Adeno-associated virus-vectored erythropoietin gene therapy for anemia in cats with chronic kidney disease

2.1 Recruitment

Four general and specialty veterinary clinics in the United States participated in the study between October 2019 and August 2021. The study was approved by the Institutional Animal Care and Use Committee at North Carolina State University (NCSU).

2.2 Adeno-associated virus vector

Codon-optimized feline EPO cDNA was cloned into an expression construct containing a chicken beta-actin promotor with a cytomegalovirus early enhancer, chimeric intron, and rabbit beta-globin polyadenylation sequence. The expression constructs were flanked by AAV2 inverted terminal repeats. Recombinant AAV serotype 1 virus vectors were generated via the triple transfection of HEK293 cells and iodixanol purification as previously described.¹⁶

2.3 Inclusion criteria

All cats enrolled were client owned and had International Renal Interest Society (IRIS) Stage 2, 3, or 4 CKD based on appropriate clinical findings (eg, small kidneys, polyuria, polydipsia), serum creatinine >1.6 mg/dL (http://www.iris-kidney.com/guidelines/staging.html), and a nonregenerative anemia with a PCV ≤25% consistent with anemia of CKD, which were confirmed during the screening visits. Chronic, concomitant diseases were stabilized before enrollment, in the opinion of the examining veterinarian, such that the cat was deemed likely to complete the study.

2.4 Exclusion criteria

Cats were excluded from study participation if they were: <2 years old, <2 kg, intended for breeding, pregnant, lactating, or had clinical pathology findings considered abnormal and clinically relevant, apart from evidence of CKD and previously identified stable chronic conditions. Specifically, cats were excluded if they had diagnosed or suspected acute kidney injury, clinically relevant or symptomatic cardiac disease, neoplasia (with the exception of suspected stable gastrointestinal lymphoma or inflammatory bowel disease), urinary tract infection, uncontrolled hyperthyroidism, uncontrolled diabetes mellitus, major surgery within the previous month or planned elective major surgery during the study period. Cats were not excluded if they had minor surgeries before Day 0 or during the study period if, in the opinion of the examining veterinarian, the surgery would not affect study assessments. Cats were also excluded if they had ever been treated with any ESA or had blood transfusions within 60 days before Day 0. Routine vaccinations had to be completed >7 days before Day 0.

2.5 Study design

The prospective, multisite, open-label study was planned to evaluate the efficacy and safety of SB-001 on CKD-associated anemia in client-owned cats and designed to support reasonable expectation of effectiveness for expanded conditional approval by the FDA. The numbers of animals were chosen to meet this requirement in a reasonable period of time.

There were 3 different treatment groups included in this study. Cats in the original dose cohort, Lo, were administered a single unit dose of 1.2 × 10⁹ genome copies (GCs) of SB-001 IM on Day 0. Subsequently, the protocol was amended, and new cats were enrolled in a second cohort, Lo ± Hi. These cats were administered a single unit dose of 1.2 × 10⁹ GC IM on Day 0 but also were administered a supplemental dose of 3.65 × 10⁹ GC on Day 42 if the PCV did not increase into the reference range or by at least 10 percentage points (pp; eg, from 19% to 29%) by Day 42. After review of the data from Lo ± Hi, the protocol was amended, and new cats were enrolled in a third cohort, Hi. These cats were administered a single dose of 3.65 × 10⁹ GC IM on Day 0 and supplemental dose of 3.65 × 10⁹ GC if the PCV did not increase into the reference range or by at least 10 pp by Day 42.

