Harry P. Erba, MD, PhD: Mary Frances, whenever I speak on PV [polycythemia vera], I’m always challenged at the end of the discussion by cases that practitioners have where they’ve done the entire evaluation and cannot find diagnostic criteria for PV or for a secondary cause. They’ve looked for sleep apnea and even congenital causes. Have you seen that kind of patient, and what advice would you give?

Mary Frances McMullin, MD, FRCP, FRCPath: Yes, there are these patients. Over the years, we’ve spent a lot of time investigating and finding new congenital mutations, and I’m aware of the fact that for many of the patients that I had 20 years ago, we have now found a series of congenital mutations. You’re still left with a group in whom you find nothing, and we’re getting to the point to say they should have extensive genetic evaluation if they’re young, particularly if they have a family history. There are still occasional patients in whom you find nothing that looked like a PV, and you do not find a mutation. That leaves you thinking, is there such a thing clearly as JAK2-negative and erythrocytosis? All the groups give criteria for that, but they have become, over time, a vanishingly small number of patients.

The other thing that I think is worth looking at is to repeat their JAK2 mutation screen. I’ve certainly investigated recently a group of these triple-negative ETs [essential thrombocythemias]. When you go back, sometimes you find low-level JAK mutations and even CALR mutations. My recommendation at the moment is that these people should look for the mutations every few years. On the final point, in some of these patients, when you follow them up over time, the erythrocytosis seems to have disappeared. Is it something that was going on in their life that we don’t know about that has gradually disappeared?

Jamile M. Shammo, MD, FASCP, FACP: I’m sorry to interrupt, but the WHO [World Health Organization] recognized this very small group of patients who would have phenotype of erythrocytosis and don’t have the JAK2 mutation. In that sense, the recommendation is, if they meet the criteria for erythrocytosis, to obtain a bone marrow biopsy, then rely on the 1 minor criteria, which is a subnormal erythropoietin level.

Again, it’s probably a very small patient population, but I want to take a moment to say that even though the WHO brought down the level of hemoglobin from 18.5 to 16.5 and from 18 to 16 in men and women, respectively, there was an oversight thinking about the iron parameters. That sometimes can complicate the picture if you have a patient who may present with thrombocytosis and may not have hematocrit or hemoglobin that’s high enough for a diagnosis of PV, but they happen to have 0 iron stores. Then, there’s your example of someone who may actually have that type of overlap. I agree with your point about doing a bone marrow biopsy, because that will allow you to assess the iron stores in that situation, and perhaps with all the clinical data together, arrive at the right diagnosis.

Mary Frances McMullin, MD, FRCP, FRCPath: We have just undertaken an extensive exercise in reviewing all the literature for the last 15 years, and then coming up with criteria for polycythemia vera. We did not think that the WHO hemoglobin levels were justified and actually stayed with the hematocrit that we had before, but you have to make the point to beware the iron-deficient patient. The other thing, on risk stratification, is when we looked at the literature, we thought that the high-risk should be people over the age of 65 rather than 60. When you look at the literature, the cutoff in all the studies is at 65 rather than 60, so we have said that high-risk is over 65, previous thrombotic events, etcetera, rather than 60. That may reflect somewhat as doctors get older, but it does seem to make sense from the evidence base.

Ruben A. Mesa, MD, FACP: My takeaway is that many of these are really continuous variables, so I think the very active, astute clinician trying to weave all of these factors together is key. With any of them—the 16.4. Is that any different than 16.6? Of course not. We all know one 60-year-old is not the equivalent of another 60-year-old. Good clinical common sense is necessary, particularly as it relates to these areas of continuous variables where the cutoffs have been statistical, but not scientific, in origin.

Harry P. Erba, MD, PhD: I’m going to put myself on the spot here with this panel of experts and say when I have a patient for whom I’ve done the entire evaluation—bone marrow, everything—and I cannot find a cause of a relatively mild erythrocytosis, usually, I will recommend a baby aspirin, but I would not phlebotomize and I definitely wouldn’t give cytotoxic therapy. Is that what you do as well?

Mary Frances McMullin, MD, FRCP, FRCPath: And follow-up.

Harry P. Erba, MD, PhD: Follow-up, yes, very important. They stay in my clinic, absolutely.

Ruben A. Mesa, MD, FACP: Again, good common sense in terms of the clinic. One, I definitely do believe there are individuals who have a mutation-negative work-up who truly still have PV. Perhaps being at a reference center, I see a higher number of these, even after we’ve done all of the other testing. You’re right; there still is always a little uncertainty with these. I’d say I particularly try to avoid any sort of cytotoxic therapy in these individuals. Phlebotomy is maybe a bit more phenotype-dependent as being a more benign therapy.

Harry P. Erba, MD, PhD: Jamile, why don’t you talk about this disease. Does it progress into anything else?

Jamile M. Shammo, MD, FASCP, FACP: Because it’s a chronic disease, and because progression perhaps happens over a long period of time, the patients may not necessarily have a great deal of symptoms to report to their physician. I think one of the most important features is to monitor the development of anemia, and obviously, most of those patients may end up being on a cytoreductive therapy. The question there: Is that due to drug or is that because of development of myelofibrosis? I find, over time, you start to reduce the doses of cytoreductive therapy and reduce the doses of Hydrea. I think that’s very important, potentially a first sign of disease progression.

Then the other thing is, obviously we need to monitor the symptoms, and we have a symptom expert here, Dr Mesa, as to spleen-related symptoms, constitutional symptoms, and also the percentage of blasts in the peripheral blood. If someone starts to have 2%, 3%, 4%, 5%, obviously that would be something that may be indicative of disease progression. We need to continue to monitor those patients pretty frequently and capture their symptomatology and signs of splenomegaly.

Harry P. Erba, MD, PhD: How common is it for PV to progress into AML [acute myeloid leukemia]?

Jamile M. Shammo, MD, FASCP, FACP: Acute evolution from PV directly to AML is extremely rare. I think I read something about 4%, perhaps. The more likely scenario is people go to myelofibrosis, and then from myelofibrosis, go to acute leukemia. The rate of myelofibrosis at about 10 years is about 10%.

Transcript Edited for Clarity