1 INTRODUCTION

Seasonal influenza, an acute respiratory infection, leads to 3–5 million cases of severe illness and 290,000–650,000 respiratory deaths globally each year.1 Influenza pandemics occur in waves, adding to the global burden. Adamantanes and neuraminidase inhibitors are two classes of antiviral medicines approved in most countries.2, 3 Baloxavir marboxil, a selective inhibitor of influenza cap-dependent endonuclease, is approved in the United States, EU, and Japan.4, 5 Neuraminidase inhibitors (oseltamivir, zanamivir, and peramivir) have antiviral activity against all influenza A and B subtypes and are effective in the treatment and prophylaxis of influenza.6, 7 Compared with oseltamivir and peramivir, zanamivir displays efficacy against the most common influenza A virus resistance mutation, H275Y.8, 9 In 2009, following the emergence of the influenza A/H1N1pmd09 global pandemic, GSK initiated a clinical development program for an aqueous formulation of zanamivir for intravenous (IV) administration at the request of the Food and Drug Administration (FDA) in the United States. In parallel, a global Compassionate Use Program (CUP) supplied zanamivir aqueous solution through an Emergency Investigational New Drug or named-patient process in the United States and local regulations including under Article 83 in parts of Europe, for IV or nebulized use to seriously ill patients with influenza infection for whom approved anti-influenza drugs were not effective or feasible.

Zanamivir powder for inhalation (RELENZA, zanamivir, GSK, United Kingdom) has been approved for >20 years for the treatment and prophylaxis of influenza A and B in adults and children ≥5 years of age in over 50 countries, including the EU, Japan, and the United States.10, 11 The clinical development program investigated IV zanamivir in adult and pediatric patients (between 6 months and 18 years of age) hospitalized with influenza.12-14 In 2019, GSK received marketing authorization for IV zanamivir in Europe for the treatment of complicated and potentially life-threatening influenza A or B virus infection in adult and pediatric (≥6 months of age) patients for influenza virus known or suspected to be resistant to anti-influenza medicinal products other than zanamivir, and/or when other anti-viral medicinal products for treatment of influenza, including inhaled zanamivir, are not suitable for the individual patient.10 The CUP was terminated following this marketing authorization.

Patient characteristics, zanamivir dosing information and safety data, including serious adverse events (SAEs), were collected from 2009 to 2019. This current report presents analysis of all available CUP data on zanamivir aqueous solution received up to January 31, 2020.

2 METHODS 2.1 CUP design and initiation

The CUP was administered via applicable country regulatory requirements, with the requesting physician considered the designated sponsor. A Physician Guidance Document was developed to support requests for IV and nebulized zanamivir on a compassionate use named-patient basis and provide guidance on dosing, administration, and patient eligibility. Similarly, the European Medicines Agency published a “Condition of Use” document to give guidance for IV zanamivir requests via Article 83. The CUP was referenced through governmental information sources, including public health influenza advice websites.15 The CUP process is summarized in Figure S1 and in Methods S1.

2.2 Dosing and administration

Recommendations for IV administration of zanamivir specified in the “Guidance for physicians” was 600 mg in a 30-min infusion, every 12 h, adjusted for age, weight (for pediatric patients), renal function, and renal replacement modality. Nebulized zanamivir was recommended at a dose of 25 mg, four times/day. The recommended duration of initial treatment was 5 days (Methods S1).

2.3 Eligibility criteria

Patients were eligible if, hospitalized and severely ill with influenza infection and not responding to authorized antiviral medicinal products (e.g., oral oseltamivir or inhaled zanamivir), or for whom drug delivery by a route other than IV was not expected to be dependable or feasible, or if there was resistance to other antiviral agents and inhaled zanamivir was not suitable.16

2.4 Data collection

Baseline and follow-up data were summarized across three databases (Table 1). The master summary tracking sheet (MSTS) provided information on patient demographics and overall medical condition, with a data cut-off of May 6, 2019. All SAEs were required to be reported to local ethics committees and regulatory authorities per regulatory agency requirements and were recorded in the GSK safety database. Data cut-off was extended to January 31, 2020, (beyond termination of IV zanamivir supplies in the CUP) to allow adequate time for additional safety reports and/or follow-up information.

