Polycythemia vera (PV) and Essential thrombocytosis (ET) progress to myelofibrosis (MF). Extramedullary hematopoeisis (EMH) is common in patients with primary or secondary MF, and can occur in the lungs. Pulmonary EMH can cause recurrent pleural effusions, pulmonary hypertension, and right heart failure with symptoms of dyspnea, cough, and fatigue. Low dose single fraction whole lung irradiation (WLI) has been utilized at our institution, and our preliminary report of 4 patients noted symptomatic improvement with no reported acute side effects. Here we report on a larger cohort of 57 patients as well as long term outcomes for 20 of those patients, including the original 4 patients.


We performed a retrospective review of 57 patients with myelofibrosis and pulmonary EMH who received single fraction WLI to a dose of 100 cGy at the Mayo Clinic from March 2001 to March 2014. Data related to the following parameters was collected: initial diagnosis, age at initial diagnosis, date of progression to myelofibrosis, initial treatment prior to radiation therapy, whole body bone marrow scan findings if available, and response to WLI. Overall survival was measured using the Kaplan Meier method. Chi-square analysis was used to evaluate predictors of response to WLI.


The median age at first WLI was 67 years (45-84 years), and 33 patients (58%) were male. Twenty-two patients (39%) had a diagnosis of primary MF, 27 patients (47%) had PV or ET, and 8 patients (14%) had another cause of secondary MF. At the time of WLI, 27 patients (47%) were on supplemental oxygen, and 3 patients (5%) were in the intensive care unit. Hydroxyurea (n=14, 25%), JAK2 inhibitors (n=9, 16%), Anagrelide (n=3, 5%), and Thalidomide and Prednisone (n=3, 5 %) were the most frequent treatments prior to WLI. EMH was confirmed on bone scan in 38 patients (67%). In the remaining 19 patients, a diagnosis of EMH was made based on clinical impression. This included symptoms of dyspnea, cough, and fatigue, echocardiographic findings of pulmonary hypertension, and in some patients recurrent pleural effusions (n=13), positive lymph node biopsy (n=2), or thoracentesis (n=1). Twenty-eight (49%) patients had other active cardiac or pulmonary conditions that likely contributed to their clinical symptoms. These patients were receiving concurrent treatment for their other conditions. In some patients there were multiple coexisting conditions.

Clinical improvement occurred in 30 patients (53%). The median time from WLI to symptomatic improvement was 10 days (1-174 days). Twenty-four patients (42%) did not have clinical improvement. Nine patients (16%) had stable symptoms, 15 patients (26%) had progressive symptoms, and 3 patients (5%) had insufficient follow up. In the group of patients with concurrent active cardiac or pulmonary conditions, 15 patients (54%) had clinical improvement following WLI. In the 29 patients who had solitary EMH, 15 (52%) patients had clinical improvement. There was no difference in response rates related to oxygen use at the time of WLI. Six patients (11%) received WLI on multiple occasions. There was no difference in the percentage of patients with positive bone marrow scans (67%) in the 2 groups. The median overall survival was 259 days for all patients. Patients who improved after WLI had a median survival of 325.5 days compared to 122.5 days for patients who did not improve.

No new hematologic abnormalities temporally related to WLI were reported. Long term follow up beyond 1 year was available for 20 patients (35%). No patients developed pneumonitis or pulmonary fibrosis that was considered related to WLI. One patient received a diagnosis of an upper esophageal squamous cell carcinoma 6 years after WLI and allogeneic stem cell transplant.


Our prior study showed WLI is safe and effective in a small number of patients with isolated pulmonary EMH from MF. The current study confirms the long term safety of this approach. Our results suggest WLI may contribute to symptomatic improvement in 1/2 of patients, even in the common clinical situation of multiple coexisting cardiac and pulmonary conditions. Repeat WLI is also well tolerated and can result in symptomatic improvement. We did not find any factors that predicted response to WLI. WLI should be considered in patients who have clinically proven pulmonary EMH and associated symptoms, even in the presence of other conditions, and can be repeated safely.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution