LAUSR

Controversy regarding the risk of non-hematologic malignancies in myelofibrosis patients still exists. We aimed to examine the association between myelofibrosis and non-hematologic malignancies. A cohort of 1,469,790 adults without a diagnosis of myelofibrosis was identified on 1 January 2007, from the electronic medical records of the largest healthcare provider in Israel. Participants were followed up until 31 December 2015, for the occurrence of myelofibrosis. All cases of myelofibrosis were adjudicated by reviewing patients’ electronic medical files. Using risk set sampling, four randomly selected controls (without myelofibrosis) were matched to each case of myelofibrosis on age, sex, ethnicity, and index date. Patients with and without myelofibrosis were followed from the index date until 31 December 2016 for the occurrence of non-hematologic malignancies based on the data from the Israel National Cancer Registry. The study included 550 patients with myelofibrosis and 2200 matched controls. Non-hematologic cancers occurred in 35 patients with myelofibrosis and 138 patients without myelofibrosis, reflecting a crude incidence rate of 27.9 and 15.3 per 1000 person-years, respectively. Myelofibrosis was independently associated with increased risk of non-hematologic malignancies with propensity score adjusted HR of 1.85 (95% CI, 1.09–3.15). No significant association was detected between myelofibrosis and the specific sites of non-hematologic malignancies. Treatment with ruxolitinib was not significantly associated with non-hematologic malignancies HR 1.36 (0.60–3.11). In conclusion, myelofibrosis appears to be associated with increased risk of non-hematologic malignancies. However, this study raises concerns about surveillance bias, suggesting that the association might be attributed to earlier detection rather than real increased risk.