Safety and Pharmacokinetics of a Four Monoclonal Antibody Combination against Botulinum C and D Neurotoxins

Botulism is caused by botulinum neurotoxin (BoNT), the most poisonous substance known. BoNTs are also classified as tier 1 biothreat agents due to their high potency and lethality. The existence of seven BoNT serotypes (A to G), which differ by 35% to 68% in amino acid sequences, necessitates the development of serotype-specific countermeasures.

immunized with pentavalent botulinum toxoid (C. Garcia-Rodriquez et al, unpublished data). Each MAb binds a nonoverlapping BoNT epitope with high affinity ( Table 1). Three of the MAbs bind all four BoNTs with an equilibrium dissociation constant (K D ) of Ͻ10 Ϫ9 M, while MAb XCD-b only binds BoNT/C and BoNT D/C. A combination of the four parental MAbs from which MAbs XCD-a, XCD-b, XCD-c, and XCD-d were derived had effective doses that protected 50% of mice (ED 50 s) challenged with 40,000 50% lethal doses (LD 50 s) of BoNT/C of 5.0 g/mouse, of BoNT C/D of 7.5 g/mouse, and of BoNT D/C of 7.5 g/mouse (C. Garcia-Rodriquez et al, unpublished data).
Subject demographic characteristics. The use of 24 research participants was planned for these studies (Fig. 1). Twenty-five were enrolled, randomized, and included in the study. Randomized participants were administered either the NTM-1634 or placebo intravenously. No deaths or adverse events (AEs) leading to study discontinuation were reported. A summary of subject disposition is presented in Table 2. Two subjects from cohort B (0.66 mg/kg) were lost to follow-up. One subject who received placebo was lost to follow-up after day 2 and the other was lost to follow-up after day 57. The subject that was lost to follow-up after day 2 was replaced in cohort B at the time of cohort C (Fig. 1). A total of 18 research participants, six per cohort, received NTM-1634 (0.33 mg/kg, 0.66 mg/kg, or 1 mg/kg), and two research participants per cohort, six in total, received placebo. The doses were chosen to (i) deliver a dose of BoNT antitoxin that exceeded the current dose of BAT, (ii) achieve a serum concentration greater than the MAb K D for BoNT, and (iii) allow measurement of the concentration of each component MAb. For example, for a 70-kg subject, these doses would provide a total neutralizing capacity of 1.85 ϫ 10 8 , 3.70 ϫ 10 8 , and 5.6 ϫ 10 8 mouse  (19 [74%]). The demographic characteristics were similar across the treatment groups except cohort C, where all the participants who received active compound were female. The demographic characteristics are summarized in Table 3.
Safety profile. No deaths or AEs leading to study discontinuation were reported. Overall, two non-drug-related serious adverse events (SAEs) were reported. One subject in cohort B (0.66 mg/kg) had an SAE of exacerbation of schizophrenia and one subject in cohort C (1 mg/kg) had and SAE of wrist fracture. Treatment-emergent adverse events (TEAEs) were reported in 19 of 25 subjects (76%), with a total of 69 TEAEs reported over the course of the study (Table 4). Cohort A (0.33 mg/kg) had four of six (66.7%) reporting 18 TEAEs. Cohort B (0.66 mg/kg) had five of six (83.3%) reporting 21 TEAEs. Cohort C (1 mg/kg) had six of six (100%) subjects reporting 20 TEAEs. The placebo group had four of seven (57.1%) subjects reporting 10 TEAEs. The most frequently reported TEAE was blood creatine phosphokinase increase, which occurred in 5 (20%) of the subjects. The next most frequently reported TEAEs (reported in two or more subjects overall) were hemoglobin decreases, hematuria, hematocrit decreases, viral upper respiratory tract infection, white blood cell count decreases, blood calcium decreases, blood potassium increases, blood sodium increases, cough, hypernatremia, myalgia, neutrophil count decreases, and proteinuria. There were no clinically significant abnormal electrocardiograms (ECGs) nor were there clinically meaningful trends identified comparing baseline vital and ECG values to those at subsequent time points. All TEAEs were considered by the investigator to be "not related" to the study drug. Antidrug antibody (ADA) results were negative for all participants with the exception of one subject who had a positive ADA result with a titer of 1.2 ng/ml for MAb XCD-b   predose on day 1. All postdose ADA results were negative for this participant. The significance of this finding is unknown, but it may be due to the setting of the cut point for ADA detection. The subject did not report any past medical history and experienced no AEs during the study. Pharmacokinetics analysis. A summary of the pharmacokinetics data is presented in Table 5. Peak concentrations for each of the antibodies, regardless of dose, were generally observed 1 to 2 h after the 1-h infusion. The peak concentrations of each of the four antibodies were also similar. After the peak, the concentrations of all four antibodies declined in a log-linear fashion, with a distinct distribution and terminal elimination phase (Fig. 2). Serum concentrations for XCD-a were quantifiable up to day

