Email this   Print this

Bridging The Gap

SHELDON H. PRESKORN, MD

There are both great opportunities and significant risks in developing novel drugs for central nervous system (CNS) diseases. The opportunities stem from advances being made in our understanding of the structure and function of the human brain as a result of neuroscience and molecular biology research. The target date for the completion of the human genome project is only a few years away. It is even now bearing fruit and those discoveries, while remarkable, will pale in comparison with those that will follow. The reason is that there are an estimated 5,000 brain-specific proteins which are the product of specific genes. With the completion of the human genome project, the genes coding for these proteins will be identified. That will in turn permit determination of their structure and in turn the development of drugs which can selectively interact with that site.

That is the good news. CNS drug discovery teams will have a vast array of targets and tools to develop novel CNS compounds. As Wall Street Week in Review might say, "Now is the time to be bullish on neuropsychopharmacology!"

What is the downside? While CNS drug discovery is firmly in the 21" century, CNS clinical research has not kept pace due to a lack of knowledge about, and tools to study, the fundamental pathophysiology of psychiatric and neurologic diseases. Precious little is known about the cause of such chronic illnesses as schizophrenia and major depression. The sheer percentage of our population affected by these illnesses and their chronicity is the reason that drugs to treat such CNS diseases are among the largest revenuegenerating drugs in many pharmaceutical companies' portfolios. However, the vagaries of drug development in this area make CNS drug discovery a highly speculative venture.

One significant problem that can be addressed is the substantial gap that has to be crossed from preclinical to clinical drug development. This gap is larger than in any other therapeutic area for several reasons. First is the lack of knowledge of the relevant pathophysiology of the disease process, making it impossible to develop appropriate animal models. Second, the vast difference between the human brain and the brain of other animals is such that it may not be possible to model these illnesses in animals even if we knew the pathophysiology. For the same reason, we cannot assess in animals many of the symptoms and signs that are the targets of the drug therapy such as hallucinations, delusions, disturbed mood and thought disorder.

Despite these problems, many companies will launch into full-scale, pivotal phase 11 studies with a new chemical entity (NCE) after the usual single- and multiple-dose phase I studies in normal volunteers to determine safety, tolerability and pharmacokinetics. This approach is highly speculative. The study may be testing the drug in the wrong condition and that is expensive from two standpoints: There is the cost of the study itself and there is the possibility that the development of what could be a useful and commercially-successful drug would be discontinued on the basis of studies done in the wrong patient population.

The problem is that conventional pivotal phase if studies are not a good way to gain understanding of the clinical pharmacology of new CNS drugs. These studies have a preselected outcome of interest (e.g. antidepressant efficacy), a rigid design which makes midcourse corrections difficult, if not impossible, and often are multisite. While the latter is usually dictated by power calculations, it means that no single investigator will likely develop much "feel" for the unique pharmacology of the drug. This approach is like limiting oneself to either hitting a home run or striking out every time.

We advocate, and have been successful with, bridging studies that span the conventional phase I and more full-scale phase II studies, performing two kinds. The first is done in normal volunteers but has as its goal a better characterization of the behavioral pharmacology of the drug in man. In essence, these studies are the phase I counterpart of the preclinical pharmacology done in animals using a variety of tools to assess the human CNS effects of the NCE. The second type is done in volunteers who have a mild variant of the initial target condition for the NCE's clinical development. We have done the latter studies in volunteers who are mildly symptomatic with schizophrenia, major depression, generalized anxiety disorder and senile dementia of the Alzheimer's type.

These studies have generally employed a double-blind, placebo-controlled, fixed-dose design, with fewer patients being enrolled. The studies are done at one or two sites to reduce the signal-to-noise problem inherent in large multisite studies. While the studies typically involve fixed doses, they are done sequentially in a manner analogous to a traditional phase I study. After each dose group has been completed, the results are assessed typically without breaking the blind and a decision is made as to whether and how to alter the original study plan to pursue an advantageous lead or avoid a potential pitfall. This approach makes such studies highly efficient from both time and cost perspectives. We have often been able to accomplish, in a fraction of the time and in a single study, what would require multiple traditional phase II studies and save millions in research expenses at the same time.

Examples of the findings from such studies include:

• Patients with the target illness could tolerate five times the dose of a NCE compared to a traditional normal volunteer.
• A NCE being considered as a possible antipsychotic might actually be better as an antidepressant.
• An immediate release preparation of a NCE on a twice-a-day schedule was not sufficiently well tolerated, despite a number of different dosing strategies, to permit an assessment of efficacy but a four-times-a-day dosing was well tolerated and had good preliminary evidence of efficacy. With that result, the company canceled the development of the immediate-release preparation and instead pursued the development of a sustained-release preparation. However, another short halflife NCE could be used once-a-day with good tolerability and preliminary evidence of efficacy, but required a period of a twice-a-day dosing to permit the development of tolerance to acute dosedependent tolerability problems. With that result, the company could pursue their pivotal phase II study with the immediate release preparation using an initiation phase with twice-a-day dosing and then a conversion to once-a-day dosing for the remainder of the study. At the same time, they could proceed with the development of a sustained-release preparation as either a backup agent or as a product extension strategy.

We are excited by the results we have had to date with such studies. You might find this approach useful in the efficient development of your promising NCE's in the CNS development arena. If so, let's talk.