Long-Term Safety of Patients with Parkinson's Disease Receiving rAAV2-Neurturin (CERE-120) Gene Transfer

Abstract

The objective of this study was to assess the long-term safety of surgically administered recombinant adeno-associated virus serotype-2 (rAAV2)-neurturin (NRTN) to patients with advanced Parkinson's disease. Publications from prior trials reported no unexpected or troubling adverse events related to the vector or transgene comprising rAAV2-NRTN. Because rAAV2-NRTN produces long-term NRTN expression, subjects were enrolled in a long-term safety assessment protocol of rAAV2-NRTN. This article presents safety data for up to 5 years, well beyond that reported in the initial publications. Data from 53 patients are included; 47 received rAAV2-NRTN bilaterally to the putamen, whereas 6 subjects received rAAV2-NRTN bilaterally into putamen-plus-substantia nigra. Patients underwent in-person safety assessments on a quarterly to bi-annual basis, including adverse event monitoring, physical and neurological examinations, brain MRI, and laboratory testing. Parkinsonian status was assessed in an unblinded fashion. Fifty-three subjects completed the long-term safety protocol. Nine nonserious adverse events (non-SAEs) in 6 subjects were deemed “possibly related” to rAAV2-NRTN by the principal investigator, whereas none were deemed “related” to rAAV2-NRTN. Over the course of long-term observation, 33 SAEs were reported in 18 subjects, all of who received rAAV2-NRTN into putamen-only; 31 SAEs were deemed not related to rAAV2-NRTN, and 2 were deemed unlikely related. Safety assessments showed no clinically relevant changes in examination, imaging, or laboratory studies. Motor status, on average, was stable or apparently modestly improved (relative to baseline) over the course of the open-label, long-term follow-up. These findings provide evidence for the long-term safety of neurturin when delivered to the putamen or the putamen-plus-substantia nigra via stereotactic surgery and rAAV2 gene transfer. They therefore supplement the safety results reported in four prior publications from the same subjects, significantly extending the safety data for gene therapy and neurotrophic factor expression targeting the brain and adding to growing evidence that rAAV vector-mediated gene therapy to the CNS can be administered safely.

Introduction

Gene transfer is a promising technique for chronic delivery of neurotrophic factors to the brain, providing the potential to improve neuronal function and protect against neurodegeneration in conditions such as Parkinson's disease (PD). Neurturin is a naturally occurring neurotrophic factor that has been shown to provide trophic support for the midbrain dopamine neurons that degenerate in PD.^1,2 Recombinant adeno-associated virus serotype-2 (rAAV2)-neurturin (rAAV2-NRTN or CERE-120) is a gene therapy vector genetically engineered to express the human gene encoding for neurturin, resulting in production of the protein at the targeted site.³ In animal models of PD, rAAV2-NRTN demonstrated neuroprotective and neuroregenerative properties.^4
–6 Four prior human clinical trials have tested rAAV2-NRTN gene transfer in advanced PD. In phase 1 and 2 human clinical trials, each 12 months in duration, rAAV2-NRTN delivered to bilateral putamen via stereotactic surgery produced no clinically relevant adverse events related to rAAV2-NRTN, and adverse events related to the surgical delivery procedure were consistent with the nature of the procedure and all resolved.^7,8 Similarly, two subsequent trials (a revised phase 1 and a controlled phase 2b) administered a three-fold higher dose to the putamen, plus additional doses to the substantia nigra, and reported a similarly favorable safety profile after 24 months and 15–24 months, respectively.^9
–11 Because a single rAAV2-NRTN treatment appears to irreversibly alter recipient cells to produce persistent, long-term trophic factor expression,^12,13 it is important to follow patients for long posttreatment durations to help identify any possible long-term safety issues. To evaluate and characterize the long-term effects of rAAV2-NRTN, we conducted long-term follow-up safety evaluations for up to 5 years' duration, well beyond what was reported in the initial publications.

