Outcome of Patients with Myelofibrosis Relapsing after Allogeneic Stem Cell Transplant: A Retrospective Study By the Chronic Malignancies Working Party of EBMT

Background: Over the last decade, there has been a significant increase in the number of patients with Myelofibrosis (MF) undergoing allogeneic stem cell transplantation (SCT). However, scarce information exists on the outcome and management of those patients who relapse following SCT. Moreover, the management of relapse occurring post-SCT is often heterogeneous and ranges from palliation to intensive salvage approaches. We therefore conducted a retrospective EBMT registry analysis of adult MF patients who relapsed following first SCT episode.

Results: A total of 1216 adult patients (997 (82%) with Primary MF (PMF) and 219 (18%) with secondary MF (sMF)) underwent 1^st allogeneic SCT between 2000 and 2010. A total of 251 patients from this cohort (206 with PMF and 45 with sMF) had conformed relapse ≥ day 30 after HSCT and were included in the analysis. Within this relapse cohort, there were 163 males and 88 females; median age was 55 years old (range 21.5-70 years). A total of 84 patients (33%) had received Myeloablative Conditioning (MAC) and 167 patients (67%) Reduced Intensity Conditioning (RIC). Regarding donor type, there were 123 matched siblings (49%) and 128 unrelated donors (51%). Acute GVHD (aGVHD) status was available for 243/251 (97%) patients; no aGVHD was evident in 143 patients, Grade I-II aGVHD 76 patients, Grade III-IV aGVHD 22 patients and 2 patients with aGVHD ungraded.

The median time to relapse after SCT was 7.1 months (range 1-111 months). The median Overall Survival (OS) from the time of relapse was 17.7 months (95% Confidence Intervals 11-24). Collectively, there was a significant difference in survival outcome for those relapsing > 7.1 months post-SCT (median survival 30.3 months post relapse) compared to those relapsing within 7.1 months following the initial SCT episode (median survival 7.9 months post relapse; p<0.001). Absence of aGVHD or grade I aGVHD only was associated with a trend towards improved survival following relapse compared to those with Grade II-IV aGVHD (p=0.12). For PMF, disease duration prior to SCT did not significantly affect outcome post relapse. Heterogeneous practice existed as regards management of the relapse episode, with considerable variation in median survival (MS) estimates. 47 patients received Donor Lymphocyte Infusions (DLI) alone (MS 76 months); 21 had chemotherapy alone (MS 23 months) whereas 14 patients had DLI combined with chemotherapy (MS 13.6 months). As regards 2^nd allografts: 53 patients underwent 2nd allograft alone (MS 23.6 months) and 26 underwent DLI and 2^nd SCT (MS 53.9). In 90 patients active management -if any- was not documented (most likely many were palliative) but represented a very poor risk group with a MS of only 4.8 months. Overall, there was a significant improvement in OS post relapse for those undergoing 2^nd SCT (n=79) versus those who did not have a 2^nd SCT (n=172; p=0.019).

Conclusions: This analysis represents the first study to define the outcome of MF patients who undergo relapse following allogeneic SCT. Treatment of relapse presents huge challenges and the heterogeneous management strategies highlighted above reflects current practice where approaches range from palliation through to intensive chemotherapy and 2^nd SCT. It is clear from this analysis that early relapse has a much worse prognosis than those who relapse later than 7.1 months post-SCT. There is a definite survival advantage for those who undergo DLI and/or a 2^nd SCT procedure, although we acknowledge that those patients undergoing a 2^nd SCT represent a highly selected group who are fit enough to undergo such intervention. Moreover, how relapse management practice will change in the era of novel therapies such as JAK inhibitors to bridge towards 2^nd SCT is currently unclear and requires evaluation in prospective studies.