TABLE 1. Patient demographic characteristics by data source (MSTS, CRF, and GSK safety database) MSTS N = 4033 CRF N = 877 GSK safety database N = 466 Agea n = 4014 n = 817 n = 422 Mean (SE) 47.3 (0.32) 44.5 (0.70) 44.3 (0.96) Min–max. 0–98 0–100 0–90 Age categorya, n (%) — — — 0 to <6 months 9 (<1) 9 (1) 5 (1) 0 to ≤6 monthsb 20 (<1) — — 6 months to ≤1 year 82 (2) 15 (2) 4 (<1) >1 to ≤2 years 42 (1) 7 (<1) 7 (2) >2 to ≤5 years 84 (2) 28 (3) 13 (3) >5 to ≤12 years 105 (3) 21 (2) 12 (3) >12 to ≤17 years 84 (2) 21 (2) 13 (3) 18 to ≤64 years 2860 (71) 600 (68) 312 (67) 65 to ≤74 years 480 (12) 82 (9) 41 (9) 75 to ≤84 years 206 (5) 32 (4) 13 (3) 85 to ≤94 years 57 (1) 1 (<1) 2 (<1) ≥95 years 5 (<1) 1 (<1) 0 Missing/unknown 19 (<1) 62 (7) 44 (9) Sex, n (%) n = 4033 n = 879 n = 466 Female 1653 (41) 355 (40) 205 (44) Male 2335 (58) 498 (57) 251 (54) Missing/unknown 45 (1) 26 (3) 10 (2) Pregnancy, n (%) 60 (4) 16 (5) 9 (4) Ethnicity, n (%) — — — Hispanic or Latino — 56 (6) — Not Hispanic or Latino — 357 (41) — Missing — 466 (53) — Geographic ancestry, n (%) — n = 879 — African American/African heritage — 35 (4) — American Indian or Alaskan native — 5 (<1) — Asian—Central/South Asian — 7 (<1) — Asian—East Asian Heritage — 9 (1) — Asian—South East Asian Heritage — 19 (2) — White—Arabic/North African Heritage — 6 (<1) — White—White/Caucasian/European Heritage — 302 (34) — Missing — 496 (56) — Abbreviations: CRF, case report form; GSK, GlaxoSmithKline; MSTS, master summary tracking sheet; SE, standard error.

A case report form (CRF) was provided to collect information on medical conditions, zanamivir dosing, treatment duration, and clinical outcomes (data cut-off: January 31, 2020). Completion and return of this form to the sponsor was encouraged but not mandatory for supply of aqueous zanamivir. Modifications made to the CRF template over the CUP minimized the collection of unusable data (Table S1).

The information collected in each data source is summarized in Table S2, with additional information regarding the databases in Methods S1.

2.5 AE reporting

Adverse events (AEs) meeting the definition of a SAE in patients who received ≥1 dose of zanamivir from the time of the first dose until 14 days after treatment completion were required to be reported. Any SAEs reported after this time were included in the GSK safety data set. SAEs were summarized separately for the pediatric population. Further description of AE reporting is provided in Methods S1.

2.6 Statistical methods

Data from the three databases (MSTS, CRF, and GSK safety) were summarized depending on data type. Frequency and percentages were reported for categorical outcomes such as AEs, SAEs, and chronic underlying illness. Mean (standard error) was reported for continuous outcomes such as age and ventilation data. A separate examination of the MSTS was performed to summarize all available information about infants and neonates aged ≤6 months.

2.7 Role of the funding source

GSK contributed to the design and collection, analysis, and interpretation of data. Authors employed by GSK participated in the writing, review, and approval of the manuscript. All authors had access to the data and approved the final manuscript for submission.

3 RESULTS 3.1 CUP global enrollment and completion of CRF

At data cut-off (May 6, 2019), requests for zanamivir treatment were received for 4,033 patients globally, with most requests from Europe (n = 3,051) followed by North America (n = 713) (Table 2). Treating physicians returned CRFs for 877 patients. The highest CRF completion rates were in Europe (n = 428) and North America (n = 379), which included 184 and 356 patients from the United Kingdom and the United States, respectively.

TABLE 2. Summary of requests for CUP enrollment by continent and country from MSTS and CRFs MSTS, n CRFs, n Asia 136 29 China 5 0 Hong Kong 95 14 Korea 1 0 Singapore 27 11 Thailand 8 4 Australia 103 32 Europe 3,051 428 Austria 12 2 Belgium 7 0 Cyprus 7 1 Denmark 67 14 Estonia 2 2 France 135 26 Germany 253 81 Gibraltar 1 1 Greece 184 30 Ireland 72 7 Italy 35 12 Latvia 5 5 Lithuania 2 1 Netherlands 63 5 Norway 17 2 Poland 1 0 Portugal 38 9 Romania 1 0 Slovakia 1 0 Spain 121 13 Sweden 1 0 Switzerland 76 33 United Kingdom 1,950 184 Middle East 27 9 Bahrain 1 0 Israel 16 6 Oman 1 1 Saudi Arabia 1 0 United Arab Emirates 8 2 North America 713 379 Canada 81 23 United States 632 356 South America 3 0 Argentina 1 0 Brazil 2 0 Abbreviations: CRF, case report form; CUP, Compassionate Use Program; MSTS, master summary tracking sheet. 3.2 Baseline characteristics

The 4,033 patients included in the CUP based on MSTS data had a mean age of 47.3 years (range: 0–98) (Table 1); over 50% were aged 40–69 years (Figure 1), with a slightly higher proportion of males (2,335 [58%] patients). Among 1,653 female patients, 60 (4%) were pregnant at the time of zanamivir request (Table 1). Based on the standalone evaluation of the MSTS for infants and neonates, at least 20 patients were aged ≤6 months, of whom 12 were born prematurely with a gestational age of 23–35 weeks.17 The 877 patients (including 16 pregnant women) in the CRF database had a mean age of 44.5 years, with more than three quarters (n = 681 [77%]) having chronic underlying illnesses or risk factors, the most common being respiratory illness recorded in 320 (36%) patients (Table 3).

