Systemic Safety of a Recombinant AAV8 Vector for Human Cocaine Hydrolase Gene Therapy: A Good Laboratory Practice Preclinical Study in Mice

Abstract

Cocaine addiction continues to impose major burdens on affected individuals and broader society but is highly resistant to medical treatment or psychotherapy. This study was undertaken with the goal of Food and Drug Administration (FDA) permission for a first-in-human clinical trial of a gene therapy for treatment-seeking cocaine users to become and remain abstinent. The approach was based on intravenous administration of AAV8-hCocH, an adeno-associated viral vector encoding a modified plasma enzyme that metabolizes cocaine into harmless by-products. To assess systemic safety, we conducted “Good Laboratory Practice” (GLP) studies in cocaine-experienced and cocaine-naive mice at doses of 5E12 and 5E13 vector genomes/kg. Results showed total lack of viral vector-related adverse effects in all tests performed. Instead, mice given one injection of AAV8-hCocH and regular daily injections of cocaine had far less tissue pathology than cocaine-injected mice with no vector treatment. Biodistribution analysis showed the vector located almost exclusively in the liver. These results indicate that a liver-directed AAV8-hCocH gene transfer at reasonable dosage is safe, well tolerated, and effective. Thus, gene transfer therapy emerges as a radically new approach to treat compulsive cocaine abuse. In fact, based on these positive findings, the FDA recently accepted our latest request for investigational new drug application (IND 18579).

Introduction

Cocaine addiction is a chronic and relapsing disorder leading to severe medical and psychosocial complications, and there are no Food and Drug Administration (FDA)-approved medications to treat it. Past research focused on various pharmacological approaches and behavioral interventions, but all showed relapse rates near 95% within a year of attempted abstinence.¹ No small-molecule drugs have had real impact on this addiction. So far, anti-cocaine vaccines have yielded only modest results.^2,3 But recent animal studies suggest that better outcomes might be obtained with an enzyme that lowers cocaine reward value by rapid drug hydrolysis, that is, a cocaine hydrolase (“CocH”).

Native human plasma butyrylcholinesterase (BChE) degrades cocaine slowly, but over the past decade, several investigators, including ourselves, were mutating BChE into an efficient CocH.^4–8 That process led to five amino acid substitutions (A199S/F227A/S287G/A328W/Y332G) conferring ∼1,500-fold higher efficiency for cocaine inactivation. But even with slow-release preparations, it is hard to envisage practical treatments based solely on direct protein injection because relapse risks remain high for years and many treatments would be needed at huge expense.⁹

Instead of enzyme injections, we observed long-term enzyme production after one-time delivery of gene transfer agent as a better strategy. The concept is to incorporate human CocH (hCocH) cDNA into an adeno-associated virus (AAV) vector to drive enzyme production. AAV8 is known for high liver transduction efficiency.¹⁰ In this study, it led to stably high levels of hCocH in the liver and blood. One intravenous (i.v.) injection of AAV8 can act for months or years, preventing i.v. cocaine from reaching brain reward centers.¹¹ High-dose gene transfer enhances CocH levels several 100-fold, making it the most abundant plasma protein after albumin.¹² This status can persist 2 or 3 years at levels that destroy circulating cocaine within seconds, greatly hindering drug access to the brain. Instead of reward, there is a stably large rise in plasma levels of a benign cocaine metabolite, ecgonine methyl ester, a hypotensive smooth muscle relaxant with no reward value.¹³

Our studies were conducted in compliance with Good Laboratory Practice (GLP) standards to support an investigational new drug (IND) application to the FDA. They were designed to address safety and tolerability of AAV-hCocH before initiating a first-ever clinical trial in patients with cocaine abuse.

Materials and Methods

Animals and ethics

One hundred twenty Balb/c mice (60 males, 18.9–23.9 g; 60 females, 17.3–20.8 g) were studied under GLP-compliant protocols. All mice were acquired from Envigo (Indianapolis, IN) under protocol A00001804-16, approved by Mayo Clinic's Institutional Animal Care and Use Committee. Experiments were conducted in accord with the Guide for Care and Use of Laboratory Animals¹⁴ in a Mayo facility approved by the American Association for Accreditation of Laboratory Animal Care. Mice at 5–6 weeks of age were housed five or fewer per cage and given standard chow diet and water ad libitum.