DISCUSSION
BoNTs are classified by the Centers for Disease Control and Prevention (CDC) as one of the highest-risk threat agents for bioterrorism due to their extreme potency and lethality, ease of production and transport, and need for prolonged intensive care (3). Thus, the development of countermeasures for all seven serotypes, including BoNT/C and BoNT/D, is a high research priority. Both Iraq and the former Soviet Union produced BoNT for use as weapons, and at least three additional countries (Iran, North Korea, and Syria) have developed or are believed to be developing BoNT as instruments of mass destruction (3,28,29). Iraq produced 19,000 liters of concentrated BoNT, more than any other biothreat agent, of which 10,000 liters were weaponized in missile warheads or bombs. The 19,000 liters represent an amount of toxin capable of killing the world's population three times over. The Japanese cult Aum Shinrikyo attempted to use BoNT for bioterrorism by dispersing toxin aerosols at multiple sites in Tokyo (3).
Botulism causes significant morbidity and mortality, and exposure of even a small number of civilians would paralyze the health care delivery system of any metropolitan area. Treatment of botulism requires prolonged hospitalization in an intensive care unit (ICU) and mechanical ventilation for up to 6 weeks. There are no current logistically feasible prophylactic agents available in the United States as medical countermeasures. Antitoxin is the only effective treatment for botulism and has been shown to reduce the duration of hospitalization, duration of mechanical ventilation, and cost of hospitalization (30). The current treatment for adult botulism is heptavalent (serotypes A to G) equine botulism antitoxin (BAT) (18). BAT is immunogenic, and hypersensitivity reactions have been reported, including serum sickness and asystole (18,31).
As an alternative and potentially safer and more effective product, we have been developing serotype-specific human or humanized MAb combinations that stoichiometrically neutralize BoNT by eliciting first-pass clearance through the liver (20,27). In addition, these antibodies are being used in diagnostic tests to identify the specific serotype of botulinum neurotoxin that was causing the symptoms and provide guidance for treatment (32,33). This study was the first in-human assessment of NTM-1634, a four-MAb combination which potently neutralizes BoNT/C, C/D, D/C, and D in rodents. The results demonstrate that single escalating doses of NTM-1634 administered intravenously into healthy subjects were well tolerated and safe. In addition, these doses also demonstrated acceptable immunogenicity profiles, with little ADA detected over the dose ranges and duration of the study. No dose-related SAEs were observed, and the AEs that were observed between the three cohorts and placebo group were similar in frequency, character, and severity. These results indicate that NTM-1634 may be a safer product than the equine polyclonal antibody BAT (18).
Previous studies with NTM-1634 have demonstrated that the effectiveness of NTM-1634 in neutralizing BoNT C/D is dependent on the presence of all four antibodies (C. Garcia-Rodriquez et al, unpublished data). This is consistent with observations for other anti-BoNT antibody combinations in development (20,24,25). It therefore follows that the duration of effectiveness of the combination will be determined by the monoclonal antibody that falls below the minimal effective levels most rapidly. The terminal t 1/2 s were similar across the range of doses for each antibody. At all doses given, however, XCD-a was cleared more rapidly than the other three, with XCD-d being the next most rapidly cleared. However, all antibodies were detectable for a minimum of 4 weeks, with the most rapidly clearing MAb, XCD-a, having a serum half-life of 11 days. This contrasts with BAT, which is an F(ab=) 2 product with short serum half-lives for BoNT/C and BoNT/D (30 and 7.5 h, respectively). The long half-life of NTM-1634 may reduce the likelihood of a relapse of botulism that has been reported with BAT (19).
