|
IPIC can be faulted for running relatively small sample CerAxon trials (though Takeda used the same sample size in its Japanese Phase III), which left them vulnerable to skews in the patient population. The meta-analysis of the two main US Phase III trials showed the 500mg dose was clinically effective, p=.03 (as was presented at the Academy of Neurology meeting in April).As I wrote previously, the MRI study (which was never intended to be a pivotal PhIII) also is supportive of CerAxon, but did not reach statistical significance due to its small sample size, population variance, and skew in the placebo group. I have spoken at length with two research neurologists who have treated large numbers of patients in Ph III citicoline trials, and who do not work for IPIC: They express complete certainty that the drug works, and that the challenge is satisfying the FDA's rigid framework for PhIII results. Even given the aforementioned placebo group fluke in the MRI trial, the contrast between drug and placebo patients (180% increase in infarct size without drug, 38% increase with CerAxon) is striking on its face....and had that placebo group skew not occurred, the p would have been at or near .05, and the NDA would have proceeded. The FDA is not alone in suffering from statistical tunnelvision, and this is a rather flimsy basis for self-congratulation amongst IPIC critics. NeuroInvestment (www.neuroinv.com) |
