The development of neurotrophic factors for treatment of Alzheimer disease (AD) and other neurodegenerative disorders has proceeded in fits and starts, in no small part due to the difficulty of delivering peptide factors where they are needed in the brain. Now, a small phase I trial of nerve growth factor (NGF) has shown that gene therapy can safely deliver the neurotrophic factor to a precisely chosen group of cholinergic neurons, and that doing so may slow cognitive decline in a handful of early AD patients.

Results from the trial, which was started in 1999 (see ARF related news story) by Mark Tuszynski and colleagues from the University of California at San Diego with collaborators at the University of California at Irvine and the University of Chicago, appeared online April 24 in Nature Medicine. The publication follows on the presentation of preliminary trial data at the Society for Neuroscience meeting in 2002.

NGF, a trophic factor for cholinergic neurons, has been studied extensively both for its role in the loss of cholinergic neurons that occurs in AD, and as a potential therapy to preserve neuronal function (see coverage of Sorrento meeting). The earliest clinical studies using NGF showed safe delivery to the CNS to be a major challenge. As a peptide, the factor didn’t cross the blood-brain barrier, and when it was directly infused into the ventricles of the brain, severe side effects ensued, including pain, weight loss, and Schwann cell migration into the spinal cord (see Tuszynski, 2002).

Tuszynski and his colleagues turned to gene therapy in attempts to deliver sufficient quantities of NGF in a local, controlled manner. Their trial involved surgical implantation of patients’ own fibroblasts which had been genetically engineered to produce human nerve growth factor. The modified cells were injected into the cholinergic nucleus basalis region of the forebrain, a region that provides input to broad areas of the cortex and has been implicated in injury-induced plasticity and cholinergic-dependent brain repair (see below).

After some initial and serious problems with the surgical procedure, the researchers safely implanted six out of eight early-stage AD patients with NGF-producing cells, and followed their progress for a mean of 22 months. The first two surgeries resulted in brain hemorrhages after the subjects, who were alert but sedated, moved during the operation, and one of the patients subsequently died. For the rest of the trial, surgery was done under general anesthesia and the patients’ heads were immobilized. Under these conditions, the surgery was successful, and no adverse effects attributed to NGF were seen.

During follow-up, two common clinical measures, the Mini-Mental Status Exam (MMSE) and the AD Assessment Scale-Cognitive subcomponent (ADAS-Cog) were used to gauge cognitive state. The rate of cognitive decline over the entire follow-up period, as measured by the MMSE, was cut in half after implantation of the NGF-producing cells compared to the year before the surgery. In the period from 6 to 18 months postsurgery, when NGF production was expected to be at its peak, the results looked even better: Two patients showed improved cognitive scores and three showed little or no decline. For the ADAS-Cog test, the researchers did not have pretreatment data for the patients, but the rate of decline in scores in the six- to 18-month postoperative period was lower compared to average rates observed in other studies. Despite the caveats that the study was small and not placebo-controlled, the authors point out that the observed reduction in disease progression was far in excess of that seen with cholinesterase inhibitors currently approved for AD.

Imaging and morphological data demonstrated that NGF was in fact exerting a neurotrophic effect on cholinergic circuits. PET scans revealed broad increases in glucose uptake by cortical neurons after surgery in four subjects who received bilateral cell implants, a reversal of the normal decline seen with time in AD patients. Increases were seen all over the cortex in regions known to be innervated by projections from the nucleus basalis, suggesting that the projections were being activated by NGF. Histochemical analysis of the brain of one subject who died 5 weeks postoperatively, showed robust NGF expression by the transplanted fibroblasts and sprouting of cholinergic axons in and around the transplant site. There was little evidence of inflammation.

In summary, the authors write that their study, along with recent clinical success with GDNF for Parkinson disease (Slevin et al., 2005; Gill et al., 2003) provides “early but suggestive” evidence that long-term growth factor treatment is well-tolerated and has the potential to improve symptoms and affect disease progression, when administered into the CNS in therapeutic doses and in a regionally restricted manner.

The logic in targeting the nucleus basalis for NGF therapy is borne out by another recent study from the Tuszynski lab, showing the importance of this region for the recovery from traumatic brain injury in rats. In a publication in Neuron on April 21, James Conner, Andrea Chiba, and Tuszynski describe the process by which rats can regain their ability to retrieve food pellets with a paw after a small motor cortex lesion impairs their reach. When the researchers damaged the cholinergic cells of the nucleus basalis with bilateral injections of the toxin 192-IgG-saporin before introducing the cortical lesion, the rats were far less able to recover paw mobility after a course of rehabilitation compared to animals with an intact nucleus basalis. By mapping the motor organization of the cortex in each group of rats, the researchers showed that the rats lacking a nucleus basalis failed to reorganize areas of cholinergic innervation around the lesioned area normally. The results show that the cholinergic function of the nucleus basalis is central to the synaptic remodeling that allows recovery from brain injury. Cholinergic deficits in the nucleus basalis are seen with normal aging and in AD, and perhaps contribute to the decreased ability of the brain to compensate for damage, whether from trauma or from the toxic processes of AD.—Pat McCaffrey


  1. I think the most interesting aspect of the recent Connor et al. article is that it builds on their earlier work, which showed that nucleus basalis cholinergic neurons were necessary for the normal changes in motor cortex representation that happen during motor learning (reaching). The current study shows that the nucleus basalis plays the same role in remodeling motor cortex in compensation for an injury. The implication drawn by the authors in their article is that damage to basal forebrain cholinergic neurons that occurs in AD may impair the brain's ability to compensate for the cortical neurodegeneration that takes place in AD. This has some important implications—drugs that enhance cholinergic function (donepezil, galantamine, etc.) seem to slow the rate of decline rather than reverse cognitive impairments in AD. (Indeed, this is what was seen with Tuszynski's NGF treatment in humans.) Some have hypothesized that the cholinergic system is involved in regulating amyloid metabolism, which has been suggested to explain the effect of procholinergic drugs on slowing decline. But a slowing in the rate of decline would also be consistent with heightened cholinergic tone allowing the brain to better adapt to the ongoing degeneration. It will be important to see if experimental lesions of cholinergic neurons in animals also affect reorganization from injuries that occur gradually or progressively.

  2. Conner et al. (2005) show in this elegant paper that the basal forebrain cholinergic system plays an essential role in cortical plasticity and functional recovery following brain injury. Thus, even partial cholinergic hypofunction may cause a disruption of cortical plasticity that may eventually limit the extent of functional recovery. These findings are of major importance by emphasizing the pivotal role of the basal cholinergic system in health and disease states including normal aging, Alzheimer disease (AD), traumatic brain injury, and so on. The authors claim that therapeutic strategies such as cholinesterase inhibitors or trophic factors may be used to increase cortical plasticity and restore functional deficits resulting from brain injury.

    Activation of the M1 muscarinic receptor could also be neuroprotective and enhance brain plasticity (Albrech et al., 2000). Thus, in addition to the strategies suggested by the authors, another alternative could be highly selective M1 muscarinic receptor agonists. We have shown that brain penetrable selective M1 muscarinic agonists via M1 muscarinic receptor activation (a) are synergistic and cross-talk with growth factors such as NGF, exhibiting neurotrophic-like activity, enhancing neurite outgrowth, and elevating secretion of the nonamyloidogenic and neurotrophic α-APPs; (b) decrease Aβ levels; (c) protect hippocampal neurons from insults such as Aβ-induced neurotoxicity, and this effect involves the Wnt signaling pathway; (d) inhibit tau protein hyperphosphorylation; and (e) improve cognitive dysfunctions in several animal models with a wide safety margin [Fisher, 2000; Fisher et al., 2003; Fisher et al., ADPD2005 (Sorrento, Italy); Farias et al., 2004]. Notably, these M1 agonists show effects similar to NGF, yet by different mechanisms, on neurite outgrowth, brain plasticity, Aβ, Wnt components, GSK-3β, and tau phosphorylation. Thus, such M1 muscarinic agonists can be regarded as low-molecular-weight CNS-penetrable compounds with neurotrophic-like activity that may be used to enhance brain plasticity, alone or in a synergistic combination with trophic factors. The M1 agonistic strategy may be useful in the treatment of AD and other diseases with cholinergic deficits, as such compounds will be less dependent on intact cholinergic innervations (a possible drawback of cholinesterase inhibitors).


    . Muscarinic acetylcholine receptors induce the expression of the immediate early growth regulatory gene CYR61. J Biol Chem. 2000 Sep 15;275(37):28929-36. PubMed.

    . M1 muscarinic receptor activation protects neurons from beta-amyloid toxicity. A role for Wnt signaling pathway. Neurobiol Dis. 2004 Nov;17(2):337-48. PubMed.

    . Therapeutic strategies in Alzheimer's disease: M1 muscarinic agonists. Jpn J Pharmacol. 2000 Oct;84(2):101-12. PubMed.

    . M1 muscarinic agonists can modulate some of the hallmarks in Alzheimer's disease: implications in future therapy. J Mol Neurosci. 2003;20(3):349-56. PubMed.

