Test of Rapamycin in Aging Dogs (TRIAD): study design and rationale for a prospective, parallel-group, double-masked, randomized, placebo-controlled, multicenter trial of rapamycin in healthy middle-aged dogs from the Dog Aging Project

Abstract

Companion dogs are a powerful model for aging research given their morphologic and genetic variability, risk for age-related disease, and habitation of the human environment. In addition, the shorter life expectancy of dogs compared to human beings provides a unique opportunity for an accelerated timeline to test interventions that might extend healthy lifespan. The Test of Rapamycin In Aging Dogs (TRIAD) randomized clinical trial is a parallel-group, double-masked, randomized, placebo-controlled, multicenter trial that will test the ability of rapamycin to prolong lifespan and improve several healthspan metrics in healthy, middle-aged dogs recruited from Dog Aging Project participants. Here, we describe the rationale, design, and goals of the TRIAD randomized clinical trial, the first rigorous test of a pharmacologic intervention against biological aging with lifespan and healthspan metrics as endpoints to be performed outside of the laboratory in any species.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11357-024-01484-7.

Introduction

Companion dogs represent a powerful translational model for aging research [1–3]. Because of their faster biological aging compared to people, dogs represent an opportunity to investigate age-related disease and factors contributing to mortality on an accelerated timeline. In addition, companion dogs are exposed to similar environmental factors as their human counterparts and experience age-related functional decline and diseases that mirror those of people [4]. In aging people, frailty, an age-associated multidimensional syndrome, is a well-documented predictor of poor health outcomes, functional deterioration, declining quality of life, and mortality [5–10], and dogs also experience a syndrome of frailty and many of these same age-related functional declines [11, 12]. Dogs also benefit from access to a relatively sophisticated veterinary healthcare system [13]. Finally, their inherent variability in size, conformation, behavior, and spectrum of disease susceptibility makes companion dogs a more suitable model for human aging than traditional laboratory species.

The Dog Aging Project (DAP) is an interdisciplinary, open-data, community science research endeavor that seeks to understand the genetic and environmental determinants of healthy aging in companion dogs, both to inform veterinary medical practice and to capitalize on the translational potential of discovery in this species [14]. The DAP has assembled a robust infrastructure to support participant engagement, scientific collaboration, data management, and biospecimen collection and analysis. Additionally, it has engaged dozens of investigators, staff, and trainees, thousands of practicing veterinarians, and nearly 50,000 participating dog-owner pairs in this productive enterprise (https://scholar.google.com/citations?user=gMu1hWQAAAAJ&hl=en&authuser=3&inst=14379318592444324147). The DAP evaluates the aging trajectory at the population, individual organism, and molecular levels, to understand not only the phenotypic variation of aging among dogs, but also the underlying mechanisms that produce diverse age-related outcomes. Deeper knowledge about the biologic basis of aging and age-related disease will create new opportunities to improve the prevention, prediction, diagnosis, and treatment of age-related diseases in people and dogs. Discovery of interventions—physical, social, dietary, pharmaceutical or otherwise—is an important target of this endeavor.

The drug rapamycin is efficacious in increasing lifespan and delaying or reversing multiple functional and molecular phenotypes of aging in laboratory mice and other short-lived species. Rapamycin inhibits the protein kinase mTOR (mechanistic target of rapamycin), an essential and central regulator of growth and metabolism that impacts a host of cellular functions [15]. Importantly, rapamycin is effective in extending lifespan and improving indices of health when initiated in middle-age and when administered transiently or intermittently [16–18]. As a nested cohort within the DAP, the Test of Rapamycin In Aging Dogs (TRIAD) randomized clinical trial is designed to evaluate the effects of rapamycin on lifespan, functional measures of aging, and age-related disease burden (i.e., “healthspan”) in healthy, mature adult to senior, medium-to-large companion dogs. A few studies support a role for mTOR in age-related conditions affecting dogs. For example, in a study of valve interstitial cells from dogs with chronic myxomatous mitral valve disease—the most common age-related heart disease of this species—transition of quiescent valve interstitial cells (VICs) to activated VICs is associated with upregulation of PI3K/AKT/mTOR, and antagonism of PI3K/AKT/mTOR reversed VIC-to-myofibroblast transition by inhibiting senescence and promoting autophagy [19]. In addition, increased expression of the mTOR/4E-BP1/eIF4E pathway has been demonstrated in canine prostatic carcinoma [20].

The TRIAD randomized clinical trial will be the first clinical trial of the effects of a pharmacologic intervention on normative aging, with lifespan as the primary endpoint and functional measures of aging as secondary endpoints. Among veterinary clinical trials, the TRIAD randomized clinical trial is unique in its size; in its centralized methods for recruitment, eligibility determination, and owner consenting and communication; and in its robust data and safety monitoring oversight. The DAP and the TRIAD investigators within this larger program have a core commitment to open data sharing within the scientific community [14].

Here, we describe the rationale and design of the ongoing TRIAD randomized clinical trial.

Background and rationale

A key tenet of geroscience—the interdisciplinary field concerned with aging biology and its impact on chronic disease and health—is that interventions targeting the molecular mechanisms of aging will delay or prevent age-related functional decline and disease [21, 22]. The mechanistic target of rapamycin is a nutrient- and stress-responsive intracellular kinase that is a critical regulator of growth, metabolism, autophagy, and aging, making it a molecular target with high potential for translational geroscience applications [15, 23]. The kinase is a component of two complexes, mTOR-complex-1 (mTORC1) and mTOR-complex-2 (mTORC2) that differ in composition, in protein targets, in signaling input, and in sensitivity to rapamycin. As a component of mTORC1, mTOR activity triggers cell growth and proliferation through intracellular pathways in response to multiple environmental nutrient and growth cues, acting as a hub for intracellular signaling so that the cell can appropriately dedicate cellular resources to align with prevailing conditions [24]. Inhibition of mTORC1 by rapamycin, the macrolide antibiotic to which mTOR owes its name, leads to downstream changes in glucose and lipid metabolism, general decreases in mRNA translation, and increased autophagy [23], all of which contribute to the extensive immunomodulatory and anti-inflammatory effects of this compound. Rapamycin, also known as sirolimus, is approved by the United States Food and Drug Administration for the treatment of certain neoplasms and for the prevention of renal transplant rejection and cardiac stent restenosis [24].

Rapamycin and its derivatives (i.e., “rapalogs”) delay aging and extend lifespan in multiple laboratory species, including four of the most critical models used in biomedical research: the budding yeast Saccharomyces cerevisiae, the nematode Caenorhabditis elegans, the fruit fly Drosophila melanogaster, and the laboratory mouse Mus musculus [15, 23]. In mice, lifelong administration of rapamycin increases lifespan, [25, 26] and its initiation in middle age appears to be equally or nearly as effective [16]. At least two studies in mice demonstrate that transient treatment with rapamycin during middle age is sufficient to robustly increase lifespan [18, 27]. One study treated C57BL/6Nia mice with rapamycin for 3 months beginning at 20–21 months of age and observed increases in life expectancy observed increases in life expectancy ranging from 39-61%, which corresponded to absolute mean lifespan extension of 9–14% [18]. In another, 22–24 month-old C57BL/6 J mice were treated with rapamycin for 6 weeks, resulting in a significant increase in lifespan of several months [27].

In addition to longer lifespan, slowing of age-related declines in cardiac [28–30], cognitive [31], renal [32], immune system [27], and intestinal [18, 33] function have been reported following treatment with rapamycin or rapalogs. With respect to the heart, treatment of mice with rapamycin for 3 months late in life is associated with significant improvements in cardiac function as assessed by echocardiography, gene expression studies, and inflammatory cytokine profiling [30]. In addition, mTORC1 inhibition might protect against or reverse certain cardiomyopathies, as suggested by studies showing benefit in murine models of left ventricular hypertrophy induced by pressure overload or thyroid hormone [34–36], myocardial ischemia/reperfusion injury [37], hypertrophic cardiomyopathy [38], and dilated cardiomyopathy [39–42]. Most recently, the results of a double blind, placebo controlled, randomized study of cats with naturally occurring subclinical hypertrophic cardiomyopathy suggested that treatment with rapamycin might prevent or delay progression of left ventricular hypertrophy [43].

Rapamycin also has benefit in mouse models of Alzheimer’s disease [44], including improvements in neuropathology [45], microvascular density and flow [46, 47], physiology [48], and cognitive performance [46–48]. Canine cognitive dysfunction syndrome is characterized by cognitive decline in several domains including memory, spatial awareness, learning, and normal behaviors; as well as disruption of the sleep/wake cycle, anxiety, and altered activity levels [49, 50]. There is growing appreciation of the suitability of dogs as a clinical [51–53] and histopathological [54–57] model for Alzheimer’s disease. Work in aging purpose-bred beagles as a model of Alzheimer’s disease has identified impairments in learning and memory that begin in middle age and are associated with amyloid-beta deposition in the brain [58]. Although age-related changes in dog cognition have been observed across a wide variety of tasks [59], declines in working memory, spatial memory, and executive functions are among the most robustly documented [51].