Methods

Fifty-three patients with advanced PD who had completed one of three clinical trials in which they received active treatment with rAAV2-NRTN have since been evaluated for longer-term safety effects in which they were followed for either 3 or up to 5 years. Forty-seven of these patients received gene therapy injections into each putamen (either in a phase 1 open-label study or in a phase 2a randomized, sham-surgery controlled study); these patients were followed for 5 years postdosing. The 6 remaining subjects received injections into each putamen-plus-substantia nigra (in a subsequent, revised phase 1 study); these patients were followed for three years postdosing via an amended long-term follow-up protocol.

Patients underwent a variety of in-person safety assessments on a quarterly to bi-annual basis, to include adverse event monitoring, physical and neurological examinations, MRI of the brain, and laboratory testing. In addition, evaluation of parkinsonian status using the Unified Parkinson's Disease Rating Scale (UPDRS) off-medication measurement of the Part III Motor Score and motor diaries was performed at these visits. Additional safety monitoring using structured phone interviews occurred at months 30 and 48. Data were collected using a fax-based case report system and stored in a database maintained and audited for accuracy and completeness by a contract research organization. Analysis of this database provided the basis for this report.

Results

Of the 56 patients participating in the trials who received rAAV2-NRTN (12 in an initial phase 1, 38 in a phase 2a (37 evaluable), and 6 in a revised phase 1), 53 enrolled in the follow-up study (12 from the initial phase 1, 35 from phase 2a, and all 6 from the revised phase 1). Patients enrolled into the sham arms of the trials are not included because the number of long-term observation cases in the sham group is small due to a high rate of dropouts after completion of the initial protocols. Of the subjects receiving rAAV2-NRTN, 41 completed the entire 5-year formal follow-up protocol, and the 6 from the revised phase 1 completed an amended 3-year formal follow-up protocol. Mean ages upon enrollment in each of the original phase 1, phase 2a, and revised phase 1 studies were 57, 60, and 50 years, respectively, and duration of PD diagnosis was 11, 9.5, and 8.8 years, respectively. The mean duration of post-rAAV2-NRTN treatment safety data reported here is 5.3 years (range 3.0–7.75 years). Six patients had received a total dose of 1.3 × 10¹¹ vg, 41 received a dose of 5.4 × 10¹¹ vg, 3 received a dose of 9.4 × 10¹¹ vg, and 3 received the highest dose of 24 × 10¹¹ vg (the latter was used for a subsequent phase 2b trial, which reported an acceptable safety profile up to 15 to 24 months).^10,11 Collectively, when considered with all earlier safety data reported for rAAV2-NRTN-treated subjects,^7

–11 these additional data represent a cumulative exposure of 299 patient-years to NRTN expression after rAAV2-NRTN administration (Table 1).

Table 1.

Summary of safety data reported for AAV2-NRTN in patients with Parkinson's disease (initial publication data plus additional data reported herein)

Study; anatomic target (protocol number)	Number treated	Original publication	Safety data initially reported (months)	Additional data here (months)	Total duration (months)	Cumulative treated subject-months
Initial phase 1; put-only (CERE-120-01)^a	12	Marks et al.⁷	12	48	60	720
Phase 2a; put-only (CERE-120-02)^b	35	Marks et al.⁸	12	48	60	2,220
Revised phase 1; put+SN (CERE-120-09)^c	6	Bartus et al.¹³	24	12	36	216
Phase 2b; put+SN (CERE-120-09)^d	24	Olanow et al.¹⁰	15–24	None	15–24	429
Total						3,585 months (299 years)

Open-label, phase 1 protocol testing two dose levels of AAV2-NRTN injected into each putamen.

Randomized, double-blind, sham-surgery controlled phase 2a trial of AAV2-NRTN injected into each putamen; 20 subjects also received sham surgeries.

Open-label, phase 1 protocol testing 2 dose levels of AAV2-NRTN injected into each putamen (one of which was 3-times higher than in two prior trials), plus injections of AAV2-NRTN injected into each substantia nigra.

