The clinical trial was approved by the Institutional Animal Care and Use Committee (IACUC approval #: DWP-IACUC-001) and was conducted in collaboration with veterinarians from four veterinary clinics. Each blood sample was delivered to the central lab and FACS analysis center for each evaluation item as requested sample form and analyzed according to each SOP, respectively.

Dogs qualified for inclusion if they were over 10 years of age, completely vaccinated, free from particular diseases, and showed reduced appetite or activity levels. Following informed consent from owners, a thorough assessment was performed, including physical and ocular exams, thoracic and abdominal radiographs, abdominal ultrasound, electrocardiography, complete blood counts (CBC), serum chemistry, electrolytes, and urinalysis. Blood sampling was performed in the morning after an overnight fast, and tests were conducted at approximately the same time of day at each hospital visit to minimize temporal variation. Flow cytometry was used to assess CD4 and CD8 levels in EDTA-anticoagulated whole blood. Dogs were excluded if they presented abnormal checkup results or a CD4/CD8 ratio below 1 in flow cytometry. After evaluation on day 0, each dog was administered a single injection of a plasmid optimized for expressing the canine GHRH gene. Post treatment all dogs were observed at the animal hospital on days 0, 30, 60, 90, 120, and 180. Assessments at each time point included CBC, serum chemistry, weight, body temperature, heart and respiratory rates, body condition score, capillary refill time, blood pressure, limb thickness, serum IGF-1, and CD3/CD4/CD8 blood levels. The Urine Protein Creatinine (UPC) ratio was assessed both before treatment on day 0 and after treatment on day 180. At each visit, the veterinarian interviewed the owner, recorded health conditions and adverse events, and completed a wellness form. The wellness form was created using insights from a study on age-related physical and functional changes in senior dogs, (19) as well as the Ohio State University Veterinary Medical Center Honoring the Bond Program Quality of Life Checklist (20) and HHHHHMM scale (21). The wellness form included assessments of activity level, exercise tolerance, mentation (attitude, alertness), play, interaction, appetite, thirst, frequency of relaxed postures, hair brightness, skin dryness, skin odor, sleep patterns, and overall quality of life. Veterinarians rated their dog’s condition on a scale from very good (5) to very poor (1) (as shown in Table 1). The clinical score, representing the total score, was computed for subsequent analysis. Apart from the clinical score, an owner questionnaire evaluated symptom changes since the last assessment, concentrating on appetite, activity, and mobility. Scores were assigned as follows: 1 indicates a significant decrease, 2 a decrease, 3 no change, 4 an increase, and 5 a significant increase. The adverse events form included counts of nausea, diarrhea, appetite loss, exercise intolerance, and the presence of urticaria, facial edema, and dyspnea. Veterinarians monitored each dog from immediately after the injection until full recovery from anesthesia, checking for any abnormalities in vital parameters and for local adverse reactions at the injection site. Follow-up evaluations were conducted on days 30, 60, 90, 120, and 180, during which veterinarians assessed the injection site for adverse reactions (including pain, erythema, granulomas, and ulcers), monitored vital signs, and reviewed hematologic test results. At each follow-up visit, owners were also surveyed regarding any observed adverse events at home. Prior to the clinical trial, an extensive owner education program informed participants about possible adverse events. The possibility of terminating the clinical trial was highlighted should severe adverse events arise. This clinical trial’s scheme is summarized in Figure 1A.

A model of the processed canine peptide GHRH model was prepared with Discovery Studio 2021 (Biovia, San Diego) and DeepView v4.1 (22). The PLS-D1000 (pAV0221) plasmid was constructed using a self-constructed vector containing the muscle-specific promoter SPc5-12, human GH 5′ untranslated region, and human GH 3′ untranslated region based on the pVAX1^tm vector (Invitrogen, Cat. No. V260-20) as a backbone, and the canine GHRH expression gene was inserted using the NcoI/HindIII sites (Figure 1B). The human processed GHRH peptide model was derived from the AlphaFold Protein Structure Database entry AF-P01286-F1 (Figures 1C,D). Residues putatively interacting with GHRHR were derived from homology with structures of human GHRH bound to receptors [7v9m.pdb (23) and 7cz5.pdb (24)]. Predicted bound structure of canine GHRHR (Uniport acc. A0A8C0MLJ3) with canine GHRH was generated with Discovery Studio 2021 and DeepView v4.1 using 7v9m.pdb as a template.