There are limited data regarding the long-term use of rapamycin in dogs. It appears to be well tolerated at doses of up to 1 mg/kg/day for up to 14 months [60, 61]. To date, the results of three randomized clinical trials of systemic rapamycin in companion dogs have been published: two primarily focused on establishing safety [62, 63] and one evaluating rapamycin as adjunctive chemotherapy for treatment of spontaneous appendicular osteosarcoma [64]. In one study [62], 24 healthy middle-aged dogs received placebo or rapamycin at 0.05 mg/kg or 0.1 mg/kg by mouth three times weekly for 10 weeks. In the second [63], 17 healthy middle-aged dogs were administered placebo or rapamycin at 0.025 mg/kg by mouth three times weekly for 6 months. In the third [64], 152 dogs with osteosarcoma were treated with 0.1 mg of rapamycin/kg by mouth four times weekly for up to 16 weeks. In all three studies, dogs treated with rapamycin experienced a low incidence of adverse events (AE), similar to that observed in control groups.

In a small, randomized, double-masked, placebo-controlled study of 24 healthy dogs [62], a statistically significant increase in fractional shortening (an echocardiographic index of left ventricular systolic performance) was documented in rapamycin- compared to placebo-treated dogs over an 11-week study period, suggesting potential cardiac benefit of this drug [62]. This finding was not repeated in a subsequent small, randomized, double-masked, placebo-controlled study of 17 healthy dogs using a lower dose but longer duration of rapamycin treatment [63].

A final consideration is that of dosing frequency and the goal of avoiding the immunosuppression and glucoregulatory dysfunction that can be associated with clinical use of rapamycin while retaining its longevity promoting effects. There is growing consensus that once-weekly dosing with rapamycin is not immunosuppressive and avoids off-target inhibition of mTORC2, the latter proposed to mediate many of rapamycin’s unwanted metabolic side effects. Once-weekly treatment in cats with naturally occurring hypertrophic cardiomyopathy produced dose-responsive transcriptomic and proteomic effects [65] and decreased progression of left ventricular hypertrophy with an adverse effect profile equivalent to that of placebo [66]. Furthermore, evidence from human studies suggests that once weekly dosing of the mTOR inhibitor RAD001 is effective while reducing risk of adverse events when compared to daily dosing [67]. Finally, in a survey-based study comparing 333 people with a history of off-label rapamycin use—with the most frequent dose being 6 mg once per week—to 172 non-users, the risk of bacterial or viral infection was no different between groups [68], indicating that once weekly lower dose rapamycin treatment might avoid its known immunosuppressive effects.

The TRIAD randomized clinical trial is designed to test definitively whether rapamycin can prolong lifespan and healthspan in dogs. A positive outcome for this trial would justify similar trials of rapamycin in people. In addition, if improvement or prevention of decline in cardiovascular health parameters are documented in a large population of dogs, results would support existing evidence from laboratory mice that rapamycin can improve age-related deterioration in cardiac function [29, 30].

Objectives and hypotheses

The primary objective of this study is to determine whether once-weekly oral administration of rapamycin increases lifespan in medium-to-large- or large-breed, mature adult or senior [69] companion dogs. As a secondary objective, this study will assess the ability of rapamycin to improve functional metrics of healthspan including physical function, neurological and neurocognitive status, and cardiovascular health. We hypothesize that compared to placebo, treatment with rapamycin will increase lifespan and reduce age-related disease burden. Finally, this study seeks to evaluate the safety of rapamycin when administered once weekly for 12 months.

Study methods

Trial design and overview

The TRIAD randomized clinical trial is a prospective, double-masked, placebo-controlled, parallel group, multicenter study. At the time of writing, TRIAD study sites include eight veterinary teaching hospitals and 11 non-academic, specialty veterinary practices in the USA, with ongoing recruitment of new clinical sites. Dogs are randomized in equal numbers to receive rapamycin or placebo by mouth once weekly for 12 months, with a subsequent 24-month post-treatment monitoring period, for a total study duration of 36 months. For the duration of the study, veterinary examinations and owner-administered at-home cognitive assessments occur at 6-month intervals. In addition, participating owners complete at-home observation surveys monthly for the first 18 months of the study period, and then every 3 months for the remaining 18 months. For all dogs, comprehensive evaluations of health indices and disease risk are documented along with assessments of physical function and frailty. Approximately one-half of all dogs will undergo comprehensive assessments of cardiovascular function at specialty cardiology clinics. The remaining half will undergo assessments of neurologic and cognitive function at specialty neurology clinics. Overall study design is illustrated in Fig. 1. All study procedures have been approved by a Data & Safety Monitoring Board (DSMB) constituted by the National Institutes on Aging, as well as by the Texas A&M University Institutional Animal Care and Use (IACUC) and Clinical Research Review Committees. Each participating practice or institution (“clinical trial site”) has agreed to adhere to the trial design and to the Animal Use Protocol approved by the Texas A&M University IACUC. Each clinical trial site has also obtained local approval of the animal use protocol, without modification, by their own IACUC or oversight body for research with privately owned animals, as required by their local policies and procedures.

Fig. 1.

General design of the Test of Rapamycin In Aging Dogs randomized clinical trial. Dogs enrolled in the Dog Aging Project (“DAP Pack”) are automatically considered for participation; alternatively, owners of dogs not enrolled in the DAP who are preferentially interested in TRIAD can complete a TRIAD eligibility pre-screening questionnaire if they wish to ascertain likely eligibility before joining the DAP. In either case, dogs must be enrolled in the DAP prior to formal eligibility assessment.

Animals

Eligible participants are client-owned, reproductively sterilized dogs of any breed background, aged 7 years or older and weighing 20 to 55 kg, inclusive, that are of typical health status for their age [70] and receive regular preventive healthcare [71]. Participation of dogs with mild, stable, age-related conditions (e.g., early, non-azotemic, non-proteinuric, non-hypertensive chronic kidney disease) for which treatment with rapamycin would not be expected to increase risk of deterioration, or for which drugs used in their treatment would not be expected to lead to drug-drug interactions with rapamycin, is allowed provided all other enrollment criteria are met. Detailed enrollment criteria are presented in Tables 1, 2, and 3.

Table 1.

TRIAD enrollment criteria. To be eligible for participation, dogs must meet all inclusion criteria. A dog that meets one or more of the exclusion criteria is excluded from participation. SBP, average session systolic arterial blood pressure

Inclusion criteria

Exclusion criteria

• ≥ 7 years of age • 20–50-kg body weight • Reproductively sterilized • Receiving regular heartworm preventivea if resident of endemic region • Vaccinated according to American Animal Hospital Association guidelines,b or demonstration of protective antibody titers for distemper, parvovirus, or canine adenovirus • In good general health as evidenced by veterinarian review of medical record and results of screening examination • Owner consents to the following: o Provide regular heartworm preventivea for study duration regardless of geographic location o Maintain vaccination or protective antibody titers for study duration o Administer study medication weekly for the duration of the 12-month treatment period o Transport dog to clinical trial site every 6 months for the duration of the 36-month study period

• History of hospitalization in the 3 months preceding screening • Treatment with drugs or supplements not allowed by study protocol • Historical or clinical findings suggestive of the following: o Progressive systemic illness o Clinically significant, chronic systemic illness o Mild, chronic, age-related systemic illness for which treatment with rapamycin might increase risk of deterioration, or for which drugs used for treatment might lead to drug-drug interactions with rapamycin • Malignant neoplasia • Heartworm infection • Trypanosoma cruzi seropositivity (dogs evaluated in Texas) > 1:40 or ≤ 1:40 if from high-risk region. Not excluded are dogs without cardiac abnormalities that are not from a high-risk region and have low (i.e., 1:20 or 1:40) T. cruzi titers that are stable or decline when rechecked in 2–3 weeks • Systemic arterial hypertension (SBP ≥ 180 mmHg or ≥ 160 mmHg with evidence of target organ damage) • Immature, mature or hypermature cataracts • Dementia or senility as determined by validated questionnaires • Temperament that prevents safe handling for examination • Known sensitivity to rapamycin • Dogs evaluated at specialty cardiology clinics: echocardiographic or electrocardiographic evidence of clinically significant heart disease, pulmonary arterial hypertension, or arrhythmia (see Table 2) • Dogs evaluated at specialty neurology clinics: clinically significant neurologic disease or sensory or mobility deficits that prevent accurate neurocognitive assessment (see Table 3)

^aAmerican Heartworm Society, Heartworm Preventives, last accessed July 2024 (www.hearwormsociety.org/preventives)

^bEllis J, Marziani E, Aziz C, Brown CM, Cohn LA, Lea C, Moore GE, Taneja N. 2022 AAHA Canine Vaccination Guidelines. J Am Anim Hosp Assoc. 2022 Sep 1;58(5):213–230

Table 2.