  3. The report by Tuszynski et al. is a fascinating study on many levels. First, the ability to deliver ex vivo nerve growth factor (NGF) via autologous fibroblasts genetically modified to express NGF for greater than 18 months into the nucleus basalis of Meynert (NBM) in a live adult human is no simple feat. The stereotaxic surgeries proved successful in six out of eight subjects, and much was learned about the need for general anesthesia for the success of this procedure. The cognitive measures employed in this study demonstrated either reduced cognitive decline or mild improvement, which is quite encouraging, especially since the sample size is quite small at this point. Clearly, this initial study has demonstrated a strong rationale for continued enrollment in this protocol.

    One of the most significant aspects of this study lies in its true translational perspective. Specifically, use of in vitro and animal models, including lesioned mice and nonhuman primates, has been performed over the past 12 to 15 years in order to bring the technology, basic mechanisms of action, and the actual data-based evaluations to a place where they could be tried successfully in demented humans. A plethora of basic science, including understanding animal models following injury and Alzheimer disease pathology, as well as technology development in the synthesis, bioactivity, and delivery of NGF and developing viral vectors for production and long-term delivery of NGF were requisite for the clinical trial.

    NGF is an unbelievably powerful growth factor. This is both a benefit as well as a potential negative aspect in terms of side effects that have been demonstrated previously with broad-scale NGF delivery. The regionally selective stereotaxic implantation of autologous fibroblasts secreting NGF into the NBM is a significant advance of this study. Thus, this report by Tuszynski et al. truly qualifies itself as originating at the laboratory bench and extending through the neurosurgery suite into the neuropsychology clinic and neuroimaging facility. In summary, the six patients that have completed the first arc of this study are literally a tip of the iceberg. The importance lies in the overall development of this program and a continued ability to demonstrate the utility of NGF delivery within a very regionally specific area (i.e., cholinergic basal forebrain), which in turn has widespread effects on the cortical cholinergic system and potential increases in CNS glucose uptake.

  4. The impressive paper by Tuszynski and colleagues represents a very promising outlook into possible future therapy of AD: They isolate patients' own fibroblasts from a skin sample, expand the cells in cell cultures, and transduce them to overexpress NGF. The authors show that upon reimplantation of these NGF producers by stereotaxic injection into the patients' basal forebrain, the progression of AD is substantially retarded. This is especially true between 6-18 months after implantation, when the potentially secreted NGF is thought to develop its full capacity of neuroprotective effects. The authors also find enhanced neural activity by PET studies in those patients investigated. Overall, as mentioned by the authors, this initial result needs to be corroborated by well-controlled clinical follow-up studies.
    I would find it very interesting to know whether the authors can find answers to the following questions:

    1. What is the rate of proliferation of expanded fibroblasts after implantation, and is there any estimation of the danger of unrestricted growth of implanted fibroblasts?

    2. What is the survival expectancy of the reimplanted fibroblasts in vivo?

    3. Is the fibroblasts' NGF secretion via the constitutive or the regulated pathway of secretion, and do the authors expect to even enhance the efficacy of the treatment if doses of secreted NGF were increased?

    4. It would be very interesting to know whether there existed a positive correlation between retardation of AD and relative increase in FDG consumption. If so, this could help to “adjust” NGF secretion?

    5. And none the least: What was the subjective evaluation of the procedure and of the outcome by the patients themselves?

    Of course, asking these questions is much easier than finding the answers ... . It will be exciting to see whether the promising trend provided by this study will be confirmed in future clinical trials.

Make a Comment

To make a comment you must login or register.


News Citations

  1. NGF Gene Therapy Trial Begins
  2. Orlando: Preliminary Results of NGF Gene Therapy Trial Are Mixed
  3. Sorrento: Trouble with the Pro’s

Paper Citations

  1. . Growth-factor gene therapy for neurodegenerative disorders. Lancet Neurol. 2002 May;1(1):51-7. PubMed.
  2. . Improvement of bilateral motor functions in patients with Parkinson disease through the unilateral intraputaminal infusion of glial cell line-derived neurotrophic factor. J Neurosurg. 2005 Feb;102(2):216-22. PubMed.
  3. . Direct brain infusion of glial cell line-derived neurotrophic factor in Parkinson disease. Nat Med. 2003 May;9(5):589-95. PubMed.

Further Reading

No Available Further Reading

Primary Papers

  1. . A phase 1 clinical trial of nerve growth factor gene therapy for Alzheimer disease. Nat Med. 2005 May;11(5):551-5. PubMed.
  2. . The basal forebrain cholinergic system is essential for cortical plasticity and functional recovery following brain injury. Neuron. 2005 Apr 21;46(2):173-9. PubMed.