Randomized Trial of Ciliary Neurotrophic Factor Delivered by Encapsulated Cell Intraocular Implants for Retinitis Pigmentosa


To evaluate the safety and effect on visual function of ciliary neurotrophic factor delivered via an intraocular encapsulated cell implant for the treatment of retinitis pigmentosa (RP).


Ciliary neurotrophic factor for late-stage retinitis pigmentosa study 3 (CNTF3; n = 65) and ciliary neurotrophic factor for early-stage retinitis pigmentosa study 4 (CNTF4; n = 68) were multicenter, sham-controlled dose-ranging studies.


Patients were randomly assigned to receive a high- or low-dose implant in 1 eye and sham surgery in the fellow eye. The primary endpoints were change in best-corrected visual acuity (BCVA) at 12 months for CNTF3 and change in visual field sensitivity at 12 months for CNTF4. Patients had the choice of retaining or removing the implant at 12 months for CNTF3 and 24 months for CNTF4.


There were no serious adverse events related to either the encapsulated cell implant or the surgical procedure. In CNTF3, there was no change in acuity in either ciliary neurotrophic factor– or sham-treated eyes at 1 year. In CNTF4, eyes treated with the high-dose implant showed a significant decrease in sensitivity while no change was seen in sham- and low dose–treated eyes at 12 months. The decrease in sensitivity was reversible upon implant removal. In both studies, ciliary neurotrophic factor treatment resulted in a dose-dependent increase in retinal thickness.


Long-term intraocular delivery of ciliary neurotrophic factor is achieved by the encapsulated cell implant. Neither study showed therapeutic benefit in the primary outcome variable.

Retinitis pigmentosa (RP) affects approximately 100,000 Americans. It is a group of retinal degenerative diseases that have a complex molecular etiology. More than 100 mutations in several genes, including rhodopsin ( RHO ), peripherin ( PRPH2 ), and PDEβ , are believed to be responsible for RP, although the genotypes of the majority of RP patients are unknown. Despite the genetic heterogeneity, patients typically experience decreased night vision early in life attributable to the loss of rod photoreceptors. While the genetic defects primarily affect rods, progressive outer retinal degeneration leads to progressive visual field loss and, ultimately, severe visual disability. Since the molecular cause underlying the retinal degeneration is not known for most patients, an approach to slow progressive loss of photoreceptors that is effective for many different genetic forms of inherited retinal degeneration would have broad applicability.

The promise of growth factors as potential therapeutics for photoreceptor degeneration was first demonstrated in 1990. Since then, many growth factors, neurotrophic factors, and cytokines have been tested in a variety of photoreceptor degeneration models, mainly by intravitreal injection of purified recombinant proteins in short-term experiments. Among them, ciliary neurotrophic factor has been shown to be the most effective in numerous animal models. However, the chronic nature of RP (years to decades) and the short-term effectiveness of purified recombinant ciliary neurotrophic factor make repetitive intraocular injections impractical.

One of the major challenges in the treatment of RP is the safe and effective local delivery of therapeutic macromolecules to the retina. The encapsulated cell technology implant (NT-501; Neurotech USA, Lincoln, Rhode Island, USA) was designed specifically to address this challenge. The encapsulated cell technology implant enables the controlled, continuous, and long-term delivery of therapeutic macromolecules, including neurotrophic factors, directly into the vitreous cavity inside the eye. In addition, encapsulated cell implants can be retrieved, thus providing an additional level of safety.

Ciliary neurotrophic factor decreases photoreceptor loss during retinal degeneration. Although its intrinsic function is not fully understood, exogenous ciliary neurotrophic factor affects the survival and differentiation of cells in the nervous system, including retinal cells. It effectively protected photoreceptors in 12 animal models of photoreceptor degeneration. Further, ciliary neurotrophic factor has passed appropriate milestones in a phase 1 human clinical study of RP. The key question for patients and clinicians is whether it can be used safely in the treatment of retinal degeneration in humans.

Two phase 2 studies were designed to demonstrate the safety profile of NT-501 in patients with early and more advanced RP, to evaluate the effect of ciliary neurotrophic factor on retinal structure and function, and to evaluate dose and primary endpoints for future studies. The primary endpoints selected here were change in best-corrected visual acuity (BCVA) at 12 months for ciliary neurotrophic factor for late-stage retinitis pigmentosa study 3 (CNTF3) and change in visual field sensitivity at 12 months for ciliary neurotrophic factor for early-stage retinitis pigmentosa study 4 (CNTF4).


Study Design

A total of 65 and 68 patients were enrolled at 13 sites in the United States for the CNTF3 and CNTF4 studies, respectively ( Table 1 ). Approvals were received from the National Institutes of Health Recombinant DNA Advisory Committee, from the Food and Drug Administration (FDA), and from the Institutional Review Board and Institutional Biosafety Committee at each site prior to enrollment. The Institutional Review Boards responsible for these studies are listed in the acknowledgment at the end of this article. Subjects signed written informed consent before determination of their full eligibility.

Table 1

Baseline Characteristics of Patients Receiving Encapsulated Cell Intraocular Implants for Retinitis Pigmentosa

Low Dose High Dose Low Dose High Dose
Male 14 (63.3%) 20 (46.5%) 10 (50.0%) 23 (47.9%)
Female 8 (36.4%) 23 (53.5%) 10 (50.0%) 25 (52.1%)
White 18 (81.8%) 37 (86.0%) 18 (90.0%) 47 (97.9%)
Non-Hispanic/Latino 19 (86.4%) 41 (95.3%) 20 (100.0%) 44 (91.7%)
Mean (SD) 41.1 (10.5) 42.0 (11) 34.9 (12) 40.2 (11.8)
Median 41.0 43.0 36.0 41.5
Range 24-59 18-67 18-58 18-59
Implant/Sham Implant/Sham Implant/Sham Implant/Sham
Mean (SD) 45.5 (11.2)/44.8 (10.4) 45.0 (10.4)/46.7 (9.1) 79.2 (7.5)/78.9 (7.2) 78.9 (6.9)/78.7 (6.4)
Median 45.9/46.1 44.9/46.8 81.0/78.8 79.5/79.5
Range 25-64/24-65 25-65/26-62 63-91/63-91 59-90/66-92
Total mac vol (mm 3 )
Mean (SD) 6.0 (0.8)/6.0 (0.7) 6.2 (1.0)/6.3 (1.1) 6.3 (1.3)/6.4 (1.4) 6.3 (0.8)/6.3 (0.8)
Median 6.1/6.0 6.0/6.1 6.1/6.4 6.2/6.2
Range 4.7-8.2/4.7-7.9 4.3-9.1/4.5-9.3 4.5-10.6/4.7-11.2 5.1-8.3/4.9-8.4
Electroretinogram (μV) a
Mean (SD) 8.32 (2.8)/8.26 (3.1) 14.3 (3.0)/12.7 (3.0) 15.4 (2.3)/17.2 (2.4) 22.2 (2.4)/22.4 (2.5)
Median 8.2/8.0 13.8/10.9 15.0/21.0 21.5/21.9
Range 2.6-65.3/1.8-58.3 3.0-84.0/3.3-88.2 3.6-65/3.7-52 5.4-149/3.9-157
Visual field sensitivity (dB)
Mean (SD) 332 (502)/323 (494) 423 (488)/444 (496) 1142 (446)/1136 (424) 1007 (429)/998 (466)
Median 202/175 210/209 1053/1095 965/966
Range 31-2307/17-2247 5-1996/2-1940 538-1885/526-1875 340-2176/220-2276

BCVA = best-corrected visual acuity (letters read by Electronic Visual Acuity); CNTF3 = ciliary neurotrophic factor for late-stage retinitis pigmentosa study 3; CNTF4 = ciliary neurotrophic factor for early-stage retinitis pigmentosa study 4; Mac vol = macular volume.

a White flash – Amplitude.

Each participant’s clinical diagnosis was consistent with retinitis pigmentosa.