Vectors

GLP-grade AAV8-hCocH vectors were produced and analyzed at the Children's Hospital of Philadelphia (CHOP) in the Clinical Vector Core for the Center for Cellular and Molecular Therapeutics. Quality control tests including visual inspection, bioburden, endotoxin level, purity, and potency were performed by the CHOP (Supplementary Table S1). AAV8-hCocH is a recombinant, replication-defective, and nonenveloped AAV of serotype 8. The viral genome (Fig. 1) has 4,013 nucleotides of single-stranded DNA, which contains AAV2 inverted terminal repeats at both ends, cytomegalovirus and ubiquitin enhancers, a chicken beta-actin promoter, the hCocH coding sequence, woodchuck hepatitis virus posttranscriptional regulatory element, and an SV40 late polyadenylation signal.¹⁵ The AAV8-hCocH vector carries a gene encoding human BChE with five amino acid substitutions, A199S/F227A/S287G/A328W/Y332G.¹⁶ This vector was co-transfected into HEK293T cells with pHELP (Applied Viromics) and pAAV 2/8 (University of Pennsylvania), as described previously.¹²

Figure 1.

Schematic view of plasmid pVIP-hCocH. The AAV8-hCocH viral genome has 4,013 nucleotides of single-stranded DNA comprising AAV2 ITRs at both ends, a CMV enhancer, a chicken beta-actin promoter, an UBC enhancer, the hCocH coding sequence, a WPRE, and an SV40 late polyadenylation signal. AAV, adeno-associated virus; CMV, cytomegalovirus; hCocH, human cocaine hydrolase; ITR, inverted terminal repeat; UBC, ubiquitin; WPRE, woodchuck hepatitis virus posttranscriptional regulatory element.

Study design

Sixty female and 60 male Balb/c mice were identified by microchip and ear notch. Treatments for all groups in both sexes are listed in Table 1. To determine dose dependency, 5E12 and 5E13 vector genomes (vg)/kg were selected based on previous efficacy studies in rodents and our proposed dosage for phase 1 clinical trial. Saline groups served as treatment controls. The AAV8-hCocH safety profile was tested in two scenarios: one with no cocaine challenge (group 1–6), the other with drug (group 7–12). Saline or AAV8-hCocH were i.v. injected on day 0. Cocaine-treated mice received once-daily intraperitoneal (i.p.) dose of 40 mg/kg cocaine (Medisca, Inc.) in 100 μL saline from day 5, five times per week for the study duration, to simulate human drug-users. Mice in groups 1–3 and 7–9 were euthanized and necropsied when transgene expression reached peak levels on day 21. Mice in groups 4–6 and 10–12 were euthanized and necropsied on day 60 to assess chronic toxicity.

Table 1.

Treatment groups

Group Number	Vector Treatment	Number of Mice	Dose (vg/kg)	Cocaine Treatment	Harvest Day
1	Saline	10 (5F+5M)	—	—	21
2	AAV8-hCocH	10 (5F+5M)	5E12	—	21
3	AAV8-hCocH	10 (5F+5M)	5E13	—	21
4	Saline	10 (5F+5M)	—	—	60
5	AAV8-hCocH	10 (5F+5M)	5E12	—	60
6	AAV8-hCocH	10 (5F+5M)	5E13	—	60
7	Saline	10 (5F+5M)	—	40 mg/kg i.p.	21
8	AAV8-hCocH	10 (5F+5M)	5E12	40 mg/kg i.p.	21
9	AAV8-hCocH	10 (5F+5M)	5E13	40 mg/kg i.p.	21
10	Saline	10 (5F+5M)	—	40 mg/kg i.p.	60
11	AAV8-hCocH	10 (5F+5M)	5E12	40 mg/kg i.p.	60
12	AAV8-hCocH	10 (5F+5M)	5E13	40 mg/kg i.p.	60

Mice in groups 7–12 began i.p. cocaine regimen for 5 days (−7 to −3) before day 0 (vector injection) and then continued for 5 days per week for the duration of the study.