2.6 Screening

The screening visit was conducted between Day −14 and Day −2. A complete history (including concomitant treatments and signalment [i.e., age, breed, sex]), physical examination, and laboratory analysis including a CBC, clinical chemistry, symmetric dimethylarginine, thyroxine, feline leukemia virus, feline immunodeficiency virus, total iron, total iron binding capacity (TIBC), urinalysis, and urine culture were completed on each cat. Cats were fasted for at least 6 hours before blood sampling. At least 5 blood pressure measurements were taken during the visit; the highest and lowest measurements were discarded, and the mean was calculated from the remaining measurements. Cats were not screened for neutralizing antibodies to the capsid because of the possible protective nature of IM administration and impracticality of this testing in veterinary clinics. The intention was to develop a product that could be used in the presence of existing antibodies; therefore, there was no reason to test for these antibodies before enrollment. Samples were collected for later analysis in case it was deemed necessary.

2.7 SB-001 administration and follow-up examination

SB-001 was administered on Day 0 via single deep IM injection in either the quadriceps or epaxial muscles, based upon clinician preference. Telephone assessments with the owner were made on Day 7 ± 1 (and Day 49 ± 1 for cases receiving the supplemental dose). If the owner reported concerns regarding the site of injection, the cat was returned to the clinic for evaluation if appropriate.

All subsequent visits were within ±3 days of the target date. Cats enrolled in Lo were scheduled for visits at Days 0, 14, 28, 56, and 70; whereas cats enrolled in Lo ± Hi or Hi were scheduled for additional visits at Days 84, 98, 112, and 126 if they received the supplemental dose. At each visit, a complete physical exam, blood pressure measurement (as described above), and CBC were completed. Documentation of all medications administered, and adverse events were documented throughout the study. Chemistry profiles were performed on Days 14, 28, 42, 70, 98, and 126, and urinalyses on Days 28, 70, and 126.

On Days 0, 28, 56, and 70 (and Day 126 for cats receiving a supplemental dose), owners were also requested to complete a quality of life (QOL) assessment in which they were asked to score their cat's general health (ie, activity level, appetite, behavior, interactivity, playfulness, grooming) on Likert scales of 1 to 10. The average item-weighted impact score (AWIS) was calculated as the average value of the question and the impact score was the sum of the questions.

Total iron binding capacity was measured at screening, Day 28, and Day 70 for all cats and on Day 126 for cats receiving a supplemental dose in Lo ± Hi and Hi. Any cat with TIBC above or iron below the reference range on Day 0 or Day 28 or both was administered iron dextran every 3 to 4 weeks, dosed according to the attending clinician's discretion, and administered IM at a different site to SB-001; additional injections could be administered at the examining veterinarian's discretion. Transferrin saturation (TSAT) was calculated by total iron divided by TIBC multiplied by 100. Iron deficiency was defined by a TSAT < 20%.

Injection site evaluations were done by the examining veterinarian on Day 0 before treatment and on Days 14 and 70, and, if possible, at any unscheduled visit that resulted in withdrawal from the study. For cats administered a supplemental dose of SB-001, that injection site was evaluated before treatment and on Day 56 and Day 126.

Predetermined reasons for withdrawal after enrollment included: owner removal of consent; severe anemia (ie, PCV ≤ 14%) or polycythemia (ie, ≥65%), adverse events prohibiting further participation in the study; treatments potentially affecting study outcome (eg, ESA, blood transfusions); owner perceived lack of therapeutic effect; welfare reasons as decided by the veterinarian; or other reasons (eg, lost to follow-up). Cats received study-related veterinary care and medication free of charge.

2.8 Assessment of efficacy and safety

Increases in PCV from screening and Day 0 were analyzed to determine the efficacy of SB-001. The screening and Day 0 values were compared to Days 14, 28, 42, 56, and 70 for all cats. For cats administered a second dose on Day 42, the screening and Day 0 values were also compared to Days 84, 98, 112, and 126. A responder was defined as a cat that achieved a PCV in the reference range (ie, >28.2%; primary goal) or an increase in PCV of >5 pp from the PCV on Day 0 (secondary goal). A nonresponder did not achieve either of these goals.

2.9 Analytical methods

All clinical pathology samples (including PCV) were sent to a central commercial laboratory (IDEXX Laboratories, Inc) for analysis.

