Extramedullary Relapse Following Reduced Intensity AlloHCT for Adult AML.

In recent studies, the incidence of extramedullary (EM) relapse post full-intensity allogeneic hematopoietic cell transplantation (AlloHCT) for AML has been reported to be approximately 5–15%. However, the incidence and clinical significance of EM relapse following reduced-intensity (RI) AlloHCT has not been studied. To better understand the incidence and predictors of EM relapse following RI-AlloHCT for AML, a consecutive case-series of 246 adult AML patients who underwent RI-AlloHCT from (inception) February 2000 through December 2008 at City of Hope were identified and selected for analysis from a prospective observational research database.

The median patient age was 55 years (range: 19–71) with a fairly balanced gender distribution (male=125, 51%; female=121, 49%). There were 108 (44%) matched related donors and 138 (56%) matched unrelated donors. 221 (90%) of patients received peripheral blood stem cells. The median time from diagnosis to RI-AlloHCT was 6.4 months (range: 0.5–124.8). Disease status at HCT: 1CR/2CR=137(56%), induction failure=40 (16%), relapse=69 (28%). Nine percent (n=22) of patients had a history of EM relapse prior to RI-AlloHCT. Twenty-eight percent of patients (n=68) received a prior auto (n=50) or allo (n=18) HCT. The majority of patients received fludarabine/melphalan conditioning n=227, 92%. GVHD prophylaxis consisted of CSA/MMF=113, 46% or Tacro/Siro=135, 54%. A total of 205 patients were evaluable for cytogenetics. Based on SWOG criteria (2000), there were 10 (5%) favorable, 118 (57.5%) intermediate and 77 (37.5%) unfavorable cytogenetics.

Initial relapse occurred in the BM in 72 (29%) patients and EM sites in 16 (6.5%) patients. The sites were: brain/CSF 5, breast 3, multiple sites 2, spine 1, liver 1, ascites 1, soft tissue 1, testes 1, and bone 1. Three (1%) had concurrent bone marrow/EM relapse. In addition, 6 (2.4%) patients had a second relapse in an EM site. The sites of second or subsequent relapse were: CSF 2, multiple sites 2, spine 1, parietal mass 1. The two-year cumulative incidence of EM relapse was 10.4% (6.8%-15.7%). Treatment for EM relapse included: combined chemo-RT 11, RT and/or chemotherapy followed by second allo/DLI 4, chemotherapy 3, palliative care 3, and RT 1. Time from Allo-HCT to initial EM relapse was 12.2 months (range: 2.8–42.9) and time from HCT to initial BM relapse 6.5 months (range: 1.0–33.3). Overall survival after initial EM relapse was 68.8% (50.3% - 81.5%) at 1-year and 50.0% (37.3% - 61.5%) at 2 years.

Of the risk factors studied (prior EM disease, prior HCT, disease status, cytogenetic risk, and age (<55, >=55)), advanced disease status (IF/RL/>3CR) (HR: 2.8, p=0.04) and unfavorable cytogenetics (HR: 5.2, p<0.01) were found to be predictive of EM relapse.

This is the first report to analyze specifically the incidence of EM relapse of AML post RI-AlloHCT. Our data show that the incidence of incidence EM relapse following RI-AlloHCT is similar to rates reported following ablative AlloHCT. By multivariable analysis, advanced disease and unfavorable cytogenetics were associated with increased risk for EM relapse, also similar to findings in the ablative setting. In conclusion, reduced-intensity conditioning does not appear to be associated with an increased risk of EM relapse. Furthermore, the median time to EM relapse exceeds median time to BM relapse and long-term survival is possible following EM relapse.

No relevant conflicts of interest to declare.