Allowed and exclusionary cardiovascular system findings at the time of screening examination, for dogs screened at specialty cardiology clinics. ECG, electrocardiography; AV, atrioventricular; VPC, ventricular premature complex; AIVR, accelerated idioventricular rhythm; HR, heart rate; LV, left ventricular; LA, left atrial; RV, right ventricular, RA, right atrial; RVOT, right ventricular outflow tract; LVOT, left ventricular outflow tract; Vmax, peak velocity; CHD, congenital heart disease; PDA, patent ductus arteriosus, VSD, ventricular septal defect

Source of findings	Allowed	Exclusionary (not exhaustive)
ECG findings	• Sinus rhythm or sinus arrhythmia • Sinus rhythm conducted with bundle branch block • Single atrial premature complexes • First-degree AV block (PR interval > 0.13 s) • Mobitz type I second-degree AV block • Single ventricular premature complexes meeting Holter criteria below	• Premature supraventricular ectopy > single complexes • Atrial fibrillation or flutter • Mobitz type II second-degree or third-degree AV block • Atrial standstill • Evidence of sinus node dysfunction • Premature ventricular ectopy warranting anti-arrhythmic therapy at screening clinician’s discretion • Premature ventricular ectopy meeting exclusionary Holter criteria below
Ambulatory ECG (Holter) findings (if indicated)	• < 50 single VPCs/24-h recording period (any breed) • Ventricular couplets, triplets, or AIVR with fastest instantaneous HR < 120 bpm if < 50 VPCs/24 h AND not of a breed (i.e., Boxer dog, Doberman pinscher, Great Dane) predisposed to cardiomyopathy • Unexpected abnormalities that do not warrant anti-arrhythmic therapy at screening clinician’s discretion • Findings appropriate for assumed physiologic setting that do not warrant anti-arrhythmic therapy at screening clinician’s discretion	• ≥ 50 VPCs/24 h, regardless of complexity and rate • Ventricular bigeminy, trigeminy, couplets, triplets or AIVR in breeds predisposed to cardiomyopathy • Ventricular couplets, triplets, tachycardia, or AIVR with fastest instantaneous HR ≥ 120 bpm (any breed) • Unexpected findings warranting anti-arrhythmic therapy at screening clinician’s discretion
Echocardiographic findings
Acquired conditions/situations	• Mitral or aortic valve regurgitation due to degenerative disease, provided LV and LA sizes and indices of LV systolic performance are normal • Tricuspid or pulmonic valve regurgitation with normal RV and RA sizes • Equivocally or mildly increased peak ventricular outflow tract velocities (RVOT or LVOT Vmax < 2.3 m/s in a non-Boxer dog or < 2.4 m/s in a Boxer dog)	• Valvular insufficiency with chamber enlargement • Primary or secondary cardiomyopathies (Supplemental Tables 1 and 2) • Cardiac tumors/masses • High probability of pulmonary arterial hypertension (Supplemental Table 3) • Vegetative endocarditis
Congenital conditions	• Mild forms of common CHD for which surgical or medical therapy is not recommended by screening clinician AND chamber sizes and indices of ventricular systolic performance are normal: o Subaortic or pulmonary stenosis (Vmax < 3.5 m/s) o Corrected PDA o Isolated small VSD o AV valve dysplasia	• Any CHD warranting surgical or medical therapy at screening clinician’s discretion • Uncorrected left-to-right shunting PDA • Corrected left-to-right shunting PDA with chamber enlargement or abnormal LV systolic performance • Right-to-left shunting PDA • Tetralogy of Fallot or other conotruncal abnormalities

Table 3.

Allowed and exclusionary comprehensive neurological findings at the time of screening examination, for dogs screened specialty neurology clinics. CADES, CAnine DEmentia Scale

Source of findings	Allowed	Exclusionary
Neurologic examination findings	• In animals without detectable gait abnormalities: o Proprioceptive deficits of one or both pelvic limbs o Abnormal cutaneous trunci muscle cutoff o Symmetrical patellar reflex deficits • Abnormal resting pupil size or pupillary light reflex caused by iris atrophy • Ventrolateral strabismus apparent on head elevation only • Mild head tilt absent other signs of vestibular disease • Mild pain on palpation of axial structures	• Abnormal level of arousal • Abnormal behavior (e.g., stereotypy, inappropriate aggression) • Gait abnormalities limiting ability to ambulate 50 feet unassisted • Gait abnormalities considered likely to alter social interaction and/or elimination behaviors • Visual impairment • Hearing impairment • Pathological nystagmus • Asymmetrical atrophy of muscles of mastication • Thoracic limb proprioceptive deficits • Moderate or severe pain on palpation of axial structures
Congenital conditions		Conditions resulting in one or more exclusionary examination findings

Dogs can enter the recruitment process for the TRIAD randomized clinical trial in one of two ways. First, all dogs enrolled in the DAP are automatically considered for TRIAD. Second, owners who are preferentially interested in TRIAD can complete a TRIAD eligibility pre-screening questionnaire as an initial step, if they wish to ascertain likely eligibility before joining the DAP. In either case, dogs must be enrolled in the DAP prior to formal eligibility assessment. Owners complete a comprehensive survey detailing their dog’s health and life experiences (https://github.com/dogagingproject/dataRelease/tree/master/Survey_Instruments/HLES) at the beginning of their participation in the DAP. In addition, owners must provide a qualifying veterinary electronic medical record. Nomination, consent, and cohort assignment of dogs participating in the TRIAD randomized clinical trial are aligned with those of the DAP and have been described in detail elsewhere [14]. Following automated eligibility assessment of the survey data provided, owners of apparently eligible dogs are asked if they wish to be considered for the TRIAD randomized clinical trial (unless they have already indicated their interest by completing the pre-screening eligibility questionnaire). Owners of apparently eligible dogs are provided a list of participating clinical trial site locations, without regard to cardiology or neurology specialty, and indicate which site they could attend for the total number (i.e., 7 in 3 years) of required in-person study visits if selected. Interested owners then submit comprehensive medical records for the preceding 3 years (or the length of time they have owned the dog if less than 3 years), which are reviewed for exclusion criteria by a veterinarian trained in the assessment of these criteria. After undergoing the informed consent process described below, owners of qualified dogs are invited to attend an in-person screening/baseline examination at the clinical trial site they indicated as most convenient. A limited number of TRIAD clinical sites offer both cardiology and neurology specialty services; in such cases, the candidate dog is offered the first available appointment without regard to specialty. This approach ensures that neither the owner nor any member of the TRIAD team dictate the secondary outcomes for which qualifying dogs will be assessed. A total of 580 dogs will be enrolled and randomized.

Informed owner consent

Informed owner consent is obtained by TRIAD trial staff prior to in-person screening and enrollment. Owners of potentially eligible dogs review a consent document and watch an informational video that provides an overview of study design, objectives, procedures, and the consent process itself. Owners are required to successfully complete a 10-question comprehension assessment and to discuss any questions or concerns related to the study with a TRIAD trial staff member.

Clinical study sites

All screening/baseline and subsequent monitoring evaluations are performed by a board-certified veterinary cardiologist or neurologist, with the provision that a trainee (i.e., veterinary resident) under the direct supervision of one of these specialists can also contribute. To date, the active trial sites are the veterinary teaching hospitals of eight participating universities (Texas A&M University, University of Georgia, Colorado State University, Iowa State University, University of Illinois, Tufts University, University of Wisconsin, and Washington State University), and 11 private, non-academic specialty veterinary practices (The Animal Cardiology Center in Brooklyn, NY; Wheat Ridge Animal Hospital in Wheat Ridge, CO; the four Blue Pearl Specialty Hospitals of Sandy Springs, GA, North Dallas, TX, Overland Park, KS, and Manhattan, NY; Metropolitan Animal Specialty Hospital in Los Angeles, CA; SAGE Veterinary Centers in San Francisco, CA; Specialists in Companion Animal Neurology in Naples, FL; Olympic Veterinary Cardiology in Everett, WA; and Pieper Veterinary in Middletown, CT). At the time of writing, recruitment of additional sites is ongoing.

Drug dosage and formulation

The target dosage of rapamycin (0.15 mg/kg once weekly) is based on prior work by our group that demonstrated significant improvement in one echocardiographic marker of cardiac function (i.e., left ventricular fractional shortening) in dogs administered the same total weekly dosage for 10 weeks [62], and established safety of rapamycin when given orally at twice this total weekly dosage for 10 weeks [62] or at one-half of this total weekly dosage (i.e., 0.025 mg/kg three times weekly) for 6 months [63]. Adverse events experienced by people treated with rapalogs have been related to trough drug concentrations in the blood [67]. Therefore, once-weekly dosing of rapamycin is expected to lead to lower trough levels than would be expected with more frequent intervals. The target weekly dosage of rapamycin used in the present study is lower than doses producing whole blood trough concentrations in dogs that are within the human immunosuppressive range [61, 72]. The duration of treatment is extrapolated from studies suggesting rapamycin’s longevity promoting effects when given transiently to middle-aged mice [18].

