TY - JOUR
AU - Group, for the rhNGF Clinical Investigator
AB - Abstract Context Nerve growth factor is a neurotrophic factor that promotes the survival of small fiber sensory neurons and sympathetic neurons in the peripheral nervous system. Recombinant human nerve growth factor (rhNGF) has demonstrated efficacy as treatment for peripheral neuropathy in experimental models and phase 2 clinical trials. Objective To evaluate the efficacy and safety of a 12-month regimen of rhNGF in patients with diabetic polyneuropathy. Design Randomized, double-blind, placebo-controlled phase 3 trial conducted from July 1997 through May 1999. Setting Eighty-four outpatient centers throughout the United States. Patients A total of 1019 men and women aged 18 to 74 years with either type 1 or type 2 diabetes and a sensory polyneuropathy attributable to diabetes. Interventions Patients were randomly assigned to receive either rhNGF, 0.1 µg/kg (n = 504), or placebo (n = 515) by subcutaneous injection 3 times per week for 48 weeks. Patients were assessed at baseline, 12 weeks, 24 weeks, and 48 weeks. Main Outcome Measures The primary outcome measure was a change in neuropathy between baseline and week 48, demonstrated by the Neuropathy Impairment Score for the Lower Limbs, compared between the 2 groups. Secondary outcome measures included quantitative sensory tests using the CASE IV System, the Neuropathy Symptom and Change questionnaire, the Patient Benefit Questionnaire (PBQ), and a global symptom assessment, as well as nerve conduction studies and occurrence of new plantar foot ulcers. Patients also were evaluated for presence of adverse events. Results Among patients who received rhNGF, 418 (83%) completed the regimen compared with 461 (90%) who received placebo. Administration of rhNGF was safe, with few adverse events attributed to treatment apart from injection site pain/hyperalgesia and other pain syndromes. However, neither the primary end point (P = .25) nor most of the secondary end points demonstrated a significant benefit of rhNGF. Exceptions were the global symptom assessment (P = .03) and 2 of 32 comparisons within the PBQ, which showed a modest but significant benefit of rhNGF (P = .05 for severity of pain in the legs and P = .003 for 6-month symptoms in the feet and legs). Conclusion Unlike previous phase 2 trials, this phase 3 clinical trial failed to demonstrate a significant beneficial effect of rhNGF on diabetic polyneuropathy. Nerve growth factor (NGF) is a protein that plays a major role in the development and maintenance of the peripheral nervous system. Nerve growth factor selectively promotes the survival of small fiber sensory neurons that mediate pain, temperature sensation, and sympathetic neurons.1 Nerve growth factor is expressed in target tissues innervated by responsive neurons, where it binds to specific high-affinity receptors and is retrogradely transported back to the neuronal cell body.2 Recent data suggest that reduced availability of NGF may play a significant role in the pathogenesis of diabetic polyneuropathy. In animals with diabetes mellitus, retrograde axonal transport of NGF is impaired,3-5 and levels of NGF messenger RNA are reduced in neuronal target tissues.6 In patients with diabetes mellitus, levels of NGF are reduced in skin biopsy specimens. Furthermore, the diminished levels of NGF in the skin correlate with decreased skin axon-reflex vasodilatation, suggesting that reduced availability of NGF from the target tissue leads to early dysfunction of small fiber neurons.7 Systemic administration of NGF prevents manifestations of neuropathy in rodent models of toxic8,9 and diabetic polyneuropathy.10 That loss of NGF might contribute toward the pathogenesis of diabetic polyneuropathy plus the demonstrated ability of exogenous NGF to prevent experimental neuropathy provided a compelling rationale for testing NGF administration in clinical studies. Two randomized, placebo-controlled trials of recombinant human NGF (rhNGF) administered to patients with polyneuropathy were initiated. In a phase 2 trial of 250 patients with diabetic polyneuropathy, improvements in signs and symptoms were seen after treatment with rhNGF, 0.1 or 0.3 µg/kg, subcutaneously, 3 times per week for 6 months.11 A second phase 2 trial of 270 patients with human immunodeficiency virus (HIV)–associated sensory neuropathy demonstrated significant improvements in neuropathic pain following 18 weeks of treatment with rhNGF, 0.1 or 0.3 µg/kg, twice a week.12 Recombinant human NGF was well tolerated with the exception of