P-196 A MATHEMATICAL COMPARISON OF A “RANDOMIZED WITHDRAWAL” CLINICAL TRIAL DESIGN AND A PARALLEL GROUPS DESIGN TO DEMONSTRATE DISEASE MODIFICATION IN ALZHEIMER’S DISEASE (AD) Scott Horton, Alexander Gutin, Suzanne Hendrix, Myriad Pharmaceuticals, Salt Lake City, UT, USA. Contact e-mail: shorton@myriad.com Background: Demonstration of disease modification in clinical studies of drugs treating different indications has been approached in many ways. Among the strategies proposed for AD are two approaches based on clinical outcomes: 1. the “randomized withdrawal” or “staggered start” designs, and 2. parallel group designs comparing slopes and severity of disease. Objective(s): This presentation demonstrates how the study de- signs described above compare to each other mathematically and under what circumstances they may be equivalent. Methods: An effect that continues after the drug is removed is reasonably defined as a disease modifying effect, but this “randomized withdrawal” design is difficult to implement due to ethical concerns. The staggered start design avoids this issue, but still has some difficulties because of the long trial period re- quired. A mathematical comparison of the randomized withdrawal design and the staggered start design is presented using the following model: Yijk(t)=i+i[min(t,T1)]+(ItT1)*{k*k+k[max(0,t-T1)]}+ijk, where t=time and T 1 =beginning of second phase of the study, k indicates a treatment change in the second phase, and Y ijk is the clinical outcome measure. Additionally, a “natural history staggered start” design based on parallel groups is proposed and compared to both the staggered start design and the randomized withdrawal design. Results: A mathematical comparison of the staggered start design to the randomized withdrawal design proves that these designs and their associated analyses are mathematically equivalent. A parallel groups design is shown to be mathematically connected to a staggered start design. The conditions under which these designs test the same hypotheses are described. A parallel groups analysis can be constructed that is mathematically equivalent to analyses from these cross-over designs. Conclusions: This paper proves that the “randomized withdrawal” and “staggered start” designs are in fact mathematically equivalent. A new “natural history staggered start” design based on parallel groups was proposed and was shown under certain condi- tions to be equivalent to the two cross-over type designs. In these conditions it is superior, because it allows measurement of the same underlying patterns of drug effect, without the operational complications (longer trial periods, drop-out bias, ethical concerns) introduced with the staggered start or random- ized withdrawal designs. P-197 THERAPEUTIC INTERVENTION IN ALZHEIMER’S DISEASE Suman Kushwaha, Institute of Human Behavior & Allied Sciences, Delhi, India. Contact e-mail: sumankushwaha@gmail.com Background: Alzheimer Disease (AD) is treatable disorder but not curable. Cholinergic deficits are found in patients of AD which is responsible for cognitive deficits & behavior problems. Early interventions with available therapeutic agents like Cholinesterase inhibitors & NMDA receptor antagonist along with symptomatic treatment & cognitive retraining of patients is essen- tial for management of patients. Aim: To Study the outcome of therapeutic intervention in AD pts. Material & Methods: Study was done in Neurobe- havior clinic, Neurology clinic. Institute of Human Behavior and Allied Sci- ences, Delhi -India. Duration of study - 2 yrs. No of pts - 36, Males 28, Females 8. All the pts were evaluated by clinical history, Neurological & Neuropsychological examination. Pts were selected by NINDS -ADARDA criteria. The MMSE score ranged from 2 - 26 points. The main presenting complaints were Memory impairment (66%), Behavior problems (33%) & Impairment of ADL (50%). Pts were categorized into Mild (16%), Moderate (21%) & Severe AD (25%). The range of duration of presentation was 4 months to 5.6 yrs. All the pts were given ChEI in 5 - 10mg dose .12 pts (33%) received NMDA receptor antagonist 5- 20 mg