Hematopoietic stem cell Transplantation in Children with very Early Onset Inflammatory Bowel Disease Secondary to Monogenic Disorders of immune-dysregulation

Abstract

Background: Very early-onset inflammatory bowel disease (VEOIBD) is defined as IBD in children under six years of age. We present outcome data of hematopoietic stem cell transplantation (HSCT) in the above children. Patients and methods: We performed a retrospective study in children under six years of age who underwent HSCT for VEOIBD with an identified monogenic disorder from December 2012 to December 2020. Results: Of the 25 children included, the underlying diagnosis was IL10R deficiency (n = 4), Wiskott-Aldrich syndrome (n = 4), Leukocyte adhesion defect (n = 4), Hyper IgM syndrome (n = 3), Chronic granulomatous disease (n = 2), and one each with XIAP deficiency, severe congenital neutropenia, Omenn syndrome, Hyper IgE syndrome, Griscelli syndrome, MHC Class II deficiency, LRBA deficiency, and IPEX syndrome. Donors included a matched family donor in 10(40%); a matched unrelated donor in 8 (32%), haploidentical in 7 (28%) (T depleted 16%, T replete with post-transplant cyclophosphamide12%). Conditioning was myeloablative in 84% ofHSCTs. We documented engraftment in 22 (88%) children, primary graft failure in 2 children (8%), mixed chimerism in 6 (24%) children with mortality in 4/6 children. Children with a sustained chimerism of > 95% did not have recurrence of any features of IBD. Overall survival was 64%, with a median follow-up of 55 months. Mixed chimerism was associated with a significantly increased risk of mortality (p-value = 0.001). Conclusions: VEOIBD caused by monogenic disorders can be offered HSCT. Early recognition, optimal supportive care, and complete chimerism are essential components to achieving survival.

Introduction

Inflammatory bowel disease (IBD) results from dysregulated intestinal immune responses in genetically susceptible patients [1, 2]. In most patients, IBD displays polygenic inheritance. However, recent advances in next-generation sequencing have revealed many cases of childhood IBD that result from monogenic inheritance in the context of inborn errors of immunity (IEI) [3]. Children under six years of age who develop IBD are categorized as having very early-onset inflammatory bowel disease (VEOIBD). Disease development prior to age two is termed the infantile subtype while neonatal disease is diagnosed in patients less than 28 days [4].

The high rate of consanguinity in India leads to an increased rate of VEOIBD, which has a reported incidence of 19.2% [5]. The complex pathophysiology and ensuing immune dysregulation in VEOIBD often result in a challenging clinical course [2]. Here we discuss a series of children with underlying monogenic IEI presenting as VEOIBD who underwent HSCT at our centre.

Patients and Methods

We performed a retrospective study, where inclusion criteria were children aged less than six years with a diagnosis of VEOIBD and a genetically proven underlying monogenic IEI, and who underwent HSCT from December 2012 to December 2020.Published guidelines and criteria were used for diagnosing VEOIBD [6, 7], and all children fulfilling the criteria underwent whole-exome sequencing. Exclusion criteria included children more than six years of age with polygenic IBD or without an identified monogenic disorder. The hospital ethics committee approved the study, and all families signed a detailed informed consent.

High resolution HLA typing was performed for the child, siblings, and parents. In case of non-availability of a fully matched family donor (MFD) and based upon the mode of inheritance of the disorder, we performed unrelated searches in all registries. A haploidentical family donor was considered in case of unavailability of a matched unrelated donor (MUD). All haplo-matched donors were screened for donor-specific anti-HLA antibodies. Peripheral blood stem cells (PBSC) was the preferred stem cell source of stem cells in view of earlier engraftment, thereby reducing the duration of neutropenia and resultant risk of sepsis and decreased risk of graft rejection as compared to bone marrow stem cells. Bone marrow stem cells were harvested in case of donor weight less than 15 kg or in case of any contraindication to PBSC collection.