The plasmid encoding synthetic canine GHRH is supplemented with 1% (w/w) poly-L-glutamic acid sodium salt and diluted to a concentration of 1 mg/mL using sterile water. Dogs were anesthetized using a mix of butorphanol (0.1–0.4 mg/kg) and midazolam (0.1–0.3 mg/kg), followed by a 1 mL subcutaneous injection of 2% lidocaine at the site. The 1 mg/mL plasmid was injected into the dog’s hind quadriceps femoris muscle, and the administration was completed immediately using an electroporation device. A previous study has reported that when plasmids are introduced into muscle cells via electroporation, the gene expression effect may persist for approximately 6 months (25). Electroporation at the muscle injection site utilized the CELLECTRA® 5P adaptive in vivo EP system and the 5P IM applicator from INOVIO Pharmaceuticals Inc., Plymouth Meeting, PA. USA at 0.5 Amps, 3 pulses of 52 milliseconds each, 1 s between pulses (26). After injection, the animals were observed by a veterinarian until they recovered from anesthesia. Afterwards, the owners were educated to observe adverse effects, and the injection site was checked for any abnormalities at each time point by veterinarians.

Blood samples were collected from the canines at baseline, and on days 30, 60, 90, 120, and 180. Blood samples were processed to isolate serum and were stored at −80 °C for subsequent analysis. Serum IGF-1 levels were measured by GCCL Co., Ltd. using immunoassay methods with the Cobas 8000® e801 module from Roche Diagnostics, Basel, Switzerland, adhering to established protocols.

The blood collected in EDTA tubes was sent to the FACS analysis center (Global R&D Center of Plumbline Life Science, Inc.) within an hour and immediately performed the experiment by the same operator using the same FACS instrument to maintain data consistency. Each used antibodies were prepared with the same lot number respectively, preventing the data variation. Additionally, the optimal usage amount for each antibody used was determined through prior testing.

Blood sample was centrifuged at 500 g for 10 min. Subsequently, RBC lysis buffer (R2035, BIOSESANG, Korea) was applied to the resulting pellet and incubated at room temperature for 10 min. Cells were washed with PBS and transferred to a FACS tube at a concentration of 1 × 106 cells per tube. Cells were washed twice with PBS containing 1% FBS and subsequently stained with antiCD3-FITC (BIO-RAD) (27) antiCD4-PECY7 (Invitrogen) (7), antiCD8-PB (BIO-RAD) (7), antiCD62L-PE (BIO-RAD) (7, 28), antiCD45RA (BIO-RAD) (7), and antiCD45RA-APC (as shown in Table 2). The T-cell subset gating strategy commenced with the identification of CD3+ CD4+ (CD4+ T cells) and CD3+ CD8+ (CD8+ T cells) populations. Naïve (CD45RA+ CD62L+), central memory (CD45RA-CD62L+), effector memory (CD45RA − CD62L−), and terminally differentiated effector memory (CD45RA+ CD62L−) T cells were identified within CD3+ CD4+ and CD3+ CD8+ subsets based on CD45RA and CD62L expression. Fluorescence was measured using a BD CANTO II instrument and analyzed using BD FACSDiva Software v9.4.

Statistical evaluations were conducted using SPSS (Windows version 26, IBM Corp., Armonk, New York) and GraphPad Prism v10 (GraphPad Software, Inc., La Jolla, California). A p-value < 0.05 denoted statistical significance. Normality was assessed with the Shapiro–Wilk test. As some variables at certain time points did not meet the assumption of normality, the Friedman test was used to analyze changes in each parameter across time points (day 0, 30, 60, 90, 120, and 180). Post hoc comparisons with day 0 were performed using Dunn’s multiple comparison test. For CD4+ naïve T cell analysis, however, an uncorrected Dunn’s test was applied. Differences in IGF-1 levels between the Low Basal Group and High Basal Group were assessed using a t-test. The variation in response rates to GHRH based on initial IGF-1 levels was examined with a chi-squared test. Data are presented as mean ± SEM, except for patient signalment variables (age and weight), which are expressed as median (range).