The CNTF3 study–specific inclusion criteria included: age 18-68 years, BCVA of 20/63-20/320 (Snellen equivalent determined with the use of an Early Treatment Diabetic Retinopathy Study [ETDRS] chart), and absence of cystoid macula edema (CME) as judged by time-domain optical coherence tomography (OCT). The CNTF4 study–specific inclusion criteria were as follows: age 18-65 years with BCVA of 20/63 or better. Patients with CME were permitted. Each eye had a mean sensitivity deviation of at least 6 dB loss of static perimetric sensitivity, on average, throughout the central 60-degree-diameter field (including non-zero points). Each eye sensitivity was to have a non-zero value for at least 30 locations, and the horizontal field extent was to be 20 degrees or greater as tested on a Humphrey field analyzer (HFAII) 30-2 test with a Goldmann V target size. The eligibility of subjects was confirmed by an independent central reading center according to standardized criteria with trained fundus photograph and OCT graders who were masked to subjects’ treatment assignment.

Safety visits were conducted at 1 day, 1 week, and 1, 3, 6, and 12 months. Blood draws for laboratory safety studies, including serum antibodies, were obtained on each visit. The primary efficacy endpoints, change in BCVA and change in visual field sensitivity, were prespecified at 12 months post implant for CNTF3 and CNTF4, respectively. Patients received either high- or low-dose NT-501 implants in 2:1 ratio in 1 eye, and a sham treatment in the fellow eye. The high dose was selected based on the dose-response effect of ciliary neurotrophic factor in the rcd1 model of retinal degeneration and was the maximum effective dose. The low dose was 50% of the minimum effective dose in the rcd1 dog model.

The original trial design approved by the FDA specified that implants be removed at 12 months (CNTF3) or 24 months (CNTF4). After the initiation of the trial, the FDA recommended that patients retain the implants at the end of the study (avoiding a second surgery). Since all patients had consented to have their implants removed at the end of the study, they were offered a choice either to keep the implant in place or to have the implant removed. For each study, patients were followed for an additional 6 months (a total of 18 months follow-up for CNTF3 and 30 months for CNTF4). At the conclusion of the trial, 16 patients in the high-dose CNTF4 study (10 with the implant in place and 6 with the implant removed) consented to a registry study with an additional year of follow-up (42 months). Those with the implant removed were also tested at 54 months.

BCVA was measured by an electronic visual acuity tester (EVA) using the ETDRS protocol. BCVA in CNTF3 was measured twice per eye on each of 3 baseline visits. Baseline 1 BCVA was used to qualify subjects and baseline 2 and 3 BCVA (average of 4 measures per eye) was used as baseline BCVA. Three BCVA measurements were taken for each subsequent visit and the average of the 3 BCVA values was used to assess the change from baseline.

Visual field sensitivity in CNTF4 was measured with the 30-2 grid using the Humphrey visual field analyzer. Visual field sensitivity was measured twice per eye on each of 3 baseline visits. Eligibility for enrollment was determined by the results of baseline 1. The average value of the 4 baseline 2 and baseline 3 examinations, each representing the sum of actual thresholds for all 76 locations, provided the baseline visual field sensitivity. Four visual field sensitivity measurements per eye were taken for each subsequent visit and the average of the 4 visual field sensitivity sums was used to assess the change from baseline. Pupil diameter was measured within the Humphrey visual field analyzer at the conclusion of each test session.

Full-field electroretinograms (ERGs) were measured at baseline and at 12 months. The procedure adhered to International Society of Clinical Electrophysiology of Vision standards, but was limited to light-adapted (30 Hz flicker and single-flash) responses.

Retinal thickness and morphology were evaluated by OCT. The fast macular thickness map protocol, a 7-mm horizontal line scan, and 6-mm vertical line scan were obtained with the Stratus OCT and software version 4.0 or higher (Carl Zeiss Meditec, Inc, Dublin, California, USA). OCT images were collected by certified technicians. The images were evaluated by masked readers at the Duke University OCT Reading Center and analyzed for average thickness at center point, total macular volume, and average thickness in 9 subfields. Pathologic findings, such as cysts, epiretinal membrane, vitreomacular traction, and choroidal neovascularization, were also recorded and analyzed.