—, not applicable; AAV, adeno-associated virus; F, female; hCocH, human cocaine hydrolase; i.p., intraperitoneal; M, male.

Vector injections and microchips

Mice were anesthetized for microchip implantation with isoflurane. Saline or vector was administered through tail vein injection. The needle catheters were prepared in-house by cutting 30-gauge needles and inserting the cut end into 15 cm of polyethylene 10 tubing with a blunt 30-gauge Luer-lock hub at the end for attaching a syringe with saline or vector. The conscious mice were placed in a restrainer, and then, a needle catheter was inserted into a lateral tail vein. Placement of the needle inside the vein was confirmed by infusing a small volume (20–30 μL) of heparinized saline. A successful injection was determined when there was no resistance during infusion, blood return after injection, or tail blanching. One hundred microliters of saline (Baxter Healthcare Corporation) was administered i.v. to mice in groups 1, 4, 7, and 10. The same saline lot was also used as diluent for the remaining treatment groups. One hundred microliters of AAV8-hCocH (Clinical Vector Core Lot A8FP1-C1720N [CHOP]) was administered i.v. to mice in groups 2, 5, 8, and 11 at 5E12 vg/kg. Mice in groups 3, 6, 9, and 12 received 5E13 vg/kg. Virus diluted in saline was administered according to group-mean body weight.

Cocaine treatment

Cocaine (40 mg/kg; Medisca, Inc.; lot no. 143370) was given i.p. in 100 μL saline (Baxter) to mice in groups 7–12. Drug was diluted in saline and dosed according to group-mean body weight.

Clinical observations

At least three times per week, mice were observed for clinical signs and general welfare. Results were recorded as study data. Cause of death was recorded for each mouse. Survival was plotted on the Kaplan–Meier survival curve, and p-values were calculated by the log-rank method.

Blood collection

Mice were bled for chemistry/enzyme assay immediately before euthanasia. Leftover chemistry samples were centrifuged at 8,000 g for 5 min to separate plasma for hCocH enzyme assay.

Measuring viral copy numbers

Total DNA was isolated from harvested tissues using QIAamp DNA Mini Kit (Qiagen, Hilden, Germany). Extracted DNA was quantitated by spectrophotometer and diluted to 10 ng/5 μL. Real-time quantitative polymerase chain reaction (QPCR) assays were run in 20 μL volumes comprising 1 × iQ SYBR Green Supermix (Bio-Rad), 10 ng of tissue DNA, and 0.1 μM of each AAV-VIP-specific primer (forward: 5′-AACGCCAATAGGGACTTTCC-3′; reverse: 5′-GGGCGTACTTGGCATATGAT-3′).¹² A standard curve was generated by serial dilutions of pAAV-VIP-hCocH plasmid DNA to quantitate tissue genome copies of AAV8-VIP-hCocH, and results were analyzed with the SDS2.3 Software (Applied Biosystems, Foster City, CA).

CocH enzyme assay

Tissues were sonicated (QSONICA; Q125) in ice-cold lysis buffer (10 mM NaPi, pH 7.4, 0.1% Tween 20) at 1:10 mass-to-volume ratio. Homogenates were clarified by 10 min centrifugation at 16,000g at 4°C. CocH activity in duplicate 50 μL aliquots of serum or tissue extracts was assayed by incubating 30 min with ³H-cocaine (50 nCi, 18 μM; Perkin Elmer, Boston, MA). Liberated ³H-benzoic acid was acidified and partitioned into toluene-based fluor for scintillation counting, as described previously.¹⁷

Tissue histopathology

Tissues collected at necropsy were placed in 10% buffered formalin and shipped to Mayo Clinic Arizona (Scottsdale, AZ) for paraffin embedding and hematoxylin and eosin (H&E) staining. Slides were blinded and examined by veterinary pathologist, Dr. Brad Bolon, DVM, PhD, at GEM-Path, Inc. (Longmont, CO).

Quality assurance

Quality assurance was performed by the Quality Assurance Consultant Paula Nuchols (Nuchols Consulting) and the Quality Assurance Representatives from the Mayo Clinic Department of Molecular Medicine in compliance with the U.S. FDA GLP for nonclinical laboratory studies, 21 CFR Part 58 (Supplementary Table S2).

