Yes. So I think there are a couple of comments -- I think there are couple of comments to be made. First of all, that -- first of all, in the ASH presentation, which, as I recall, was among the N of equal to 33 patients. This is Dr. Tefferi's ASH presentation. He had, as you said, 2 patients who had a Grade 2 hyperbilirubinemia and did not have any patients that were reported, at that time, at least, with Grade 3 or above hyperbilirubinemia. And there were no obvious patients with other LFT abnormalities. I would comment that, number one -- and that was quite different that the EHA presentation. Still, no Grade 3 or Grade 4 LFT abnormalities associated with the ET study -- sorry, no Grade 4. There were a couple of Grade 3 ALTs and 1 AST and no Grade 3 or Grade 4 [indiscernible]. I think the really big difference in these studies was, first of all, that the EHA data, that was a data cut of around May of 2013. And I don't recall the exact number of months on average that patients had been treated. But it was up to -- roughly up to 2.5 years and on average, around 14 months, if memory serves me correct. And so that's a really quite a bit of longer than Dr. Tefferi have been studying patients. And of course, remember also that the EHA data for sure was everything that was reported regardless of attribution et cetera, and regardless of baseline data. So I think it is certainly possible that there will be differences on a -- from individual disease basis. And it's certainly possible that there may be, for example, a different level of baseline abnormalities. For example, in myelofibrosis, we know that in myelofibrosis, patients have extramedullary hematopoiesis. That is their bone marrow isn't working and so they push out their production of red cells and white cells and so forth to the liver and spleen. And so they may have a much higher incidents of baseline abnormalities, and some of those may go away. So I think it's -- we're just speculating now. I think it's a little too hard to make a real firm comment about the differences in myelofibrosis patients versus ET in the reactions to imetelstat. It's too much apples and oranges at this point.