Non-osteopenic Bone Pathology After Allo-hematopoietic Stem Cell Transplantation in Patients with Inborn Errors of Immunity

Purpose There is a lack of data on post-HSCT non-osteopenic bone pathology specifically for children with inborn errors of immunity (IEI). We collected data on non-osteopenic bone pathology in children with IEI post-HSCT over two decades in a large tertiary pediatric immunology center. Methods Descriptive study with data analysis of bone pathology in allo-HSCT for IEI was performed between 1/1/2000 to 31/12/2018 including patients alive at follow-up to July 2022. Records were analyzed for bone pathology and risk factors. Exclusion criteria included isolated reduced bone density, fractures, and skeletal anomalies due to underlying IEI and short stature without other bone pathology. Bone pathologies were divided into 5 categories: bone tumors; skeletal dysplasia; avascular necrosis; evolving bone deformities; slipped upper femoral epiphysis. Results A total of 429 children received HSCT between 2000 and 2018; 340 are alive at last assessment. Non-osteopenic bone pathology was observed post-HSCT in 9.4% of patients (32/340, mean 7.8 years post-HSCT). Eleven patients (34%) had > 1 category of bone pathology. Seventeen patients (17/32; 53%) presented with bilateral bone pathology. The majority of patients received treosulfan-based conditioning (26/32; 81.2%). Totally, 65.6% (21/32) of patients had a history of prolonged steroid use (> 6 months). Pain was the presenting symptom in 66% of patients, and surgical intervention was required in 43.7%. The highest incidence of bone pathologies was seen in Wiskott-Aldrich syndrome (WAS) (n = 8/34; 23.5%) followed by hemophagocytic lymphohistiocytosis patients (n = 3/16; 18.8%). Conclusion Non-osteopenic bone pathology in long-term survivors of allo-HSCT for IEI is not rare. Most patients did not present with complaints until at least 5 years post-HSCT highlighting the need for ongoing bone health assessment for patients with IEI. Children presenting with stunted growth and bone pathology post-HSCT should undergo skeletal survey to rule out development of post-HSCT skeletal dysplasia. Increased rates and complexity of bone pathology were seen amongst patients with Wiskott-Aldrich syndrome. Supplementary Information The online version contains supplementary material available at 10.1007/s10875-023-01465-z.


Introduction
Allogenic hematopoietic stem cell transplantation (Allo-HSCT) is the primary curative treatment for many inborn errors of immunity (IEI), with increasing numbers of patients treated over the last two decades. Improved quality of care pre-and post-HSCT has led to an improved overall survival rate for patients treated with HSCT, with improved management of early post-HSCT complications and longterm surveillance for earlier detection and management of late effects [1,2]. Late effects such as pulmonary and cardiac complications are well described [3,4] but late effects resulting in bone abnormalities are not well described, particularly in patients with IEI. Osteopenia has been reported among 25-50% of pediatric and adult recipients of stem cell transplant for malignant disease [5][6][7]. Multiple factors contribute to osteopenia and osteoporosis in this cohort including previous chemotherapy/radiotherapy for management of the underlying disease, conditioning chemotherapy, and polypharmacy, in particular the use of steroids for management of graft versus host disease (GvHD) or autoimmunity [8,9].
Osteochondroma has been reported post-HSCT in patients with underlying malignant disorders and was noted in 5-20% of patients who received radiotherapy for malignancy such as neuroblastoma or in the context of total body irradiation (TBI)-based conditioning for HSCT [10][11][12]. Exposure to radiotherapy was considered a significant risk factor for the development of this benign bone tumor [10,13,14].
Great Ormond Street Hospital (GOSH) is one of two supra-regional UK centers delivering HSCT for children with IEI. A long-term follow-up program for children post HSCT is in place, allowing for systematic monitoring and data collection over time. Here, we report all non-osteopenic bone pathology in children with IEI who underwent HSCT in our center between 2000 and 2018.