Data analysis

Data were analyzed with GraphPad Prism 7 Software (GraphPad Software, Inc.) and are presented as mean values ± standard error of the mean. Survival analysis was performed on long-term study groups and plotted on the Kaplan–Meier survival curve. Multiple group comparisons were tested with one-way analysis of variance (ANOVA) with p < 0.05 considered statistically significant.

Results

Motility, clinical observations, and pathology

To see if AAV-hCocH gene therapy would affect survival rates, the Kaplan–Meier analysis was run on mice of both sexes with or without cocaine and vector treatment. With no drug challenge, all mice survived till the end of the study, day 60, but cocaine-challenged females began dying after 21 days of cocaine. By day 60, all vector-free females had died. But treatments with AAV-hCocH at 5E12 vg/kg or 5E13 vg/kg rescued all of them (Fig. 2A). In male mice, 60% given only cocaine-injection survived till the end of the study, and, like the females, the vector treatment at both dosages greatly improved the survival rates (Fig. 2B). Evidently, AAV-hCocH had no negative impact on survival in male or female mice but prevented cocaine-induced death in the long run. Signs and general mouse welfare were monitored at least three times per week. Those with i.v. AAV-hCocH injection remained clinically healthy for the duration. No test article–related early deaths, clinical signs of toxicity, aberrant food consumption, or change in body weights were detected.

Figure 2.

Survival of AAV-hCocH-treated mice with and without cocaine challenge. The Kaplan–Meier survival curves of female (A) and male (B) mice (n = 5 per group). Live mice are plotted as vertical lines. Log-rank tests were used for statistical analysis.

Cocaine is a potent stimulant of locomotor activity as it blocks reuptake of dopamine, serotonin, and norepinephrine in presynaptic brain neurons. But during our study, we found that AAV-hCocH greatly reduced cocaine-induced hyperactivity in all treated mice. Locomotion recorded 5 min after cocaine injection was scored at four activity levels: normal, mild, moderate, and severe. Before vector treatment, all mice responded to cocaine injection with mild hyperactivity but that effect diminished 10 days after low-dose vector treatment and 2 days after high dose (Fig. 3). Across all 60 days of study, the treatments (low and high dose) substantially lowered cocaine-induced hyperactivity, while unprotected mice (saline group) responded reliably with mild-to-moderate hyperactivity immediately after a cocaine injection (Fig. 3). These results strongly support our hypothesis that AAV-hCocH gene transfer can drive CocH expression to levels that obliterate cocaine stimulation.

Figure 3.

Cocaine-induced hyperactivity in treated and untreated mice. Female (A) and male (B) mice (n = 5) were i.p. injected with cocaine (40 mg/kg) 5 days a week from day 5 to 60. AAV8-hCocH at the dosage of 5E12 or 5E13 vg/kg was i.v. injected on day 0. Saline with equal volume served as vehicle control. Hyperactivities were evaluated three to four times a week at 5 min after cocaine injection. Clinical changes were graded at four levels: normal, mild, moderate, and severe. i.p., intraperitoneal; i.v., intravenous.

Blood chemistry

At high vector dose, AAV8 hepatotropism is known to induce a short-lived release of circulating liver enzymes.¹⁸ But long-term cocaine abuse leads to serious liver injury, and the body is flooded with other toxins that the damaged liver cannot filter out.¹⁹ We deemed it important to see if gene transfer with hepatotropic viral vectors would enhance or reduce cocaine's liver toxicity. To explore that issue, we tested three key plasma biochemical markers of liver inflammation: alkaline phosphatase (ALP), alanine aminotransferase (ALT), and aspartate aminotransferase (AST) at the necropsy end point (day 60). With no cocaine challenge, all mice, including saline control and vector-treated groups (low and high doses), had normal levels of ALT, AST, and ALP (Fig. 4). With daily i.p. injection of 40 mg/kg cocaine, mice without hCocH vector showed a large rise in all three plasma markers. In sharp contrast, mice with active vector showed no liver damage from cocaine intake and retained normal levels of all liver-damage markers (Fig. 4). In short, AAV8-hCocH caused no detectable liver damage on its own, and it prevented liver damage from sustained cocaine exposure.