For each participant, the appropriate dose is calculated based on body weight at the screening/baseline examination. The trial uses six weight range groups: 20–25 kg, 25.1–30 kg, 30.1–35 kg, 35.1–40 kg, 40.1–45 kg, and 45.1–55 kg. For each dog, study drug dose (mg) is calculated as 0.15 mg/kg multiplied by the upper limit of the appropriate weight range (kg), rounded to the nearest dose achievable with available whole tablet sizes (i.e., 0.4, 1.2, and 2.4 mg). Dogs receive a combination of tablets at each treatment to achieve the appropriate dose. Therefore, for dogs receiving rapamycin, the range of doses delivered is 0.138 to 0.180 mg/kg. Placebo-treated dogs receive a combination of tablets that contain no active ingredients and are visually identical to the rapamycin tablets indicated for their assigned weight range group. The study drug dose is not adjusted if a dog’s weight group assignment changes during the 12-month treatment period.

Both rapamycin and placebo formulations are pharmaceutical grade, enteric-coated, and prepared by a single provider (TriviumVet, Waterford, Ireland) for use in companion animals [43] in accordance with World Health Organization Good Manufacturing Practices guidelines [73]. Study drugs are periodically tested by the manufacturer to ensure purity and consistency. Authorization to use this formulation of rapamycin (marketed as “RapaVet”) has been granted by the United States Food and Drug Administration Center for Veterinary Medicine. The formulation being used is subject to an Investigational New Animal Drug application for indications other than longevity, and the data from this trial are not expected to be used for those other label claims. The human medical rapamycin preparation Rapamune (Pfizer, New York, NY) was released in 1999, and numerous approved generic formulations are now available. Proprietary and publicly available data on RapaVet demonstrate similar pharmacokinetics to these previously available pharmaceuticals [74].

Drug administration

Placebo or rapamycin is administered by mouth once weekly for 52 consecutive weeks. Rapamycin is formulated as a delayed-release enteric-coated tablet, with the placebo being inert, visually identical, and color- and size-matched. All treatments are administered at home by the dog’s owner on Wednesday mornings, with or without food according to the dog’s usual feeding schedule. Owners are instructed to place the tablet(s) on the back of the dog’s tongue to discourage chewing and disruption of the enteric coating. Tablets containing the appropriate study drug are provided to the owner in blister packs contained within a pre-assembled dosing kit, each containing a six-month supply of study drug. To maintain masking of the dispenser, dosing kits are designated with a code name that corresponds to weight group and study drug assignment, as described below (“Randomization, allocation, and masking methods”).

Owners receive monthly online surveys to query medication administration compliance during the 1-year treatment period. Additionally, they maintain paper medication logs, which are returned to the clinical trial site at each visit during the 1-year treatment period. Any unused study drug is also returned at these visits so that remaining tablets can be inventoried and discrepancies in expected residuals can be reviewed and recorded by TRIAD trial staff before proper disposal of the unused study drug.

Adverse response to drug administration

Dogs with suspected intolerance to the study drug (e.g., vomiting after administration) are managed by dividing the dose of study drug into two roughly equal portions (i.e., as is achievable without splitting tablets), which are subsequently administered as morning and evening doses. Administration of the study drug is discontinued if intolerance persists. Temporary suspension of study drug administration for up to three consecutive doses is implemented when an acute illness for which rapamycin is considered contraindicated is diagnosed, when laboratory testing abnormalities likely to be rapamycin-related occur, or when rapamycin administration might affect disease progression or management. In these instances, if the condition or laboratory testing abnormality persists for 4 weeks or longer, or recurs when the study drug is re-initiated, the study drug is permanently discontinued. The study drug is also discontinued if one or more of the following occur: serious adverse events (SAE) deemed likely rapamycin-related; severe or chronic illnesses (e.g., immune-mediated disease) that might be exacerbated by rapamycin therapy; or illnesses for which the required treatment (e.g., chemotherapy) or monitoring might be complicated by concurrent rapamycin therapy. Owners can also choose to discontinue study drug administration, study participation, or both, at any time for any reason. If discontinuation of the study drug is necessary, all scheduled monitoring continues for the remainder of the 3-year study period when possible and masking of the investigator and dog owner to treatment group is maintained. Data from dogs for which the study drug is discontinued will be included in the TRIAD intention-to-treat outcome analysis.

Owner engagement

Consistent follow-up is encouraged by regular communication between study staff and dog owners, and by regular automated email communication. At the time of each enrolled dog’s visit to a clinical study site, the subsequent 6-month visit is scheduled by the site personnel. Dates and times of scheduled appointments are relayed to TRIAD study staff who contact owners in advance of those appointments to arrange delivery of biologic sample collection kits and to provide appointment reminders. Should an owner become unable to attend a scheduled appointment, study staff work with the clinical study site to facilitate rescheduling; in rare instances that an appointment cannot be rescheduled in a timely window (e.g., due to serious illness on the part of the owner), clinical trial staff work with the owner to schedule the next planned recheck in the correct window of time so that participation can continue.

Randomization, allocation, and masking methods

Randomization is done using a permuted block design with variable block sizes ranging from 1 to 8 and a 1:1 treatment-allocation ratio and is stratified by clinical trial site and sex. Randomization sequences were generated prior to the beginning of the study by the Clinical Trial Statistician using a commercial computer program (Sealed Envelope Ltd. 2022. Create a blocked randomization list. [Online] Available from: https://www.sealedenvelope.com/simple-randomiser/v1/lists [last accessed 5 Mar 2024]). Based on treatment assignment and body weight group, each enrolled dog is assigned to one of 12 code name groups (1–12), which represent the 12 possible study drug (n = 2) and weight group (n = 6) combinations. Assignment of code names to each of the 12 drug-weight-group combinations was performed at the outset of the trial and is non-sequential. The key to code name assignment for each drug-weight-group combination is accessible only to six individuals (i.e., three Texas A&M University Veterinary Pharmacists, the TRIAD Randomization Project Manager, the DAP Project Manager, and a Logistics Team Assistant), charged with either study drug kit construction or ensuring study integrity and participant safety. These individuals have no direct involvement with patient screening or enrollment, case management, or data analysis. Upon determination that a screened dog is eligible, the randomization table of the screening clinical trial site is consulted by two people: one of the Texas A&M University Veterinary Pharmacists and the TRIAD Randomization Project Manager, the latter responsible for study drug inventory management and reporting. Each determines the dog’s assignment independently and then confirms with the other that the same result was obtained. Upon confirmation of assignment to rapamycin or placebo, the dog’s weight group is determined, and the appropriate code name group is assigned and recorded in a secure data management system (Research Electronic Data Capture; REDCap).

Each clinical trial site designates one to three team members to serve as Dispensing Contacts who are responsible for the handling and dispensing of pre-assembled dosing kits, and for maintaining documentation of study drug usage for participants enrolled at their site. These staff members are trained by the TRIAD Randomization Project Manager to ensure that study drug handling, dispensing, and record keeping are performed accurately. The dispensing contacts are the only people at each clinical trial site with knowledge of the dosing kit code name assigned to each participant at their respective sites. The dispensing contacts do not have access to the code name key, and so remain blinded to the treatment assignment of enrolled dogs. Upon randomization of a newly enrolled participant, the TRIAD randomization project manager communicates this information to the participating clinical trial site’s dispensing contact. The corresponding pre-assembled dosing kit is shipped to the clinical trial site to be prescribed by the site’s veterinarian and dispensed to the participant. The randomization project manager and each site’s dispensing contact carefully monitor allocation of dosing kits and periodically reconcile their records. Additionally, upon receipt of the pre-assembled dosing kit, owners report the code name on the kit to REDCap where it is compared against what was assigned, to confirm that the correct kit was dispensed.

Study investigators, owners, study monitors, and statisticians are masked to treatment allocation. Access to randomization assignments is limited to five of the six individuals described above (i.e., all but the logistics team assistant). In the event of a veterinary emergency, unmasking occurs by the TRIAD randomization project manager revealing the dog’s treatment assignment to the necessary parties, which can include the dog’s owner, the dog’s attending veterinarian(s), and/or the clinical trial liaison to the DSMB.

Study outcome measures

Outcome measures will be evaluated using data from all randomized dogs.

Efficacy

With respect to efficacy, the primary outcome measure will be survival time from randomization, evaluated in the intention-to-treat subset (defined below in “Planned statistical analyses”). The null hypothesis that rapamycin has no effect on survival in companion dogs will be tested against the alternative hypothesis that rapamycin has a positive effect on survival, with positive effect defined as an increase in mean lifespan. Incidence of cancer diagnoses and death, incidence of infectious disease diagnoses, incidence of cataract formation, and incidence of multimorbidity (i.e., diagnosis of more than one age-related disease) will be documented.

Secondary efficacy outcome measures will include physical function, cardiovascular health, and neurocognitive health. For each secondary outcome, the null hypothesis that rapamycin has no effect on the outcome will be tested against the alternative hypothesis that rapamycin has positive effect on the outcome, with positive effects defined as reduction of disease burden.