Patient Characteristics
Records for patients with IEI who underwent allo-HSCT at GOSH between 1/1/2000 to 31/12/2018 were analyzed. Inclusion criteria included patients who were alive at last follow-up (up to 1st of July 2022). Exclusion criteria included patients with reduced bone density, fractures, scoliosis, and short stature without any other underlying non-osteopenic bone pathology or non-osteopenic bone pathology related to underlying IEI. Patient records were checked for non-osteopenic bone pathology and potential risk factors: age, gender, underlying IEI, transplant, intensity of conditioning regimen, stem cell source, onset and duration of steroid therapy, onset of hormonal replacement therapy (HRT) including gonadal hormones, gonadotropinreleasing hormone analogues (GnRHa) and growth hormone (GH), body mass index (BMI) in centiles at onset of bone disease, and donor engraftment at time of onset of bone disease and last follow-up. BMI > 91% was considered overweight. Written informed consent was obtained from all parents.

Statistical Analysis
This is a descriptive analysis of patients who developed non-osteopenic bone pathology. Statistical analysis was performed using SPSS, version 24. Descriptive analyses were performed using median, mean, minimum, and maximum.

Patient Demographics, Disease Category, and HSCT Characteristics of Patients with Non-osteopenic Bone Disease
In the overall cohort, 9.4% of the patients (32/340) developed non-osteopenic bone pathology. Patient demographics, disease category, and HSCT characteristics are described in Supplementary table 1. The mean age at first transplant was 2.74 years (range: 0.39-13.8 years), and there was a male predominance (71.8% (23/32) of the patients). Highest rates of bone pathology were observed in patients with WAS followed by HLH. Patients (10%) underwent HSCT for WAS of which 23.5% (8/34) developed bone pathology, 4.7% of patients underwent HSCT for HLH of which 18.8% (3/16) developed bone pathology suggesting a disproportionate incidence. On the contrary, the rates of bone pathology were less than 10% among the two most frequent diagnoses: SCID and CGD. Furthermore, no bone pathology was recorded among patients with cluster of differentiation 40 ligand (CD40L) deficiency, dedicator of cytokine 8 d (DOCK8) deficiency, immune dysregulation, polyendocrinopathy, enteropathy X-linked (IPEX), or activated PI3K delta syndrome (APDS). Table 1 shows a breakdown of IEI diagnoses for the overall cohort and rates of bone pathology in each IEI.
Three patients received either a second HSCT (n = 2) or DLI (n = 1) for management of graft loss or mixed chimerism. Conditioning regimen included Treo-or Bu-based conditioning in 25 and 2 cases, respectively. Seven patients received a fludarabine/ melphalan (Flu/Mel) conditioning regimen. Only one patient received localized radiotherapy ahead of HSCT for treatment of malignancy (P16). Peripheral blood stem cells (PBSCs) were the main stem cell source in 19 transplants followed by bone marrow (BM) in 10 patients and cord blood in 5 patients. Seventeen (53.1%) of patients received a 10/10 matched unrelated donor (MUD), 1 received a matched family donor (MFD), 14 (43.7%) received 7-9/10 mismatched unrelated donor (MMUD) transplants, and amongst the 2 haploidentical transplants, one received a haploidentical TCR alpha/beta depleted HSCT followed by genetically modified T cell add-back as part of a clinical trial (NCT02065869), while the other received a OKT3 conditioning. Grade I/II acute GvHD (aGvHD) occurred in 50% of patients (16/32), grade III/IV in 5 patients, and 28% of patients (9/32) developed chronic GvHD (cGvHD). Patients (81%; 26/32) received steroid therapy, with 21 (66%) patients receiving prolonged steroid therapy for > 6 months (cumulative pre-and post-HSCT steroid exposure). Twelve patients (37.5%) were overweight (BMI > 91%), and short stature was present in 13 patients (40%) at the time of diagnosis of bone disease. Fourteen patients (43%) had gonadal dysfunction, of which 5 patients received hormonal therapy pre-bone pathology.
Patients (34%; 11/32) had more than one bone pathology. Mean time to developing first bone pathology was 7.8 years post-transplant (range: 2 to 16.2), with a mean follow-up of 10.6 years (range: 4.6-15). Twenty-one patients (65.6%) complained of pain ahead of diagnosis. Table 2 shows detailed description of patients with non-osteopenic bone pathology.

Bone Tumors
While one patient developed a malignant bone disease (P16), 10 patients developed benign bone tumors. At the age of 18 years, he developed a high-grade osteosarcoma in the distal left femur (the same site of previous lymphoma) requiring above knee amputation. Twelve months later, CT chest confirmed localized pulmonary metastases which were surgically resected. Six months later, he developed disease recurrence at the amputation site requiring further resection. He is now 21 years old and receiving palliative care with metastatic lung lesions. He remains 100% donor engrafted.

