Dose-limiting toxicity with hydroxyurea is certainly a problem for some patients with PV. For example, some individuals may have neutropenia (with or without adequate control of the hematocrit or platelet count) at the hydroxyurea doses that they are able to tolerate. The  worst-case scenario involves the patient who is slightly neutropenic, but with platelet counts that are still too high and who still requires phlebotomy to control their hematocrit level. Other patients on hydroxyurea may achieve the target hematocrit level, but their platelet counts are too high and they are neutropenic. It then becomes like robbing Peter to pay Paul. 

Some retrospective studies have reported an association between hydroxyurea resistance or intolerance and an increased risk of transformation to primary myelofibrosis or acute myeloid leukemia. The magnitude of the leukemogenic potential of hydroxyurea in PV is controversial and it is difficult to study. If there is an effect, I do not believe that it is a dramatic one, but I do tell my patients with PV that there is sufficient evidence for concern regarding the long-term safety of hydroxyurea. Without question, hydroxyurea use is clearly linked to the development of cutaneous lesions, such as leg ulcers. There have also been reports of skin carcinomas with prolonged hydroxyurea treatment. 

There is limited evidence regarding the leukemogenic potential of second-line therapies for PV. Pegylated interferon may cause autoimmune disorders, mood disorders such as depression, and elevated hepatic enzyme levels. Busulfan, which is not commonly used for PV, may cause severe cytopenia and is likely more leukemogenic than other second-line therapies. Ruxolitinib may be associated with a slightly increased risk of shingles and nonmelanoma skin cancers; however, it is challenging to single out the adverse effects of this agent because most patients on ruxolitinib were previously treated with hydroxyurea, sometimes for long periods.