self-limited injection site hyperalgesia and other pain-related syndromes. This article reports the results of a large randomized, double-blind, placebo-controlled, phase 3 study of rhNGF administered to patients with diabetic polyneuropathy to determine safety and efficacy during a 12-month period. Methods Patient Characteristics From July 1997 through May 1999, 84 US study sites recruited a total of 1019 patients aged 18 to 74 years with either type 1 or type 2 diabetes mellitus who were being treated with either insulin or oral hypoglycemic medications. Inclusion criteria were a sensory peripheral polyneuropathy attributable to diabetes mellitus as determined by a neurologist, and abnormal results of nerve conduction studies in 1 or more attributes of 2 or more nerves studied (≥99th percentile or ≤1st percentile, as appropriate) and abnormal results of quantitative sensory tests for cooling and/or heat pain thresholds (≥90th percentile but with obtainable thresholds). Patients were excluded if they had unstable glycemic control, other neurological diseases that might mimic features of neuropathy, proximal asymmetric neuropathy, cranial neuropathies, mononeuropathies, and clinically significant systemic diseases other than diabetes mellitus that might have confounded interpretation of the study results, such as peripheral vascular disease, history of cancer within the past 6 months, significant renal or hepatic disease, nondiabetic risk factors for neuropathy, or use of capsaicin within the month preceding the trial. Patients also were excluded if they previously had participated in a clinical trial of rhNGF. Every patient signed an informed consent document. Internal review boards reviewed and approved the protocol at each institution. Treatment Protocol Patients were randomly assigned in equal proportions to 1 of 2 treatment groups, according to a randomization schedule prepared by a study biostatistician to achieve equal sample sizes for the treatment groups overall and within each diabetes mellitus type (type 1 or 2). The randomization schedule was generated by an Interactive Voice Response System that assigned cartons of drug labeled with a unique double-blind identification number. Eligible patients were randomized to receive either rhNGF, 0.1 µg/kg, or placebo administered 3 times per week for 12 months, subcutaneously, through a 28-gauge needle and 1-mL syringe. Genentech, Inc (South San Francisco, Calif) provided rhNGF as a single-dose preparation formulated at a concentration of 0.1 mg/mL in 20 mmol/L of acetate solution with 136 mmol/L of sodium chloride. Each vial of placebo equivalent contained a 20 mmol/L-acetate solution with 300 mmol/L of sodium chloride. To assist with blinding, the hypertonic concentration of the sodium chloride solution in the placebo was chosen to mimic the injection site hyperalgesia associated with rhNGF administration.11,12 In addition, a separate physician apart from the examining neurologist at each site inquired about adverse events. Patient Monitoring and Evaluation Patients completed 2 screening visits and a randomization visit during which they provided a medical history; underwent a complete physical and neurological examination; routine laboratory screening (hematologic and blood chemistry analyses), including a complete blood cell count, percent glycosylated hemoglobin (hemoglobin A1C), SMA20, serum thyrotropin levels, and serum thyroxine (free T4) levels; vitamin B12 and folate levels; serum creatine kinase levels; an electrocardiogram; and a urinalysis. Women of childbearing potential had a serum sample drawn for a human chorionic gonadotropin test. At each of the pair of screening visits, patients underwent quantitative sensory testing, nerve conduction studies, and were evaluated using the Neuropathy Impairment Score (NIS),13 Neuropathy Symptoms and Change (NSC),11 and a monofilament test (a gross measure of touch pressure on the sole of the foot). During the double-blind treatment period, patients were seen by the site investigators every 28 days, during which the study drug and injection calendars were collected and dispensed and adverse events were monitored. At weeks 12, 24, and 48, the patients again were evaluated using the NIS, NSC, the quantitative sensory tests, nerve conduction studies, and the monofilament test. At month 12, these studies were performed twice within 10 days, as they were during the screening period, and the results were averaged to decrease variability at the beginning and end of the treatment period. At every third month, hematologic