The chemotherapy used in conditioning was predominantly myeloablative, particularly in conditions with underlying immune dysregulation. Myeloablative conditioning (MAC) included Thiotepa/Treosulphan/Fludarabine or Fludarabine/Busulphan in the case of matched related donors. Anti-thymocyte globulin (ATG) was added in the case of MUD donors. Haploidentical HSCT with post-transplant cyclophosphamide (PTCy) at a dose of 50 milligram/kilogram (mg/kg) on Day + 3 and + 4, included conditioning with Fludarabine/Melphalan. Children with underlying severe combined immune deficiency underwent HSCT with reduced-intensity conditioning (RIC), including Fludarabine/Treosulphan. When TCR alpha beta depletion was employed, conditioning was myeloablative with Fludarabine/Treosulphan/Thiotepa/ATG. GVHD prophylaxis included tacrolimus, and methotrexate on day + 1, 3, 6 for bone marrow and day + 1, 3, 6, 11 for peripheral blood stem cells grafts. A short tandem repeat sequence by a polymerase chain reaction was used in case of same-sex donor for assessment of donor chimerism and Fluorescence in situ hybridization in case of sex mismatched HSCT. Donor chimerism was assessed on engraftment around D + 17, followed by D + 30, + 60, +90, 6 months and 1 year post HSCT. The National Institute of Health (NIH) Consensus guidelines were utilized for scoring GVHD [8].

Statistical Analysis

Data were analysed using the IBM SPSS version 25.0 (SPSS, Inc., Chicago, IL). Continuous variables were represented as mean ± S.D and the non-normal data as median + interquartile range. The categorical data was represented as percentages. The comparison of normally distributed and non-normally distributed continuous variables was done using the independent sample t-test and Mann – Whitney U test. Categorical variables were compared using the Chi-squared or Fisher’s exact test. The statistical analysis was carried out at a 95% confidence interval (CI). P value < 0.05 was considered statistically significant.

Results

Patient demographics

A total of 25 children with VEOIBD underwent HSCT at our centre. The age ranged from 1 to 93 months with a median of 17 months. The male: female ratio was 2:1 with a median follow-up of 55 months. The underlying diagnosis was IL10R deficiency in 4 (16%), Wiskott-Aldrich syndrome (WAS) in4 (16%), Leukocyte adhesion defect (LAD) in 4 (16%), Hyper IgM syndrome in 3 (12%), Chronic granulomatous disease (CGD) in 2 (8%), and one (4%) each with X-linked inhibitor of apoptosis protein (XIAP)deficiency, severe congenital neutropenia with ELANE mutation, Omenn syndrome, Hyper IgE syndrome with DOCK8 mutation, Griscelli syndrome, MHC Class II deficiency, lipopolysaccharide-responsive and beige-like anchor protein (LRBA) deficiency, and immune dysregulation, polyendocrinopathy, enteropathy, X-linked (IPEX) syndrome. In addition, VEOIBD was of the neonatal type in 4 (16%), childhood type in 3 (12%), and infantile type in 18 (72%) children.

Pre-HSCT events

All children had features suggestive of IBD at diagnosis, including recurrent episodes of loose stools, recurrent gastrointestinal bleeding, and diffuse abdominal pain. All children had failure to thrive with weight for age below the third centile. The child with LRBA deficiency received Abatacept till the start of conditioning. The child with IPEX syndrome was on multiple hormone replacements, including thyroid supplements and hydrocortisone which were continued. The co-morbidities documented before HSCT included pneumonia in 9 (36%), perianal ulcers in 8 (32%), meningitis with Klebsiella pneumoniae in 1 (4%) child, and multiple endocrinopathies in the child with IPEX syndrome.

Graft source and conditioning.

Types of donors were matched family donor in 10 (40%), MUD in 8 (32%), T depleted haploidentical HSCT with TCR alpha/beta depletion in 4 (16%), and T replete haploidentical HSCT with PTCy in 3 (12%) children. Peripheral blood stem cells were the predominant source of stem cells in 18 (72%) HSCTs, with bone marrow in 4 (16%) and umbilical cord blood unit in 2 (8%). Conditioning regimen was MAC in 21 (84%) and RIC in 4 (16%) HSCTs. The median CD34 dose for the matched family and matched unrelated donor HSCTs was 5 × 10⁶/kg recipient body weight. In the T cell depleted haploidentical HSCTs, the median CD34 dose was 10 × 10⁶/kg recipient body weight. In the T cell replete haploidentical HSCTs, CD34 dose was infused at a median of 7 × 10⁶/kg, while maintaining a CD3 cell dose above 1.5 × 10⁸/kg recipient body weight.

