Isolated Extra-Medullary Relapse of Acute Leukemia after Allogeneic Stem-Cell Transplantation; Different Kinetics and Better Prognosis Than Systemic Relapse.

Allogeneic stem-cell transplantation (SCT) is a potentially curative treatment for acute leukemia, however relapsing disease is the major cause of treatment failure after SCT. Isolated extramedullary relapse is considered a rare event and its characteristics and prognosis are less defined than in systemic relapse. We analyzed outcomes of 356 consecutive patients (pts) with AML/MDS (n=277) and ALL (n=79) given SCT over an 8-year period in a single institution. This was a relatively high-risk group with only 34% of pts been in CR1. 68% of transplants were myeloablative, 54% of the donors were HLA matched siblings, 39% unrelated and 7% haploidentical and cord blood. With a median follow-up of 30 months (range, 1–103), 158 patients are alive; 75 died of treatment related causes, 149 patients relapsed (123 of them have died). The overall and disease-free survival rates were 37% (95CI, 31–43) and 31% (95CI, 25–36), respectively. The cumulative incidence of relapse was 47% (95CI, 42–54). Seventeen pts had an isolated extramedullary relapse (11.4% of all first relapses after SCT); CNS (n=6), mediastinum (n=2), pleura (n=2), skin (n=2), breast (n=2), bone (n=1), other soft tissue masses (n=2). Isolated extramedullary relapse occurred later after SCT [median 14 months (range, 1–38)] than systemic relapse [median 3 months (range, 1–59), p=0.002]. Pts with an isolated extramedullary relapse were younger than pts with systemic relapse, median age 38 and 46 years, respectively (p=0.02). Isolated extramedullary relapse occurred more commonly in ALL (23% of relapses) than MDS/AML (8%, P=0.02). There was a trend for higher incidence in pts transplanted in remission (16% of relapses) than in pts transplanted in active disease (8%, p=0.09). There was no relation to donor or conditioning type. Interestingly, among 14 pts having isolated extramedullary relapse more than 3 months after SCT, 11 had a history of chronic GVHD (79%), compared with 29 of 70 pts with systemic relapse (41%, p=0.01). Pts with isolated extramedullary relapse were most often treated with systemic chemotherapy (including intrathecal chemotherapy in CNS relapse) followed by radiation therapy when feasible and DLI in the absence of active GVHD. Pts with systemic relapse were treated with chemotherapy and DLI or a second SCT. Among the 17 pts with isolated extramedullary relapse, 12 achieved CR (71%), 5 remained in continuous CR, 7 relapsed again, 2 systemic (both died), 5 extramedullary (2 of them are in long-term remission). Among the 132 pts with systemic relapse, 35 achieved CR (27%), 14 died in remission, 7 remained in continuous CR, 14 relapsed again, 9 systemic (all died), 5 extramedullary (1 of them is in long-term remission). After a second relapse, only pts with an extramedullay relapse survived long-term. The median survival after relapse was 2.5 and 20 months after systemic and isolated extramedullary relapse, respectively, and the 2-year OS rates were 7% and 38%, respectively (p=0.0002). The CNS was the most common site of extramedullary relapse. In all, 9 pts (AML-4, ALL-5) had an isolated relapse, 6 first and 3 second relapse after SCT. Eight pts achieved CR with intrathecal therapy, 7 were given cranio-spinal irradiation and only one recurred in the CNS. Four are in continuous CNS and systemic remission 6,13,43 and 43 months after relapse. In conclusion, isolated extramedullary relapse of acute leukemia after SCT is relatively common, especially in ALL. It occurs later than systemic relapse and more commonly in pts with chronic GVHD, suggestive that GVL is able to prevent systemic recurrence but is less successful in the extramedullay sites. When treated aggressively with a combination of chemotherapy, radiation therapy and immunotherapy prognosis is better with isolated extramedullary relapse and long-term survival is feasible.