Figure 4.

Plasma levels of liver-damage markers. Plasma ALT (A), AST (B), and ALP (C) levels determined 60 days after i.v. injection of AAV8-hCocH at 5E12 or 5E13 vg/kg. Results are means ± standard error of the mean (n = 5). *p < 0.05, versus female vehicle control; ^##p < 0.01, ^###p < 0.001, versus male vehicle control. ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; F, female; M, male.

Vector biodistribution

Biodistribution of AAV8-hCocH vector in mouse tissues was determined by QPCR analysis in relevant organs collected from each mouse at 21 and 60 days post-vector administration. Liver, the key tissue, showed higher levels than all others, such as the brain, spleen, lung, kidney, heart, quadriceps muscle, and gonads (Fig. 5). Dose-dependent levels of vector DNA were found in the livers of injected mice at both 21 and 60 days, and the concentrations remained steady across that interval (Fig. 5).

Figure 5.

Biodistribution of AAV8-hCocH after i.v. injection in mice. Gene copy numbers of AAV8-hCocH DNA in tissues including the liver, brain, spleen, lung, kidney, heart, quadriceps muscles, and gonads were shown at 21 days (A) and 60 days (B) after viral injection (5E12 or 5E13 vg/kg). Copy numbers of AAV8-hCocH are reported as copies/100 ng of total DNA (n = 5).

In parallel, we quantified cocaine-hydrolyzing activity in other sites. Highest levels were in blood plasma and the liver in both low- and high-dose vector groups. The levels remained similar on days 21 and 60 (Fig. 6). Roughly 100-fold lower but still readily detectable activities were seen in the brain, spleen, lung, and kidney (Fig. 6). Considering that the vector genome was limited to the liver, trace activity in tissue homogenates likely reflected blood contamination during sample collection.

Figure 6.

Enzyme levels of CocH in plasma and tissues after AAV8-hCocH i.v. injection. Enzyme activities of CocH in plasma and tissues including the liver, brain, spleen, lung, and kidney were determined at 21 days (A) and 60 days (B) after i.v. injection of AAV8-hCocH at 5E12 or 5E13 vg/kg (n = 5). Samples collected from the heart, quadriceps muscles, and gonads were not tested in observation of low vector genome copies in these tissues. Mean concentrations are expressed as mU/mL of tissue extract.

Histopathology

No macroscopic changes were identified at necropsy nor were there vector-related lesions in any of the tested organs, bone marrow, brain, gonad, heart, kidney, liver, lung, skeletal muscle, spinal cord, or spleen, at either 21 or 60 days post-vector injection (Table 2). Treatment with cocaine alone was associated with increased incidences of epicardial calcification (lesions characterized by superficial fibrosis and mineral deposits) and hepatocyte degeneration and mineralization (Table 2). In the heart, cocaine exposure was linked to substantial epicardial calcification, ranging in severity from mild to moderate and prominently affecting the right ventricle. In the liver, cocaine exposure was associated with hepatocyte degeneration from mild to moderate, with affected cells that were large, pale, and often had large nuclei with intranuclear cytoplasmic inclusions. Administering AAV-hCocH reduced the incidence and intensity of these cocaine-induced pathologies (Table 2). Cardiac calcification in vector and cocaine co-treated groups ranged in severity from mild to moderate, and hepatocyte mineralization was improved from minimal to mild severity.

Table 2.

Summary of pathological findings

	Day 21 post-viral injection				Day 60 post-viral injection
	Saline	AAV-hCocH	Saline+Cocaine	AAV-hCocH+Cocaine	Saline	AAV-hCocH	Saline+Cocaine	AAV-hCocH+Cocaine
	n = 10	n = 10	n = 10	n = 10	n = 9	n = 10	n = 8	n = 9
Bone marrow	0	0	0	0	0	0	0	0
Brain	0	0	0	0	0	0	0	0
Gonad	0	0	0	0	0	0	0	0
Lung	0	0	0	0	0	0	0	0
Skeletal muscle	0	0	0	0	0	0	0	0
Spinal cord	0	0	0	0	0	0	0	0
Spleen	0	0	0	0	0	0	0	0
Kidney (protein casts)	0	1Min	1Min	0	0	0	0	0
Heart (mineral)	1Mod	0	1Min	0Min	1Min	1Min	1Min	0Min
			1Mild	2Mild			1Mild	0Mild
			3Mod	2Mod			2Mod	0Mod
			1Sev	0Sev			1Sev	1Sev
Liver (mineral)	0	0	3Min	1Min	0	0	2Min	3Min
			2Mild	2Mild			1Mild	1Mild
Liver (multifocal)	0	0	0	0	0	0	2Mild	0
							2Mod