and blood chemistry analyses were performed as described above and anti-NGF antibody levels were checked. Site investigators were trained to perform the neurological evaluation, quantitative sensory testing, and nerve conduction studies in a uniform manner at joint training sessions. The raw data from the nerve conduction studies and the quantitative sensory testing were monitored to ensure that test performance, stimulus response characteristics, and results met predefined testing criteria. Symptom Assessment and Neurological Evaluation Overall neuropathic impairment as determined by neurological examination was assessed using a subset of the NIS that specifically assesses impairments in the lower limb (NIS-LL),13 providing an overall score based on a graded battery of muscle strength, reflexes, and sensory testing performed by the study site neurologist. Three different symptom assessments were used: (1) The NSC, a physician-completed questionnaire, which tallies number, severity, and change of symptoms, evaluates motor, autonomic, large fiber, and small fiber sensory nerve function in all extremities. (2) A global symptom assessment (1 question) administered at the conclusion of the treatment period queried patients on a 7-point scale as to whether their neuropathy symptoms had changed from pretreatment to posttreatment. (3) The Patient Benefit Questionnaire (PBQ) assessed quality of life and activities of daily living. Quantitative Sensory Tests Sensory perception was quantified using the CASE IV System (WR Medical Electronics Co, Stillwater, Minn). Three sensory modalities were measured, including cooling detection threshold (CDT), an intermediate response of pain from graded heating pulses (HP: 5.0) (both measures of small-diameter fiber function), and vibratory detection threshold (VDT, a measure of large-diameter fiber function), using techniques and algorithms previously described.14-18 A 4, 2, and 1, stepping with null stimuli algorithm was used for the CDT and VDT testing.17 Thresholds were expressed as stimulus steps,1-25 as units of displacement (VDT), change in degrees (CDT), and as normal deviates. Percentiles and normal deviates were estimated for age, sex, anatomical site, and other physical characteristics based on a previous normative study of more than 300 patients.14-18 Nerve Conduction Studies Nerve conduction studies were performed or supervised by American Board of Electromyography–certified physicians at each site, according to a standardized uniform protocol, and monitored at a quality assurance center. For each patient, recordings were obtained from the sural sensory and peroneal motor nerves as well as the ulnar motor and sensory nerves. Bilateral sural nerve studies were performed first, and if the responses were present, the peroneal and ulnar nerve studies were performed on the side that had the highest sural amplitude or on the left side if sural responses were absent. Only unilateral measures were included in the evaluation. Unobtainable potentials were assigned a value of 0 for amplitude measurements and a missing value for latency measurements. Sensory action potentials were recorded with surface electrodes at all centers. Lower limb temperature was controlled within and across participating centers. End Points for Efficacy The primary efficacy variable was the change in the patient's neuropathy after 48 weeks, compared with baseline as measured by the NIS-LL (range, 0-85). A patient was classified as having worsened if the NIS-LL change from baseline was 2 points or more, considered to have improved if the NIS-LL change was 2 points or less, or classified as having no change from baseline. These criteria were based on the recommendations for a clinically meaningful change in a clinical trial defined for the total NIS by the Peripheral Nerve Society Consensus Committee.19 The primary efficacy variable was evaluated with an intent-to-treat analysis. Patients who terminated the study prematurely were followed up through month 12 whenever possible. When it was not feasible to obtain the month 12 assessments, the last observation was carried forward. Patients who dropped out of the trial because of an adverse event or noncompliance and for whom no follow-up data were available were classified as having worsened. Secondary efficacy variables included the change from baseline in the following measures: CDT, HP:5.0, NIS-LL sensory subscore, NSC (the change in the sensory severity score and in the number of sensory symptoms), PBQ, occurrence of a new foot ulcer, NIS-LL mean