Efficacy, safety, and tolerability of Kedrion 10% IVIG in primary immunodeficiency

Kedrion IVIG 10% is a 10% (weight/volume) IgG (Human) solution for intravenous (IV) injection derived from US plasma donors and produced by Kedrion, S.p.A. at their manufacturing facilities in Bolognana (Lucca), Italy.

The drug product contains approximately 100 mg/mL of human plasma protein, at least 96% of which is IgG. The majority of the IgG protein (>90%) is present in a monomer plus dimer form. IgG molecules may interact forming aggregates and polymers. The product specification limits for polymers and aggregates is ≤3%. The maximum IgA content is 50 μg/mL. Other excipients include glycine (functioning as a stabilizer), sodium chloride, and water for injection. The product contains no preservatives.

Kedrion IVIG 10% has a pH between 4.0 and 5.5, with a mean osmolality of 287 mOsmol/kg (range: 267–296 mOsmol/kg; n = 15 clinical lots).

The Kedrion 10% Immune Globulin (Human) manufacturing process consists of the following steps that have been validated for their ability either to remove or inactivate viral pathogens:

Kedrion introduced an additional purification step (the first is the II+III precipitation), with sodium caprylate precipitation, whose efficacy in precipitation of contaminants and impurities is well known (Parkkinen et al. 2006). Sodium caprylate precipitates α and β globulins, lipids, kallikrein, and other contaminants. The precipitation with sodium caprylate has been shown to remove procoagulant activity and reduces the content of Coagulation Factor XIa by more than 99% (evaluated by thrombin generation assay).

After sodium caprylate contact, the solution is filtered and concentrated. It then undergoes solvent/detergent treatment. The solution is further purified by ion-exchange chromatography using a strong anion exchange resin that binds IgG. The recovered IgG solution is then diafiltered with a solution of saline and glycine and the pH is lowered to 4.0–4.5. It is then subjected to a specific step of viral removal, which is nanofiltration through a 20 nm filter. The nanofiltered solution is finally concentrated to the target protein concentration of 90–110 g/L. The final formulated IgG is bulk sterile filtered and aseptically filled into sterile, depyrogenated glass vials. Lastly, the final product undergoes low-pH incubation (a specific step of viral inactivation) and quarantine.

Kedrion IVIG 10% is derived from a minimum of 1000 plasma donors, thus, is expected to contain a broad spectrum of antibody specificities.

This was a prospective, open-label, single-arm, multi-centre, historically controlled, Phase III study measuring the efficacy, PK, safety, and tolerability of Kedrion 10% IVIG in patients with hypogammaglobulinemia and antibody deficiency. Prior to study enrollment, the diagnosis of PID was confirmed and included a review of infection history, serum IgG concentration prior to IgG therapy, and antibody responses to vaccination with polysaccharide and protein vaccines. The study design required at least 40 evaluable subjects aged 2–70 years, with well-defined PID and documented agammaglobulinemia or hypogammaglobulinemia, to be treated for 12 months to determine the rate of serious bacterial infections, and up to 30 patients were to be assessed for PK of the study product.

Upon consenting and screening, eligible subjects received IV infusions of the study product at the same dose and interval as used for their previous IVIG maintenance therapy. Kedrion IVIG 10% was administered at a dose of 300–900 mg/kg (of body weight) every 21 or 28 days (±4 days; depending on their pre-study IVIG treatment schedule) for a period of 12 months. The dose regimen remained unchanged throughout the study period unless there was a medically justified need to modify it. The total subject follow-up was approximately 14 months, during which subjects underwent 12 months of treatment, 13 treatment visits (subjects on 28-day infusion schedule) or 17 treatment visits (subjects on 21-day infusion schedule), and a follow-up visit 3 or 4 weeks after the last study drug infusion.

