Outcomes Following Allogeneic Hematopoietic Stem Cell Transplantation for Childhood Cerebral Variant Adrenoleukodystrophy: A Modern, Single-Institution Experience

Abstract 3543

X-linked adrenoleukodystrophy (ALD) is a peroxisomal disorder affecting 1 in 21,000 males. Approximately 40% of patients manifest the devastating childhood-onset cerebral variant (ccALD), characterized by rapid progression to death from cerebral demyelination and ensuing neurologic dysfunction. To date, only allogeneic hematopoietic stem cell transplantation (HSCT) has been shown to alter the natural course of ccALD; additionally, the use of peri-HSCT N-acetylcysteine (NAC) has been shown to enhance survival in boys with extensive pre-HSCT cerebral disease. We reviewed characteristics of as well as survival and neurologic functional outcomes for 60 consecutive boys undergoing allo-HSCT for ccALD at the University of Minnesota between 2000 and 2009; to our knowledge, this analysis represents the largest, single-institution experience in HSCT for ccALD.

Seventeen (28%) boys were diagnosed due to a family history; 37 (62%) were diagnosed secondary to signs of disease; 6 (10%) had an unknown instigation for diagnosis. All (100%) had magnetic resonance imaging (MRI) evidence of cerebral involvement at the time of HSCT. The median age at HSCT was 8.7 years (range 4.0 to 23.3 yrs). MRI severity scoring was according to the system published by Loes (0 = no disease on MRI, 34 = maximal involvement); the median pre-HSCT Loes score was 9 (range 2 to 19). Clinical neurologic involvement was graded per a neurologic function scale (NFS, 0 = no neurologic findings, 25 = maximal neurologic findings) spanning the domains of hearing/communication, vision, gait/movement, swallowing, incontinence, and the presence of seizures; the median pre-HSCT NFS was 1 (range 0 to 14).

Eighteen (30%) had a related donor, while 42 (70%) received an alternative graft. Twenty-eight (47%) were marrow recipients; 32 (53%) received umbilical cord blood. The preparative regimen was busulfan/cyclophosphamide-based in 28 (46%) boys; cyclophosphamide/total body irradiation-based in 16 (27%); and reduced-intensity conditioning in 16 (27%). Since 2005, all patients (34 boys, 57%) received peri-HSCT NAC.

Forty-seven (78%) are alive at the time of analysis; median follow-up among survivors is 3.7 years (range 0.7 to 9.6 yrs). Death was from disease progression in 5 boys, graft failure in 3, infection in 2, and 1 patient each died from acute graft-versus-host disease (aGvHD), hemorrhage, and hemolytic anemia. Eleven (18%) boys experienced grade II-IV aGvHD; 4 (7%) developed grade III-IV aGvHD.

In univariate survival analysis of pre-HSCT factors, Loes score was statistically significant [pre-HSCT Loes < 10, 89% 5 year survival (95% CI: 70% - 96%); Loes ≥ 10, 60% (34% - 78%); p = 0.03]. A trend toward better survival was seen for patients with no pre-HSCT neurologic signs of disease [NFS = 0 pre-HSCT, 91% (69% - 98%); NFS ≥ 1, 66% (46% - 81%); p = 0.08] and for those who received NAC [NAC exposed, 80% 1 year survival (72% - 95%); no NAC exposure, 73% (52% - 86%); p = 0.08]. Graft source, age at HSCT, conditioning regimen, presence of adrenal insufficiency, nor reason for diagnosis impacted survival. Of note, of 8 boys with pre-HSCT Loes < 10 who received a related allograft, all (100%) are alive at the time of analysis.

In univariate analysis of pre-HSCT factors on post-HSCT clinical neurologic outcomes, both pre-HSCT Loes score and NFS demonstrated significant impact. For those with a pre-HSCT Loes < 10, the most recent NFS assessment (median 2.9 yrs, post-HSCT, range 0.2 to 8.3 yrs) showed a median change of 0 points (range -1 to 20) when compared to the pre-HSCT score; while for those with a pre-HSCT Loes ≥ 10, most recent NFS (median 2.1 yrs, range 0.8 to 6.6 yrs) showed a median change of 7.5 points (range 0 to 23 points); p < 0.01. For those with pre-HSCT NFS of 0, the most recent NFS (median 2.2 yrs, range 0.2 to 8.3 yrs) a median change of 0 points (range 0 to 8) was observed; while for those with a pre-HSCT NFS ≥ 1, most recent NFS (median 2.3 yrs, range 0.8 to 7.2 yrs) showed a median change of 7 points (range -1 to 23); p < 0.01.

In summary, we report a large, recent, single-institution experience of allo-HSCT for ccALD. While HSCT in early disease yields encouraging functional outcomes and survival in comparison to untreated ccALD, additional modifications will be necessary to achieve optimal results for patients transplanted with advanced disease. Finally, this data continues to support the need for early diagnosis of those affected by this devastating disease.