Randomized, double-blind, sham-surgery controlled phase 2b trial of AAV2-NRTN injected into each putamen (3-times higher than phase 2a trial), plus injections into each substantia nigra; 27 subjects also received sham surgeries.

AAV2, adeno-associated virus serotype-2; NRTN, neurturin; put, bilateral putamen; SN, bilateral substantial nigra.

Adverse events

A total of 184 adverse events, of any severity, were reported by 44 subjects. Of these, 29 events in 17 subjects were categorized by the investigators as severe in nature. With the exception of 9 adverse events (4 mild, 4 moderate, and 1 severe) in 6 subjects that were assessed to be “possibly related” to rAAV2-NRTN, none of the other adverse events were deemed related to the treatment. One subject developed 3 episodes of psychosis, 3 subjects developed 1 episode each of dyskinesia, and 1 subject each reported dizziness, insomnia, and altered mood. Table 2 catalogs these adverse events reported over the course of the long-term assessment.

Table 2.

Adverse events reported during the five years of follow-up

Event type	Number of subjects	Number of events	Description of events
Adverse events of any severity	44	184	Neurological (32)Musculoskeletal (26)Psychiatric (22)Injury (16)Infectious (14)Gastrointestinal (12)Dermatological (9)Renal (9)Respiratory (7)Vascular (6)Metabolic (5)Ocular (5)Neoplastic (2)Urological (2)Cardiac (1)Hematological (1)Other (15)
Severe adverse events	17	29	Musculoskeletal (6)Neurological (5)Infectious (4)Psychiatric (4)Injury (3)Gastrointestinal (1)Metabolic (1)Ocular (1)Renal (1)Other (3)
Adverse events deemed possibly related to rAAV2-NRTN treatment	6	9	Dyskinesia (3)Psychosis (3 events in same subject)Dizziness (1)Insomnia (1)Mood alteration (1)

rAAV2-NRTN, recombinant adeno-associated virus serotype-2–neurturin.

There were 33 events in 18 rAAV2-NRTN-treated subjects over the course of the long-term follow-up observations that met the conventional criteria for serious adverse events (SAEs). Ten subjects experienced 1 adverse event each, 3 subjects experienced 2 adverse events, 3 subjects experienced 3 adverse events, and 2 subjects experienced 4 adverse events. Time of onset of the SAEs relative to treatment with rAAV2-NRTN ranged from 12.3 to 60.3 months. Thirty-one of these events were deemed by investigators as not related to rAAV2-NRTN treatment, and two were classified as unlikely related (an episode each of psychosis and gait disturbance, with onset occurring 38.2 and 57.6 months postdosing, respectively). In the revised phase 1 trial (which involved an increased rAAV2-NRTN dose to putamen plus additional injections into the substantia nigra), only two adverse events (neck pain and cervical spine stenosis) were noted during the additional one-year follow-up. Both occurred in the same subject and were deemed not related to treatment or surgical procedure. No SAEs were reported in the revised phase 1b study in which subjects received rAAV2-NRTN in the putamen and substantia nigra.

Medical and neurological examinations, imaging, and laboratory studies

Physical examinations, conducted every 6 months, revealed no clinically significant changes beyond those expected with progressive PD. Subjects underwent an MRI scan of the brain at month 24, with additional MRIs acquired if clinically indicated. No abnormalities related to the rAAV2-NRTN treatment were noted. Subjects underwent testing of complete blood count, chemistry, and liver and renal function every six months for the first two years and annually thereafter for the duration of the follow-up monitoring. No clinically significant laboratory abnormalities occurred.