During the treatment and follow-up periods, patients underwent the following evaluations: adverse event monitoring and recording of concomitant medications, review of subject diaries, physical exam (prior to each infusion and at the follow-up visit), multiple vital signs assessments during the infusion and immediately after it was completed, and routine laboratory safety tests (hematology, blood chemistry, and urinalysis completed before each infusion and at the follow-up visit). Measurements of direct anti-globulin (DAT/Coombs) were performed before and after the first infusion, before and 24 hours after the fifth infusion, and at the follow-up visit. Viral safety testing was performed several times during the study and at its conclusion. PK assessments on 31 subjects were undertaken before and after the fifth study infusion.

The predefined primary efficacy endpoint was the incidence of acute serious bacterial infections (defined by the US Food and Drug Administration (FDA) as bacterial pneumonia, bacteremia/sepsis, osteomyelitis/septic arthritis, bacterial meningitis or visceral abscess) per subject during the 12-month treatment period (US FDA 2008). Serious adverse events (SAEs) potentially qualifying as primary endpoint events were adjudicated by the study Data and Safety Monitoring Board. Formal statistical analysis (noninferiority test) was applied for the occurrence of acute serious bacterial infection per subject. An upper 99% one-sided confidence limit of <1 per subject per year for the incidence of acute serious bacterial infection was predefined as an indicator of acceptable efficacy. To estimate the annualized event rate and develop the appropriate one-sided 99% upper confidence limit, a generalized linear models SAS procedure for Poisson regression PROC GENMOD was used. This procedure involves estimating a scale parameter of over-dispersion while it permits investigation of covariates, and it also allows for differential follow-up of subjects.

Eligible patients were males or females aged 2–70 years, with PID who required treatment with IVIG. In addition to documented agammaglobulinemia or hypogammaglobulinemia, documented antibody deficiency was required. Subjects had to have received 300–900 mg/kg of a licensed IVIG therapy at 21- or 28-day intervals for at least 3 treatment cycles prior to study enrollment; if receiving subcutaneous immunoglobulin (SCIG), subjects had to switch to IGIV for 3 infusion cycles to establish a steady state prior to receiving Kedrion IVIG 10%. All subjects had to have at least 2 documented IgG trough levels of ≥ 500 mg/dL, obtained at 2 scheduled infusions (21 or 28 days) within 12 months (including one within 6 months) prior to study enrollment. Other immunoglobulin products were prohibited for 1 infusion cycle (21 or 28 days) prior to the first infusion of Kedrion 10% IVIG and until the completion of the first follow-up visit.

To participate in the study, all subjects or parents/guardians (in case of pediatric subjects, defined as 2–17 years of age at study entry) provided written informed consent. In addition, pediatric subjects completed assent forms as appropriate per local regulations.

Subjects receiving blood products other than human albumin or human immunoglobulin within 12 months prior to enrollment and subjects who had participated in another clinical study within 3 weeks prior to enrollment were excluded. The use of corticosteroids (oral or parenteral daily dose of ≥0.15 mg/kg/day of prednisone or equivalent), other immunosuppressive drugs or chemotherapy were prohibited.

Patients were excluded from the study if they had any of the following: secondary immunodeficiency; newly diagnosed, previously untreated PID; dysgammaglobulinemia; isolated IgG subclass deficiency; history of repeated reactions or hypersensitivity to IVIG or other injectable forms of IgG; IgA deficiency with known antibodies to IgA; lifetime history of any thrombotic event; significant protein-losing enteropathy, nephrotic syndrome, or lymphangiectasia; acute infection within 7 days prior to screening; known history of, or having tested positive at enrollment for, human immunodeficiency virus (HIV) type 1, HIV type 2, hepatitis B virus, hepatitis C virus, hepatitis A virus, or parvovirus B19; history of epilepsy or severe migraines, or transient ischemic attack; profound anemia or persistent severe neutropenia or lymphopenia; liver function tests (transaminases) above 2.5 times the upper limit of normal; presence of a severe chronic disease or other condition precluding safe participation in the study. Other gender-specific exclusion criteria included current or planned pregnancy, breastfeeding, or refusal to use adequate contraception for women of childbearing potential.