A total of 49 elderly dogs (≥10 years old) without a diagnosis of any specific disease were initially recruited for the study. Nineteen dogs were excluded based on predefined criteria, resulting in 30 dogs being enrolled (Figure 1A). The exclusion criteria comprised a confirmed diagnosis of autoimmune or other specific diseases, participation in another clinical trial within the preceding 6 months, a CD4-to-CD8 ratio ≥1, and inability to participate for a period exceeding 4 months. Table 3 presents the demographic information of the canine subjects. The study population included 13 castrated males, 10 spayed females, 6 intact females, and 1 intact male. The median age of the population was 11 years (range: 10–16), with a median weight of 5 kg (range: 3.2–41.0). The most common breeds for the study were Maltese (n = 6), Poodle (n = 6), and Pomeranian (n = 4).

Clinical scores, derived from owner-completed questionnaires on days 0, 30, 60, 90, and 180, yielded mean values of 56.17 ± 1.46, 61.00 ± 1.35, 62.87 ± 1.59, 64.67 ± 1.19, 65.03 ± 1.47, and 65.47 ± 1.10, respectively. Twenty seven of 30 treated dogs (90%) showed improvement in clinical score. A Friedman test indicated a significant change in clinical scores over time (p < 0.0001). Dunn’s multiple comparison test revealed significant increases in clinical scores from day 0 at day 60, 90, 120, and 180 (all p < 0.001).

Evaluation of appetite, activity, and exercise intolerance scores using the Friedman test revealed significant temporal differences for all three parameters (all p < 0.0001). Dunn’s multiple comparison test showed that appetite change scores were significantly higher at days 60, 90, and 180 (all p < 0.05) compared with baseline; activity change scores were significantly higher at days 60, 90, 120, and 180 (all p < 0.05); and exercise intolerance scores were significantly higher at days 60 and 180 (all p < 0.05). Scores for all parameters began to increase on day 30 and remained elevated through day 180 following injection of the GHRH-encoding plasmid. Mean clinical score, appetite change score, activity change score, and exercise intolerance change score by time point are presented in Table 4 and Figure 2.

Friedman test analysis revealed no significant temporal differences in body weight, left forelimb thickness, or right forelimb thickness (all p > 0.05) following injection of the GHRH-encoding plasmid (Table 5; Figure 3). For the hindlimbs, significant overall changes were observed in the left hindlimb (p = 0.0018) and right hindlimb (p = 0.0011). Dunn’s multiple comparison test indicated that both the left and right hindlimb thicknesses significantly increased at day 180 compared with baseline (p = 0.0168 and p = 0.0209, respectively).

To investigate the effects of canine GHRH administration on the GHRH/GH/IGF-1 axis, we measured serum IGF-1 levels pre- and post-treatment. The average basal serum IGF-1 level in a group of 30 dogs was 125.81 ± 21.01 ng/mL. Mean serum IGF-1 concentration following administration of GHRH encoding plasmid did not increase over time when analyzed in bulk (Figure 4). Analyzing further, we ranked the dogs based on their basal IGF concentration at Day0. The dogs were stratified to two groups post-hoc based on a baseline IGF-1 concentration of 90 ng/mL. The first group of 13 dogs had the mean basal IGF concentration of “51.55 ± 4.70 ng/mL” (Low Basal Group) while the other 17 dogs had the mean basal IGF concentration of “182.59 ± 30.56 ng/mL” (High Basal Group). These basal IGF-1 levels between the groups were statistically significant (t-test; p = 0.001). In the Low Basal Group, 84.6% (11/13) dogs exhibited an increase in IGF-1 levels after the administration of GHRH encoding plasmid. In contrast, in the High Basal Group, 76.5% (13/17) dogs showed a decrease in IGF-1 levels following the administration of GHRH encoding plasmid.

In addition, when classified into two groups based on small dogs (less than 10 kg) in the classification by body weight of the American Veterinary Medical Association, dogs weighing less than 10 kg tended to maintain the canine baseline IGF-1 (90 ng/mL) after GHRH plasmid administration. On the other hand, in medium to large dogs (dogs weighing 10 kg or more), the baseline IGF-1 was 246.68 ng/mL, which was 2.74 times higher than the canine baseline IGF-1 (90 ng/mL), but tended to decrease after GHRH plasmid administration (up to 25% decrease) (see Supplementary Figure 1).