2.10 Pre-existing antibodies

Any animal that required the supplemental dose on Day 42, was evaluated post hoc for AAV1 neutralizing antibodies using stored samples collected during the screening process. The assay was performed by the University of Pennsylvania Gene therapy program as per confidential standard operating procedure.

2.11 Animal welfare

All owners signed an informed consent document detailing the known risks involved in the study. Cats remained in the care of their owners. They were checked regularly by the examining veterinarian. The owner, the examining veterinarian, or an investigator was able to withdraw the cat from the study at any point if it required any medical procedure or therapy not authorized by the study protocol or for any welfare concern. The protocol required that any cat with severe anemia (ie, PCV ≤ 14%) or polycythemia (ie, ≥65%) be medically managed based on current standard of care. Adverse events defined as any undesirable reaction or adverse effect observed after the administration of SB-001 were recorded and appropriate medical therapy was provided.

2.12 Data analysis

All tests of significance, unless otherwise stated, were performed at alpha = .05, 2-sided. All data calculations were generated using SAS version 9.4 (SAS Institute, Cary, North Carolina). The primary effectiveness variable to evaluate the improvement of anemia was PCV values. All other variables were considered secondary.

For the primary variable, PCV (%), the Day 0 values were compared to Days 14, 28, 42, 56, 70, 84, 98, 112, and 126. Descriptive statistics were calculated for each study day and for the changes from Day 0 to each subsequent study day. For each cohort, possible differences between the changes from Day 0 were assessed by repeated measures analysis of variance modeling. The models included study day as a fixed effect and site as a random effect. The normality of the residuals from each model was investigated using the Shapiro-Wilk test. If the P value was ≤.01, then the values were ranked before generating the model.

Changes in QOL variables from Day 0 were calculated for each cat. Descriptive statistics were evaluated for each visit and for the change from Day 0 to Day 28, Day 56, and Day 70 (and Day 126 if applicable). A change in QOL (ie, AWIS and impact scores) was evaluated between responders and nonresponders using analysis of variance with response as a fixed effect. These questions were also evaluated at each time point compared to Day 0 as to whether they were improved (+1), unchanged (0), or decreased (−1) and the cases were given a sum total score at each time point.

For overall survival, the median time to death was calculated using the Kaplan-Meier method and survival curves were calculated by cohort, IRIS stage, all cats, each responder definition, and by IRIS stage within each cohort and all cats. For each analysis, possible differences in the median time to death were assessed by the log-rank test.

For the response definition, the number and percentage of responders/nonresponders and the 95% confidence interval were calculated by cohort and overall. Descriptive statistics were prepared for responders/nonresponders for body condition score at Day 0, serum iron status at Days 0, 28, 42, and 70 during the study and the number of medications. Possible differences between responders and nonresponders were assessed by analysis of variance.

3.1 Demographics

Thirty-two cats were screened for the study; 9 (25%) did not meet the inclusion criteria for the following reasons: higher than allowed PCV (n = 4); feline immunodeficiency virus positive (n = 1); cardiac failure (n = 1); blood transfusion <60 days before Day 0 (n = 1); euthanasia before Day 0 (n = 1); or death by natural causes (n = 1). A total of 23 cats met the criteria for inclusion based on protocol compliance at 4 different sites: NC State Veterinary Hospital, Raleigh, North Carolina (n = 10); Friendship Hospital for Animals, Washington, DC (n = 7); VCA Sacramento Veterinary Referral Center, Sacramento, California (n = 4); and Bulger Veterinary Hospital, Lawrence, Massachusetts (n = 2). The 23 cats enrolled in the study were mostly domestic shorthair cats, with a larger percentage of males than females (Table 1). The mean age of enrolled cats was 13.8 ± 4.57 years. There were 2 CKD Stage 2 cats, 12 CKD Stage 3 cats, and 9 CKD Stage 4 cats.

