Study design of a prospective, open-label study comparing increased infusion parameters of a 20% subcutaneous immunoglobulin (IgPro20) in patients with primary immunodeficiency

Background: Immunoglobulin G (IgG) replacement therapy can be administered subcutaneously (SCIG) using an infusion pump or by manual push. Both methods have shown similar serum IgG trough levels for the same total weekly/monthly dose as well as safety and tolerability profiles. The currently approved infusion parameters for the SCIG Hizentra® (IgPro20, in primary immunodeficiency (PID) in the United States are volumes of ≤25 mL and flow rates of ≤25 mL/h per injection site. Previously, clinical studies and case reports had demonstrated use of higher infusion parameters than those approved, but safety and tolerability of these have not been systematically evaluated. In the absence of regulatory guidance, we have developed a novel prospective clinical study applying forced upward titration design to evaluate the safety and tolerability of high infusion parameters of IgPro20 in PID patients using pump-assisted and manual push techniques (NCT03033745). Results: A total of 45 patients were planned for inclusion. Primary endpoints were defined as the proportion of patients successfully infusing certain infusion parameters (responder rate). The study included three cohorts (n=15 planned per group): 1) Pump-assisted Volume Cohort with weekly infusions at volumes of 25, 40 and 50 mL per injection site; 2) Pump-assisted Flow Rate Cohort with weekly infusions at flow rates of 25, 50, 75 and 100 mL/h per injection site, and 3) Manual Push Flow Rate Cohort with 2 to 7 infusions at flow rates 30, 60 and 120 mL/h per injection site. Each infusion parameter level was tested for 4 weeks, after which those who successfully infused at the current level (responders) were switched to the next level. Responder rate, safety and tolerability were assessed. Study results will be available in early 2020. Conclusions: This study applied a novel and rigorous prospective evaluation of individual safety and tolerability levels of pump-assisted and manual push SCIG at higher infusion

parameters than currently approved in the United States and demonstrated their safety and tolerability.

Background
Primary immunodeficiency (PID) is primarily treated with immunoglobulin G (IgG) replacement therapy, administered either intravenously (IVIG) or subcutaneously (SCIG) (1)(2)(3)(4). In clinical practice, infusion pumps are used for SCIG administration, usually delivering up to 25 mL of IgG into the injection site, with weekly infusions taking up to 2 hours depending on individual dose and flow rate (1,5). Although clinical experience has demonstrated that larger volumes and flow rates can be successfully administered using an infusion pump (3), safety and tolerability of higher infusion parameters have not been tested in a rigorous clinical study environment using forced upward titration design. An alternative to the infusion pump is the manual push technique, which offers selfadministration by the patient in smaller, more frequent doses using a syringe and butterfly needle (5,6). Manual push infusion is reported to be effective; have an acceptable safety profile; and provide treatment flexibility, independence, and satisfaction (7). Infusion times are much shorter (5 to 20 minutes) with manual push technique and fewer injection sites are used as compared with weekly pump-assisted infusions (5,8,9). Both techniques, manual push and pump-assisted infusion, have previously shown similar serum IgG levels for the same dose and overall safety and tolerability profiles, although the comparative incidence of adverse events (AEs) varies between studies (5,(9)(10)(11). Therefore, the choice of SCIG administration using either pump-assisted or manual push infusion can be tailored according to the individual patient's profile and preferences (6,12). IgPro20 (Hizentra®, CSL Behring, King of Prussia, PA, USA) is a ready-to-use formulation of polyvalent SCIG (highly purified [≥98% purity] IgG) approved for the treatment of PID globally (13). Currently, a volume of up to 25 mL per injection site and a flow rate of up to 25 mL/h per injection site are approved for IgPro20 in the United States for the treatment of PID (14). However, higher infusion volumes (60 mL per injection site) and flow rates (>60 mL/h per injection site) of 20% SCIG products have been demonstrated to be well tolerated in both pediatric and adult patients, as well as in obese adults (9)(10)(11)15). Higher infusion paremeters of SCIG products are important because, in general, patients on SCIG replacement therapy tend to prefer shorter infusion times (achievable with higher flow rates) and fewer needle sticks per infusion (requiring higher volume per individual injection site) (16,17).
In order to fully characterize the current landscape regarding clinical studies conducted with pump-assisted and manual push infusion, a systematic search of the currently published literature was performed. The PubMed and EMBASE databases were searched from January 2009 until April 2019 for full text publications reporting actual treated patients with PID, using the following search terms: "subcutaneous immunoglobulin" or "subcutaneous immune globulin" or "SCIG" or "IGSC "manual administration" or "rapid push" or "manual" and "primary immunodeficiency" or "primary immune deficiency." In total, 15 studies were identified (3, 5, 6, 9-12, 15, 18-24), of which 8 studies met the inclusion criteria ( Figure 1) (5, 6, 9-12, 15, 20). Among the studies included in the analysis, a range of SCIG therapies were administered to patients with PID using manual push or pump-assisted techniques (5, 6, 9-12, 15, 20). The duration of the treatment varied across the studies, ranging from 28 to 45.5 months. Mean IgG levels reported were similar between push and pump techniques and were also similar across the studies (5, 6, 9-12, 15, 20). The comparative rate of infusion site reactions reported in the literature for both manual push and pump-assisted SCIG administration vary (5,10,20). This is most likely due to differences in study design, scales, timepoints, and report methodology.
Across the identified studies, the average flow rates using a manual push technique ranged from 12 to 60 mL/h per injection site (1,9,10,20) and for pump-assisted techniques at a volume of up to 50 mL per injection site (6) and an infusion flow rate of up to 20 to 50 mL/h per injection site (5,(25)(26)(27). The studies reported no major safety concerns, as most AEs were local reactions mostly of mild nature. Although these clinical studies have demonstrated the safety and effectiveness of both the frequent use of higher infusion parameters than currently approved for IgPro20 in PID in the United States and the two infusion techniques (pump-assisted or manual push) in clinical practice, safety and tolerability of these has not been evaluated in a systematic manner in a prospective clinical trial where every patient is required to test their individual comfort limits regarding flow rate and/or volume per injection site.
Of note, guidance from several major regulatory agencies is available for standard registration studies evaluating the efficacy, safety, and pharmacokinetics of IgG products in PID (28)(29)(30). However, such guidance and a standardized approach are lacking for studies evaluating infusion parameters of IgG products that are higher than those approved for existing IgG products. The present study was specifically designed to obtain safety and tolerability data of infusion parameters of IgPro20 higher than those currently approved in patients with PID for both pump-assisted and manual push techniques for regulatory purposes.