Numerals indicate numbers in mouse groups. Tissue damage was graded as follows: Min, minimal; Mild, mild; Mod, moderate; Sev, severe.

H&E stains showed marked increased mineralization in the heart and liver tissues of cocaine-exposed control groups, whereas the vector treatment groups showed uniformly healthy morphology (Fig. 7). Thus, short-term (21 days) or long-term (60 days) cocaine administration alone was associated with increased incidences of epicardial calcification and hepatocyte mineralization. Coadministration of AAV-hCocH substantially reduced these heart and liver lesions and also limited cocaine-induced toxicity.

Figure 7.

Histological sections of the heart and liver in prolonged cocaine-exposed mice with and without AAV8-hCocH treatment. Mice were i.p. injected with cocaine (40 mg/kg) 5 days a week from day 5 to 60. AAV8-hCocH vectors were i.v. injected at the dose of 5E13 vg/kg on day 0, and saline injection serves as the vehicle control. Heart and liver tissues were collected on day 21 (A) and day 60 (B) for hematoxylin and eosin staining.

Discussion

The studies presented here were performed to satisfy the FDA concerns expressed at a recent pre-IND meeting regarding a proposed clinical trial focused mainly on the impact of modified human BChE with AAV vector on the short-term and long-term toxicity and viral vector biodistribution. Our study demonstrated that neither of these concerns was evident in preclinical models. Results from the present study conducted under GLP conditions showed that i.v. injection of AAV8-hCocH is safe and well tolerated in normal mice and in cocaine-exposed mice. Accordingly, a proposed clinical trial has now been approved by the FDA to proceed to patient accrual under IND.

Gene therapy using recombinant AAV is attractive because of efficient and prolonged transgene expression in target tissues.²⁰ AAV serotype 8 has shown a significantly greater liver transduction efficiently than those of other serotypes, which resulted in efforts to develop this virus as a gene therapy vector for hemophilia,²¹ human immunodeficiency virus,²² rare diseases,²³ and now cocaine addiction. The next step toward human application of rAAV8-mediated gene transfer to human diseases requires systemic safety data. Over the past 10 years, our group performed a series of dose-efficacy and safety experiments in rats and mice for AAV8-hCocH without finding any viral vector–related toxicity.^24–27 The present study was undertaken to seek further systemic safety data of these high vector doses to define the upper dose limit in the future human trials.

Cocaine dependence is a complicated, destructive, and often chronic illness that is difficult to treat. Cocaine is one of the most addictive drugs because of its immediate and powerful rewarding effects. Often, cocaine-dependent individuals experience difficulty in abstaining due to cognitive impairment from repeated cocaine use, strong use–related social and environmental cues, and high levels of life stress. Development of treatments for cocaine dependence has been complicated by the tendency for abusers not to complete treatment programs and their propensity for relapse. Considering these challenges, we advocate a new treatment option using AAV gene therapy to attain long-term protective effects with a single shot. Animal studies show promise in successfully treating cocaine dependence. Combination of AAV-hCocH gene therapy with cognitive behavioral therapy may bring best results for treating cocaine addiction with respect to patient retention and relapse prevention after abstinence. More research is needed to test treatment programs to better understand and treat cocaine addiction.

We previously compared the therapeutic efficacy of AAV8-hCocH gene transfer with the administration of hCocH protein.^11,28 Efficacy studies included measures of cocaine half-life in untreated mice and rats versus animals that received either directly injected CocH protein or CocH gene transfer. These studies involved tail vein injection of benzene ring-labeled 3H-cocaine and subsequent evaluation of carefully timed plasma samples to determine levels of residual cocaine and benzoic acid (key reaction product of BChE-driven cocaine hydrolysis). With gene transfer stably expressing CocH, cocaine's in vivo half-life was very short (95% elimination in 4 min).