Cats were enrolled by order of presentation into the following treatment groups: Cats in Lo (n = 6) received a single dose on Day 0. Nine cats were enrolled in Lo ± Hi; 3 responded to the single dose on Day 0 and did not require the supplemental dose, whereas 5 cats received both the Day 0 and Day 42 dose. Eight cats were enrolled in Hi; 1 received the supplemental dose on Day 42 but was euthanized because of uremic crisis before it was reevaluated.

3.2 Change in PCV

For all cats combined, there was a significant (P < .04) increase in PCV from Day 0 to Day 28 (mean increase 5.67 ± 6.36 pp, n = 21), Day 42 (mean 7.57 ± 8.87 pp, n = 21), Day 56 (mean 9.59 ± 11.0 pp, n = 17), and Day 70 (mean 13.1 ± 13.5 pp, n = 14; Figure 1).

The cohort of Lo cats (n = 6) exhibited a 6.33 pp increase in PCV (P = .04) by Day 28 (Table 2).

Individual cats in Lo ± Hi (n = 9) demonstrated increases in PCV but the mean PCV of the cohort was not significantly different from baseline at any timepoint. When data from cats administered a supplemental dose at Day 42 was evaluated independently, the PCV did not increase significantly from baseline or from Day 42 (mean increase of 6.60 ± 6.61 pp from Day 42 to Day 70, n = 5, P = .1). Attrition of this study sample might have impacted the statistical analysis: 6 cats alive at Day 56, 5 alive at Day 70, 4 at Day 84, 3 at Days 98 and 112, and 2 at Day 126.

A significant increase in PCV from Day 0 was noted in the cohort of Hi cats (n = 8) at Day 28 (P = .05), Day 42 (P = .01), and Day 70 (P = .004) (Table 2).

3.3 Response

For all cats, the response rate to injection of SB-001 (Figure 2) was 86% (n = 19/22; 95% CI, 65%, 97%) defined as achieving the primary or secondary goal at any time from Day 28 to Day 70. Only 4 cats failed to achieve any response, 1 of which was euthanized on Day 14 and not included in the response analysis. For cats exiting the study early, data from the last visit were carried forward regardless of the day of exit.

Cats reaching the primary goal at Day 42 had a higher mean body condition score at Day 0 (4.5 vs 2.75 out of 9, P = .02).

3.4 Response by cohort

Five of 6 cats in Lo achieved either the primary (n = 2) or secondary (n = 5 including the 2 meeting the primary goal) goals at any time between Day 28 and Day 70. Two cats from Lo achieved the primary goal by Day 28 and maintained this magnitude of response until the end of the study. In 1 of the 3 cats achieving only the secondary goal, the response was noted at presentation for euthanasia and might have represented hemoconcentration. In the other 2 cats, the secondary goal was not sustained beyond Day 28. One of the Lo cats failed to respond at any time point.

Six of 8 cats in Lo ± Hi achieved the primary goal and maintained it until either the end of the study (n = 3) or death (n = 3). In 1 additional cat, the secondary goal was met at presentation for euthanasia and might have represented hemoconcentration. A response was not achieved in 2 Lo ± Hi cats. Four out of 5 cats receiving a supplemental dose achieved the primary goal and 1 failed to respond. One cat was euthanized on Day 14 and was not included in the analysis of response.

An overall response rate of 7/8 (88%, 95% CI, 47%, 100%) was seen in Hi, with 6 cats meeting the primary goal and 1 cat meeting only the secondary goal at any time from Day 28 to Day 70. In 5 of these cats, the primary goal was achieved by Day 28 and sustained until the end of the study (n = 3) or death (n = 2). In addition, 1 cat achieved and sustained the primary goal and 1 cat achieved and sustained the secondary goal by Day 42 while 1 cat failed to respond.

Stable inflammatory bowel disease or GI lymphoma was present in 2 Lo (1 responder, 1 nonresponder), and 2 Lo ± Hi (both responders) cats.