PK evaluation was performed at or after infusion 5 of the study to wash out residual previous IVIG or SCIG products. Blood samples were taken before the infusion and at: 30 minutes, 2 hours, 24 hours, 3 (± 1) days, 7 (± 1) days, 14 (± 1) days, and either 21 (± 1) days or 28 (± 1) days following infusion. Samples were tested for the concentration of total IgG, IgG subclasses and specific antibody titers against tetanus toxoid, Streptococcus pneumoniae, Hemophilus influenzae type b, and measles. PK calculations related to total IgG and specific IgG antibodies included the following: maximum serum concentration (C_max), minimum serum concentration (C_min), the time to reach maximum concentration (T_max), IgG elimination half-life (T_1/2), the area under the concentration-time curve over one dosing interval (AUC_0-t), the area under the concentration-time curve up to the last measurable concentration (AUC_0-tau), the area under the concentration-time curve to infinity (AUC_0-inf), and terminal elimination rate constant (λ_z). In addition, for total IgG, the following were analyzed: total body clearance (CL), volume of distribution at steady state (weight adjusted) (V_ss), terminal exponential volume of distribution (weight adjusted) (V_z), incremental recovery (IR), and mean residence time (MRT). The degree of fit (R²) of the slope of the regression line had to be ≥0.80 for the reporting of t_1/2, λ_z, and AUC_0-inf, V_ss, V_z, and MRT.

Between November 2012 and July 2013, 49 subjects were screened and 45 were enrolled (39 in the United States and 6 in Canada) to receive Kedrion 10% IVIG for 12 months. Enrollment at the 11 participating sites (9 in the United States and 2 in Canada) was initiated following appropriate Institutional Review Board/Research Ethics Board approval. Prior to enrollment, investigators verified that all patients strictly met inclusion and exclusion criteria. All patients had documented antibody deficiency and carried the diagnosis of agammaglobulinemia, hypogammaglobulinemia, or common variable immunodeficiency (Table 1).

The enrolled population was predominantly Caucasian (91%); 71% were males and 29% females. Nine subjects (20%) were in the pediatric age group (aged 4–17 years at screening) and 36 (80%) were adults, including 9 elderly subjects (Table 1).

Patients were treated prior to the study for a mean of 8.9 years and a median of 8.2 years (Table 2). Dosing intervals were 21 days for 6 patients and 28 days for 39 patients. All 6 patients on the 21-day interval, and 38 of 39 patients on the 28-day interval, completed the study, for a total completion rate of 98%. Historical IgG levels prior to each subject’s initial replacement therapy were abnormal with a mean of 340 ± 218 mg/dL and a median of 363 mg/dL (Table 2). IgG trough levels taken at least twice during the 12 months prior to study entry (during prior replacement therapy) exceeded 500 mg/dL in all subjects. Mean IgG trough levels prior to study initiation were 923.9 mg/dL with a median of 921mg/dL during commercial IVIG treatment, with a mean and median IVIG dose of 501.7 mg/kg and 466 mg/kg, respectively. Interestingly, most subjects (25% or 56% of the total population) received a dose ≤500 mg/kg per infusion interval. During the study, the mean dose administered remained stable, 513.3 ± 146.4 mg/kg (median 506.9 mg/kg) at study entry and 526.2 ± 171.6 mg/kg (median 519.0 mg/kg) at study end.

Only 2 acute serious bacterial infections occurred during the study (in 2 adult patients, both with episodes of bacterial pneumonia), yielding an overall rate of acute serious bacterial infections of 0.048 per person per year, with an upper limit of the 99% one-sided confidence interval (CI) of 0.11 (generalized linear models procedure for Poisson regression) (Table 3), demonstrating the efficacy of Kedrion 10% IVIG. No acute serious bacterial infections occurred in the pediatric subgroup during the 12-month treatment period.

Efficacy was also demonstrated using secondary endpoints during the 1-year treatment period (Table 4) including:

The majority of the enrolled subjects (78%) experienced at least 1 infection other than acute severe bacterial infections (referred to as “other infections” in subsequent text), with 5 subjects (11%) experiencing 1, 9 subjects (20%) each experiencing 2 or 3, and 12 subjects (27%) experiencing 4 or more other infections (Table 5). See Table 6 for categories of infections seen and the percentage for each. The overall incidence of other infection episodes was 2.884 per patient, per year.

