This is Elliott. On hepatitis C, we have a big commitment to remain and become a leader in the field following upon our legacy in virology, HIV and our now blockbuster drug, BARACLUDE of hepatitis B. And we began 10 years ago with the thought that multiple mechanisms would be needed. Clearly, the field in science is changing drastically. There will be less of a role, nevertheless, a role for interferons, but a primary focus on all oral agents. We believe that it will require 2 new direct-acting antivirals at a minimum. We, therefore, feel that the 5A, with all its properties and its advancement and about 1,000 patients' safety profile, its efficacy will be a good contributor to that future regimen and do believe that a nuke, like the one that we acquired or about to acquire from an Inhibitex will be an excellent partner. It is true that this year will unmask a lot of data, nucleotide with ribavirin in genotype 1. Nobody knows exactly how that experiment is going to unfold. We feel that could be improved upon with an agent like 5A and the nucleotide. We are collaborating with the Pharmasset, Gilead, nuke and 5A. That data will be available at EASL this year. The particular nuke that we have added now to our portfolio, and I should mention, we have several ongoing BMS-owned all-oral regimens with a protease, a non-nuke, a protease and a 5A and different combinations of ribavirin. And now with our nuke, after the acquisition of Inhibitex, we have the combination of 5A as a major target, as well as some collaborations with the next-generation medicines. The 12-week data on -- of the nucleotide on top of standard of care should be available towards the end of the year and, hopefully, at the Liver meeting by the end of the year, at EASL, the INX Inhibitex drug on 7 days with ribavirin will be discussed, we hope. We were particularly attracted to this particular nuke because of the new platform that's used to, I think, circumvent the problems that people have had in this area for toxicity before. We got comfortable by the fact that we saw no mitochondrial toxicity, which we think's been the root of problems in the past. We have a strategy to use lower doses because of its synergy perhaps with ribavirin, but importantly with other agents such as 5A. And this year, we'll be trying to find the right dose of that compound with our combinations for the goal of having an all-oral pan-genotypic regimen. Charlie?