change from baseline, nerve conduction change from baseline (sural and ulnar sensory amplitudes and distal latency), and the global assessment questionnaire. Evaluable patients were defined as those without major protocol violations who had the required end point measures from the baseline and blinded treatment periods, who completed the full treatment regimen, and who were determined to be in at least 80% compliance with the study medication schedule as measured by self-report and inventory of returned vials. Statistics This study was designed to enroll 1000 patients (500 in each arm) who were to receive treatment for 12 months. Based on the assumption that the shift in location after 12 months of treatment would be at least as large as what was observed in the 6-month phase 2 clinical trial, simulations demonstrated that with 425 evaluable patients per arm, this study should have at least 80% power to detect a significant difference between groups. Five hundred patients per arm were selected assuming a 15% dropout rate. Continuous variables are summarized as means (SDs). Discrete data are summarized as frequency and percentage. Comparability of the 2 groups with regard to demographic and baseline characteristics was assessed using analysis of variance for continuous variables and χ2 test for discrete variables. Between-group comparison of the primary efficacy end point was performed using the Cochran-Mantel-Haenszel statistic using mean scores. Between-group comparisons of the secondary efficacy end points were performed using the Wilcoxon rank-sum test. The comparison of the incidence of new foot ulcers was performed using the Cochran-Mantel-Haenszel statistic. Results Patient Characteristics Of the 1019 patients enrolled in this study, 515 received placebo and 504 received rhNGF. Of these, 836 were determined to be evaluable according to the criteria defined (see "Methods"): 442 received placebo and 394 received rhNGF. There were no significant differences in the demographic or clinical characteristics between the 2 groups at baseline (Table 1). Nearly 75% of the patients in both groups had type 2 diabetes mellitus. Overall, 180 patients (17.7%) prematurely discontinued the study drug: 86 (17.1%) in the rhNGF treatment group and 54 (10.5%) in the placebo group. Forty-eight (9.5%) of those in the rhNGF group discontinued because of adverse events, compared with 20 (3.9%) in the placebo group (Figure 1). Adverse Events The adverse event profile was very similar to that in the phase 126 and phase 2 clinical trials. Three adverse events were notably more frequent in the rhNGF group than the placebo group: injection site pain/hyperalgesia (67.2% vs 11.6%), myalgia (6.2% vs 2.2%), and peripheral edema (5% vs 1.2%). Injection site pain/hyperalgesia was the most frequent adverse event and led to the discontinuation of the drug in 22 patients (4%), all from the rhNGF group. Eleven patients died during the study: 7 in the placebo group and 4 in the rhNGF group. Approximately 19% of the patients experienced serious adverse events, which were balanced in distribution across the 2 treatment groups. None of the deaths or serious adverse events were attributed to the drug and were generally unremarkable in view of the patient population. One rhNGF-treated patient tested positive for anti-NGF antibodies at 6 months but was negative at 9 months. Efficacy Efficacy of rhNGF was assessed by comparing end points pretreatment and posttreatment. Change in the NIS-LL score served as the basis of the primary end point. In the intent-to-treat analysis, all patients without available month 12 values were considered to have worsened. Overall, 31% of the patients improved, 38% worsened, and 31% remained unchanged (Table 2). With this analysis, the rhNGF treatment group appeared to do worse than the placebo group (P = .04). When the discontinuation penalty was removed and patients without available 48-week values were categorized on the basis of the last available NIS-LL value carried forward, there were no significant differences between the treatment groups (P = .25). When restricted to the evaluable subset, the analysis of the primary end point also showed no significant differences (P = .32). No beneficial effect of rhNGF vs placebo was observed for the majority of the secondary end points, including the quantitative sensory tests (CDT, HP:5.0, and VDT), the NSC, the sensory subscore of the NSC, nerve conduction testing, the incidence of foot ulcers, and the primary components of the PBQ (Table 3 and Table 4). With regard to all the