Harry P. Erba, MD, PhD: What are specifically the goals of treatment? Now, when I was training back around 20 or 25 years ago, my attending physician said, “Well, you keep the hematocrit less than 45%.” Or, if you had a physician who liked hemoglobin, it was hemoglobin less than 15%, but typically, it was a hematocrit less than 45%. The data were really epidemiologic data, some preclinical studies looking at blood viscosity. It wasn’t until the CYTO-PV study published by the Italians and Marchioli in the New England Journal of Medicine that we had clinical trial data saying that keeping the hematocrit less than 45% was important.

They did a study in patients with PV who were on phlebotomy therapy and low-dose aspirin, and many of them—over half of them—were getting hydroxyurea, as well. They randomized them into 2 populations: a population that had tight control of the hematocrit, less than 45%; or a population with less stringent control, 45% to 50%. They asked the question. The endpoint was what the rate of death was due to cardiovascular disease or major thrombotic events. And actually, the study was stopped because of poor accrual. But they had enough accruals, such that they got an answer. At 3 years, there was a statistically significant difference in the rate of death due to cardiovascular disease and thrombotic events. It was 10% in the patients between 45% and 50% and around 2.9%, so almost 4-fold less incidence in patients who had the tighter control of the hematocrit.

That seemed to settle the issue in a very well-done randomized trial. However, in a multivariate analysis, they looked at other factors, and lo and behold, white count also turned out as important. The way they did that was by comparing the thrombotic events of patients who had a white count less than 7000 mm³ to those between 7000 and 8500 mm³; between 8500 and 11,000 mm³; and then above 11,000 mm³. And above 11,000 mm³, they saw a statistically significant increase in the number of thrombotic events. There was a trend for more thrombotic events above that. Some have taken this information to say, “Well, you need to keep the white count less than 11,000 mm³.” I’m not sure if that’s the right conclusion. What does the panel think? Is it at 11,000 mm³ you’re fine, but at 11,100 mm³, change therapy? What do you think?

Rami Komrokji, MD: I’m not sure if it’s that clear of a cutoff. But I think part of the study did not distinguish whether the reduction in the hematocrit was also an effect of the cytoreductive therapy, decreasing the leukocytosis, and other effects. I think, no doubt, the leukocytosis is important, but I’m not sure if we really know the exact number. I like the numbers to be within a normal range, to be honest, when I’m treating patients if I’m offering them cytoreductive therapy. It’s something we look at, but I don’t think those are conclusive data. When I trained, which was a little bit after you did, we also distinguished between males and female. So, we used to get a goal of 42% for females and 45% for males. Again, I think some of the patients will tell you when they need treatment. The symptoms are, in some patients, related to the numbers. I have patients who will call and say, “I think I need a phlebotomy. My symptoms are coming back.” Another important point is also how frequently you look. Because sometimes, if you are checking the counts every 6 months or 4 months, you don’t know in between that what was caused by white blood cell count or the hematocrit. The more you look, the more you get tighter control on the counts.

Jamile M. Shammo, MD, FASCP, FACP: I think that there were 2 other studies that showed leukocytosis can actually be related to increased risk of thrombosis, independent from the hematocrit level. I think even Tefferi had a study where he tried to come up with a different score, and they looked at a white cell count greater than 15,000 mm³. There is no uniformity as to what white cell count should you be concerned at. This is 11,000 mm³, the other study is 15,000 mm³, and there’s a trial also looking at leukocytosis where there were some patients on the order of 13,000 mm³. So, when I look at the leukocytosis, I think of it as a reflection of myeloproliferation, that perhaps this patient needs to be paid attention to a little bit more than someone else who has a nice-looking CBC [complete blood count], perfectly controlled with hydroxyurea. That’s what I do.

Harry P. Erba, MD, PhD: Srdan?

Srdan Verstovsek, MD, PhD: That begs the question that in the otherwise low-risk patient with PV who has a high white cell count, would the high white cell count alone—assume that you control the red blood cell count with phlebotomies, and symptoms are not there—be a reason to initiate therapy if we have no uniformity in retrospective analysis? We are not sure; that’s why no guidelines are there. But what we do agree on is that progressive leukocytosis is not good for the patient.

Rami Komrokji, MD: Right.

Harry P. Erba, MD, PhD: I think that’s a very key point. It’s not 1 point in time.

Srdan Verstovsek, MD, PhD: You have to follow it. It’s a chronic disease. It’s the chronic change. If it’s changing and keeps going up to 20,000 or 25,000 mm³, I don’t feel comfortable anymore. I would usually introduce cytoreductive therapy, and then the goal would be to normalize it.

Ruben A. Mesa, MD, FACP: It’s really a spectrum, and clearly it’s a crude marker. I don’t believe there’s a phenotypic or biologic difference between the patients who manifest leukocytosis and those who do not. I think Jamile’s point is an excellent one. The second point is that we really don’t have data yet. All that being the case, what does it mean in terms of therapy? Everyone interprets it in a variety of ways. Does that mean you just control it to that threshold of risk? I doubt that’s good enough. I think that’s a bit of a manifestation.

I, too, all things being equal, will at least have a discussion with the patient who has leukocytosis and polycythemia vera about whether there should be cytoreduction. I will be transparent that I don’t know for certain that it’s helpful, but I have that concern. I view with polycythemia vera that conservative approaches—phlebotomy and aspirin alone as the standard—have largely risen out of a fear of hydroxyurea that has made us most conservative. How do we limit the usage of hydroxyurea? If we had a great up-front therapy for polycythemia vera, then there probably would be no differentiation in that threshold.

Harry P. Erba, MD, PhD: Maybe this is a bad comparison, but I do think about the data we have in sickle cell disease with hydroxyurea, and people like to talk about how hydroxyurea increases hemoglobin F levels. But let’s not forget the way those studies were done: The doses of hydroxyurea were actually adjusted to have the patients on it achieve a low normal white blood cell count. Maybe the lower incidence of vaso-occlusive crises is also due to control of the leukocytosis, which has been shown to be, like in PV, a risk factor for vaso-occlusive events. It fits together. I think what I’ve heard the panel say is yes, leukocytosis may be very important for the risk of thrombotic events, but in terms of a specific threshold, we can’t identify one.

Transcript Edited for Clarity