Engraftment, chimerism and graft failure

Engraftment by D + 14 to + 21 was documented in 22 (88%) children. One child with Omenn syndrome died before engraftment due to carbapenem-resistant gram-negative sepsis. We confirmed primary graft failure in 2 children (8%); one child with Wiskott-Aldrich syndrome undergoing a haplo-HSCT with PTCy and another child with IL10R deficiency who underwent a T cell depleted haplo-HSCT.

Mixed chimerism with a drop in donor chimerism after D + 45 was documented in 6 (24%) children. The underlying diagnosis in the above six children included IL10R deficiency in 2, and one each with WAS, MHC Class II deficiency, LAD and DOCK8 deficiency. Three children, each with LAD, DOCK8 deficiency, and IL10R deficiency, succumbed to sepsis following drop in chimerism. In the child with LAD, a drop in chimerism to 50% was documented with the septic event; while in the child with DOCK8 deficiency, recurrence of loose, frequent stools with streaks of blood correlated with a chimerism of 77%. He improved transiently with supportive care, however, succumbed to a septic event. In the child with IL10R deficiency, drop in chimerism to 60% correlated with rapid deterioration, with frequent blood-stained loose stools, metabolic acidosis and eventual decompensated septic shock.

One child with IL10R deficiency was stable with mixed chimerism following four whole blood donor lymphocyte infusions. Donor chimerism was detected to be 97% at 3 months post HSCT. The child was clinically stable, and hence was on follow up. Subsequently, chimerism dropped to 93.4% at 5 months post HSCT, and first DLI at a dose of 1 × 10⁵/kg recipient body weight of donor T cells were infused. Further whole blood DLIs at graded doses were performed at chimerism levels of 75% (6 months), 68.5% (7 months) and 45% (9 months). Donor chimerism checked at 11 months post HSCT was 95% and child was clinically well.

One child with WAS underwent a haplo-HSCT following secondary graft failure of a MUD graft. The conditioning regimen during the first HSCT included fludarabine/busulphan/anti-thymocyte globulin. Conditioning regimen for the second HSCT included fludarabine/melphalan/PTCy. This patient did not engraft and family opted not to attempt a third transplant and continued best supportive care. The baby succumbed to the illness.

One child with MHC class II deficiency underwent a successful second matched sibling donor HSCT with a MAC regimen including fludarabine/treosulphan/thiotepa and the same donor. The first HSCT was conditioned with a reduced intensity regimen including fludarabine and treosulphan.

Supportive care

All children had documented hypokalemia and hypomagnesemia and were on regular corrections. All children were on parenteral nutrition with 50% dextrose, amino acids, and SMOF lipid infusions, and an elemental diet was allowed orally. In addition, we used continuous nasogastric feeds with semi-elemental feeds for all four children with IL10R deficiency. All children were on antifungal prophylaxis with micafungin or voriconazole based on the underlying condition and prior documented fungal infection, and antiviral prophylaxis with acyclovir.

Post HSCT complications

We documented neutropenic enterocolitis in 16 (64%) children, with Acinetobacterbaumanii and Klebsiellapneumoniae isolated in 2 (4%) children each. In addition, hematochezia during the neutropenic period was noted in the child with LAD type III. Seizures and posterior reversible encephalopathy syndrome (PRES) were documented in 2 (4%) children. All children recovered from the above complications except one with Omenn syndrome, who died of Acinetobacter sepsis before engraftment. Two children (4%) with Hyper IgM syndrome had documented Cryptosporidium parvum in stool and responded to nitazoxanide.

