Osteonecrosis after Allogeneic HSCT in Children with High-Risk or Relapsed ALL Treated in the ALL-SCT 2003 Trial

Introduction: Allogeneic hematopoietic stem cell transplantation (HSCT) is widely performed in children with high-risk acute lymphoblastic leukemia (ALL) in first or subsequent remission. Cure rates in these high-risk patients are approaching 70% but are compromised by the burden of sequelae of both anti-leukemic treatment and HSCT with serious long-term impact on quality of life. One of the most prevalent and debilitating side effects is osteonecrosis (ON) but data on incidence and risk factors of ON after allogeneic HSCT are still scarce.

Methods: Children and adolescents with ALL enrolled in the prospective ALL-SCT-BFM 2003 trial were eligible for this analysis. ON assessment was performed by means of adverse event and routine follow-up reporting. In patients in whom ON have been reported additional data were retrieved from the treating center via an ON-specific questionnaire.

Results: 359 patients could be analyzed, median age at HSCT was 10 years (range 0.5-23), median time from HSCT until last follow-up was 5.7 years (range 0.1-11.7). 231 (64%) patients were male, 171 (48%) patients were transplanted in first complete remission (CR), 156 (43%) patients in second CR and 32 (9%) in >CR2. 300 (85%) patients received total body irradiation (TBI) and etoposide as conditioning regimen, whereas only in 55 (15%) patients chemotherapy conditioning was administered. The main stem cell source was bone marrow (247; 69%). 260 (73%) patients developed acute graft versus host disease (aGvHD), 34 (10%) of grade 3-4. The 2-years cumulative incidence (CI) of chronic GvHD (cGvHD) was 26% (SD 3%), and of extensive cGvHD 14% (SD 2%).

Until last follow-up, 29 patients were diagnosed with symptomatic ON (sON), the CI of sON at 5 years after HSCT was 7% (SD 1%) and median time from HSCT to ON diagnosis was 15 months (range 1-80). 10 (38%) patients were diagnosed with ON within one year after HSCT, 7 (27%) in the second year and 9 (35%) thereafter. No patient was diagnosed with sON later than 6.7 years from HSCT. In four (13.8%) of those patients, sON was already present prior to HSCT. Most patients (14; 48%) presented with ON in the knees followed by hips (13; 45%) and feet (13; 45%), but ON in the shoulders (5; 17%) were also reported. In the majority of patients with ON (22; 76%), more than one joint was affected.

In univariate analysis, the CI of ON at 5 years after HSCT in children under 10 years of age was 2% (SD 1%), in children aged 10-14 years 10% (SD 3%) and 13% (SD 4%) in adolescents 15 years or older (p=0.005). Moreover, the risk to develop ON in children presenting with ON prior to HSCT (CI of ON 58% SD 19% vs. 7% SD 2% without ON prior to HSCT, p<0.001) and developing cGvHD (hazard ratio (HR) 1.8; 95% CI, 1.2-2.7; p=0.009) were significantly higher. There was no difference between male and female patients (CI 8% SD 2% vs. 5% SD 2%, p=0.34) and between patients in first or later remission, the types of conditioning regimen, stem cells sources and aGvHD. In multivariate analysis, age at HSCT (10-15 years vs. <10 years HR 8.178, p=0.007; >15 years vs. <10 years HR 7.614, p=0.017) and diagnosis of sON prior to HSCT proved as significant independent risk factors for the development of ON, whereas all other factors had no significant impact.

Conclusions: Osteonecrosis occurred with a CI of 7% after allogeneic HSCT in children and adolescents with ALL enrolled in the ALL-SCT 2003 trial. Besides diagnosis of symptomatic ON prior to HSCT, age perpetuates as the most important prognostic factor for the development of ON. Notably, CI of ON after allogeneic HSCT appears to be in a smilar range as in children and adoelscents undergoing conventional antileukemic treatment. The impact of chronic GvHD in the development of ON after allogeneic HSCT requires further study.