what could go wrong?, page-46

I hope this helps you guys a bit, it's communication I had with Pete back in the day with regards to the trial design and Erbitux. Good to see the robust discussions still continuing on these threads:

#######

The primary objective of the Phase III is to show that HA-Irinotecan is clincally superior to irinotecan

We are not using single agent irinotecan in our Phase III trial but using HA-irinotecan as part of the FOLFIRI regimen. This is driven by the ability to recruit which, in turn, is a reflection of current clincal practice. Our Clinical Advisory Board told us that single agent iri would be very difficult to recruit. FOLFIRI does perform a little better than single agent iri but, if our Phase II data is real, HA-irinotecan performs substantially better. We are basing our stats on a 3m median progression for the control group (FOLFIRI) and powering to see a 6 week difference for the FOLF(HA)-Iri group. There was a 12 week benefit in the Phase II -obviously we do not know whether HA-Irinotecan in the FOLFIRI setting will provide a further benefit, i.e. maintaining the 12 week difference. Regardless, we are powering the study for a much smaller difference than that seen in the Phase II.

The vast majority of patients (~90%?) receive FOLFOX +/- Avastin in 1st line mCRC and then, on progression, generally move to an irinotecan containing regimen. Merck and BMS are certainly trying to push Erbitux more into first line and, indeed, the wtKras patients will be the ones for whom this will be considered. That said, there is still the issue of skin rash which puts some patients off from using Erbitux.

We have excellent preclinical data that shows that HA-irinotecan + Erbitux is better than irinotecan + Erbitux. We will consider a phase II trial in this setting once we have proof-of-principle from the Phase III - this would allow us to present a publication to clincians but not to promote the drug for this indication until our label was changed to include it. We considered using HA-I in combination with Erbitux for the Phase III but it would have been very expensive, especially after the CRYSTAL study (~$80m) and would have taken longer (more patients, longer time to endpoints).

We will recruit wtKras and mutant Kras (there is no difference in their responses to iri, only Erbitux). Remember that in many markets Erbitux is too expensive to use and we shoudl be able to recruit a good mix. We will look at the patients according to Kras status as a prospectively defined analysis in the SAP.

The dose and schedule of irinotecan in HA-Irinotecan is identical to th current irinotecan label and we have an identical adverse event profile which suggests that HA-Iri may be able to be incorporated into any iri containing regimen without fear of additional unexpected adverse events.

The important part is recruitment and, despite some of the changes in therapy, there will still be a large number of patients who will be receiving iri second line, mostly in the FOLFIRI regimen. The clinical strategy to maximize the sales of HA-irinotecan will unfold after we have proven that HA-Irinotecan is better than Irinotecan. The majority of iri is used second line in any case (and it has been that way for some time) and if Erbitux does take more ground in 1st line, we should assume that HA-Irinotecan + Erbitux can play in that space too.

#######