Survival analysis of haematopoietic cell transplantation for childhood cerebral X-linked adrenoleukodystrophy: a comparison study

Summary

Background

Favourable outcomes have been reported for patients with childhood cerebral adrenoleukodystrophy (CCALD) who had received haematopoietic cell transplantation (HCT) at the early stage of cerebral involvement. However, comparative data for non-transplanted CCALD patients are limited. We analysed survival of CCALD patients who had not received HCT and, in a subgroup with early cerebral disease, compared survival in those who underwent HCT with those who did not.

Methods

Retrospective survival analyses were done on 283 CCALD patients identified at the Kennedy Krieger Institute who had not received HCT, focusing on a 30-member early stage cerebral subgroup whose neurological disability and MRI severity scores matched those in a 19-member transplanted subgroup previously reported. A Kaplan-Meier survival curve and log-rank test were used for survival analysis and to estimate the difference between the survival probabilities of the groups with statistical significance set at α=0·05.

Findings

Mean age at onset of symptoms in the entire 283 non-transplanted group was 7 years (SD 2 years). 131 (46%) patients died during the mean follow-up period of 5·9 years (5·3) at a mean age of 12·3 years (4·9). 5-year survival was 66%. The 5-year survival probability of 54% in the early stage group was significantly poorer (χ²=7·47, p=0·006) than the 5-year survival of 95% in the transplanted group with early stage cerebral disease.

Interpretation

HCT done in the early and progressive stages of CCALD is beneficial, and our data support the recommendation that transplantation be offered to patients in the early stages of CCALD.

Introduction

X-linked adrenoleukodystrophy (ALD) is caused by a defect in the gene ABCD1, which is located at Xq28 and codes for a peroxisomal membrane protein.¹ It is panethnic and affects around 1 in 20 000 males.² There are two major neurological phenotypes: the rapidly progressive cerebral phenotype in which there is an intensely inflammatory demyelination that most frequently affects the parietal-occipital lobes,³ and adrenomyeloneuropathy which presents as a slowly progressive paraparesis in adults due to distal axonopathy that affects most severely the long tracts in the spinal cord. The cerebral inflammatory phenotype is most frequent in boys younger than 11 years. This phenotype is referred to as childhood cerebral adrenoleukodystrophy (CCALD), and affects about 35% of all male ALD patients.⁴ Adolescent and adult cerebral inflammatory phenotypes also occur but are less frequent.⁴ In a series of 167 patients, the mean age of onset of CCALD was 7·2 years (SD 1·7, range 2·75–10).⁵ The initial manifestations might be subtle, but they were followed by increasing deficits in behaviour, cortical function, vision, auditory discrimination, coordination, and motor function, as well as seizures, often leading to an apparent vegetative state in 1·9 years (SD 2; range 0·5–10 years. Mean age at death was 9·4 years (range 5·1–19 years).⁵

Substantial published work indicates that haematopoietic cell transplantation (HCT) benefits patients with CCALD if the procedure is done at an early stage6, 7, 8, 9, 10, 11, 12, 13, 14 and it is now considered the standard therapy by most clinicians. Aubourg and colleagues⁶ were the first to report the beneficial effect of HCT in a patient with early stage CCALD. Shapiro and co-workers,⁷ in a follow-up study of 12 patients, showed that the beneficial effect was maintained for 5–10 years. The most comprehensive study of HCT in CCALD is that of Peters and colleagues,⁸ who provided follow-up data for 126 patients who had received HCT between 1982 and 1999. Although the overall 5-year and 8-year survival probabilities of the analysed group were both 56%, a much more favourable outcome (ie, 5-year survival probability of 92%) was seen in a 25-member subgroup.⁸ The distinctive feature of this subgroup was that neurological involvement at time of HCT was at an early stage. Peters and colleagues used two criteria to assign patients to this group: score of 0 or 1 on a 4-point neurological deficit score, or a score of less than 9 on the 34-point ALD MRI severity scale developed by Loes and colleagues.¹⁵ 53% of patients who had a neurological deficit score of 0 or 1 at HCT remained neurologically stable 5 years after the procedure.

Although the data strongly suggest that HCT done in the early stage of the illness has a favourable effect, they do not exclude the possibility that this group with early stage disease might represent a distinct phenotype category with a more favourable prognosis even without HCT. Placebo-controlled studies of HCT have not been done, and the encouraging results in early stage cerebral patients combined with the severity of untreated CCALD make such studies unlikely to be ethically permissible in the future. Assessment of the efficacy of HCT thus depends on comparison with previous controls. As we reviewed the available data for the natural history of untreated CCALD, we noted that the available survival probability analyses were based on the total CCALD group,8, 16 without reference to the degree of severity of the illness at time of diagnosis. Yet, favourable results with HCT have been noted only in patients with early stage cerebral disease. This observation prompted us to do a retrospective survival analysis of a cohort of 283 CCALD patients who have been followed up at the Kennedy Krieger Institute and who had not received HCT. We compared these data with the survival analysis of transplanted patients reported by Peters and colleagues.⁸ Particular emphasis was placed on the survival analysis of 30 non-transplanted patients whose degree of neurological disability matched a 19-member subgroup of early stage cerebral disease transplanted patients who had a favourable 5-year survival probability of 95%.