secondary end points, neither the rhNGF group nor the placebo group worsened during the course of the study but either remained unchanged or numerically improved compared with baseline values. The global assessment score did demonstrate a statistically significant but modest benefit from treatment with rhNGF (P = .03 by O'Brien rank sum). Overall, 22% of patients receiving rhNGF subjectively believed that their neuropathy had worsened during the course of the study as compared with 26% of those receiving placebo (Table 5). In contrast, 45% of the rhNGF group and 37% of the placebo group believed that they had improved. The activities of daily living score and the quality of life measure components of the PBQ failed to show any significant difference between the groups (P = .80; Table 3). Only 2 of 32 treatment comparisons did show a significant benefit in favor of rhNGF, and this might be expected by chance. These were the severity of pain in the leg (rhNGF improved 11.1 vs 8.6 points for the placebo; P = .05) and the 6-month symptoms in leg measure (rhNGF improved 13.8 vs 11.1 points for the placebo, P = .003). Most components of the PBQ showed some numerical improvement at month 12 in both treatment groups. There were no categories in which the placebo group did significantly better than the rhNGF group. Blinding To determine whether this study was adequately blinded, patients and the examining neurologists were asked at month 12 to identify the treatment received. Forty-nine percent of patients correctly indicated that they were receiving placebo, and 33% correctly indicated that they were receiving rhNGF. Similarly, 42% of the time neurologists correctly identified the treatment as placebo, and 34% of the time they correctly identified rhNGF based on the presence or absence of efficacy. In addition, patients 26% of the time and neurologists 40% of the time indicated that they could not guess the treatment type. Although the rates of correct identification were less than 50%, the availability of an "unknown" category resulted in a statistically significant association between the actual treatment received and the opinion about the treatment received (P<.05). Comment This multicenter study of 1019 patients with diabetic polyneuropathy demonstrated that rhNGF administration is safe but failed to show a significant benefit of rhNGF treatment when administered at a dose of 0.1 µg/kg, subcutaneously, 3 times per week for 12 months. The primary efficacy end point did not show any significant differences favoring the rhNGF treatment group, although both groups showed a trend toward improvement in this measure during the course of the study. Similarly, there were no significant differences between the 2 groups in most of the secondary end point measures. The only exceptions were the global symptom assessment and 2 individual measures within the PBQ that showed a significant benefit from rhNGF. The neuropathy in both treatment groups failed to progress during the course of the study, and most measures suggested improvement in both treatment groups. Large epidemiological studies, such as the Rochester Diabetic Neuropathy Study,20 have demonstrated that in the short term, individual measures of neuropathy may have a variable course and transient improvements in subjective measures are common. Progression of neuropathy may be clearly demonstrated over 2 years according to the Rochester Study, but it is less clear in 1-year increments. Nevertheless, given the large number of patients enrolled in this study, the lack of significant neuropathy progression is surprising. Our patient population may have differed from the Rochester population in that they had better glycemic control. The mean serum hemoglobin A1C for our patients was 8.7%, compared with 11.2% for the Rochester patients with neuropathy. Improvement of the placebo group during the course of the study, as was the case here, makes it far more difficult to demonstrate superiority of the drug to the placebo. If the ability of small doses of rhNGF to reverse neuropathy is not robust or if rhNGF is only capable of preventing the progression of neuropathy, it might not be possible to detect a significant beneficial effect in a 1-year trial. It is unknown whether NGF should be expected to improve neuronal function or just prevent progression of neuropathy in a clinical setting. Each of the preclinical experimental models demonstrated that NGF has the ability to prevent neuropathy, not reverse it.8-10 Moreover, the doses of NGF used in those animal studies ranged