Evolving Bone Disease
Twelve patients developed other bone pathologies: genu valgum deformity (n = 10), scoliosis (n = 1), and synostosis (n = 1). Genu valgum deformity developed at a median Other evolving bone disease included scoliosis (P5) and proximal radio-ulnar synostosis (P18). P5 received 2 HSCT (Bu-based myeloablative 9/10 cord HSCT followed by Flu/ Mel 9/10 mMUD PBSC HSCT for management of Griscelli Syndrome (2 months after first HSCT)). He developed aGvHD and required steroids for 3 weeks only. He had evidence of gonadal failure and was overweight. He was initially diagnosed with a left fibular osteochondroma 11 years post-HSCT then presented with back pain. Radiology confirmed a diagnosis of scoliosis 12.5 years after HSCT, which has not required corrective surgery. P18 received Bu/Flu MUD BM for CGD. She did not develop GvHD, but she had prolonged steroid use (> 12 months) pre-HSCT. She had associated hypothyroidism and gonadal failure. She presented with inability to rotate her left forearm 6.6 years post-HSCT. Radiology confirmed proximal radio-ulnar synostosis. Both patients had 100% donor engraftment at last follow-up (14 and 10 years, respectively).

SUFE
Six patients (18.75%) had SUFE at a median of 8.5 years post-HSCT (range: 5.4-11 years). Underlying diagnoses included SCID (n = 4; P9, P15, P17, P19) and WAS (n = 2; P6, P22). All six patients received Treo-based conditioning with 4 MUD and 2 MMUD HSCT and BM being the stem cell source for 4/6 patients. P15 received DLI for slipping chimerism. Patients (50%; 3/6) had aGvHD. Three patients received steroid therapy, but P15 had a prolonged course of > 6 months for management of extensive chronic GvHD with nephrotic syndrome (5 years of steroid therapy). Interestingly, all patients had normal BMI apart from P9 who was overweight at time of diagnosis of bone disease. P17 received GnRHa for early puberty, and P22 received GH therapy for short stature. All patients developed SUFE at an age ≤ 10 years and 83% (5/6) presented with bilateral SUFE. All patients required a surgical procedure.