dose. Antipsychotics were given to 15(41%)pts. The medical co-morbid condition like hypertension , diabetes etc were treated accordingly. The follow up period was 6 - 18 months. Results: Behavior improved in 8 pts (22%) .MMSE score improved by an average 4 points in 22 (61%)pts. ADL improved in 7 (19.4%) pts. These improvements were seen in mild to moderate group. There was overall im- provement in quality of life of AD patients and their caregivers. Conclusion: As we can not cure the AD, we should give the advantage of available therapeutic options to all pts as there is improvement in all the deficient domains ie Cognitive, Behavior & ADL which will be helpful in managing the patients. The therapeutic intervention should be started in early course of disease to get the maximum benefit. P-198 HUPERZINE A AS A TREATMENT FOR ALZHEIMER’S DISEASE John T. Little 1 , Sally Walsh 2 , Paul Aisen 1 , 1 Georgetown University Hospital, Washington, DC, USA; 2 University College Hospital Galway, Galway, Ireland. Contact e-mail: jtl106@gunet.georgetown.edu Background: Huperzine A is a natural cholinesterase inhibitor and NMDA antagonist derived from the Chinese herb Huperzia serrata. There is evidence that huperzine A may compare favorably in symptomatic efficacy to currently available treatments. Additionally, huperzine A has antioxidant and neuropro- tective properties that suggest that it may be useful as a disease-modifying treatment for Alzheimer’s Disease (AD). Objective(s): To present the current status of the development of huperzine A as a treatment for AD, including progress in the U.S. Phase II trial. Methods: Pre-clinical and clinical studies supporting the use of huperzine A in Alzheimer’s disease will be reviewed, and the design and status of the multi-center, double-blind, placebo-controlled huperzine A therapeutic trial ongoing in the United States will be presented. Results: Preclinical data and preliminary clinical studies indicate that huper- zine A is a highly selective acetylcholinesterase inhibitor with NMDA antag- onism and neuroprotective properties that may contribute to its efficacy in AD treatment. A randomized controlled Phase II clinical trial is now under way at 30 U.S. sites. The primary aim of the trial is to compare change in cognition in subjects with mild to moderate AD after 16 weeks of treatment with huperzine A (at a dose of 200mcg bid or 400mcg bid) to placebo. Secondary aims include assessment of behavior, function and global clinical status after 16 weeks of treatment, with continuing follow-up for a total of 48 weeks. The target enrollment is 210 subjects, providing 90% power to demonstrate the anticipated cognitive benefit. As of January, 2007, 170 subjects have been enrolled. Though most subjects enrolled after failing to tolerate standard cholinesterase inhibitors, to date only 26 of 170 subjects enrolled have dis- continued treatment prior to the final 48 week study visit. Conclusions: Pre-clinical and clinical studies are encouraging regarding a possible role for huperzine A in the treatment of Alzheimer’s disease. The current Phase II study has a high study retention rate, suggesting good tolerability of the drug. The completion of the U.S. study should indicate the efficacy of huperzine A in AD, and may lead to the inititiation of Phase III pivotal trials. P-199 USE OF ANTI-CHOLINESTERASE INHIBITORS AND MEMANTINE IN A POPULATION-BASED STUDY OF INCIDENT AD CASES: PREVALENCE OF USE, CHARACTERISTICS, AND RELATIONSHIP TO MORTALITY Michelle M. Mielke 1 , JoAnn Tschanz 2 , Maria Norton 2 , Chris Corcoran 2 , Peter Rabins 1 , Martin Steinberg 1 , Michelle Carlson 1 , Robert C. Green 3 , John Breitner 4 , Kathleen Welsh-Bohmer 5 , Constantine G. Lyketsos 1 , 1 Johns Hopkins University, Baltimore, MD, USA; 2 Utah State University, Logan, UT, USA; 3 Boston University School of Medicine, Boston, MA, USA; 4 University of Washington, Seattle, WA, USA; 5 Duke University, Durham, NC, USA. Contact e-mail: mmielke1@jhmi.edu Background: Randomized clinical trials have demonstrated the efficacy of cholinesterase inhibitors and memantine for patients with AD. However, S161 Posters: P-196