Results
As this study is ongoing, full results will be presented in the future.

Discussion
The present study aimed to address the gap in the existing landscape of standardized and/or recommended by regulatory guidance documents study designs that evaluate the safety and tolerability of IgG infusions in IgG replacement therapy. Namely, we have developed a trial of high infusion parameters testing weekly doses of IgPro20 of up to 100 mL per week, compared with an average United States dose of 50 mL (10 grams) per week. IgG replacement therapy (administered as either IVIG or SCIG) has become the cornerstone of treatment for PID, and as an immunomodulatory agent for an increasing number of autoimmune and inflammatory disorders (1-4). As such, there are wellcharacterized regulatory frameworks for evaluating efficacy, safety, or pharmacokinetics of these products, including specification that their evaluation in defined subpopulations such as pediatric, geriatric, and/or obese patients is an essential requirement (28)(29)(30).
These recommendations facilitate the design of studies investigating these products.
However, there are currently no such requirements or guidance on the investigation of various infusion parameters, such as volume and flow rate per injection site, or for various infusion techniques, such as pump-assisted or manual push administration. Lack of any regulatory guidance or data from other studies with forced upward titration design on this topic is a limitation to further improve the convenience of administration of SCIG therapies.
In pivotal (registration) IgG clinical studies, patients are allowed to test flow rate and volume parameters based on typical clinical practice within the range specified by study protocol and at levels generally considered tolerable to patients. However, these studies do not routinely investigate infusion parameters that are higher than those used conventionally, which may reduce the time needed for injection and the number of injection sites. The latter two characteristics reflect patients' preferences on SCIG and lead to a better health-related quality of life for patients with PID (16,17). A high, stable health-related quality of life is important both as a separate goal, especially in PID patients who need life-long IgG replacement therapy, and as a means of increasing patients' treatment compliance (31).
Another key question requiring discussion is how to define the minimum proportion of study patients achieving and tolerating a certain infusion parameter that would be acceptable for regulatory agencies to include the parameter as a generally recommended infusion regimen within the product labeling. This may vary by geographic region. For example, the European Union often defines the highest limits for injection volume or rate as that which is "individually tolerated," whereas United States prescribing information includes specific definitions for minimal and maximal injection volumes and flow rates, supported by clinical research. This is further complicated when the initial infusion parameter recommendations need to be updated based on the clinical evidences which shows possibility of alteration of the infusion parameters.
During interactions with the United states Food and Drug Administration regarding updates to the recommendations for IgPro20 infusion regimens, it became clear that the usual design of extension studies, which repeat pivotal study design over a longer study duration to collect long-term safety and efficacy data would not be a satisfactory method for investigating acceptability of higher infusion parameters than approved. To provide data qualifying for a label update, the study design instead required the application of a forced upward titration for every study participant. The range of volumes and flow rates in our study was selected based on extensive information from clinical studies and almost 8 years of market use of IgPro20, as well as other 20% SCIG products.
In addition to the investigation of infusion parameters higher than those currently approved, this is also the first clinical study of IgPro20 infusion using the manual push technique, which will evaluate several features of this technique in comparison with pumpassisted administration.
As a separate topic, there are currently very few infusion pumps for subcutaneous IgG products on the market that allow delivery of high volumes (more than 50-60 mL) at high flow rates (>100 mL/h). Future development of SCIG infusion techniques, especially with increasing use of IgG therapy for autoimmune conditions where IgG doses are significantly higher than in immune replacement, should be facilitated by new pump models that can deliver larger volumes at faster rates.