from 1 to 10 mg/kg, administered between 3 and 7 times a week, in contrast to rhNGF administered at 0.1 µg/kg, 3 times a week in our trial. Thus, the dose chosen for this study may be at the threshold or below the minimum dose needed to demonstrate efficacy. Painful myalgias and arthralgias seen at higher doses in phase 1 clinical trials restricted dosing to less than 1 mg/kg. However, in this study, only 0.1 µg/kg was used as the lowest effective dose in the phase 2 trial and chosen because there were no significant differences between 0.1 and 0.3 µg/kg doses in the phase 2 trial with regard to efficacy, and the lower dose was thought to be less likely to cause adverse events. Given inherent differences between the trials, a threshold dose might be efficacious in one trial and not in another. It also is possible that our assessment was inadequate to detect a beneficial effect. All sensory measurements were performed at the toe. Considering the advanced neuropathy of our study population, improvement might have been detectable at more proximal locations where sprouting fibers are more likely to grow, and we might have missed it by restricting our examination to the big toe. In addition, the NIS-LL is not sensitive to small fiber sensory dysfunction, the neuronal population most likely to respond to rhNGF. This measure was chosen with the agreement of regulatory agencies since it was a validated instrument that can demonstrate a clinically meaningful change in functional status over time. Alternatively, it is possible that the rhNGF did not get to the target tissue in adequate concentrations or that the drug is not effective at ameliorating diabetic neuropathy. Two phase 2 clinical trials, one in patients with diabetic polyneuropathy and the other in patients with HIV neuropathy, using comparable doses, demonstrated significant improvement in the neuropathy over an even shorter time. Why was this trial different? Direct comparisons with the phase 2 clinical trials have been unrevealing. Demographic characteristics were similar between the 2 trials with the exception of older age (maximum age of 59 years in phase 2 vs 75 years in our study) and a greater proportion of patients with type 2 diabetes mellitus in phase 3 (50% in phase 2 vs 75% in our study). When we eliminated patients who would not have qualified for the phase 2 clinical trial (based on age and quantitative sensory testing percentiles), the data still failed to show a beneficial effect of rhNGF. We also failed to detect a significant treatment effect when the data were examined at 3 months and 6 months and when subsets were examined according to age, sex, diabetes type, baseline hemoglobin A1C levels, and baseline NIS-LL score. One other important difference was that patients with proliferative retinopathy were excluded from the phase 2 trial but not the phase 3 trial. Patients without proliferative retinopathy are less likely to have severe neuropathy. A final point for consideration is the possibility that the drug may not have been identical in the 2 sets of studies. Although the structure of the final protein was identical, the manufacturing process was altered prior to the phase 3 trial. The new rhNGF was in a solution with a slightly different molarity of acetate and sodium chloride. It also was formulated at a very different concentration (2 mg/mL for phase 2 and 0.1 mg/mL for phase 3). The 2 drugs demonstrated similar bioactivity in PC12 cell cultures, but they were not compared in animal studies. Interestingly, several patients who participated in phase 2 and received the phase 3 rhNGF through open-label extension studies reported that the injection site hyperalgesia felt different. In fact, injection site hyperalgesia was reported in more than 90% of the phase 2 patients and only 67% in our study. Conclusion In summary, this phase 3 clinical trial confirmed the safety of rhNGF demonstrated in previous studies but failed to demonstrate a beneficial effect of rhNGF in treating diabetic polyneuropathy. References 1. Apfel SC. Clinical Applications of Neurotrophic Factors. Philadelphia, Pa: Lippincott-Raven Publishers; 1997. 