While we have not directly compared the potency of NTM-1634 to BAT for BoNT/C or BoNT/D, the potency of a BoNT/A MAb combination (NTM-1631) under development is 400-to 600-fold more potent than BAT in mouse protection studies with BoNT/A1, and a three-MAb combination to BoNT/F more than 150 to 450 times more potent than BAT (25, 27) on a weight basis. The lowest dose of NTM-1634 studied (0.33 mg/kg) would deliver a total BoNT/C neutralizing capacity of 1.85 ϫ 10 8 mouse LD 50 s of BoNT/C based on the preclinical data described above. This compares to a single dose BoNT/C neutralizing capacity for BAT of 3.0 ϫ 10 7 mouse LD 50 s, or more than 6-fold more. This greater potency is expected, since polyclonal antisera rarely have more than 1% of the total IgG directed against the target antigen, compared to ϳ100% for recombinant IgG. The combined safety, potency, and long half-life of NTM-1634 make it possible for potential intramuscular or subcutaneous administration after exposure and prior to the development of symptoms, which is not possible with BAT. These features would make NTM-1634 simpler to administer in a mass casualty scenario. The high potency and long half-life would also permit administration of NTM-1634 for the prevention of botulism as an alternative to vaccination. This is important, since there is no longer a vaccine available to prevent types C and D botulism.
In summary, this study provides an early evaluation of the pharmacokinetics and safety profile of NTM-1634. The highly potent protection in animal models and the safety profile and long half-life of Ͼ1 month in humans demonstrate the potential utility of NTM-1634 for treatment of botulism due to serotypes C and D. Thus, further clinical development of NTM-1634 for the treatment and prevention of BoNT intoxication due to serotypes C and D is warranted.

MATERIALS AND METHODS
This was a phase 1, single-center, placebo-controlled, double-blind dose escalation study to evaluate the safety, pharmacokinetics characteristics, and immunogenicity of NTM-1634 in healthy adults. This first in-human study consisted of three cohorts (A, 0.33 mg/kg; B, 0.66 mg/kg; C, 1 mg/kg) of eight subjects each. Each subject received a single i.v. infusion of NTM-1634 or placebo administered over 1 h. The placebo was normal saline.
NTM-1634. NTM-1634 consists of four human single-chain Fv (scFv) MAbs isolated from yeast display libraries constructed from the antibody variable region genes of humans immunized with pentavalent botulinum toxoid (C. Garcia-Rodriquez et al, in preparation). The heavy-and light-chain variable region genes of the scFv were inserted into vectors containing the human kappa and IgG1 constant region genes to create human IgG1/kappa MAbs, and four separate stable CHO-K1 cell lines were established. The IgG1 isotype was selected on the basis of studies showing the importance of Fc receptor engagement for the synergy in potency seen when the MAbs are combined (J. D. Marks, unpublished data). Each MAb (XCD-a, XCD-b, XCD-c, and XCD-d) was expressed and purified individually using protein A and ion-exchange chromatography and then combined in equimolar amounts to create NTM-1634.
Subjects. The study enrolled a total of 25 healthy adult volunteers. Subjects in all cohorts participated in the study for approximately 21 weeks, including a 4-week screening period, a 3-day inpatient stay, and approximately 17-week follow-up study after drug administration. Eligible research participants were healthy male or healthy, nonpregnant, nonlactating females between the ages of 18 and 45 years.
Subjects had body mass indices of between 18.5 and 30 kg/m 2 , negative illicit drug screens, and adequate venous access for the infusion. Subjects were considered study ineligible if they had a history of a chronic medical condition that would interfere with the accurate assessment of the objectives of the study, a history of severe allergic reaction to any type of medications, bee stings, food, or environmental factor or reaction to immunoglobulins, positive serology for HIV, HBsAg, or hepatitis C virus (HCV) antibodies, were pregnant or breastfeeding, were previously exposed to BoNT and active drug, or have alcohol dependence. Research subjects who had received any monoclonal antibody in the past or any antibody or blood products, treatment with another investigational drug within 28 days of dosing, donated blood within 56 days of enrollment, or use of H1 antihistamines or beta-blockers within 5 days of dosing were also excluded. Additional exclusion criteria included a marked baseline prolongation of QT/QTc interval, clinically significant electrocardiogram, systolic blood pressure of Ͼ140 mm Hg or diastolic blood pressure Ͼ90 mm Hg, resting heart rate Ͻ50 or Ͼ100 beats per min, or oral temperature of Ն38°C.