The cumulative incidence of acute GVHD at D + 100 was 4/25 (16%). Acute grade IV gut GVHD was recorded in two children with CGD. Both the children were treated with multiple lines of treatment including methylprednisolone, ruxolitinib and mycophenolate mofetil. Additionally, one child received etanercept. Both the children succumbed to grade IV gut GVHD. Grade I/II skin GVHD was documented in two children, which responded to steroids. In both these children, dose of calcineurin inhibitors were optimized and topical steroid ointment was also used.

Chronic mouth, skin, and eye (NIH score 2) were noted in 2 (4%) children. Calcineurin inhibitors were continued and both children responded to prednisolone continued for over 18 months at a dose of 1 mg/kg/day. They tolerated gradual steroid taper once clinical improvement was noted.

We documented cytomegalovirus reactivation in 6 (24%) children. Two children were diagnosed with Leukocyte adhesion defect, and one each with CGD, WAS, Hyper IgM syndrome and XIAP deficiency. Cytomegalovirus reactivation with a maximum viral copy number up to 5000 was detected after D + 30 post HSCT in all children and responded to 14 days of ganciclovir at a dose of 5 mg/kg/dose twice a day in all children. None of the above children had a drop in chimerism in relation to viral reactivation.

Survival and mortality

Overall survival in the cohort was 64%, with a median follow-up of 55 months (Fig. 1). Among the 9 (36%) children who died, the underlying diagnosis was CGD in 2, WAS in 2, IL10R deficiency in 2, and one each with LAD, DOCK8 deficiency, and Omenn syndrome. Causes of death included primary graft failure in 2 (4%), secondary graft failure in 4 (8%), grade IV gut GVHD in 2 (4%) and pre-engraftment gram-negative sepsis in 1 (2%). Mixed chimerism was associated with a significantly increased risk of mortality (p-value = 0.001).

Fig. 1.

Kaplan Meier survival curve depicting overall survival of 64% with a median follow up of 55 months

Follow-up

Children with chimerism > 95%, sustained after 6 months post HSCT, had normal formed stools without any evidence of blood, and did not have recurrence of any clinical features based on validated IBD disease scoring systems.

The patient, disease, and HSCT details have been represented in Table 1.

Table 1.