Overall, of the 602 infusions, 97 (16%) were temporally associated with an AE (defined as AEs occurring during or within 72 hours after an infusion of the study drug) (Table 7). The upper bound for the 95% CI for the frequency of infusions temporally associated with an AE was 20.4%. Therefore the primary safety objective of demonstrating that less than 40% of infusions had 1 or more infusion-related AE was met.

Less than one-third (14 subjects) of the total population experienced treatment-related AEs (53 events). Of those, 7 episodes in 4 subjects were assessed as “certainly related”, 5 episodes in 3 individuals as “probably related” and 41 episodes in 11 subjects as “possibly related” with the study drug (Table 8).

A total of 8 SAEs occurred in 4 subjects during the study period. None were assessed as related to the study treatment and none resulted in withdrawal from the study. No clinically relevant changes from baseline were detected in mean hematology or blood chemistry parameters at the end of the treatment period or at the last follow-up.

The PK population included 31 subjects, of whom 6 were on a 21-day schedule and 25 were on a 28-day schedule. PK parameters for total IgG are shown in Table 9. Mean C_max was 1980 ± 279 mg/dL for 21-day subjects, 1950 ± 357 mg/dL for 28-day subjects, and 1960 ± 339 mg/dL for all PK subjects. Mean AUC_0-t was 26 300 ± 4790 day*mg/dL for 21-day subjects, 33 500 ± 5020 day*mg/dL for 28-day subjects, and 32 100 ± 5690 day*mg/dL for all PK subjects. The median T_max was 0.16 days (range, 0.14–1.07 days) and 0.13 days (range, 0.07–1.06 days) for the 21-day or 28-day cycles, respectively. The calculated serum half-life (T_1/2, mean ± SD) was 22.5 ± 4.5 days for the 21-day subjects, 37.1 ± 11.1 days for the 28-day subjects, and 35 ± 11.6 days for all PK subjects. Their T_1/2 values corresponds to CL rates of 2.09 ± 0.45 mL/day/kg (21-day schedule), 1.55 ± 0.44 mL/day/kg (28-day schedule), and 1.66 ± 0.49 mL/day/kg (all PK subjects). Mean V_ss was 0.75 ± 0.23 and 0.79 ± 0.22 dL/kg for 21-day and 28-day subjects, respectfully. Mean apparent MRT was longer for 28-day subjects compared with 21-day subjects (52.4 ± 15.4 vs 32.3 ± 6.29 days, respectively).

The mean T_1/2 values for IgG subclasses and specific antibodies were generally the same as for total IgG, and demonstrated that the IgG subclasses and specific antibodies were not differentially affected by the manufacturing process.

Mean trough IgG levels at baseline (immediately prior to the first study IVIG infusion) were 976 ± 253 mg/dL and 917 ± 195 mg/dL for 21-day and 28-day subjects, respectively (Figures 1 and 2). Mean ± SD trough IgG levels were 1037 ± 282 mg/dL in 21-day subjects, and 865 ± 174 mg/dL in 28-day subjects after steady state was achieved, and 846 ± 230 mg/dL and 844 ± 163 mg/dL in the 2 subsets, respectively, at the follow-up visit. Five patients had trough IgG levels below 500 mg/dL transiently during the study period. In 2 of the 5 patients, this was a one-time recording and levels returned to above 500 mg/dL with no dose adjustments. In the remaining patients, low levels were recorded more than once, and in 2 subjects, there was a dose increase. In one patient this could be explained by the great swings in trough levels recorded during, as well as prior to enrollment, while on a stable dose of a commercial product (not shown). The second patient had 2 consecutive infections that required hospitalization (Table 10) and the third patient missed their infusion schedules by a week or more (Table 11). These cases illustrate the importance of recording and consistently adjusting doses before, during, and after serious infections in patients who are not compliant or have unstable IgG trough levels.