Main Session
Sep
30
PQA 09 - Hematologic Malignancies, Health Services Research, Digital Health Innovation and Informatics
3611 - Impact of CNS-Directed RT before Allogeneic Hematopoietic Cell Transplantation (alloHCT) among Patients with Acute Leukemia and CNS Involvement
Presenter(s)
Maryam Ebadi, MD - University of Washington, Seattle, WA
M. Ebadi1, T. Gooley2, J. S. Appelbaum2, V. Venur2, S. J. Lee2, R. P. Ermoian3, M. E. M. Percival2, A. Rashidi2, R. D. Cassaday2, and Y. D. Tseng1; 1Department of Radiation Oncology, University of Washington/ Fred Hutchinson Cancer Center, Seattle, WA, 2Clinical Research Division, Fred Hutchinson Cancer Center, Seattle, WA, 3Department of Radiation Oncology, University of Washington, Seattle, WA
Purpose/Objective(s):
CNS involvement by acute leukemia portends worse outcomes after alloHCT. As leukemia cells are radiosensitive, moderate doses of RT may improve tumor control. We hypothesized that CNS-directed RT prior to alloHCT may decrease risk of CNS relapse (CNS-R) and improve progression free survival (PFS) among patients with a history of CNS leukemia.Materials/Methods:
Eligible patients in this retrospective single institution analysis included adult patients with acute leukemia that underwent alloHCT (2013-2024) and had a history of CNS involvement by CSF flow cytometry at any time between diagnosis and alloHCT. CNS-R, PFS, and OS were estimated from alloHCT. Predictors of CNS-R and PFS were determined using Cox regression.Results:
Among 112 patients (50% AML, 46% ALL, 4% mixed phenotype; median 49 years [range 20-76]), 64 (57%) were male. 21 (19%) patients received CNS-directed RT (CSI 76%, whole brain [WBRT] 19%, spine 5%) prior to alloHCT (RT cohort). The median dose was 12 Gy (range 10.8-24) for CSI and 24 Gy (range 23.4-24) for WBRT. CSI was delivered in tandem with alloHCT and the median time interval between WBRT and alloHCT was 5 months (range 0.8-10.5). Compared to the cohort with no CNS-directed RT (no-RT cohort), the RT cohort was younger (median 41 vs 50 years), more frequently received myeloablative conditioning (71% vs 56%), and tended to have more adverse features including high disease risk index (DRI; 52% vs 22%), transplant in second or higher remission (57% vs 28%) and active systemic disease at the time of HCT (14% vs 8%). Notably, 6 patients (all in the RT cohort) had active CNS disease prior to alloHCT. 71 (78%) patients in the no-RT cohort received TBI (38 myeloablative, 33 non-myeloablative) compared to 17 (81%) patients in the RT cohort (13 myeloablative, 4 non-myeloablative). The no-RT cohort received more IT therapies (median 6 [range 1-22], 4 [0-17] prior to HCT) compared to those in the RT cohort (median 4.5 [1-22], 3 [1-22] prior to HCT). With a median follow-up of 53 months (range 5-137) for survivors, 8 CNS-Rs occurred, 6 concurrent with or after systemic relapse (systemic-R). Median time to CNS-R was 13 months (range1-60). There were no differences between the RT and no-RT cohorts in CNS-R (2-year CNS-R 5% vs 6%), PFS (2-year PFS 31% vs 37%), or OS (2-year 51% vs 48%). On univariate analysis, CNS-directed RT was not associated with CNS-R (HR 0.63, 95%CI 0.08-5.2, p=0.66). Multivariable analysis (MVA) was not performed given the small number of CNS-Rs. Adjusting for disease status at alloHCT and active CNS disease at alloHCT, CNS-directed RT was not associated with PFS (HR 1.35, 95%CI 0.68-2.69, p=0.40) on MVA.Conclusion:
Within this modest cohort, CNS-directed RT did not impact CNS-R or PFS. This may in part be due to most CNS-Rs occurring during or after a systemic-R, highlighting the importance of adequate systemic control to prevent CNS-R.