Cherie,

Before I post retorts to your writings, let me first state that we have both made the treatment decisions we have based on our assessments of the data available and our individual cases of MS. The point of these discussions is to share information to help others in their treatment decisions without the bias we have due to the treatment decisions we've made.

It would be statistically impossible for the Tysabri effectiveness results to be due to confounding factors. The Tysabri trials had almost 2000 patients which would eliminate and small sample bias and there was a Tysabri/Avonex comparison versus Avonex alone arm. The other factor you choose to ignore is that 6% of Tysabri patients developed neutralizing antibodies making the drug ineffective for them. 6% of the TYsabri patients were effectively on placebo and the drug still reduced relapses by 70% relative to placebo.

Avoiding relapses is "pleasurable"? *edit* Avoiding relapses is necessary maybe not sufficient but definitely necessary. Besides nothing about this disease is pleasurable.

Using the PML outcomes from the trials to paint PML outcome expectations is flawed since in the trials, PML wasn't being looked for and was mistreated by adding further immunnosuppression. These patients never had a chance. In the commercial setting, Doctors are watching for PML and trying to remove the immune supression when detected. It certainly doesn't mean that PML won't always remain a serious infection but it does mean that the outcomes won't necessarily be the same.

Please don't quote PML risk numbers that are incorrect. You scare uninformed patients taking the drug and scare away others that are evaluating their treatment decisions. The 3 cases of PML that occurred in the trials don't belong in the PML numerator as we are trying to assess PML incidence in the commercial setting with the commercial restrictions. So we've got 4 PML cases for the numerator. Your patient numbers in the denominator are also incorrect as they are as of Sep 31, 2008. In the 3 months since then more patients (probably 1300 per month) have reached the 1 year, 18 month, 2 years and over treatment threshold. The other confounding items in making this calculation is that 3 of the PML cases occurred between 14 and 17 months and one at 26 months and that only 1 of the cases occurred in the US TOUCH program even though 60% of the patients on the drug are in the US. Do you use the 1 year patient numbers or the 18 month patient numbers given the 14 month cases? Does the TOUCH program provide further PML risk reduction through tighter patient admittance criteria and clooser monitoring? There are a great deal of unanswered questions and certainly a great deal of discussion needed above and beyond a dismissive comment that I'm wrong and the risk is about what the label says.

Chris