Motor function

Long-term motor function showed no evidence of accelerated deterioration. The uncontrolled nature of these data does not permit conclusions regarding efficacy but does support the safety of the intervention. Compared with baseline, the mean UPDRS motor score in the off-medication state decreased by 14.9% (from 41.8 to 33) for the subjects originally enrolled in the initial phase 1 study and increased by 3.2% (from 38.6 to 38.8) for the subjects treated in the phase 2 study at the final 5-year assessment time point. The 24-hour, self-reported home motor diaries suggested a mean decrease in “off time” of 1.9 hours/day for the original phase 1 subjects and 3.0 hours/day for phase 2a subjects. Mean increase in “on time without troubling dyskinesia” was 2.4 hours/day and 1.5 hours/day for phase 1 and phase 2 subjects, respectively. Similar formal motor assessments are not available for the six revised phase 1 subjects (putamen-plus-substantia nigra dosing). However, based on discussions with the site investigators, none of the subjects' PD symptoms were markedly different from what was reported at 24 months postdosing,⁹ which showed modest improvements relative to predosing baseline scores.

Discussion

These long-term follow-up safety data complement those previously reported for stereotactically administered rAAV2-NRTN,^7

–11 thereby significantly extending the duration of human postdosing safety data that exist for rAAV2-NRTN specifically and rAAV-mediated gene therapy to the brain generally. Building on the shorter-term evaluations of the initial publications, this long-term follow-up assessment demonstrates the persistence of an excellent safety profile for several years. In addition to the findings from the 53 subjects in this report, an additional 24 subjects who had been administered higher-dose rAAV2-NRTN to both putamen and substantia nigra and followed for 15–24 months demonstrated no clinically significant adverse effects attributed to rAAV2-NRTN.¹²

Importantly, all available preclinical and autopsy data confirm that NRTN continues to be expressed in these subjects. Animal data from several gene therapy programs, including rAAV2-NRTN (for review, see Bartus et al.¹³) have consistently shown long-term (i.e., several years), continuous transgene expression, with no evidence of toxicity, inflammation, or any cell duress or loss. Moreover, data from patients enrolled in these rAAV2-NRTN trials,^12,14 as well as several others enrolled in a related rAAV2-NGF Alzheimer's trial, provide direct evidence for the persistence of long-term, bioactive protein expression in human brain after injection of an rAAV2 vector containing a CAG promoter and neurotrophic factor human transgene. Careful histological evaluation found no evidence of a decrease in NRTN expression or any negative consequences to neurons in the region of NRTN expression.^12,14

The doses administered in the rAAV2-NRTN trials were well within the range shown to be effective in animal models of PD. For example, in the most recent AAV2-NRTN trials,^9
–11 a putaminal dose of 1.0 × 10¹² vg was used, compared with a dose of 2.3 × 10¹⁰ vg shown to be effective in 6-OHDA rats⁵ and a dose of 5.13 × 10¹¹ vg shown to be effective in MPTP monkeys.⁶ When interspecies differences in size of putamen or striatum are taken into account (e.g., human putamen: 4000 mm³; primate putamen: 700 mm³; rat striatum: 27.5 mm³), the human dose was 43 times higher than the rat “scaled, human-equivalent dose” and nearly twice as great as the monkey “scaled, human-equivalent dose.” Moreover, the human rAAV2-NRTN dose is greater than 3 of the 4 doses planned for the ongoing rAAV2-GDNF trial. Perhaps just as importantly, the concentration of expressed protein within the sphere of targeted putamen is estimated to be approximately 8.3 × 10⁸ vg/mm³ (based on dose delivered and projected spread of protein within putamen), which exceeds that projected for three of the four rAAV2-GDNF doses, with only the highest dose approaching that of rAAV2-NRTN, as a result of dilution of vector from CED-induced wider distribution throughout the putamen.

No SAEs were reported for patients in extended observation after combined rAAV2-NRTN administration to the putamen and substantial nigra. Although the number of subjects we report here who received injections into both the putamen and substantia nigra is small (n = 6), their data are supplemented by the original 24-month safety report⁹ and the additional cohort of 24 patients who were treated in the subsequent phase 2b trial.^10,11 The lack of any safety issues in any of these human studies is important; for before conducting these studies, safety concerns were raised about targeting the substantia nigra with a gene therapy vector and/or expressing NRTN in or near the substantia nigra or other brain stem sites, based on certain reports in animals.^15
–17 The absence of any safety issues several years after persistent NRTN expression in both putamen and substantia nigra confirms the safety data collected in animals specifically with rAAV2-NRTN ^14,18 and therefore provides greater confidence that properly designed preclinical studies testing doses appropriately scaled to account for volumetric, species differences in targeted structures can help predict safety outcomes in human gene therapy trials.