Safety

Outcome measures of interest will include the proportions of dogs requiring discontinuation of study drug, developing adverse clinical and laboratory test findings, or developing other AE. An AE is defined as any unfavorable or unintended observation that occurs subsequent to use of the study drug, regardless of whether it might be related to the study drug. An SAE is defined as any health experience or event that is fatal or life-threatening, is permanently disabling, or requires inpatient hospitalization. Classification of AEs and SAEs is based on a modified version of the Veterinary Cooperative Oncology Group Common Terminology Criteria for Adverse Events [75]. Monthly for the first 18 months of the study period and then every 3 months for the remaining 18 months, owners of participating dogs complete at-home observation surveys, which are designed to document potential AEs. Owners are asked to keep a drug diary during the 1-year treatment period and to submit copies of the diary at 6-month intervals. Survey responses and diary submissions are monitored regularly by study staff. Data collected during veterinary visits at baseline and every 6 months for the duration of the 36-month study, including data from detailed patient history, physical examination, clinical measures of cardiovascular and neurocognitive function, and biochemical and urine analysis, are evaluated with respect to safety.

Exploratory

Data from hematology, serum biochemistry, and urinalysis, obtained at each veterinary visit, are used to evaluate rapamycin’s effects, if any, on clinical pathology parameters. In addition, blood markers of inflammation are evaluated from samples obtained at scheduled veterinary visits. As chronic inflammation is a major contributor to age-related declines in health [76, 77], rapamycin’s positive effects on health and longevity might be due, in part, to its association with reduced inflammation [78–80].

Concurrent therapies

At screening, dogs receiving one or more of the following are excluded: medications that indicate an exclusionary disease or condition; drugs with known or suspected pharmacokinetic or pharmacodynamic interactions with rapamycin; drugs known or suspected to increase the risk of adverse effects of rapamycin; drugs with potential adverse effects that overlap with those of rapamycin; or drugs proposed to slow biological aging (e.g., rapalogs, metformin, resveratrol). Routine endoparasite and ectoparasite preventives, some medications used situationally or intermittently for behavior modification (e.g., trazodone), and medications not meeting the exclusionary criteria above and that are used in the management of mild, stable, non-exclusionary conditions are allowed at the time of screening. A complete annotated list of allowed therapies will be included as supplementary data in the final reporting of trial outcomes.

Once enrolled, dogs that develop a condition or disease that requires initiation of a new medication are not removed from the trial—even if the disease or condition would have been exclusionary at screening—unless the attending veterinarian or owner requests removal, or the TRIAD Clinical Monitoring Team determines that removal is in the best interest of the dog due to possible increased risk of harm from continuation of rapamycin. To evaluate the risk of harm, for each newly initiated drug, a veterinary clinical pharmacologist on the TRIAD team interrogates the biomedical literature and publicly available drug interaction databases for potential interactions with rapamycin, including pharmacokinetic interactions, mechanistic overlap, or potentiation of AEs. The TRIAD Clinical Monitoring Team weighs this information in determining if the dog should be removed from the trial. If the team agrees that the study drug can be continued when the new medication is initiated, owners are instructed not to initiate the new drug on the same day as study drug administration to prevent confounding of AEs from the new drug, the study drug, and the new disease or condition. If the study drug is discontinued, follow-up monitoring (i.e., recheck examinations and at-home owner surveys) is continued. At all study visits and during completion of monthly at-home surveys, owners of participating dogs are prompted to report all medications and supplements given at any point in the study, which are recorded.

Procedures and schedule of events

Schedule of events

All enrolled dogs undergo a 12-month treatment and monitoring period and a subsequent 24-month monitoring-only period, for a total study duration of 36 months. A full schedule of events is presented in Table 4.

Table 4.

TRIAD schedule of study events. SBP, systemic arterial blood pressure; ECG, electrocardiogram; CADES, CAnine DEmentia Scale; CBC, complete blood count; CCDRS, canine cognitive dysfunction rating scale; PBMCs, peripheral blood mononuclear cells; (x), if indicated

Dogs of consenting owners that qualify based on comprehensive pre-screening medical records review undergo a screening evaluation at a participating clinical trial site. At this visit (study days -14 to -7), a medical history is obtained, and dogs undergo physical examination and indirect systemic arterial blood pressure measurement. Biospecimens for hematology, serum biochemical analysis, Dirofilaria immitis antigen testing, and urinalysis are collected from all dogs and shipped overnight on ice to a centralized reference laboratory (Texas Veterinary Medical Diagnostic Laboratory, College Station, TX, USA). For dogs evaluated at a clinical trial site in Texas, additional blood is obtained for measurement of serum antibodies to Trypanosoma cruzi. In addition to these tests, dogs enrolled at specialty cardiology sites undergo surface electrocardiography and a full transthoracic echocardiographic examination, while dogs enrolled at specialty neurology sites undergo standard neurologic examination, neurocognitive testing [81], and administration of a canine cognitive dysfunction survey (CAnine DEmentia Scale [CADES]) [82–84]. Finally, the owners of all dogs complete an additional survey for canine cognitive dysfunction (Canine Cognitive Dysfunction Rating Scale) [85] and a health-related quality of life instrument [86] prior to enrollment.

Enrollment decisions are made by a group of veterinarians that comprise a Clinical Monitoring Team, who meet weekly to review the results from the previous week’s screening examinations and to determine whether dogs meet all eligibility criteria.

Following enrollment and randomization, included dogs are re-evaluated at six-month intervals (i.e., at months 6, 12, 18, 24, 30, and 36 post-enrollment) as described for the screening/baseline evaluation, for a total of 3 years, with the goal of all evaluations occurring ± 2 weeks of the target visit date.

Indirect systemic arterial blood pressure measurement

At each visit, indirect systolic arterial blood pressure (SBP) is measured in a quiet environment following an acclimation period of at least 10 min and before all other procedures, in a manner conforming to guidelines of the American College of Veterinary Internal Medicine [87]. Measurements are taken by a trained individual using Doppler ultrasonography or oscillometry, with a blood pressure cuff placed on the base of the tail, crus, or antebrachium. For each measurement session, the first measurement is discarded. Five subsequent, consecutive measurements are recorded, and their average taken as the SBP value for that session. If the session SBP is ≥ 160 mm Hg, a second acclimation period of at least 60 min is allowed and measurements are repeated. The lowest average session SBP is considered with reference to enrollment criteria and when assigning a new diagnosis of systemic arterial hypertension in enrolled dogs. The diagnosis and classification of systemic arterial hypertension follow American College of Veterinary Internal Medicine guidelines [87]. Blood pressure cuff size and anatomic placement site are recorded and maintained across study visits within a given dog.

Electrocardiography

For dogs enrolled at specialty cardiology sites, a lead II surface ECG is performed at each study visit by a trained individual with dogs restrained in right lateral recumbency using standard limb electrode placement. Average heart rate and underlying rhythm are recorded, as well as any abnormalities of rhythm or conduction. At screening, if premature ventricular ectopy that does not prompt the overseeing clinician to initiate antiarrhythmic therapy is identified on ECG, an ambulatory ECG (Holter) monitor is applied for 24 h, with the results used to inform enrollment decisions. At the attending study veterinarian’s discretion, Holter monitor placement might also be performed for dogs in which electrocardiographic abnormalities develop over the course of the study.

Echocardiography

For dogs enrolled at specialty cardiology sites, full transthoracic echocardiographic studies with simultaneous electrocardiography are performed at each study visit by a board-certified veterinary cardiologist, or a trainee under the supervision of a veterinary cardiologist. Dogs are restrained in lateral recumbency and assessed using standard ultrasonographic equipment outfitted with phased-array transducers. All examinations consist of two-dimensional, M-mode, color Doppler, and spectral Doppler echocardiographic evaluations using standard right parasternal, subcostal, and left parasternal imaging planes [88, 89].

Prior to the start of the study, study clinicians are provided detailed guidelines that were developed a priori by a panel of study cardiologists and are designed to standardize echocardiographic evaluations and diagnostic and clinical definitions across study sites. Lists of all echocardiographic parameters to be measured and calculated at each study visit are provided in Tables 5 and 6. Echocardiographic measurements are taken from at least three cardiac cycles and the mean recorded for each. Criteria used in the diagnosis of dilated cardiomyopathy and assessment of pulmonary arterial hypertension probability are presented in Supplemental Tables 1, 2 and 3. A complete, referenced description of definitions used for the assessment of cardiac chamber size and function is also available from the corresponding author on request.

Table 5.

