Please email ft@pharmawire.com or call us at Americas: +1 212-500-1384 or Europe: + 44 (0)20 7059 6251 for further information on Pharmawire and how to receive more articles like the one below.
--------------------------------------------------------------------------------------------------------
Myriad Genetics’ Alzheimer’s candidate Flurizan has a much larger sample size which will hopefully improve its chance of overcoming statistical noise, when compared to Neurochem’s Phase III Alzhemed trials, physicians involved in the trial said. The larger sample size will hopefully overcome statistical noise, which is a key factor affecting Neurochem’s current Phase III Alzhemed trials.
Neurochem announced this Monday that its Alzheimer’s drug, Alzhemed, failed Phase III US trials. The company announced that high variations between more than 1,000 patients at trial locations in Canada and the US have “complicated” the statistical analysis. The FDA recognized the difficult issues surrounding a trial of this magnitude, with its significant site effect and the large number of covariates identified during the modeling process.
Neurochem shares plummeted 60% to a five-year low of USD 2.38 this morning. The company has also recently announced it has USD 80m in cash.
Myriad shares on the other hand, took a hit when the news on Neurochem’s failed trials came to light, but have now stabilized at USD 42.82. The stock had a 52-wk high and low of USD 48.20 and 23.15.
Dr Gary Kennedy, director of geriatric psychiatry at Montefiore Medical Center in New York City, and consultant to Myriad Genetics on the Flurizan trials, explained that the main advantage that Myriad Genetic’s Flurizan trials has over the Alzhemed trials, comes down to a basic principle of sample size to overcome statistical noise.
Myriad’s Phase III US trials, on the other hand, are the largest placebo-controlled study ever undertaken of an investigational drug for the treatment of Alzheimer’s disease, with a total of approximately 1,600 patients enrolled. This overshadows Neurochem which had recruited 950 patients with mild-to-moderate Alzheimer’s Disease.
A larger sample size usually increases the power of a trial and decreases the probability of making a false-positive error.
Jacobo Mintzer, professor of Psychiatry, Neurology, Physiology and Neuroscience at the Medical University of South Carolina, and an investigator who worked on both Myriad’s Flurizan and Neurochem’s Alzhemed trials, also agreed that it will be very difficult to demonstrate an absence of an event that will occur. ”In order to demonstrate the absence of that, you need a large group of patients in the trial,” he said.
To detect a disease modifying effect, either the trial will have to be conducted over a longer period of time, or the design has to increase the number of patients enrolled, he explained. For each variable added, additional subjects are needed to generate statistical power. Therefore, a large sample size will decrease the effect of random variation on the estimate of a treatment effect, he explained.
In almost all trials, you have to take into account the ”site-to-site” and ”patient-to-patient” variation, said Kennedy. Mintzer also agreed that one way to counter the regional ”site-to-site” variability will be a large patient population. He also added both Myriad and Neurochem will have to develop a better statistical model to interpret the data. ”The FDA cannot act in the absence of a well established statistical model,” he explained.
Myriad has a USD 1.84bn market cap.
--------------------------------------------------------------------------------------------------------
Pharmawire publishes real-time news and data on the most price sensitive issues in the global pharmaceutical market. Built for financial professionals, Pharmawire covers product approvals, litigation, licensing deals and M & A. Pharmawire provides clients with articles such as the one above in real-time via an online platform and personalized email and BlackBerry alerts. For further information on Pharmawire please email ft@pharmawire.com or call us at Americas: +1 212-500-1384 or Europe: + 44 (0)20 7059 6251