In conclusion, the findings from the present longitudinal study provide the largest body of human evidence to date for the long-term safety of neurturin specifically and potentially neurotrophic factors generally. They also argue that the process of ectopic expression in the brain of transgene proteins (apart from the pharmacology of the specific transgene) after gene transfer can be applied safely. Along with autopsy data demonstrating accurately targeted, biologically active, long-term NRTN expression in the human brain,^12,19 they add to the growing evidence that rAAV-mediated gene therapy to the CNS offers a safe, effective, and practical means for long-term transgene expression as a possible therapeutic approach to treating human brain disorders.¹³

Footnotes

Acknowledgments

The CERE-120 Study Group is composed of a wide variety of investigators who contributed a range of different activities toward the CERE-120 clinical program. They are listed according to the following subgroups:

Principal investigators: Jill Ostrem, Mark Stacy, Leo Verhagen, Catherine Cho, Joseph Jankovic, Matthew Stern, Ray L. Watts, Jerold Vitek, Milind Deogaonkar, Stewart Factor, John Nutt, and Michele Tagliati.

Study coordinators: Robin Taylor, Lisa Gauger, Rebecca McMurray, Peggie Smith, Ricardo Renvill, Christine Hunter, Lina Shinawi, Heather Maccarone, Jeff Worrell, Yvette Ellis, Patricia St. Marie, and April Wilson.

CERE-120 steering committee members: Warren Olanow, Andres Lozano, Anthony Lang, Karl Kieburtz, and Charles Davis.

Neurosurgeons: Paul Larson, Philip Starr, Roy Bakay, Ron Alterman, Nicholas Boulis, Dennis Turner, Barton Guthrie, Richard Simpson, and Gordon Baltuch.

CERE-120 SAB members: Eugene Johnson, Jeffrey Kordower, Jude Samulski, and Fred H. Gage.

Key Ceregene personnel involved with various aspects of the CERE-120 clinical program: Joao Siffert, Jeff Ostrove, David Weiner, Christopher Herzog, and Dan Lee.

Sangamo Biosciences supporting personnel: Dale Ando and Shelley Wang.

Additionally, all authors acknowledge with appreciation the advice and suggestions received from Inder Verma (on Ceregene's SAB) and Floyd Bloom (on an independent CERE-120 Steering Committee) during different phases of the CERE-120 program, as well as the competitive financial grant support received for the CERE-120 trials from the Michael J. Fox Foundation for Parkinson's Research.

Author Disclosure

In 2013, Sangamo Biosciences, Inc., acquired all of Ceregene's intellectual property, including the rights to develop and market CERE-120. Although the authors and certain members of the CERE-120 Study Group received different forms of remuneration from Ceregene during the initial conduct of this study (e.g., reimbursement for conducting the trial, consulting fees, or salary), before or at the time of the Sangamo acquisition, no subsequent payments have been provided by Ceregene. R.T.B., T.B., and C.H. temporarily received consulting compensation from Sangamo Biosciences, but those relationships have since been suspended, well before work on this publication began, and none of those consulting activities were related to any work involving this publication or the long-term follow-up data presented.

References

Kotzbauer

, Lampe

, Heuckeroth

, et al. Neurturin, a relative of glial-cell-line-derived neurotrophic factor. Nature, 1996; 384:467–470.

Horger

, Nishimura

, Armanini

, et al. Neurturin exerts potent actions on survival and function of midbrain dopaminergic neurons. J Neurosci, 1998; 18:4929–4937.