The impact of canine GHRH therapy on immune cells was also analyzed. CD3+ T-cell percentages in leukocytes were recorded at days 0, 30, 60, 90, 120, and 180, with respective mean ± SEM values of 83.12 ± 1.82, 81.31 ± 2.78, 91.31 ± 0.84, 89.44 ± 1.14, 90.69 ± 1.08, and 92.45 ± 0.87. To evaluate the variation in CD3+ T cells from day 0, fold change analysis was utilized, with findings depicted in Figure 5. Administration of a plasmid encoding GHRH resulted in a statistically significant change in the CD3⁺ T cell population over time (Friedman test, p < 0.0001), with post hoc analysis revealing significantly higher values at days 60, 120, and 180 compared with baseline (p < 0.05 for each).

Naïve CD4+ T-cell percentages in leukocytes were quantified at day 0, 30, 60, 90, 120, and 180, yielding means ± SEM of 33.33 ± 2.36, 31.53 ± 2.33, 35.87 ± 2.52, 38.70 ± 2.76, 36.85 ± 3.13, and 34.01 ± 2.35, respectively. A Friedman test revealed a significant change over time (p = 0.0255), with post hoc analysis (uncorrected Dunn’s test) showing a significant increase at day 90 compared with baseline (p < 0.05).

The proportion of naïve CD8+ T cells within leukocytes was evaluated across the same period, yielding percentages of 45.96 ± 3.98, 49.30 ± 3.93, 50.94 ± 3.72, 56.10 ± 3.91, 57.02 ± 3.59, and 54.86 ± 3.80 (mean ± SEM). A Friedman test revealed a significant change over time (p < 0.0001), and post hoc analysis (Dunn’s multiple comparison test) showed significant increases from day 90 through day 180 compared with baseline (all p < 0.005).

Additionally, Naïve CD8+ T-cell counts showed an upward trend by day 120 in both the IGF-1 Low Basal Group and High Basal Group (Supplementary Figure 2).

Safety evaluations involved monitoring skin reactions at the drug application site, analyzing CBC and serum chemistry for abnormalities, verifying UPC, and assessing adverse event reports from owners during the 180-day clinical trial period. No dogs exhibited abnormal skin symptoms at the site of drug administration. One subject showed mild hind limb tremors following GHRH plasmid injection, which resolved without intervention.

Over the 180-day observation period, five treated dogs exhibited atypical CBC results. Dog No.14 showed mild anemia with a hematocrit (HCT) of 36.6% (reference range [RR]: 37.3–61.7%) on day 30; values at all other time points were within the RR. On day 120, two dogs (No.22 and No.29) exhibited mild thrombocytopenia with platelet counts of 126 K/μl and 120 K/μl, respectively (RR: 148–484 K/μl). Dog No.30 exhibited severe thrombocytopenia with a platelet count of 22 K/μl (RR: 148–484 K/μl) but no clinical symptoms. One dog (No.25) showed pancytopenia (HCT 33%, RR: 37.3–61.7%; PLT 35 K/μl, RR: 148–484 K/μl; WBC 3.95 K/μl, RR: 5.05–16.76 K/μl) without clinical symptoms. In all five dogs, hematologic values returned to within the RR at subsequent evaluations without any treatment.

During each evaluation, serum chemistry parameters were assessed, including total protein, albumin, globulin, ALT, AST, ALP, GGT, total bilirubin, amylase, lipase, glucose, total cholesterol, triglyceride, BUN, creatinine, calcium, phosphorus, Na+, K+, Cl−, and venous pH. Results indicated that ALP levels increased in two dogs, while one dog demonstrated elevated ALT levels. No further abnormalities were observed beyond these instances. In one case of elevated ALP, levels doubled the upper normal limit between day 30 and 180. In another case, ALP levels tripled the upper threshold between day 60 and day 120. In both cases, ALP values returned to within the reference range by day 180 without any treatment. For the dog with elevated ALT, levels rose to double the upper limit between day 90 and day 180. No abnormalities were found in the UPC test performed on the 30 senior dogs participating in the study on day 180.

Review of adverse event forms revealed that 11 dogs (36.7%) experienced vomiting and 9 dogs (30.0%) experienced diarrhea during the 180-day period. Across the study, the 9 dogs with diarrhea had an average of 4.8 episodes each, and the 11 dogs with vomiting had an average of 2.5 episodes each. Two dogs exhibited transient anorexia accompanied by diarrhea on the same day, and one dog showed transient anorexia with reduced vigor.