A subset of all high-dose CNTF4 patients (n = 10) from a single center (Dallas, Texas, USA) were evaluated on the 12-month visit by spectral-domain OCT (Spectralis HRA + OCT; Heidelberg Engineering, Heidelberg, Germany). The images of horizontal midline scans were exported to data-analysis software (Igor Pro; WaveMetrics, Inc, Portland, Oregon, USA) and segmented to identify the Bruch membrane/choroid boundary, the ellipsoid zone (inner/outer segment border), and the inner nuclear layer/outer plexiform layer boundary. Using the locations of these boundaries, a masked reader defined the receptor outer segment plus retinal pigment epithelium (OS+) as the distance between the ellipsoid zone and Bruch membrane/choroid boundaries. The outer nuclear layer was the distance between the inner nuclear layer/outer plexiform layer and ellipsoid zone boundaries. In order to avoid possible complications from CME, segment thicknesses were determined 1.7 mm (6 degrees) nasal and temporal to the fovea and compared between high-dose ciliary neurotrophic factor–implanted and sham-treated eyes. Since spectral-domain OCT scans were not available at baseline, comparisons were between the implanted and sham eyes at 12 months.

Study Treatment

The ciliary neurotrophic factor–secreting, encapsulated cell implants, designated NT-501, are 6 mm long with 1 mm diameter and are constructed of a semi-permeable polymer outer membrane. The low-dose implants released 5 ng/day and the high-dose implant released 20 ng/day prior to implant.

Statistical Analysis

The efficacy analysis was performed on an intent-to-treat basis among all subjects. Since all subjects completed the study as planned, the last-observation-carried-forward method for missing data was not used. For change in BCVA, ERG, and visual field sensitivity, the within-group and between-group comparisons were based on a paired t test. Clinical response rates were compared between groups using a 2-sided Fisher exact test. For retinal thickness change as measured by OCT, the overall comparison among treatment medians was assessed using the Kruskal-Wallis test. Pair-wise differences between treatment medians were assessed using the Wilcoxon rank sum test.

Evaluation of Encapsulated Cell Implants After Removal

Immediately upon removal, the devices were placed into Endo-SFM conditioned medium (GIBCO BRL, Gaithersburg, Maryland, USA) at 37 C, 5% CO 2 , 95% humidity for 24 hours. The rate of ciliary neurotrophic factor secretion was determined using a commercial enzyme-linked immunosorbent assay kit (R&D Systems, Minneapolis, Minnesota, USA).


Study Patients

Between January 8, 2007 and October 31, 2007, 65 patients and 68 patients were enrolled into the CNTF3 and CNTF4 studies, respectively, and were randomly assigned to study treatment. Groups were balanced for demographic and baseline ocular characteristics ( Table 1 ). All patients completed the 12-month primary endpoint follow-up and no patients dropped out of the study.

Safety Profile

Cumulative adverse events for the 12-month study period (CNTF3 and CNTF4) are summarized in Table 2 . The most frequent adverse event was miosis, measured on the Humphrey field analyzer in 25.6% of CNTF3 and 31.3% of CNTF4 patients assigned to receive the high-dose implant. Unequal pupil sizes were also reported by many of these patients. Although neither the field technicians nor the patients knew whether the implant was causing miosis or dilation, the unequal pupil sizes could possibly have interfered with masking. No serious adverse events related to the NT-501 implant or surgical procedures were reported during the 12-month study period. No treatment-related severe adverse effects, including retinal detachment, endophthalmitis, intraocular pressure (IOP) increase, or choroidal neovascularization (CNV), were reported. Neither ciliary neurotrophic factor nor antibodies against it was detected in the serum. Likewise, no antibodies against the encapsulated cells were detected.

