption phase was characterized by a geometric mean (CV %) C_max of 2080 ng/mL (25.4%) reached at median T_max of 2.87 hours (supplemental Figure 7A and supplemental Table 4). Multiple-dose PK was characterized by steady-state trough concentrations on D29, D57, and D85 (supplemental Figure 7B). The geometric mean (CV %) C_trough on D85 was 804 ng/mL (67.5%).

Adverse events (AEs) were reported in 85.7% of patients receiving leniolisib and in 90.0% of the placebo group; these events were mostly grade 1 (74.2%; Table 2). Study drug–related AEs occurred in 8 patients; the incidence was lower in the leniolisib arm (23.8%) than in the placebo arm (30.0%). Transient alopecia was reported in 2 of 21 patients receiving leniolisib. In total, 5 patients reported a serious AE, with none judged as related to study medication. No deaths were reported within 30 days of the end of trial, and no AEs led to discontinuation of study treatment.

Skin rash is a reported class effect of PI3Kδ inhibitors and was observed with leniolisib in healthy subjects at doses of 70 and 140 mg twice daily (<10% of subjects) as well as in patients with primary Sjögren syndrome at 70 mg twice daily (55% of patients).^32,33 Here, 1 patient reported maculopapular rash (grade 1), which was not suspected to be related to leniolisib treatment. Immune-related AEs reported with the use of other PI3Kδ inhibitors for hematologic malignancies were not seen.^32,34 

Grade 3 or greater neutropenia is another class effect of PI3Kδ inhibitors seen in cancer treatment trials but was not seen in the present study.^32,34 We observed transient neutropenia in the leniolisib group, with a nadir on D15 (mean count, 2.204 × 10⁹/L) that recovered by D85. Four patients treated with leniolisib had neutrophil levels <1.0 × 10⁹/L at a single visit, but none had infections, and at the subsequent visit counts were >1.0 × 10⁹/L for all of them. Four of the 21 patients in the leniolisib arm developed asymptomatic mild neutropenia by D85 (0.90 × 10⁹/L-1.30 × 10⁹/L). Mean neutrophil level at D85 of all 21 patients receiving leniolisib was 2.94 × 10⁹/L (range, 0.90 × 10⁹/L-6.50 × 10⁹/L). Mild and transient neutropenia was previously observed in healthy subjects treated with 70 mg leniolisib twice daily.

In patients with APDS, leniolisib given orally at a dose of 70 mg twice daily over a period of 12 weeks reduced lymphadenopathy and increased naïve B-cell percentage significantly more than placebo, meeting both primary endpoints. Other key disease outcome measures, including spleen size, lymphocyte subsets, and cytopenias also improved. IgM levels decreased to healthy limits in most patients receiving leniolisib over 12 weeks. Changes in health-related quality of life measures were not statistically significant. However, investigator narratives reported a majority (70.0%) of patients receiving leniolisib (and 44.4% receiving placebo) had improvements in activity and energy levels.

We observed favorable changes in lymphocyte subsets attributed to improvements in trafficking, development, and maintenance of these cells. Mechanisms of defective EBV/CMV control in patients with APDS, such as elevated transitional B cells, which are thought to be entry points and reservoirs for EBV, and increased senescent CD8⁺ T cells, thought to be dysfunctional effectors, were all notably improved.¹³ 

Leniolisib was well tolerated. Many of the reported AEs were present at baseline. One patient in the leniolisib arm contracted asymptomatic COVID-19. Investigators reported more study drug–related AEs in the placebo group than the leniolisib group. Skin rash was reported in 1 patient receiving leniolisib but was judged to be fungal and not related to study treatment by the investigator, in contrast to effects reported with other PI3Kδ inhibitors used in non-APDS populations.^32,33 This finding is consistent with that reported in part 1, in which we observed no skin rashes.²⁵ We hypothesize that the AEs commonly associated with PI3Kδ inhibitors in other populations may be minimized in patients with APDS as their baseline hyperactive PI3Kδ is brought within a physiologic range and not inhibited completely by leniolisib.

This trial has several limitations. Firstly, the sample size was small, particularly for the naïve B-cell analysis, and immunophenotyping data were not available for all subsets for all patients. However, the supportive naïve B-cell analysis with the entire PD analysis set was consistent with results from the smaller data set. The sample size was decreased further because of protocol deviations: the analysis was based on treated patients with no relevant protocol deviations rather than intent-to-treat. The limited sample size might have also affected the randomization of disease complications and prior treatments; although most characteristics are similar between the 2 groups, bronchiectasis and gastrointestinal complications were more prevalent in the placebo group. Interestingly, the higher frequencies of these complications in the placebo group did not appear to correspond to disease impairment because the placebo group had less impairment on the Work Productivity and Activity Impairment plus Classroom Impairment Questionnaire than the leniolisib group (supplemental Figure 5). Regarding lung disease specifically, it is notable that patients with APDS who were well managed with adequate health care (eg, IRT) from early childhood did not develop significant bronchiectasis. Conversely, those patients who had clinically significant lung problems including surgical pulmonectomy and bronchiectasis did well on leniolisib and completed the 12-week clinical trial.

Secondly, the health-related quality of life surveys have not been validated in children or patients with inborn errors of immunity, which may contribute to nonsignificant results. Thirdly, conducting an international trial during a global pandemic entailed complications, including changes or limitations in performing assessments and procedures at predetermined time points. Finally, outcomes were evaluated over only a 12-week period, which was chosen to limit the amount of time that patients in the placebo group were without therapy. Washout periods were required for patients receiving antiproliferative, immunosuppressive, or B-cell depletion therapies before the trial and ranged from 6 weeks to 6 months, depending on the agent. Combined with the 12-week trial, patients in the placebo arm could have been without treatment for 4 to 8.5 months. Given the progressive nature of APDS, it would not have been ethical to withhold treatment from patients in the placebo group for longer. However, 12 weeks was insufficient to assess clinically meaningful measures, such as decreases in infection frequency, hospitalization, antimicrobial use, and continued need for IRT; these will be evaluated in the future. Pediatric trials in younger children with APDS are also planned.