Echocardiographic parameters measured at all visits for dogs screened and followed at specialty cardiology clinics. 2DE, two-dimensional echocardiography; Lax, long-axis view; Sax, short-axis view; 4C, four-chamber view; 5C, five-chamber view; SMOD, measurement taken using Simpson’s method of discs; LA, left atrial; LV, left ventricular; RA, right atrial, RV, right ventricular

Echo mode	View(s)	Parameter	Units	Definition
2D	Right parasternal Lax, 4C	LAD2DE,Lax	cm	LA maximum diameter
		EDV	cm	LV end-diastolic volume (SMOD)
		ESV	mL	LV end-systolic volume (SMOD)
	Right parasternal Lax, 5C	AoD2DE,Lax	cm	Aortic diameter
	Right parasternal Sax	IVSd2DE,Sax	cm	IVS thickness at end-diastole
		LVIDd2DE,Sax	cm	LV internal diameter at end-diastole
		LVPWd2DE,Sax	cm	LV posterior wall thickness at end-diastole
		LVIDs2DE,Sax	cm	LV internal diameter at end-systole
		LA2DE,Sax	cm	LA transverse diameter
		Ao2DE,Sax	cm	Aortic diameter
		RPAd	cm	Minimum diastolic diameter of right pulmonary artery
		RPAs	cm	Maximum systolic diameter of right pulmonary artery
	Left apical 4C	EDV	mL	LV end-diastolic volume (SMOD)
		ESV	mL	LV end-systolic volume (SMOD)
M-mode (MM)	Right parasternal Sax	IVSdMM,Sax	cm	Interventricular septal thickness at end-diastole
		LVIDdMM,Sax		LV internal diameter at end-diastole
		LVPWdMM,Sax	cm	LV posterior wall thickness at end-diastole
		IVSsMM,Sax	cm	Interventricular septal thickness at end-systole
		LVIDsMM,Sax	cm	LV internal diameter at end-systole
		LVPWsMM,Sax	cm	LV posterior wall thickness at end-systole
		EPSS	mm	MV E-point-to-septal separation
	Left apical 4C	TAPSE	cm	Tricuspid annular plane systolic excursion
Spectral doppler	Sax	PV Vmax	m/s	Peak pulmonary systolic flow velocity
	Subcostal or left apical 5C	AV Vmax	m/s	Peak aortic systolic flow velocity
	Left apical 4C	MV Emax	m/s	Peak transmitral E wave velocity
		MV DTE	msec	Transmitral E wave deceleration time
		MV Amax	m/s	Peak transmitral valve A wave velocity
	Left apical 5C	IVRT	msec	Isovolumic relaxation time
	Any	TR Vmax	m/s	Peak velocity of tricuspid insufficiency
Pulsed wave tissue doppler	Left apical 4C	E′lat	m/s	Peak velocity of early diastolic lateral mitral annulus motion
		A′lat	m/s	Peak velocity of late diastolic lateral mitral annulus motion
		S′lat	m/s	Peak velocity of systolic lateral mitral annulus motion

Table 6.

Echocardiographic parameters calculated at all visits for dogs screened and followed at specialty cardiology clinics. See Table 5 for definitions of measured source parameters. 2D, two-dimensional echocardiography; SMOD, measurement taken using Simpson’s method of discs; Lax, long-axis; Sax, short-axis; BW, body weight; BSA, body surface area

Echo mode	Parameter	Units	Definition	Calculation
2D	EF (SMOD)	%	Ejection fraction (SMOD)	(EDV-ESV)/EDV
	LAD:AoD2DE,Lax		2D left-atrial-to-aortic dimension ratio in long axis	LADLax/AoDLax
	LA:Ao2DE,Sax		Left-atrial-to-aortic dimension ratio in short axis	LASax/AoSax
	FS2D	%	Left ventricular fractional shortening (2D)	(LVIDd – LVIDs)/LVIDd
	EDVIBSA (SMOD)	mL/m2	End-diastolic volume indexed to BSA	EDV/BSA, where BSA (m2) = 0.101 × BW (kg)2/3
	ESVIBSA (SMOD)	mL/m2	End-systolic volume indexed to BSA	ESV/BSA, where BSA (m2) = 0.101 × BW (kg)2/3
	EDVIBW (SMOD)	mL/kg	End-diastolic volume indexed to BW	EDV/BW (kg)
	ESVIBW (SMOD)	mL/kg	End-systolic volume indexed to BW	EDV/BW (kg)
	RPAD index	%	Right pulmonary artery distensibility index	([RPAs – RPAd]/RPAs) × 100
M-mode (MM)	nLVIDdMM,Sax	cm/kg0.299	LVIDd normalized to body weight	LVIDd (cm)/weight (kg)0.299
	nLVIDsMM,Sax	cm/kg0.387	LVIDs normalized to body weight	LVIDs (cm)/weight (kg)0.387
	FSMM,Sax	%	Left ventricular fractional shortening (M-mode)	(LVIDd – LVIDs)/LVIDd
Spectral Doppler	E:IVRT	m/s2	Ratio of maximum transmitral E wave velocity to isovolumic relaxation time	MV Emax/IVRT

In addition to qualitative assessments of cardiac structure, the measurements and calculations described below are used to make enrollment decisions for dogs screened at specialty cardiology sites. Maximum end-systolic left atrial dimension (LAD) and aortic valve diameter (AoD) are measured from right parasternal long-axis and right parasternal long-axis outflow views, respectively, and used to calculate the left-atrial-to aortic root ratio (LAD:Ao^2DE,Lx), as previously described [90]. Left ventricular internal dimensions are measured from a right parasternal short-axis view using M-mode echocardiography at end-diastole (LVIDd) or end-systole (LVIDs). For dogs of non-sighthound breeds, normalized left ventricular dimensions are calculated using the following formulae [91]: normalized LVIDd (LVIDdN) = LVIDd (cm)/body weight (kg)^0.299; normalized LVIDs (LVIDsN) = LVIDs (cm)/body weight (kg)^0.387. For sighthounds, body weight exponents of 0.323 and 0.390, respectively, are instead used in these calculations [92]. Left ventricular fractional shortening (FS) is calculated as ([LVIDd − LVIDs]/LVIDd) × 100. Left ventricular internal volumes are estimated using single-plane Simpson’s method of discs, applied to images taken from right parasternal long-axis or left apical views at end-diastole (EDV) or end-systole (ESV), and used to calculate left ventricular ejection fraction (EF) as (EDV − ESV)/EDV. In addition, ESV is indexed to body weight (ESVI^BW) and body surface area (ESVI^BSA) using the following formulae [91, 93]: ESVI^BW = ESV (mL)/body weight (kg); ESVI^BSA = ESV (mL)/body surface area (m²), where body surface area = 0.101 × body weight (kg)^2/3. All participating clinicians are provided an electronic spreadsheet prepopulated with these formulas to facilitate accurate performance of the echocardiographic calculations described above, and calculations are checked for accuracy by the lead study cardiologist prior to final enrollment decisions.

Neurological examination

All dogs undergo a basic neurological evaluation as part of the physical examination at each visit. For dogs enrolled at specialty neurology sites, at each veterinary visit, a comprehensive neurological examination including assessment of mentation, posture, gait, postural reactions, spinal reflexes, cranial nerves, and hyperesthesia, is performed by a board-certified veterinary neurologist or a trainee under supervision of a veterinary neurologist.

Neurocognitive assessments

At 6-month intervals, owners are instructed to complete an at-home survey (Canine Cognitive Dysfunction Rating Scale) [85] designed to identify clinical cases of canine cognitive dysfunction. At scheduled times throughout the calendar year, owners are instructed to administer two at-home cognitive tasks designed to assess short-term working and spatial memory. Briefly, in the first of these assessments, leashed dogs are presented an array of three containers—each containing a food reward—to be visited sequentially. At two of the three locations, the dog is allowed to consume the reward immediately. Owners are then instructed to release the dog from a central location and to record the first container approached, as a measure of memory for location of the sole remaining reward [94]. In the second assessment, dogs watch as a food reward is placed into one of two containers, each placed at a different location equidistant from the dog. Owners are then instructed to release the dog after a variable delay, and to observe and record the accuracy of the first search location [95, 96].

For dogs enrolled at specialty neurology sites, an additional neurocognitive assessment is performed at each visit by a board-certified veterinary neurologist or a trainee under supervision of a veterinary neurologist. This assessment involves administration of a standardized sustained gaze test (a measure of attention), during which the evaluator performs five trials, as previously described [81]. Video recordings from each session are submitted to a team of study veterinarians, who review each test for accuracy of completion and record the timed results of valid trials.

Every 6 months, approximately coinciding with veterinary visits, owners of dogs enrolled at specialty neurology sites also complete the validated 17-item CADES instrument [82], designed to collect information about spatial orientation, social interactions, sleep–wake cycles, and house soiling behaviors along a continuum of severity up to and including a clinical diagnosis of canine cognitive dysfunction syndrome [83, 84]. A trained individual assists owners in completing the CADES instrument at the baseline/screening visit, whereas subsequent surveys—identical apart from being administered electronically—are completed at-home by participating owners.

Physical activity and mobility assessments

For all dogs, annual completion of the DAP Health and Life Experience Survey provides owner-reported information about general activity level including interest in activity, vigor of activity, and type(s) of activity commonly pursued. For all dogs, owners are also instructed to complete several at-home measurement and mobility activities [97] every 6 months. These activities are designed to document gait speed and loss of muscle mass, important components of frailty assessments in people [5, 98]. Briefly, owners follow pictorial and video instructions to measure their dogs’ body dimensions (e.g., leg length, body length, thigh circumference) using a flexible tape measure, and report each measurement in an online portal. Owners also measure and report the time taken for their dogs to traverse a standard distance on-leash, to traverse the same distance off-leash, and to climb a set of stairs (if safely available) off-leash, again using an online portal for reporting.