Age distribution for patients enrolled on CUP from the MSTS. *Does not include standalone analysis of infants and neonates aged ≤6 months. CUP, Compassionate Use Program; MSTS, master summary tracking sheet

TABLE 3. Summary of chronic underlying illness and risk factors in ≥1% of patients receiving zanamivir on the CUP registered in the CRF database Chronic underlying illness/risk factor, n (%) CRF (N = 877) Any illness or risk factor 681 (77) Respiratory 320 (36) Tobacco use 190 (22) Chronic obstructive pulmonary disease 76 (9) Asthma 71 (8) Chronic lung disease 54 (6) Chronic supplemental oxygen 15 (2) Rheumatology and immunology 295 (34) Any immunocompromisea 168 (19) Leukemia/lymphoma 129 (15) Organ/bone marrow transplantation 76 (9) Rheumatoid arthritis 19 (2) HIV/AIDS 16 (2) Vasculitis 14 (2) Gastrointestinal disease 155 (18) Morbid obesity 96 (11) Malnutrition 27 (3) Cirrhosis/chronic liver disease 23 (3) Crohn's disease/inflammatory bowel disease 14 (2) Endocrine disease 137 (16) Diabetes mellitus 132 (15) Oncology 128 (15) Current cancer or cancer treatment with 1 year 128 (15) Cardiovascular 126 (14) Coronary artery disease 67 (8) Arrythmia 40 (5) Congestive heart failure 31 (4) Cardiomyopathy 20 (2) Renal disease 78 (9) Chronic renal insufficiency 64 (7) End stage renal disease: hemodialysis 17 (2) Neurology 43 (5) Seizure disorder 22 (3) Stroke/cerebral vascular disease 17 (2) Newborn prematurity 10 (1) Abbreviations: AIDS, acquired immune deficiency syndrome; CRF, case report form; CUP, Compassionate Use Program; HIV, human immunodeficiency virus. 3.3 Zanamivir formulation, dosages, and dosing duration

A summary of zanamivir treatment captured in the MSTS and CRF databases is provided in Table 4. The majority (≥95%) of patients received zanamivir via the IV route. Based on MSTS data, an additional treatment course (beyond the initial 5-day course) was requested for nearly 25% of patients (range: 1–75 days).

TABLE 4. Summary of zanamivir treatment for all patients registered to the MSTS and CRF databases MSTS (N = 4,033) CRF (N = 877) Route of administration, n (%) n = 4,033 n = 877 IV 3,879 (96) 823 (94) Nebulized 144 (4) 33 (4) Both IV and nebulized 9 (<1) 11 (1) Missing 1 (<1) 10 (1) Treatment duration, days — n = 820 Mean (SE) — 6.9 (0.17) Duration of dose, n (%) n = 4,033 — 5 days 3,072 (76) — ˃5 days 961 (24) — Additional courses of treatment, n (%)a n = 961 — 1 (5-day initial course + 5 days) 767 (80) — 2 (5-day initial course + 10 days) 147 (15) — ≥3 (5-day initial course + ≥ 15 days) 47 (5) — Dose adjustment due to change in renal function, n (%) — n = 877 Yes — 231 (26) No — 341 (39) Missing — 305 (35) Zanamivir treatment stopped prematurely, n (%) — n = 877 Yes — 155 (18) No — 547 (62) Missing — 175 (20) Reason for stopping zanamivir prematurely, n (%) — n = 155 AE — 57 (6) Death — 3 (<1) Treating physician's discretion — 64 (7) Decision by patient or proxy — 10 (1) Other — 9 (1) Missing — 12 (1) Abbreviations: AE, adverse event; CRF, case report form; IV, intravenous; MSTS, master summary tracking sheet; SE, standard error. 3.4 Clinical outcomes recorded in the CRFs

Among the 877 patients in the CRF database, 30% (n = 267) recovered or their condition resolved, 36% (n = 320) had not recovered or their condition was not resolved at the time of CRF completion, 26% (n = 231) died, <1% (n = 1) were recovering/resolving, and no outcome was recorded for 7% (n = 60) patients. Three patients <6 months of age reported SAEs (pneumonitis, cardiopulmonary failure, and lung disorder) with a fatal outcome.

3.5 Serious adverse events

Overall, 466 patients reported ≥1 SAE based on data included in the GSK safety database with