Response to the letter by Langabeer on "Retrospective screening for Philadelphia-negative myeloproliferative neoplasms in patients with cerebral infarctions as revealed using the revised 2016 World Health Organization diagnostic criteria".

TO Dr. LANGABEER: Thank you for your interests in our observations and good suggestions. Thromboembolism can precede overt presentation of a myeloproliferative neoplasm (MPN) by a few years. In this context, your suggestions on the molecular testing in stroke patients may be reasonable. Mayo Clinic has screened JAK2V617F in patients with 664 patients with non-splanchnic thrombosis, including 136 stokes. JAKV617F was found in only 6 patients, and the mutant allele burden was low in all instances (2.2–7.5%). Based on these observations, they concluded that mutation screening was not warranted as part of the hypercoagulable work-up in the absence of MPN [1]. In addition, JAK2V617F seems to be found in normal population [2]. Furthermore, currently early detection of MPN driver gene mutations does not confer therapeutic implications in stroke patients with normal hematologic features, because available hematologic intervention such as cytoreductive therapy is indicated only in patients with overt thrombocytosis or erythrocytosis. Accordingly, we need to pay more attention to hematologic alterations at the time of diagnosis and regular follow-up of hematologic tests in patients with stroke rather than routine molecular screening, at least in daily clinical practice. For the selected cases of embolic stoke or cerebrovascular sinus thrombosis with otherwise undetermined source, molecular screening for MPN may be reasonable even in the absence of abnormal hematologic findings [3]. Collectively, molecular screening for MPN in stoke patients is still an open question, requiring prospective investigations and discussions.