Patients, disease and HSCT details of children included in the cohort n = 25

Patient number	Diagnosis	Type of IBD	Age at the time of HSCT	Conditioning regimen	Source of stem cells	Donor type	Engraftment/ Chimerism	Follow up/ Months since HSCT
Patient 1	LAD 1	Infantile	9 months	FLU, TREO	PBSC	MFD	Engrafted, MC, late graft rejection	Died/ 7 months
Patient 2	CGD	Infantile	2-year 1 month	FLU, BU, ATG	PBSC	MUD	Engrafted, CC	Died of grade IV GVHD/ 2 months
Patient 3	LAD 3	Neonatal	2 year 5 months	FLU, TREO	PBSC	MFD	Engrafted. CC	Alive/ 41 months
Patient 4	XIAP deficiency	Neonatal	1 month	TT, FLU, TREO, ATG	PBSC	Haplo HSCT with TCR alpha/beta depletion	Engrafted, CC	Alive/ 30 months
Patient 5	LAD 1	Neonatal	6 months	TT, FLU, TREO, ATG	PBSC	Haplo HSCT with TCR alpha/beta depletion	Engrafted, CC	Alive/ 29 months
Patient 6	CGD	Infantile	17 months	FLU, MEL	PBSC	Haplo HSCT with PTCy	Engrafted, CC	Died of grade IV GVHD/ 4months
Patient 7	Severe congenital neutropenia with ELANE mutation	Childhood	10years	TT, FLU, TREO	PBSC	MFD	Engrafted, CC	Alive/ 22 months
Patient 8	LAD 2	Neonatal	4 months	TT, FLU, TREO, ATG	PBSC	MUD	Engrafted, CC	Alive/ 11 months
Patient 9	Hyper IgM syndrome	Childhood	5 years	FLU, BU	Bone marrow	MSD	Engrafted, CC	Alive/ 89 months
Patient 10	Omenn syndrome	Infantile	5 months	FLU, TREO	Bone marrow	MSD	Died before engraftment	Died of sepsis/ D + 8
Patient 11	WAS	Infantile	16 months	FLU, MEL	PBSC	Haplo HSCT with PTCy	Engrafted, CC	Alive/ 72 months
Patient 12	Hyper IgM syndrome	Infantile	5 years	FLU, BU, ATG	PBSC	MUD	Engrafted, CC	Alive/ 86 months
Patient 13	Hyper IgE syndrome with DOCK8 mutation	Infantile	12 months	FLU, TREO	Bone marrow	MSD	Engrafted, MC	Died of sepsis/ 30 months
Patient 14	WAS	Infantile	2 years 9 months	FLU, BU, ATG	PBSC	MUD	Engrafted, CC	Alive/ 56 months
Patient 15	Hyper IgM syndrome	Childhood	5 years	TT, FLU, TREO	PBSC	MSD	Engrafted, CC	Alive/ 53 months
Patient 16	WAS	Infantile	2 years 1 month	FLU, BU, ATG (1st HSCT) FLU, MEL (2nd HSCT)	PBSC PBSC	MUD Haplo HSCT with PTCy	Engrafted, MC Primary graft failure	Alive Died of sepsis/ D + 32
Patient 17	LRBA deficiency	Infantile	21 months	TT, FLU, TREO	Bone marrow	MSD	Engrafted, CC	Alive/ 28 months
Patient 18	MHC class II defect	Infantile	2 years 8 months	FLU, TREO (1st HSCT) TT, FLU, TREO (2nd HSCT)	PBSC	MFD	MC, Rejection Engrafted, CC	Alive Alive/ 29 months
Patient 19	WAS	Infantile	2 years 8 months	FLU, BU	PBSC	Haplo HSCT with PTCy	Primary graft failure	Died of sepsis/ D + 35
Patient 20	IPEX syndrome	Infantile	17months	FLU, BU, ATG	Umbilical cord blood unit	MUD	Engrafted, CC	Alive/ 92 months
Patient 21	Griscelli syndrome	Infantile	23 months	FLU, TREO, ATG	Umbilical cord blood unit	MUD	Engrafted, CC	Alive/ 88 months
Patient 22	IL 10R deficiency	Infantile	12 months	TT, FLU, TREO, ATG	PBSC	Haplo HSCT with TCR alpha/beta depletion	Engrafted, CC	Alive/ 53 months
Patient 23	IL 10R deficiency	Infantile	2 years	TT, FLU, TREO, ATG	PBSC	MUD	Engrafted, MC	Died of sepsis/ 4 months
Patient 24	IL 10R deficiency	Infantile	8 months	TT, FLU, TREO, ATG	PBSC	Haplo HSCT with TCR alpha/beta depletion	Primary graft failure	Died of sepsis/ D + 30
Patient 25	IL 10Rdeficiency	Infantile	10 months	TT, FLU, TREO	PBSC	MFD	Engrafted, MC	Alive/ 11 months

HSCT = hematopoietic stem cell transplantation, TT = thiotepa, TREO = treosulphan, FLU = fludarabine, BU = busulphan, ATG = anti-thymocyte globulin, CC = complete chimerism, MC = mixed chimerism, MSD = matched sibling donor, MFD = matched family donor, MUD = matched unrelated donor, Haplo HSCT = haploidentical HSCT, PBSC = peripheral blood stem cells, GVHD = graft versus host disease, LAD = leukocyte adhesion defect, CGD = chronic granulomatous disease, WAS = Wiskott-Aldrich syndrome, IPEX = immune dysregulation, polyendocrinopathy, enteropathy, X-linked syndrome, XIAP = X-linked inhibitor of apoptosis protein, LRBA = lipopolysaccharide-responsive and beige-like anchor protein

Discussion

The study reports outcome data of HSCT in children with monogenic VEOIBD, with overall survival of 64%. Mixed chimerism was documented in 24% of children, which was significantly associated with a poor outcome (p value = 0.001). Complete resolution of IBD was observed in children with sustained complete donor chimerism.

