This was an open label, non-blinded pilot study of 10 CDM dogs receiving the study peptide, OPT501, either intravenously or subcutaneously for up to 24 weeks. As a pilot study, there were no placebo or control groups. The study (I-013432) was approved by the Center for Veterinary Medicine, a branch within the US Food and Drug Administration. Clinical and disease management changes were recorded, and Th40 cell levels determined before, during, and after treatment. The dogs were housed at home with their owners and no dietary restrictions, instructions regarding physical activity, or other restrictions were in place. That is, the owners treated and fed their dogs per their normal schedules.

The study was reviewed and approved by the Colorado State University Clinical Review Board (protocol #640). Ten CDM subjects were recruited through the Colorado State University Veterinary Teaching Hospital in Fort Collins, CO, or at the Montclair Animal Clinic in Denver, CO. All subject owners were consented prior to enrollment of their dog. All experiments were carried out in accordance with ARRIVE guidelines and relevant regulations.

OPT501, a 15-amino acid peptide was synthesized as an acetate salt at AmbioPharm, Inc., North Augusta, SC, in a cGMP facility. The certificate of analysis confirmed that the peptide was 97% pure. It was received as a lyophilized powder and was stored at -80°C until use. For short term storage, up to 3 months, the powder was stored at -20°C. For intravenous (i.v.) and subcutaneous (s.c.) administration, OPT501 was dissolved in sterile Phosphate Buffered Saline (PBS), pH 7.2.

Treatment of CDM dogs was preceded by a safety and tolerability study done through a CRO, WuXi Biologics, where OPT101 (human version of the drug) was administered via i.v. infusion over 30 minutes to both rats and healthy dogs. OPT101 was well tolerated, and only moderate adverse events (injection site reactions related to histamine release; the higher doses were associated with decreases in blood pressure and heart rate, which returned to normal 1 hour after the end of infusion) were observed at doses 25–50 times higher than the doses used in this study.

Subjects treated with OPT501 via the i.v. route received 30-minute infusions of 1–2 mg/kg on Monday, Wednesday, and Friday the first week then once weekly for 8 weeks. Subjects treated with OPT501 via the s.c. route received a dose of 2.0 mg/kg on Monday, Wednesday, and Friday the first week followed by 2.0 mg/kg twice weekly, Monday/Friday, or 4 mg/kg once per week for up to 24 weeks. Some subjects received OPT501 s.c. at a dose of 4 mg/kg twice weekly for up to 24 weeks. Some dogs continued to receive weekly s.c. injections for up to 18 months. Adverse events: Other than injection site irritation with s.c. dosing, no OPT501 related adverse events were observed in the CDM dogs. Changes in insulin dosage were done under advisement of the attending veterinarian.

Venous blood was drawn into 2 ml heparinized blood collection tubes for preparation of peripheral blood mononuclear cells (PBMC). Blood was drawn into EDTA tubes for complete blood count (CBC) and red top tubes for chemistry panels. CBC and chemistry panels were processed at each veterinary clinic’s associated laboratory service (CSU: Veterinary Diagnostic Laboratory; Montclair: Antech, Inc.).

Normal ranges for the CSU Veterinary Diagnostic Laboratory are: Lymphocytes – 1000-4,800/µl; Neutrophils – 2.6–11 x 10³/µl; Platelets – 200–500 x 10³/µl; Alkaline Phosphatase – 15–140 IU/L; Alanine Aminotranferase – 10–90 IU/L; Cholesterol – 130–300 mg/dl; Fructosamine – 210-350 µmol/L; Glucose – 70–115 mg/dl.

Normal ranges for the Antech laboratory are: Lymphocytes – 690-4,500/µl; Neutrophils – 2.06-10.6 x 10³/µl; Platelets – 170–400 x 10³/µl; Alkaline Phosphatase – 5–131 IU/L; Alanine Aminotranferase – 12–118 IU/L; Cholesterol – 92–324 mg/dl; Fructosamine – 136-350 µmol/L; Glucose – 70–138 mg/dl.

Systemic Inflammatory Index calculation: The Systemic Inflammatory Index (SII (19, 20);) is derived by multiplying the total number of neutrophils and platelets then dividing by the total number of lymphocytes ((neutrophils x platelets)/lymphocytes).