Leniolisib is not the only PI3Kδ inhibitor that has been investigated for treatment of APDS. A phase 2 trial of the inhaled agent nemiralisib was completed in 5 patients but further studies were not pursued (NCT02593539).³⁵ A phase 3 trial of the oral inhibitor seletalisib was terminated;³⁶ published phase 1b and open label extension studies describe moderate efficacy and AEs that led to treatment discontinuation.³⁷ The reasons behind the differences in efficacy and tolerability between seletalisib and leniolisib are not immediately clear but may involve inhibitor structure and isoform selectivity.

Overall, treatment with leniolisib demonstrated targeted therapy of a rare but biologically relevant immune dysregulation and immune deficiency with a significant benefit over placebo with respect to the coprimary endpoints, reducing lymphadenopathy and increasing the percentage of naïve B cells by D85, and was well tolerated in patients with APDS.

The authors thank the patients who took part in the study and their families. The authors also acknowledge the local physicians who cared for these patients: Jack H. Bleesing, Cincinnati Children’s Hospital Medical Center and University of Cincinnati College of Medicine, Cincinnati, OH; Zeynep Yesim Kucuk, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH; Jeffrey T. Trost, Stephen G. Diamantoni and Associates Family Practice, Quarryville, PA; Diana K. Baker, University of Iowa Carver College of Medicine, Iowa City, IA; Polly J. Ferguson, University of Iowa Carver College of Medicine, Iowa City, IA; Amy M. Scurlock, University of Arkansas for Medical Sciences and Arkansas Children’s Hospital, Little Rock, AR; Christa S. Zerbe, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD; Ivan Kingyue Chinn, Baylor College of Medicine and Texas Children’s Hospital, Houston, TX; Jill M. Powers, Grand River Medical Group, Dubuque, IA; Janna A. Kroiss, Aurora Family Medicine, Two Rivers, WI; John M. Routes, Medical College of Wisconsin, Milwaukee, WI; Yasmin W. Khan, Vanderbilt University Medical Center, Nashville, TN; Ramsay L. Fuleihan, Columbia University Irving Medical Center and New York-Presbyterian Morgan Stanley Children’s Hospital, New York, NY; Lee S. Clore, Jr, Allergy and Asthma Specialists, PSC, Owensboro, KY; Gary I. Kleiner, University of Miami Miller School of Medicine, Miami, FL; T. Prescott Atkinson, The University of Alabama at Birmingham, Birmingham, AL; Cecilia P. Mikita, Walter Reed National Military Medical Center, Bethesda, MD; Christopher V. Macomb, Walter Reed National Military Medical Center, Bethesda, MD; John Hansen-Flaschen, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA; Ganga R. Ranasuriya, University of Pittsburgh Medical Center Susquehanna Health, Lung Center, Williamsport, PA; Patricia A. Takach, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA; Maryam Ali Al-Nesf Al-Mansouri, Hamad Medical Corporation and Qatar University College of Medicine, Doha, Qatar; Saad Mujeb Al-Shareef, King Faisal Specialist Hospital, Riyadh, Saudi Arabia; Randa Arnaout, King Faisal Specialist Hospital, Riyadh, Saudi Arabia; and Clara Mosa, ARNAS Civico Di Cristina Benfratelli Hospital, Palermo, Italy. The authors thank Anuj Kashyap, Lavenda Chirombo Kluczynski, and Alanvin Orpia for their technical assistance, and Amy FitzPatrick, Carolyn Keating, and the staff at PRECISIONscientia for data collation and manuscript preparation.

This work was supported by the Division of Intramural Research, National Institute of Allergy and Infectious Diseases, National Institutes of Health. The opinions expressed in this presentation are the authors’ own and do not reflect the view of the National Institutes of Health, the Department of Health and Human Services, or the US government. Novartis Pharmaceuticals AG and Pharming Group NV provided funding for the clinical trials, and Pharming Healthcare, Inc, provided funding for the manuscript preparation by PRECISIONscientia.

Contribution: V.K.R., S.W., A. Šedivá, A.P., C.S., A. Shcherbina, N.C., T.C., V.A.D., A.T., Y.Z., E.K., J.K., V.L., Y.R., and G.U. performed research and collected data; K.R. and K.K. were involved in the design of the research; J.B. and A.R. were involved in oversight of the data reports; and S.M.H. and M.J.L. maintained oversight of the NIH research.

Conflict-of-interest disclosure: A. Šedivá is a consult for Octapharma, Takeda, and Pharming Group NV. A. Shcherbina receives honoraria from and is a consult for Octapharma, CSL Behring, and Novartis AG. E.K. is an employee of Leidos Biomedical Research, Inc. V.A.D. is a consultant for and/or receives honoraria from AstraZeneca, Kedrion, Takeda, CSL Behring, Pfizer, and Pharming Group NV. V.A.D. also receives honoraria from Pfizer, and research funding from Takeda. K.K. is an employee and shareholder of Novartis Pharma AG. K.R. is an employee of Novartis Pharma AG. A.R. and J.B. are current employees and stock option holders of Pharming Group NV, and J.B. holds individual stock in NeoClone. The remaining authors declare no competing financial interests.

Correspondence: V. Koneti Rao, Room 10/12C106, 10 Center Drive, Bethesda, MD 20892; e-mail: .