3.5 Quality of life

The AWIS and the impact score were higher on Day 28 for responders (Table 3). Most responders had an increased sum total score (Table 4). Many owners struggled with completing the questionnaires, especially as the study progressed. The responses to the questionnaires became more difficult to evaluate as the animals developed progressive CKD and comorbidities. It was not uncommon for 1 or more clinical variables to improve even when the cat's overall condition and QOL deteriorated.

3.6 Serum iron status

Iron status was not different between responders and nonresponders at Day 70. Iron deficiency (TSAT < 20%) at screening, Day 28, and Day 70, was noted in 2, 8, and 7 responders, and 1, 3, and 1 nonresponders respectively.

3.7 Pre-existing antibodies

The 5 Lo ± Hi cats that received the supplemental dose were analyzed post hoc and negative for AAV1 neutralizing antibodies at screening. No other cats were evaluated for neutralizing antibodies.

3.8 Medications potentially affecting feline EPO response

Medications associated with hyporesponsiveness to ESAs were administered to both responders and nonresponders (defined at Day 70) during the study period. Specifically, renin-angiotensin-aldosterone system (RAAS) inhibitors (enalapril maleate, telmisartan) were administered to 2 responders and 3 nonresponders. Aluminum hydroxide was administered to 7 responders and 5 nonresponders.

3.9 Adverse events

Most adverse clinical signs noted in the study population could be attributed to the CKD or comorbidities. The events often reflected the target population (ie, elderly cats with CKD). Comorbidities were identified in responders and nonresponders; however, nonresponders received significantly more concurrent medications (excluding iron) than responders at Day 42 (P = .01, by primary goal; P = .04, by secondary goal) and on Day 70 (P = .01 by primary goal) suggesting they might have had more severe disease.

Polycythemia developed in 3 cats (1 in Lo, 2 in Hi): 2 cats developed a PCV of 66% and 57%, respectively on Day 70, and 1 additional cat peaked at 53% on Day 85. None of these cats required treatment for hypertension. The cats with polycythemia were followed after the study end, to Days 82, 125, and 135 with final PCVs of 67%, 63%, and 53%, respectively. None of these cats required therapeutic intervention for polycythemia at any time.

Moderate-to-severe hypertension (ie, systolic blood pressure ≥160 mm Hg) was observed in 12 of 23 cats at Day 0. Antihypertensive medications were administered to 4 of these cats before enrollment, initiated in an additional 6 of these cats and reinstated in 1 cat who had previously been treated but the treatment had been discontinued giving a total of 10 cats treated for hypertension by or after Day 0. An additional 6 cats developed hypertension during the study; only 5 cats remained normotensive. Hypertension with signs of encephalopathy, including seizures, was noted in 4 cats (3 responders by primary goal, 1 responder by secondary goal). The 4 cats that developed signs of encephalopathy were hypertensive at Day 0 (Day 0 mm Hg: 178, 184, 187, and 184, respectively). Once hypertension was confirmed and situational hypertension excluded, all hypertensive cats in the study were treated with either amlodipine or telmisartan (agent and dose determined by clinician discretion); except for 1 cat that could not be orally medicated. Most cats remained in the same or moved to a lesser IRIS hypertension substage during the study, including the cats that developed signs of encephalopathy. Exacerbation of hypertension did not appear to directly correlate with the dose of SB-001 or the increase in PCV.

No adverse events were noted in association with SB-001 injection. Injections were well tolerated; cats did not need additional sedation for the injection.

3.10 Survival

Median days alive for cats completing Day 28 was 130.5 (range, 34-454; n = 22). Overall median survival time for all cats was 126 days (n = 23). There was no difference between median survival of cats responding by primary goal at any time from Day 28 to Day 70 (141.5 days, n = 14) vs nonresponders (63 days, n = 8; P = .12). Median survival times were not different between the 3 cohorts (Figure 3) or between cats of different IRIS CKD stage.