PBMCs were prepared using Ficoll Paque Plus (Cytiva, Marlborough, MA) according to the manufacturer protocol. Resulting PBMC were stained in PBS (Cytiva, Marlborough, MA) containing 0.5% bovine serum albumin and 2 mM EDTA (both from Sigma-Aldrich, St. Louis, MO). Cells were stained with anti-dog CD4, anti-dog CD3 (Bio-Rad, Hercules, CA; cat# MCA1038APC and MCA1774F, respectively), and anti-human CD40 (produced in-house; clone G28-5). Analysis was done using FlowJo software (Becton Dickinson, Franklin Lakes, NJ). Cells were first gated on live cells then on CD3. CD4⁺CD40⁺ percentages of total CD4⁺ in the CD3 gate are reported for dogs as previously described [8].

All human subjects, female and male, >18 years old, with or without stage 3 type 1 diabetes, were recruited at the Barbara Davis Center for Diabetes and gave consent under Colorado Multiple Institutional Review Board protocol no. 01-384. The duration of disease in the T1D subjects was 6 weeks to 40 years. Pre-T1D subjects were part of a TrialNet study (10) and met the criteria (high-risk HLA and/or being a first degree relative of a subject with T1D) for inclusion.

Non-obese diabetic (NOD) and Balb/c mice (Taconic Biosciences, Germantown, NY) were housed at an AAALAC-approved facility. All animal experiments were performed under an IACUC-approved protocol (#00529) and adhered to the NIH Public Health Service Policy on Humane Care and Use of Laboratory Animals.

Blood was drawn from mice into heparinized syringes and from humans into heparinized blood collection tubes. PBMC preparation, cell staining, gating and analysis was done exactly as described above except that the following antibodies were used: Anti-human CD3, cat# 130-113-134 (Miltenyi Biotec, Waltham, MA), anti-human CD4, cat# 25-0049-T100 (Cytek Biosciences, Fremont, CA), anti-human CD40, clone G28-5 (produced in-house), anti-mouse CD3, cat# 12-0031-82 (Thermo Fisher Scientific, Waltham, MA), anti-mouse CD4, cat# 130312 (BioLegend, San Diego, CA), and anti-mouse CD40, clone 1C10 (produced in-house).

C-peptide was measured in plasma retrieved from PBMC separation utilizing an ELISA kit from MilliporeSigma (St. Louis, MO; cat# EZCCP-47K). This kit did not state a normal range for dogs. The samples were collected without instructions for fasting or postprandially and the data is therefore exploratory as the values can vary with/without meals.

Three-dimensional peptide folding was done using the on-line peptide folding prediction program PEP-FOLD3, available at the RPBS web portal (https://mobyle.rpbs.univ-paris-diderot.fr) (21).

Statistical analysis was performed using the Prism program from GraphPad (San Diego, CA) and the data is expressed as means (SEM). T-tests, unpaired or paired, and one-way ANOVAs were used as indicated in the figure legends and p-values of < 0.05 were considered significant.

OPT101 for human use was derived from the known interaction site between CD154 and CD40 to include three of the five amino acids described by crystallography to be critical for CD40 interaction (12). Canine CD154 and murine CD154, the sequence from which OPT101 was derived (12), are both 260 amino acids in length. The peptide, OPT501, was designed as a 15-mer analogous to the drug product OPT101 (Figure 1A). Within the canine total protein sequence, the 15-mer was generated by beginning at Valine (V) 136 and continuing through Asparagine (N) 150 (Figure 1A). The KGYY shown by crystallography to be involved in the binding of the CD154 protein to CD40 was included (Figure 1A; bold amino acids). The canine sequence differs at four amino acids when directly compared to OPT101: a Glutamine (Q) to Arginine (R) substitution, a Lysine (K) to Proline (P) substitution, a Methionine (M) to Isoleucine (I) substitution, and a Lysine (K) to Serine (S) substitution (Figure 1A). In prior studies, specific residues within OPT101 were assessed for their contribution to peptide activity in disease-relevant models (12). Crystallography data reported by Bajorath et al. (18) indicated that Tyrosine in positions 9 and 10 (positions 144 and 145 in the whole protein) were crucial for CD154 and CD40 interaction. However, our functional peptide data showed that Tyrosine in position 9 is crucial for biological function in prevention of diabetes while Tyrosine in position 10 is dispensable (12) (Figure 1A). These observations identify KGY as a conserved core motif within this region, which the canine CD154 sequence contains (Figure 1A). Given the amino acid differences, we compared the likely models of OPT101 and OPT501 using the PEP-FOLD 3 program for three-dimensional peptide modeling (21). The models predict that OPT101 adopts a helical segment that is not likely in OPT501, whereas the models predict that the Proline in OPT501 may facilitate a tighter fold of the peptide (Figure 1B; Proline in red). However, the models predict a KGYY-containing planar region in both peptides consistent with potential interaction with CD40 (Figure 1B).