TEVA Pharmaceuticals recently tested a BChE protein version similarly modified for high activity in cocaine hydrolysis. That protein, TV-1380, is a form of human serum albumin fused at its amino terminus to the carboxyl-terminus of a truncated and mutated BChE. The pure protein was given intramuscularly to patients after extensive toxicology for proof-of-concept testing. Those trials²⁹ consisted of randomized, blinded phase 1 studies conducted to evaluate the safety, pharmacokinetics, and pharmacodynamics of TV-1380 after single and multiple administrations in healthy subjects. TV-1380 proved safe and well tolerated with a half-life of 43–77 h and showed a dose-proportional increase in systemic exposure. Consistent with preclinical results, ex vivo cocaine hydrolysis, TV-1380 activity rose in a dose-dependent manner, and there was a direct relationship between ex vivo cocaine hydrolysis and TV-1380 serum levels. There was no evidence that TV-1380 affected the heart rate or electrocardiography intervals. Thus, TV-1380 and similarly modified cholinesterases appear to offer a safe means to accelerate cocaine hydrolysis with little or no adverse impact.

To date, there has been no clinical research on AAV8-hCocH. We are proposing a first-in-human study with that vector encoding modified BChE. Considerable human experience in phase 1 and 2 clinical trials has demonstrated the safety and efficacy of i.v. AAV8 vector encoding various other therapeutic transgenes.^21,30 In 2011 and 2014, Nathwani et al. (sponsored by St. Jude Children's Research Hospital and University College London) reported long-term safety and efficacy of AAV8-Factor IX (FIX) in patients with severe hemophilia B after i.v. infusion of AAV8 vectors in human subjects.^31,32 That study had 10 patients: 2 patients at 2 × 10¹¹ vg/kg, 2 patients at 6 × 10¹¹ vg/kg, and 6 patients at 2 × 10¹² vg/kg. A single i.v. vector infusion in each patient led to dose-dependent increase in protein expression at 2–11% of the normal value of FIX (1 IU/dL). A maximal rise of transgenic protein was reached at 2–4 weeks after infusion. In some patients, FIX levels declined as ALT and AST rose, possibly due to immune rejection of transduced cells. But a short course of prednisolone returned liver enzymes to baseline and stabilized FIX levels.

In a follow-up 2014 study, Nathwani et al. reported FIX levels at 1–6% of normal over a median period of 3.2 years, and the highest dose level was most effective at reducing bleeding episodes in these patients. To date, the only reported side effects from i.v. infusion of AAV8 in human subjects were mild elevations in ALT and AST, effectively controlled by a short course of glucocorticoid therapy. All effects were asymptomatic and transient. No serious side effects were observed.

Overall, data from our laboratory and others provide a strong basis to conclude that a cocaine-hydrolyzing enzyme in adequate dose can drive a large drop in “reward value” of a given cocaine dose. Our studies on rats and mice prove that a reward from very large cocaine quantities can be brought low enough to eliminate all cocaine-seeking behavior. We do not anticipate equally dramatic effects in human users at low doses of viral vector. Nonetheless, as vector dosage escalates in clinical trial, rising BChE levels should weaken both the reward value and the toxicity of cocaine. Such effects should enable more users to resist relapse. The results of the current GLP compliant studies demonstrated that i.v. administration of AAV8-hCocH is generally well tolerated and provides long-term beneficial effects with cocaine exposure, which supports its advancement into human phase 1/2 clinical trial for patients with cocaine addiction.

Author Disclosure

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Funding Information

This research was supported by a Translational Avant-Garde Award from the National Institute on Drug Abuse (DA42492) and the Mayo Foundation for Medical Research. It was also supported by Regenerative Medicine Minnesota, by Mayo Clinic CTSA (UL1 TR002377), and by the National Center for Advancing Translational Science.

Supplementary Material

Supplementary Table S1

Supplementary Table S2