Study design. Subjects were randomized into three cohorts of eight subjects and admitted to the study unit on day Ϫ1. Infusion occurred on day 1, and subjects remained at the clinic until discharge on day 2 (25 h after the infusion had ended). Follow-up visits occurred on days 3,4,8,15,29,43,57 For each dosing cohort, the first two subjects were randomized 1:1 active/placebo so that one of the first two subjects received active treatment and the other control. The treatment assignment of the remaining six subjects in each cohort was 3:1 active/placebo. The randomization list was generated by the unblinded study biostatistician and transferred to the unblinded study pharmacist prior to the start of the study. The study staff participating in the administration of the study product and assessment of subjects were not aware of the contents of the i.v. bag. Drug and placebo appeared identical.
The study drug was manufactured by XOMA Corporation (Berkeley, CA) and filled by Althea Technologies (San Diego, CA). Assays developed for NTM-1643 characterization and release were qualified using biological testing, including a MAb-specific enzyme-linked immunosorbent binding assay (ELISA) which used recombinant BoNT/C protein domains (22, 34 and G. Manzanarez et al., unpublished data), developed and qualified by Ology Bioservices, and an in vivo quadruple antibody protection assay in mice, developed and qualified in collaboration with SRI International (Menlo Park, CA). The investigational product was formulated as a 5-mg/ml, clear colorless, sterile aqueous solution in a pH 6.0 buffered vehicle without any preservatives. The drug product was supplied in 2-ml, pyrogen-free, type 1 glass vials. Placebo was a sterile, nonpyrogenic isotonic solution of 0.9% sodium chloride injection, USP grade, and water for injection. The placebo did not contain preservatives, bacteriostatic or antimicrobial agents, or added buffer. Normal saline was used to dilute the NTM-1634 for i.v. infusion and was supplied in single-dose plastic containers.
Safety analyses. The safety endpoints were the occurrence of serious adverse events (SAEs) following administration of NTM-1634 to the final follow-up visit, the occurrence of adverse events (AEs) from administration of NTM-1634 to day 57, the occurrence of changes from baseline in physical examination, vital signs, and clinical safety laboratory values following administration of NTM-1634 to the final follow-up visit, and the occurrence of changes from baseline in ECG parameters after administration of NTM-1634 on day 1 (day of infusion). Immunogenicity was assessed by determining the presence of human anti-human antibodies.
Pharmacokinetics analyses. Four immunoassays were developed to measure the concentrations of XCD-a, XCD-b, XCD-c, and XCD-d in serum samples. A bridging electrochemiluminescence (ECL) assay was used (22,27,34). Biotinylated (b) and ruthenylated (ru) BoNT/C domains were used in the assay as the capturing and detecting reagents, respectively. The assay uses the bivalent binding capability of the antibodies to form a bridging complex with biotinylated-domain and ruthenylated-domain to generate ECL signals for the measurement of the target antibody concentration in serum. Six assays were developed using the same format for each antibody assay. ECL signals generated from captured immune complexes formed when one arm of the MAb bound to a biotinylated domain and the other arm to the ruthenylated domain were detected by a Meso-Scale Discovery (MSD)SECTOR Imager 6000 and reported in ECL units. Calibration standards and quality control samples were prepared by spiking known amounts of each MAb into human serum. With the minimum required dilution of 1:10, the lower limit of quantitation (LLOQ) was 10 ng/ml, with a quantifiable range of 10 to 2,000 ng/ml. Immunogenicity assays were developed and validated to detect anti-XCD-a, XCD-b, XCD-c, and XCD-d antibodies in human serum. Bridging ECL assays used labeled MAbs for the capture and detection of antidrug antibodies (ADAs) in serum. A mixture of biotinylated and ruthenylated XCD-a, XCD-b, XCD-c, and XCD-d MAbs were incubated with human serum, and ECL signals were generated when one arm of an ADA bound to a b-MAb and the other arm bound to a ru-MAb. Samples that generated a signal above the screening cut point (SCP) were considered positive and confirmed with the use of a competitive assay in which the serum samples were incubated with and without nonlabeled MAbs. Samples with ECL signals above the confirmation cut point were considered positive and further analyzed for titers.
Statistical analyses. Pharmacokinetics (PK) parameters for each of the four monoclonal antibodies of NTM-1634 (XCD-a, XCD-b, XCD-c, and XCD-d) were calculated from the serum concentration-time data using noncompartmental techniques (Phoenix-WinNonlin version 6.3.0; Pharsight Corporation, St. Louis,