CDM subjects were recruited, owners consented, and the demographic data is presented in Table 1. All participants could be classified as mid-life or older (median: 9.5 years, range: 6–13 years). Disease duration ranged from one month to 48 months (Table 1). The two patients at one month since diabetes onset were considered new onset. All subjects experienced unregulated blood glucose despite on-going insulin treatment with weekly averages ranging from 350 to 450 mg/dl glucose. Unlike human type 1 diabetes patients, the CDM patients, all of which were treated with Vetsulin™, had difficulty regulating glucose even on high dose insulin. At diagnosis blood glucose was high with glucosuria and three dogs had urine ketones. Urine ketosis follows ketonemia in poorly regulated diabetes and can lead to diabetic ketoacidosis, which requires close monitoring and intensive care. Five of the dogs (50%) had cataracts.

CDM subjects were treated as described in methods and Table 2. Blood was drawn and PBMC prepared on the first day of treatment prior to administration of OPT501, then at various timepoints throughout the treatment schedule. Healthy control (HC) dogs had mean Th40 cell levels of 22.5%, which is comparable to HC mice and HC humans (no autoimmune disease, no infectious disease, no cancer; Figure 2A). Compared to HC dogs, all CDM subjects had significantly elevated Th40 percentages prior to treatment as reported previously (8) (Figures 2A, B; p< 0.0001). CDM Th40 percentage was significantly higher than human type 1 diabetes and diabetic NOD mice (Figure 2B; p = 0.0019 and p = 0.0178, respectively). After six days of treatment, the Th40 cell percentage began to decline with a further decrease on days 12 and 16 (Figure 2A; p = 0.0331, p< 0.0001, and p< 0.0001, respectively). By day 21 the percentage was the same as those of HC dogs (Figure 2A; p< 0.0001). Importantly, Th40 cell percentages did not decrease further and remained steady throughout the study.

We determined that total lymphocytes in CDM were significantly increased compared to HC (Figure 2C; p<0.0001). Treatment with OPT501 reduced total lymphocytes to normal (HC) levels (Figure 2C; p<0.0001). We originally reported that CDM dogs experienced a high systemic inflammatory index (SII), which is derived by multiplying the total number of neutrophils and platelets then dividing by the total number of lymphocytes (8). Treatment with OPT501 reduced SII significantly (Figure 2D; p=0.0080).

Alkaline phosphatase (ALP) is an enzyme found in liver, bone, kidney, and intestine (22–24). The primary function is protein breakdown, specifically in liver; in bone it is involved in regulating the mineralization process and Vitamin D metabolism (23). High ALP can reflect liver and/or bone damage, and it is associated with risk of cardiovascular disease including atherosclerosis, dyslipidemia, and hyper-cholesterolemia in humans (25). Alanine Aminotransferase (ALT) converts alanine into pyruvate and produces glutamate, a neurotransmitter. Increases in ALT and ALP are associated with primary and reactive liver disease and are frequently monitored during drug administration as an indicator of hepatotoxicity (26). In the CDM patients, ALP was above the canine normal range in nine of 10 subjects (90%) with four (40%) being >900 U/L/H (Figure 3A). ALT was above normal range in six dogs (60%; Figure 3B). Following OTP501 administration, ALP was reduced in all dogs, achieving normal range in two (20%) dogs (Figure 3A; p = 0.0145). ALT was reduced in all OPT501-treated dogs, with only three remaining out of normal range, however the change did not reach significance (Figure 3B; p = 0.0506). These changes suggest that the study drug is not hepatotoxic and positively impacts hepatic changes seen in diabetics.