Discussion
Here, we describe the largest cohort of patients with IEI and non-osteopenic bone pathology post-HSCT with follow-up extending to 22 years. Regular follow-up of these patients has demonstrated that non-osteopenic bone pathology is not rare and is present in 9.4% of our cohort of patients. Mean time to develop non-osteopenic bone pathology was 7.8 years with one-third of our cohort developing more than one bone pathology at different time points and more than half diagnosed with bilateral disease. Although the majority of patients (81%) received steroids peri-HSCT with 66% patients (21/32) receiving prolonged steroids for > 6 months (cumulative pre-and post-HSCT steroid exposure), nonosteopenic bone pathology presented years after steroid exposure, and a fifth of patients developed non-osteopenic bone pathology despite no steroid exposure. Pain was the main presenting feature (69% of patients), and 43% of patients required surgical intervention at the time of analysis, with more awaiting procedures. Chronic pain and the need for (repeat) orthopedic surgical interventions have a huge impact on quality of life of a growing child [16].
Specific bone pathology such as the development of osteochondroma has been associated with the use of radiotherapy in malignancies [11]. None of our 10 children who developed osteochondroma underwent radiation as part of their treatment or conditioning regimen, pointing to other factors that might have contributed to the development of bone tumors. Some of the risk factors previously identified in the malignant HSCT setting [12,14] were present in our cohort, such as age at HSCT < 3 years (100% of our osteochondroma cohort respectively) and male gender (90%), whereas other reported risk factors such as use of growth hormone replacement therapy, or busulfan-based conditioning regime were not found in our osteochondroma cohort. We reported 3% AVN in our cohort, with more than half of patients developing bilateral AVN. Previous groups have reported a similar occurrence rate between 2 and 15% of patients post-HSCT depending on donor choice (increased incidence in MUD HSCT) and increased/prolonged exposure to steroids, ciclosporin, and mycophenolate mofetil [6,[17][18][19][20]. In line with these known risk factors, most patients developing AVN in our cohort had increased steroid exposure and an unrelated donor transplant. Of note, while almost all our patients had received a prolonged course of steroids of > 6 months before developing AVN, 2 patients (P13, P19) developed AVN after only receiving a short steroid course (≤ 1 month). Active surveillance is needed even in patients with limited exposure to steroids. Genu valgum deformity was seen among 3% of our patients and was associated with restricted mobility. Moreover, 40% of our patients have required surgical intervention. As previously reported [21], we noted a possible association between genu valgum deformity and obesity with 40% of patients overweight. Early lifestyle intervention for patients with increasing BMI is required to limit the risk for diabetes, hypertension [22], and bone pathology. Although being overweight is a known risk factor for SUFE [23], interestingly, none of the patients with SUFE in our cohort were overweight at time of diagnosis apart from one and all of our patients developed SUFE at a younger age (≤ 10 years) compared to previous reports that SUFE is an adolescent disease associated with obesity [23,24]. All patients with SUFE in our cohort required surgical intervention with the majority (83%) having bilateral SUFE.
We identified 6 patients with skeletal dysplasia post-HSCT, 5 of which had either diaphyseal dysplasia (n = 1) or similar radiographical finding of spondyloepimetaphyseal dysplasia as described by Botto et al. All had stunted growth, and all underwent HSCT < 2 years of age. In contrast to their findings, we did not identify a specific predisposition for skeletal dysplasia in HLH patients post-HSCT. Risk factors for the development of skeletal dysplasia post-HSCT in IEI could include specific exposure to peri-HSCT chemotherapy at an early age (including cumulative steroid exposure pre and post) impacting on bone development and growth. We would recommend that all children presenting with stunted growth post-HSCT to undergo radiological skeletal survey to rule out the development of skeletal dysplasia.
Our results showed increased rates of non-osteopenic bone pathologies among patients with WAS and HLH: 23.5% and 18.8% respectively, compared to ≤ 10% overall. Patients with WAS developed multiple forms of bone pathologies including osteochondroma, genu valgum, SUFE, AVN, and skeletal dysplasia. We could not attribute this to (prolonged) steroid usage alone. Other IEI requiring prolonged steroid therapy, such as patients with CGD [25], had low frequency of bone disease (2.3%) despite being the 2nd most frequent indication for HSCT. Other factors could be responsible for WAS patients' increased propensity to develop bone disease, in particular osteochondroma. WAS protein (WASp) is a key regulator of the actin cytoskeleton in hematopoietic cells [26] and in osteoclasts. The action of WASp is required for the organization of actin which is essential for osteoclast activity and bone resorption as demonstrated in vivo and in vitro studies [27][28][29]. Interestingly, Caffey disease, another bone disease associated with hyperostosis and massive subperiosteal new bone formation, has been reported among patients with WAS, independent of HSCT [30,31]. Bone pathology found in WAS patients pre-and post-HSCT might point to a specific WASrelated etiology causing defective osteoclast activity. Aside from the insults of chemotherapy and an intrinsic defect in the osteoclast cytoskeleton, there may be a pathophysiological predisposition of long-term survivors with a genetic defect in the WAS gene to have a higher incidence of various non-osteopenic bone disease, and thus, this cohort of patients should be actively followed up for bone abnormalities.

Conclusion
Non-osteopenic bone pathology in long-term survivors of allo-HSCT for IEI is not rare. Most patients did not present with complaints until at least 5 years post-HSCT highlighting the need for ongoing bone health assessment. We recommend annual structured musculoskeletal examination, with targeted imaging for reported joint/bone pain, abnormalities in gait/movement, bone deformity, or short stature to allow for early detection and management of non-osteopenic bone pathology. Further studies are also required to evaluate osteopenic bone disease among this cohort of patients, better understand the mechanism and risks factors for both non-osteopenic and osteopenic bone pathology in specific IEI context, which would require a large prospective study with international collaborators. Author Contribution RE and MK conceptualized the study; ZMG did the data collection; RE, MK, and ZMG wrote the first draft; TS, JB, SA, CB, AW, and WQ contributed to patient care and reviewing and finalizing the draft.
Data Availability Data and material are available.

Declarations
Ethics Approval Research was ethically approved.

Consent to Participation
Informed consent was taken from parents/ legal guardians together from patients as appropriate.

Consent for Publication
Informed consent was taken from parents/legal guardians together from patients as appropriate.

Conflict of Interest The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.