Biological sample collection, processing, and shipping

Prior to each study visit, owners are mailed a study-specific biologic sample collection kit, which they are instructed to bring to their scheduled study visit. These kits comprise detailed collection, processing, and shipping instructions for veterinary staff; pre-labeled biospecimen collection and shipping containers; pre-paid shipping labels; and an ice pack to be used for sample shipment. Each clinical trial site is provided a small number of back-up kits, to be used if an owner fails to bring the kit to a study visit.

Blood is collected by venipuncture and placed immediately into collection tubes containing lithium heparin (2–4 mL of blood), EDTA (7.5 mL of blood), or no additive (8 mL of blood). Samples containing lithium heparin or EDTA are gently inverted immediately after filling. Approximately 0.25 mL of EDTA-containing whole blood are placed onto a blood spot card (Whatman 903™ Protein Saver Card, GE Healthcare, Cardiff, United Kingdom) for subsequent analysis and allowed to dry at room temperature for at least 1 h prior to placement in a plastic bag containing a desiccant for shipment. Blood samples in collection tubes containing no additive or lithium heparin are allowed to sit at room temperature for 5 to 10 min prior to centrifugation. Serum or plasma from these centrifuged samples are decanted and placed in pre-labeled cryogenic tubes for shipment. Urine samples are collected from a voluntarily voided stream, cystocentesis with or without ultrasound guidance, or direct urinary bladder catheterization, in order of preference. Within 1 h of collection, a minimum of 3 mL of urine are placed in a pre-labeled collection tube, containing no additives, for shipment. Fecal samples for microbiomic analyses are collected from a voluntarily voided sample, or through the use of a fecal loop or gloved finger placed in the rectum. Within 1 h of collection, fecal samples are placed in pre-labeled collection tubes (PERFORMAbiome•GUT PB-200, DNA Genotek, Ottawa, Ontario, Canada) and shaken vigorously according to manufacturer instructions prior to shipment. A minimum of 100 mg of hair is clipped from the dog and placed directly into a pre-labeled trace-metal free bag for shipment.

All biospecimens are shipped overnight on ice to a centralized reference laboratory (Texas A&M Veterinary Medical Diagnostic Laboratory, College Station, TX, USA), where they are analyzed on-site (complete blood count, chemistry profile, and urinalysis) or processed (including separation of peripheral blood mononuclear cells [PBMCs] from EDTA-containing whole blood), aliquoted, and shipped to other laboratories for analysis. A full list of biospecimens to be collected, as well as each specimen’s intended use(s), is presented in Table 7. Molecular profiles generated for TRIAD participants will mirror those used in another Dog Aging Project nested cohort, the Precision cohort. Details of the analyses of these samples will be described in a forthcoming manuscript addressing this cohort.

Table 7.

Biospecimens collected and their intended use(s). CBC, complete blood count; IL, interleukin; TNF, tumor necrosis factor

Biospecimen collected	Sample tube/container	Additive	Sample type derived	Sample purpose/assay
Venous blood	2-mL lavender top	EDTA	Whole blood	CBC
	1-mL lavender top			DNA deep sequencing
	1-mL lavender top			Biobanking
	500-µL lavender top			Blood rapamycin levels
	3-mL lavender top		Peripheral blood mononuclear cells	DNA/RNA extraction
				Flow cytometry
				Epigenetics
				Biobanking
	4-mL red top	None	Serum	Serum biochemistry panel
				Heartworm antigen test
				Biobanking
				C-reactive protein
	4-mL red top			Ultra-sensitive cardiac troponin-I
				S100 calcium-binding protein A12
				Cytokine panel (IL-2,−6,−8, TNF⍺)
				Lipoprotein profile
				N-methylhistamine
				Piloting of assays and trainee use
	2-mL green top	Lithium heparin	Plasma	Metabolomics
				Biobanking
Urine	10-mL yellow top	None	Urine	Urinalysis
Hair	Plastic bag	None	Hair	Hair toxicology
Feces	PERFORMAbiome•GUT PB-200	Proprietary stabilization liquid	Fecal microbial DNA	Fecal microbiome

Blood rapamycin quantification

Rapamycin concentration and activity will be measured in a subset of blood samples obtained at visits during the treatment period (i.e., at visits 2 or 3). In a prior placebo-controlled study of rapamycin in companion dogs [63], dried blood spot cards were collected from four rapamycin-treated and three placebo-treated dogs on a treatment day immediately before and 1-, 2-, 6-, and 24-h post-treatment. Rapamycin concentrations were later assayed using high-performance liquid chromatography tandem mass spectrometry [99]. Using this approach, rapamycin was detected in the blood of all rapamycin-group dogs with peak values at 2-h post-treatment, while rapamycin was not detected in the blood of any placebo-treated dog, which indicates that dried blood spot cards are a suitable system for detecting whole blood rapamycin concentration in companion dogs. For analysis of rapamycin activity, PBMCs will be shipped to the University of Wisconsin-Madison. Evidence of rapamycin activity will be assessed by immunodetection of post-translational modification by phosphorylation of established downstream mTOR protein targets [100].

At-home electronic owner surveys

Monthly during the first 18 months of the study and every 3 months for the remainder of the study, owners complete at-home surveys designed to prompt reporting of potential AEs or changes in dog behavior. (Supplemental File) These surveys are monitored regularly by study staff and used to generate AE reports. Validation analysis of owner-reported health conditions compared to diagnoses described in the veterinary medical record has been performed for 308 dogs in the DAP. Medical records included ≥ 85% of dogs’ lifespan. Gwet’s AC1 was used to test for between-instrument concordance, with values above 0.6 reflecting substantial agreement [101, 102]. We calculated 90% reproducibility between owner reports and medical record diagnoses, with very strong concordance.

End-of-life assessment

For dogs who die—whether the manner of death is euthanasia or unassisted death—during the study period, owners are asked to complete an end-of-life survey [103], designed to extract information regarding the dog’s health condition over the weeks preceding death and any new diagnoses made since the most recent veterinary visit. Date of death and apparent cause of death are recorded for all dogs. Owners are also asked to consent to a routine necropsy, performed by a board-certified veterinary pathologist, who generates reports that include gross and histopathologic findings and interpretation. These reports are used to substantiate cause of death, when possible.

Statistical methods and considerations

Power and sample size calculations

A total of 580 dogs (290 rapamycin-treated and 290 placebo-treated) will be enrolled and randomized. Based on our criteria for enrollment and simulations based on mortality data from 855 dogs within the age and weight range of the present study [104], the TRIAD randomized clinical trial will have greater than 80% power to detect as small as a 6% difference in mean total lifespan, with greater than 80% power to detect an 11% difference in survival within the 3-year study timeframe. Mortality and survival analysis predicts that approximately one-third of dogs in the placebo group will die within 3 years of enrollment. To determine the sufficiently powered sample size, we simulated time-to-event data using the Gompertz hazard function, defined as $\mu \left(x\right)=\alpha e^{\beta x},$ where $\mu (x)$ is instantaneous mortality $\mu$ at age $x$, $\alpha$ is the baseline hazard rate, and $\beta$ is the age-related rate of increase in mortality. Mortality data from these 855 dogs yielded $\alpha$=0.0066 and $\beta$=0.29 using the R package MortalityLaws (R Core Team [2021]. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/). We assumed that the intervention with rapamycin improves life expectancy by reducing the baseline Gompertz hazard rate. We also assumed age distribution of 50% of dogs between age 7 and 8 years, 25% between age 8 and 9 years, and 25% between age 9 and 10 years. Incidentally, this matches the age distribution of the 150 dogs currently enrolled. Given the simulated time-to-event data, all analyses were conducted with log-rank tests using the survdiff function in the R package Survival.

Planned statistical analyses for the primary outcome (lifespan)

The primary analyses will be based on an intention-to-treat (ITT) dataset, defined as all dogs who were randomized to a treatment group after completing a screening exam. At established time points during the trial, the randomized treatment assignments for each dog will be linked to the dog’s outcome and covariate data. The ITT dataset will be analyzed with adjustments as necessary to account for multiple hypothesis testing. The primary outcome of the trial is survival time from randomization, assessed at 3 years. We will use Kaplan–Meier curves to summarize the survival for each treatment group. We will perform formal inference using a Cox proportional hazards regression model adjusting for age and the randomization strata. We will test the null hypothesis of no effect of rapamycin on survival using a 1 degree of freedom test. The same analyses will be performed for the per protocol (PP) subset, defined as all dogs from the ITT dataset who (1) received at least 48 doses of study drug during the 1-year treatment period; (2) continued to participate in the trial throughout the 2-year monitoring period or died during the monitoring period; and (3) whose status as alive or dead was known at the end of the 3-year study period. To the extent that inferential tests are performed as part of these comparisons, they will be two-sided, and precise p values will be reported.

With respect to safety, AEs and SAEs will be analyzed using cumulative risk, defined as the observed proportion of dogs with at least one AE up to time t divided by the total number of dogs in each intervention group. Adverse events and SAE will also be analyzed by modeling each as count data using Poisson regression, so that multiple AE will be considered beyond the first observed AE. The observation time for such analyses will be the 3-year study period.