Elevated serum cholesterol is associated with CDM (27) and cardiovascular disease (CVD) including canine atherosclerosis (28–30). High serum cholesterol is not surprising in human type 2 diabetes but triglyceride and LDL increases can also be associated with human type 1 diabetes (31). Nine of ten CDM subjects (90%) had elevated cholesterol, and the tenth was in the higher half of the normal range (Figure 3C). OPT501 treatment reduced cholesterol in all subjects with nine subjects (90%) moving to normal range (Figure 3C; p = 0.0004). Although the dogs were not fasted prior to blood collection, a recent study showed no impact of feeding on cholesterol concentrations in healthy dogs (32). The lowered cholesterol outcome suggests an alternate indication for OPT501 in hyper cholesterol associated canine disease.

Fructosamine is a compound that forms from glycation reactions between free glucose and primary amines followed by isomerization via an Amadori reaction (33, 34). High fructosamine reflects continuously dysregulated glucose levels and is used as a measure of disease management. Therefore, we compared pre-treatment fructosamine with that measured at study completion. In all CDM subjects, there was an average 1.8-fold reduction in fructosamine over the course of the OPT501 treatment, and five of nine subjects (55.6%) had values in the normal range by the end of the study (Figure 4A; p< 0.0001). Using a fructosamine to HbA1c conversion formula (HbA1c = 0.017 × Fructosamine + 1.61) (35), all subjects had HbA1c equivalents above 10% before treatment (Figure 4B). OPT501 treatment reduced HbA1c equivalents significantly, with five of nine (55.6%) achieving levels under 7% and three achieving the American Diabetes Association desired goal for human diabetes of 6.5% (Figure 4B; p< 0.0001).

In all dogs, glucose was measured prior to treatment and post treatment by standard blood chemistry panel. A continuous glucose monitor (CGM) was placed in five of the dogs two weeks prior to treatment and maintained throughout the treatment period. Prior to treatment, all dogs had blood glucose concentrations substantially above normal limits (median: 398 mg/dl, range: 283–616 mg/dl; Figure 5A). At the end of treatment all subjects had significantly lower blood glucose, with four of the ten (40%) achieving normal range (Figure 5A; p< 0.0001). None of the dogs demonstrated hypoglycemia. Representative CGM data from one female and one male subject are shown (Figures 5B, C, respectively). Blood glucose levels were collected for 14 days prior to treatment, during, and post final treatment. Daily averages for the defined periods are shown. In both subjects, OPT501 treatment significantly reduced daily glucose levels that were maintained for two weeks post final dose (Figures 5B, C; all p< 0.0001). Daily insulin was recorded by eight owners during the treatment period. Based on consultation with their veterinarian, average daily insulin use was reduced in all subjects by eight weeks (Figure 5D; p = 0.0001). Two subjects reduced from 10 U twice daily (20U/day) to 1 U/day; a 95% reduction in insulin.

The American Diabetes Association reports that C-peptide can be detected in up to 29% of human type 1 diabetes, even 31–40 years after disease onset (36), indicating low level residual β-cell function. When disease duration was shorter, the C-peptide values were higher; higher residual C-peptide also was associated with later age at onset (36). With controlled inflammation, β-cell mass regeneration and insulin production could be increased. There are no reports for normal C-peptide levels in canine subjects. C-peptide was measured in four dogs treated with OPT501 by the i.v. route. Two of the four CDM dogs had been diagnosed less than one month and another had been diagnosed 3.5 years prior to study enrollment (Table 1). All four dogs had residual levels of C-peptide, and three were low (Figure 6A). Insulin and C-peptide have a 1:1 molar ratio in serum (37, 38). When the CDM dogs were treated with OPT501, C-peptide levels increased significantly (Figure 6A; p = 0.0091). The individual C-peptide trajectories for each dog are shown (Figure 6B).

Anecdotally, most owners (non-blinded) reported that their dogs were more energetic and generally seemed to have improved health while on OPT501 and no owner reported any infections or other adverse events during OPT501 treatment.