Conclusion
In conclusion, this study is the first prospective study that applied a new rigorous forced upward titration study design to evaluate individual safety and tolerability levels of pumpassisted and manual push SCIG infusion parameters. Results of this study may be used to seek regulatory approval for higher infusion SCIG parameters and new administration methods.

S t u d y d e s i g n
This was a multicenter, open-label, parallel-arm, non-randomized, study applying forced upward titration design to evaluate safety and tolerability of infusion parameters of IgPro20 at levels higher than those currently approved in patients with PID (NCT03033745). Patients were enrolled in one of the following three cohorts: Pumpassisted Volume Cohort, Pump-assisted Flow Rate Cohort and Manual Push Flow Rate Cohort. Each included patient meeting cohort-specific inclusion criteria, detailed below.
The study contained a screening period of up to 28 days (4 weeks) followed by an active

Objectives and endpoints
The primary objective of the study was to determine responder rates under escalating infusion parameters. In the pump-assisted cohorts, a responder was defined as a patient who administered ≥3 valid infusions at a given infusion parameter level. In the Manual Push Flow Rate Cohort, a responder was defined as a patient who administered the minimum prespecified number of valid infusions for a given infusion parameter level, as specified in Table 1. In pump-assited infusions, these constitute a volume two-fold higher and a flow rate fourfold higher than currently approved for IgPro20 in PID in the United States. The manual push technique is not currently approved. Secondary objectives were to evaluate the safety and tolerability of pump-assisted and manual push IgPro20 infusions at higher infusion parameters (volume and flow rate) than those currently approved. The assessment of serum IgG concentrations under various IgPro20 infusion parameters was an exploratory objective of the study.
An infusion was considered valid if ≥95% of the prescribed dose was used without any interruptions or decrease of the infusion parameter under test (including possible interruption for technical reasons) and where no prohibited concomitant medication was administered ( Table 2).  The actual infusion volume is ≥95% of the planned infusion volume for that infusion day Flow rate The actual infusion duration is not more than 1 minute or 10% (whichever is larger) longer than the calculated infusion duration based on the desired flow rate (ie, 0.5, 1, or 2 mL/min) and the actual volume infused eDiary, electronic diary The rate of total treatment-emergent adverse events (TEAEs) and local reactions per patient and per infusion as well as the severity, duration, and time of onset of local reactions, including tolerability of infusion parameters, ie, percentage of infusions without severe local reactions, were the secondary endpoints. Patients with hypersensitivity to IgPro20, ongoing serious bacterial infections at the time of screening or with other significant medical conditions, and who received therapy or were under drug or medication abuse ≤1 year before the study were excluded from the study. The patients were assigned to a cohort based on investigators discretion and not randomized.

Availability of Data and Materials
The study protocol and results are available on clinicaltrials.gov  Individual patient data sharing statement.docx