Table 2

Adverse Events at 12 Months in Patients Receiving Encapsulated Cell Intraocular Implants for Retinitis Pigmentosa

Adverse Events/Eye Disorders CNTF3 CNTF4
Low Dose (n = 22) High Dose (n = 43) Low Dose (n = 20) High Dose (n = 48)
Intraocular pressure increase a 0 (0.0%) 1 (2.3%) 0 (0.0%) 1 (2.1%)
Eye hemorrhage b 2 (9.1%) 0 (0.0%) 1 (5%) 0 (0.0%)
Photopsia 0 (0.0%) 1 (2.3%) 0 (0.0%) 4 (8.3%)
Miosis 1 (4.5%) 11 (25.6%) 0 (0.0%) 15 (31.3%)
Cataract c 1 (4.5%) 2 (4.7%) 0 (0.0%) 2 (4.2%)
Choroidal neovascularization 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%)
Wound leaks or erosion 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%)
Endophthalmitis 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%)
Implant extrusion 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%)
Retinal detachment 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%)

CNTF3 = ciliary neurotrophic factor for late-stage retinitis pigmentosa study 3; CNTF4 = ciliary neurotrophic factor for early-stage retinitis pigmentosa study 4.

a Intraocular pressure increase (24-31 mm Hg) usually lasted a few days to a few weeks and pressure returned to normal at the next scheduled visit without medical intervention.

b Related to the surgical wound and recovered with no sequelae within 10 days.

c Worsening of a pre-existing cataract (mild).

Visual Acuity Changes

No significant changes in visual acuity were observed in ciliary neurotrophic factor–treated or sham-treated eyes in CNTF3 and CNTF4 patients ( Table 3 ).

Table 3

Summary of Changes From Baseline at 12 Months in Patients Receiving Encapsulated Cell Intraocular Implants for Retinitis Pigmentosa

Endpoint CNTF3 CNTF4
Low-Dose Implant/Sham High-Dose Implant/Sham Low-Dose Implant/Sham High-Dose Implant/Sham
Change in best-corrected visual acuity
Month 12
Mean (SD) −2.9 ± 11.3/−2.3 ± 12 −1.3 ± 9.6/−3.2 ± 10.5 −0.5 ± 5.0/0.5 ± 4.7 0.7 ± 4.1/1.0 ± 4.6
Range −48-11/−47-11 −25-14/−45-15 −16-7/−15-6 −7-12/−9-11
P value .650 .275 .497 .646
Change in total macular volume (mm 3 )
Month 12
Mean (SD) 0.09 ± 0.23/−0.05 ± 0.17 0.23 ± 0.58/−0.02 ± 0.18 0.22 ± 0.21/−0.05 ± 0.38 0.43 ± 0.37/−0.05 ± 0.27
Range −0.4-0.4/−0.5-0.2 −1.2-2.2/−0.5-0.3 −0.2-0.7/−1.2-0.8 −0.3-1.7/−1.3-0.8
P value .148 <.001 .001 <.001
Change in electroretinogram (μV) (geometric means)
Month 12
Mean (SD) 1.10 ± 1.5/0.99 ± 1.48 0.78 ± 1.5/0.89 ± 1.73 1.03 ± 1.62/0.99 ± 1.48 0.79 ± 1.52/0.87 ± 1.73
Range 0.5-2.9/0.6-2.3 0.2-1.4/0.1-2.1 0.4-2.9/0.6-2.3 0.2-1.4/0.1-2.1
P value .347 .129 .776 .242
Change in Humphrey visual field sensitivity (dB)
Month 12
Mean (SD) 4.1 ± 109.5/4.7 ± 101.4 −98.4 ± 165.3/−14 ± 101.5 1.4 ± 174.6/16.8 ± 165.4 −164.3 ± 114.6/−67.1 ± 104.2
Range −172-282/−213-374 −487-242/−294-286 −265-354/−191-423 −489-95/−322-152
P value .97 .001 .137 <.001

CNTF3 = ciliary neurotrophic factor for late-stage retinitis pigmentosa study 3; CNTF4 = ciliary neurotrophic factor for early-stage retinitis pigmentosa study 4.

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Jan 9, 2017 | Posted by in OPHTHALMOLOGY | Comments Off on Randomized Trial of Ciliary Neurotrophic Factor Delivered by Encapsulated Cell Intraocular Implants for Retinitis Pigmentosa

Full access? Get Clinical Tree

Get Clinical Tree app for offline access