Planned statistical analyses for the secondary outcome (healthspan)

We will test the hypothesis that rapamycin is effective in broadly delaying and reducing the incidence of age-related diseases and conditions, including cancer diagnoses, infectious diseases, sensory loss, chronic kidney disease, and degenerative joint disease, conditions that also increase with age in human beings. At established time points during the trial, the randomized treatment assignments for each dog will be linked to the dog’s outcome data and covariate data. Cox proportional hazards will be used to identify associations between treatment assignment and healthspan outcomes. Some continuous data items will be converted to categorical variables to uncover associations between treatment assignment and specific disease burden. Clinically relevant measures of hematology, clinical chemistry, and urinalysis panels will also be analyzed using mixed effects regression models. A compilation of functional frailty markers including changes in customary physical activity, mobility, strength, unintentional weight loss, and decline in thigh circumference will be assessed for each TRIAD dog annually, using a mixed effects regression model fit to the time-dependent outcomes.

The comprehensive assessments collected longitudinally provide a remarkable framework to understand how aging manifests in the animals of the cohort. Physiological measures, age, and sex will be tested as covariates using regression analyses to determine the extent to which they account for the variance in morbidity and mortality. Multiple factor analysis will be used to examine associations and variation among individuals. Novel insights identified in this approach will be further investigated within the larger DAP dataset and used to explore the biology of association in independent ancillary studies.

Planned interim statistical analyses

Two interim analyses for superiority will be conducted: the first after 200 dogs have been randomized and have either completed the 3-year study period or met the study endpoint of death during the 3-year study period, and the second after 400 dogs have been randomized and have either completed the 3-year study period or met the study endpoint of death during the 3-year study period. The final analysis will be done after 580 dogs have been randomized and have either completed the 3-year study period or have met the study endpoint of death during the 3-year study period. An O’Brien-Fleming boundary will be used so that consideration will be given to stopping the trial after 200 dogs have been randomized if the difference in survival is statistically significant at the significance level of 0.0052; the same consideration will be given at the second interim analysis if survival is different at the significance level of 0.0141. The final analysis will be conducted at the significance level of 0.0451 to maintain the overall type-I error rate of 0.05. The data from the first and second interim analyses will be reviewed by the DSMB. The DSMB will make a recommendation to continue the study as is, modify, or terminate due to the treatment arm being demonstrably better than the placebo arm. There will be no stopping boundaries for futility.

Clinical monitoring and safety oversight

Given their age and body weight range, included dogs are in the latter portion of their expected lifespans, and therefore, at high relative risk of death and development or progression of various age-related conditions during the 3-year study period. To ensure detection of AEs, dogs are monitored every 6 months by a trial veterinarian. In addition, owners are prompted monthly to report possible AEs and are provided a medication and observation diary to record dog experiences. Owners are also instructed to spontaneously report experiences at any time and are provided the contact information for a trial veterinarian who is available by phone after hours for emergencies involving study participants. A clinical monitoring team meets weekly to review data collected from veterinary visits, owner surveys, and reports of AEs, to verify that the trial is conducted in compliance with planned procedures. Safety oversight is provided by a National Institute on Aging-appointed DSMB that is charged with monitoring study progress, AEs, and SAEs, and that meets a minimum of every 6 months or ad hoc at their discretion.

Data management methods

Data collection is performed by staff of individual clinical trial sites. Each site’s participating veterinarian is responsible for ensuring the accuracy, completeness, legibility, and timeliness of reported data. Hardcopies of checklists and study examination forms are provided as source document worksheets for recording data for each dog. Digital copies of these completed forms, owner-completed medication logs (at applicable visits), as well as the electrocardiogram and echocardiogram reports (for dogs evaluated at specialty cardiology clinics), and sustained gaze videos (for dogs evaluated at specialty neurology clinics) are sent from each clinical trial site to the DAP via email for upload into a secure, web-based electronic data capture platform (Research Electronic Data Capture; REDCap), hosted at the University of Washington [105, 106].

Hardcopies of documents provided to the clinical trial site or generated at the clinical trial site are retained by the clinical trial site until the completion of the study. Hardcopies are secured in the patient’s hardcopy record and stored in a secure location. At the conclusion of the study, these documents will be sent to the DAP Clinical Coordinating Center for secure storage.

Future data and biological sample sharing

The DAP is an Open Science project, with a core commitment to make all results, raw data, and biospecimens collected available to the research community. DAP data are updated annually and made available on the DAP’s research data repository [14], following completion of a data use agreement (accessible at https://pubs.dogagingproject.org). Access to DAP data does not require collaboration with DAP researchers. Because the TRIAD randomized clinical trial is a double-masked, randomized, placebo-controlled trial, data that could potentially lead to unmasking or otherwise bias the outcome of the study will not be included in the annual public release of DAP data until the trial’s conclusion. A subset of biospecimens obtained from TRIAD dogs are stored in the DAP Biobank [107]. These biospecimens will be made available to qualified researchers, provided that the proposed research is in the public interest and the researchers commit to providing the DAP with any data obtained from sample analysis to allow its addition to the DAP data repository. Like the data release, biospecimens that could potentially lead to unmasking or otherwise bias the outcome of the TRIAD randomized clinical trial will not be available from the DAP Biobank until the conclusion of the study.

Trial status at time of writing

At the time of writing, 252 dogs have been screened for inclusion at one of the clinical trial sites. Of these, 158 have met all eligibility criteria and have been enrolled and randomized. One hundred and thirty-one and 27 of these dogs were enrolled and are being followed at specialty cardiology and neurology clinics, respectively.

Discussion and conclusions

The ongoing Test of Rapamycin in Aging Dogs (TRIAD) randomized clinical trial described here is a first-of-its-kind large-scale study designed to test as a primary outcome the ability of rapamycin to prolong lifespan in companion dogs. This prospective, parallel-group, double-masked, randomized, placebo-controlled, multicenter clinical trial will also assess as secondary outcomes the ability of rapamycin to improve several healthspan metrics in healthy, middle-aged dogs.

We expect some degree of heterogeneity in the response to rapamycin, given that dogs in this cohort are not of a single breed and are unrelated. Our power calculations for lifespan effect are based on data from an equivalently comprised cohort of dogs and account for heterogeneity among individuals, at least as far as aging is concerned. Based on work performed in mice, we expect the size of rapamycin’s effect on lifespan to be large enough to overcome any illness-based variation in response. During the study, detailed medical records are collected from all dogs, and study exclusion criteria should eliminate dogs with overt medical conditions at the start of the trial. Nonetheless, we will not be able to avoid clinically silent or mild conditions. Dogs completing the TRIAD randomized clinical trial remain in the DAP, meaning that collection of medical record information extends beyond the 3-year study period. Conditions that emerge after completion of the trial will become known and as those data are acquired, post-hoc analyses of rapamycin-treated subgroups will be undertaken to investigate whether differences in existing or emerging health conditions among individuals might be associated with differences in the response to rapamycin.

Given the translational relevance imparted by companion dog morphologic and genetic variability, risk for age-related disease, and diverse environmental exposures, the TRIAD randomized clinical trial will provide insight into the feasibility and utility of future studies designed to evaluate potential lifespan-prolonging interventions in human beings. It is our hope that the TRIAD randomized clinical trial will be a foundation for future pseudo-pragmatic clinical trial design, recruitment strategies, logistic implementation, and data analytic and integrative approaches. We anticipate that results will provide avenues for other geroscience interventions in animals and human beings, including non-pharmacologic, environmental, pharmacologic, nutritional, biologic, and gene- and cell-based therapies.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

AE

Adverse events

CADES

CAnine DEmentia Scale

DAP

Dog Aging Project

DSMB

Data & Safety Monitoring Board

IACUC

Institutional Animal Care and Use Committee

mTOR

Mechanistic target of rapamycin

mTORC1

MTOR-complex-1

PBMCs

Peripheral blood mononuclear cells

SAE

Serious adverse events

SBP

Systolic arterial blood pressure

TRIAD

Test of Rapamycin in Aging Dogs

Funding

The Dog Aging Project is supported by grant UI9AG057377 (AEC, KEC, VF, GA, BGB, BB, BD, AF, KKF, ANG, SNH, MI, KI, SAK, RLM, JOP, SMS, AR, DELP, MK,) from the National Institute on Aging, a part of the National Institutes of Health, and by private donations.

Data availability

These data will be held from public release until the completion of the trial and will then be housed on the Terra platform at the Broad Institute of MIT and Harvard.

Declarations

IACUC approval declaration

The University of Washington IRB deemed that recruitment of dog owners for the Dog Aging Project and the administration and content of the DAP surveys are human subjects research that qualifies for category 2 exempt status (IRB ID no. 5988, effective 10/30/2018). All study-related procedures involving privately owned dogs were approved by the Texas A&M University IACUC, under animal use protocols 2021-0317 CAM (effective 12/8/2021 to present) and 2018-0368 CA (effective 12/19/2018 to 12/18/2021).

Conflict of interest

The authors declare the following conflicts of interest: MK is a scientific advisor for TriviumVet; DELP is a consultant for WndrHLTH.

Disclaimer

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.