Gasmi

, Herzog

, Brandon

, et al. Striatal delivery of neurturin by CERE-120, an AAV2 vector for the treatment of dopaminergic neuron degeneration in Parkinson's disease. Mol Ther, 2007; 15:62–68.

Herzog

, Dass

, Holden

, et al. Striatal delivery of CERE-120, an AAV2 vector encoding human neurturin, enhances activity of the dopaminergic nigrostriatal system in aged monkeys. Mov Disord, 2007; 22:1124–1132.

Gasmi

, Brandon

, Herzog

, et al. AAV2-mediated delivery of human neurturin to the rat nigrostriatal system: Long-term efficacy and tolerability of CERE-120 for Parkinson's disease. Neurobiol Dis, 2007; 27:67–76.

Kordower

, Herzog

, Dass

, et al. Delivery of neurturin by AAV2 (CERE-120)-mediated gene transfer provides structural and functional neuroprotection and neurorestoration in MPTP-treated monkeys. Ann Neurol, 2006; 60:706–715.

Marks

Jr. , Ostrem

, Verhagen

, et al. Safety and tolerability of intraputaminal delivery of CERE-120 (adeno-associated virus serotype 2-neurturin) to patients with idiopathic Parkinson's disease: An open-label, Phase I trial. Lancet Neurol, 2008; 7:400–408.

Marks

, Bartus

, Siffert

, et al. Gene delivery of AAV2-neurturin for Parkinson's disease: A double-blind, randomised, controlled trial. Lancet Neurol, 2010; 9:1164–1172.

Bartus

, Baumann

, Siffert

, et al. Safety/feasibility of targeting the substantia nigra with AAV2-neurturin in Parkinson patients. Neurology, 2013; 80:1–4.

10.

Olanow

, Bartus

, Baumann

, et al. Gene delivery of neurturin to putamen and substantia nigra in Parkinson disease: A double-blind, randomized, controlled trial. Ann Neurol, 2015; 78:248–257.

11.

Bartus

, Kordower

, Johnson

Jr. , et al. Post-mortem assessment of the short and long-term effects of the trophic factor neurturin in patients with α synucleinopathies. Neurobiol Dis, 2015; 78:162–171.

12.

Bartus

, Kordower

, Johnson

Jr. , et al. Post-mortem assessment of the short and long-term effects of the trophic factor neurturin in patients with α synucleinopathies. Neurobiol Dis, 2015; 78:162–171.

13.

Bartus

, Weinbert

, Samulski

. Parkinson's disease gene therapy: Success by design meets failure by efficacy. Mol Ther, 2014; 22:487–497.

14.

Bartus

, Brown

, Wilson

, et al. Properly scaled and targeted AAV2-NRTN (neurturin) to the substantia nigra is safe, effective and causes no weight loss: Support for nigral targeting in Parkinson's disease. Neurobiol Dis, 2011; 44:38–52.

15.

Manfredsson

, et al. Tight long-term dynamic doxycycline responsive nigrostriatal GDNF using a single rAAV vector. Mol Ther, 2009; 17:1857–1867.

16.

Manfredsson

, et al. Nigrostriatal rAAV-mediated GDNF overexpression induces robust weight loss in a rat model of age-related obesity. Mol Ther, 2009; 17:980–991.

17.

, Bankiewicz

. Safety evaluation of AAV2-GDNF gene transfer into the dopaminergic nigrostriatal pathway in aged and parkinsonian rhesus monkeys. Hum Gene Ther, 2009; 20:1627–1640.

18.

Herzog

, Brown

, Kruegel

, et al. Enhanced neurotrophic distribution, cell signaling and neuroprotection following substantia nigral versus striatal delivery of AAV2-NRTN (CERE-120). Neurobiol Dis, 2013; 58C:38–48.

19.

Bartus

, Herzog

, Chu

, et al. Bioactivity of AAV2-neurturin gene therapy (CERE-120): Differences between Parkinson's disease and nonhuman primate brains. Mov Disord, 2011; 26:27–36.