2. Rohrer H, Heumann R, Thoenen H. The synthesis of nerve growth factor (NGF) in developing skin is independent of innervation. Dev Biol.1988;128:240-244.Google Scholar 3. Jakobsen J, Brimijoin S, Skau K, Sidenius P, Wells D. Retrograde axonal transport of transmitter enzymes, fucose labeled protein and nerve growth factor in streptozocin-diabetic rats. Diabetes.1981;30:797-803.Google Scholar 4. Schmidt RE, Grabau GG, Yip HK. Retrograde axonal transport of 251 nerve growth factor in ileal mesenteric nerves in vitro. Brain Res.1986;378:325-336.Google Scholar 5. Hellweg R, Raivich G, Hartung HD, Hock C, Kreutzberg GW. Axonal transport of endogenous nerve growth factor (NGF) and NGF receptor in experimental diabetic neuropathy. Exp Neurol.1994;130:24-30.Google Scholar 6. Fernyhough P, Diemel LT, Brewster WJ, Tomlinson DR. Deficits in sciatic nerve neuropeptide content coincide with a reduction in target tissue nerve growth factor messenger RNA in streptozocin-diabetic rat. Neuroscience.1994;62:337-344.Google Scholar 7. Anand P, Terenghi G, Warner G, Kopelman P, Williams-Chestnut RE, Sinicropi DV. The role of endogenous nerve growth factor in human diabetic neuropathy. Nat Med.1996;2:703-707.Google Scholar 8. Apfel SC, Lipton RB, Arezzo JC, Kessler JA. Nerve growth factor prevents toxic neuropathy in mice. Ann Neurol.1991;29:87-89.Google Scholar 9. Apfel SC, Arezzo JC, Lipson L, Kessler JA. Nerve growth factor prevents experimental cisplatin neuropathy. Ann Neurol.1992;31:76-80.Google Scholar 10. Apfel SC, Arezzo JC, Brownlee M, Federoff H, Kessler JA. Nerve growth factor administration protects against experimental diabetic sensory neuropathy. Brain Res.1994;634:7-12.Google Scholar 11. Apfel SC, Kessler JA, Adornato BT, Litchy WJ, Sanders C, Rask CA.and the NGF Study Group. Recombinant human nerve growth factor in the treatment of diabetic polyneuropathy. Neurology.1998;51:695-702.Google Scholar 12. McArthur JC, Yiannoutsos C, Simpson DM. et al. A phase II trial of nerve growth factor for sensory neuropathy associated with HIV infection. Neurology.2000;54:1080-1088.Google Scholar 13. Dyck PJ, Litchy WJ, Lehman KA, Hokanson BA, Low PA, O'Brien PC. Variables influencing neuropathic endpoints. Neurology.1995;45:1115-1121.Google Scholar 14. Dyck PJ, Bushek W, Spring EM. et al. Vibratory and cooling detection thresholds compared with other tests in diagnosing and staging diabetic neuropathy. Diabetes Care.1987;10:432-440.Google Scholar 15. Dyck PJ, Zimmerman I, Gillen DA, Johnson BS, Karnes JL, O'Brien PC. Cool, warm and heat-pain detection thresholds. Neurology.1993;43:1500-1508.Google Scholar 16. Gruener G, Dyck PJ. Quantitative sensory testing: methodology, applications, and future directions. J Clin Neurophysiol.1994;11:568-583.Google Scholar 17. Dyck PJ, O'Brien PC, Kosanke JL, Gillen DA, Karnes JL. A 4, 2, and 1 stepping algorithm for quick and accurate estimation of cutaneous sensation threshold. Neurology.1993;43:1508-1512.Google Scholar 18. Dyck PJ, Zimmerman IR, Johnson DM. et al. A standard test of heat pain response using CASE IV. J Neurol Sci.1996;136:54-63.Google Scholar 19. Dyck PJ.for the members of the Peripheral Nerve Society. Diabetic polyneuropathy in controlled clinical trials. Ann Neurol.1995;38:478-482.Google Scholar 20. Dyck PJ, Davies JL, Litchy WJ, O'Brien PC. Longitudinal assessment of diabetic polyneuropathy using a composite score in the Rochester Diabetic Neuropathy Study cohort. Neurology.1997;49:229-239.Google Scholar 21. Cohen S, Levi-Montalcini R, Hamburger V. A nerve growth stimulating factor isolated from sarcomas 37 and 180. Proc Natl Acad Sci U S A.1954;40:1014-1018.Google Scholar 22. Kessler JA, Black IB. Nerve growth factor stimulates the development of substance P in sensory ganglia. Proc Natl Acad Sci U S A.1980;77:649-652.Google Scholar 23. MacLean DB, Bennett B, Morris M, Wheeler FB. Differential regulation of calcitonin gene related peptide and substance P in cultured neonatal rat vagal sensory neurons. Brain Res.1989;478:349-355.Google Scholar 24. Diamond J, Holmes M, Coughlin M. Endogenous NGF and nerve impulses regulate the collateral sprouting of sensory axons in the skin of the adult rat. J Neurosci.1992;12:1454-1466.Google Scholar 25. Jackson GR, Apffel L, Werrbach-Perez K, Perez-Polo JR. Role of nerve growth factor in oxidant-antioxidant balance and neuronal injury. J Neurosci Res.1990;25:360-368.Google Scholar 26. Petty BG, Cornblath DR, Adornato BT. et al. The effect of systemically administered recombinant human nerve growth factor in healthy human subjects. Ann Neurol.1994;36:244-246.Google Scholar
TI - Efficacy and Safety of Recombinant Human Nerve Growth Factor in Patients With Diabetic Polyneuropathy: A Randomized Controlled Trial
JF - JAMA
DO - 10.1001/jama.284.17.2215
DA - 2000-11-01
UR - https://www.deepdyve.com/lp/american-medical-association/efficacy-and-safety-of-recombinant-human-nerve-growth-factor-in-0jlj0cmXEB
SP - 2215
EP - 2221
VL - 284
IS - 17
DP - DeepDyve
ER -