Over 50 genes have been implicated in VEOIBD [9]. In addition, a recent publication has identified a novel exon deletion in the IL10RA gene, thus expanding the diagnostic pool [10]. Few groups have published data on HSCT in VEOIBD, including XIAP mutation and IL10R defects [11–13].

Myeloablative conditioning regimen was reported to be superior to RIC regimens [14, 15]. Worth et al. reported HSCT outcomes with a reduced toxicity conditioning with alemtuzumab, fludarabine, and treosulphan/melphalan in children with intractable enteropathy. In the above study, 26 children underwent 27 HSCTs, where 19 patients are alive at a median follow up of 51 months, and sustained improvement in enteropathy was noted in 17 of these children [16]. Rao et al. reported successful HSCT in IPEX syndrome with a resolution of symptoms in 4 patients with an alemtuzumab-based RIC regimen [17]. We used a MAC regimen in all patients in our cohort except for those with severe combined immune deficiency.

Early HSCT is crucial to alleviate IBD features and improve extraintestinal manifestations in children with monogenic VEOIBD. Patient selection is key to an optimal outcome. Lekbua et al. have identified colitis at an early age, failure to thrive, and organomegaly as risk factors for poor outcomes in children with VEOIBD and XIAP mutation and recommend early HSCT in this sub-group of patients [18]. Children with XIAP deficiency can continue to have persistent IBD and autoimmune features post a successful HSCT as well [19, 20].

Pre HSCT stabilization in conjunction with pediatric gastroenterologists is essential. Exclusive enteral nutrition and amino acid-based elemental formula have been reported as first-line treatments for VEOIBD [21, 22]. In addition, several targeted therapies have been studied as a bridge to allow HSCT in children with VEOIBD and inborn errors of immunity (IEI). Bakhtiar et al. have highlighted the need for targeted therapies prior to HSCT in patients with PIDD and immune dysregulation, in view of worse outcomes when patients underwent HSCT with advanced stages of poorly controlled infections and autoimmunity [23]. Abatacept is an effective CTLA4 agonist in children with CTLA4 and LRBA deficiencies [24]. Shouval et al. demonstrated that one of the pathogeneses of intestinal inflammation in IL10R defects includes loss of IL10 signalling through increased production of IL1 and the subsequent activation of CD4 + T cells [25]. Anti-IL1 therapy, including anakinra and canakinumab, has been used in sick children with IL10R defects. Perianal disease has been reported to be common in VEOIBD, particularly with IL10- and IL10- receptor signalling defects [26, 27]. Surgical procedures, particularly diverting colostomies, have been used in children with IBD [27, 28]. In our cohort, the child with LRBA deficiency was on Abatacept until HSCT, which has been stopped since. One child with an IL10R defect had undergone a colostomy before diagnosis in view of perianal disease, and was taken up for HSCT with the stoma.

Conclusion

VEOIBD caused by monogenic disorders can be offered HSCT. Children with complete chimerism of > 95% did not have any IBD related symptoms post-HSCT, and only those with mixed chimerism manifested gastrointestinal symptoms. Early recognition and optimal supportive care are essential components to achieving survival. Understanding the pathophysiology of these complex disorders would likely improve disease-free survival further.

Abbreviations

VEOIBD

Very early onset inflammatory bowel disease

HSCT

Hematopoietic stem cell transplantation

GVHD

Graft versus host disease

MUD

Matched unrelated donor

MFD

Matched family donor

MAC

Myeloablative conditioning

RIC

Reduced intensity conditioning

CMV

Cytomegalovirus

EBV

Epstein Barr virus

ATG

Anti thymocyte globulin

IEI

Inborn errors of immunity

Authors’ contributions

SKM, RU, RR conceptualized the design of the study and wrote the manuscript; VVS, RC, HV helped with data collection and data entry; IJ helped with proof reading; BR did the statistical analysis.

Funding

None.

Declarations

Conflict of Interest

None.

Conflict of Interest

There are no conflicts of interest.