Celiac disease (CD), the most common chronic enteropathy, is caused by breakdown of peripheral tolerance toward wheat gluten proteins and related prolamins, whose elimination from the diet leads to healing of intestinal lesions (1). The relevance of T cells in promoting mucosal injury was further supported by a report on the adoptive transfer of CD from an affected donor to the recipient after unmanipulated allogeneic hematopoietic stem cell transplantation (HSCT) (2). Conversely, resolution of CD has been previously described in 2 patients given allogeneic HSCT from an human leukocyte antigen (HLA)-identical sibling for concomitant hematological diseases (3,4). Here, we report on 2 additional patients who experienced full recovery from CD after receiving HSCT following a myeloablative conditioning regimen for β-thalassemia major. Moreover, with the aim of obtaining insights into the development of gluten tolerance, we performed an extensive immunological study, including characterization of dendritic cells (DC), regulatory T cells, cytokine profile, and antigen-specific T-cell response.
A boy (patient 1), born from nonconsanguineous parents known to be carriers of thalassemia trait, was prenatally diagnosed to have β-thalassemia major. Iron chelation therapy and regular transfusions were started at 1 year of age. One year later, following the appearance of dental enamel hypoplasia, the patient underwent serological screening for CD by searching for anti-endomysium antibodies (EMA), which turned out to be positive, and then endoscopic duodenal biopsy that confirmed the presence of the characteristic mucosal lesions (Fig. 1A) (5). A strict gluten-free diet (GFD) was then established. At 4 years of age, the patient underwent HSCT from an HLA-compatible unrelated donor after a myeloablative conditioning regimen including busulfan, thiotepa, and fludarabine. Graft-versus-host disease (GvHD) prophylaxis consisted of cyclosporine-A, short-term methotrexate, and pretransplant antithymocyte globulin. The patient engrafted and recovered normal cell blood counts, with transfusion independence and full donor chimerism. He did not develop acute or chronic GvHD; cyclosporine-A was withdrawn 1 year after HSCT. A strict GFD was maintained throughout the period of the immunosuppressive therapy; then, both serology for EMA and duodenal biopsy were performed and gave normal results (Fig. 1B). Hypothesizing that HSCT may have also cured CD, 1 year after HSCT, gluten was introduced in the diet with monitoring of clinical conditions and serological markers scheduled every 6 months and duodenal biopsy scheduled after 1 and 2 years. In detail, the patient introduced >20 g of gluten per day.
Patient 2 refers to a girl who received the diagnosis of β-thalassemia major at 9 months of age. She was given regular transfusional support and iron-chelation therapy. During childhood, a progressive weight and height growth decline became evident. At the age of 14, she experienced abdominal discomfort and episodes of cramps and diarrhea. Serological screening for CD resulted positive, and endoscopic duodenal biopsy showed the characteristic mucosal lesions (5). A strict GFD was started leading to the disappearance of abdominal symptoms and serologic hallmarks. At 17 years of age, she underwent HSCT from an HLA-identical sibling, after the same myeloablative regimen of patient 1. Full, stable donor-recipient chimerism was obtained and the patient recovered normal blood cell counts with transfusion independence. Neither acute nor chronic GvHD developed; cyclosporine-A was discontinued 1 year after HSCT. One year after the allograft, EMA test and duodenal biopsy were performed and both proved consistent with CD under GFD. Following the procedure adopted for patient 1, a gluten-containing diet (GCD) (>20 g/day) was then introduced, and the same strict clinical and diagnostic monitoring was performed.
During the 5 years of follow-up after HSCT, a permanent negativity of all clinical, serological, and histological (Fig. 1C for patient 1) features of CD was evident in both cases despite gluten exposure. Specifically, a complete recovery from the conditions that led to the diagnosis of CD, that is, tooth enamel hypoplasia for case 1 and intestinal symptoms for patient 2, was observed together with a return to a fully active life.
BIOLOGICAL SAMPLES AND IMMUNOLOGICAL STUDIES
For the immunological study, we also enrolled 11 active patients with CD (F/M 8/3, mean age 37 years, range 19–48), 8 of which also after a course of GFD with proved mucosal recovery, and 9 dyspeptic patients (F/M 6/3, mean age 38.5 years, range 22–56) with normal duodenal mucosa at histology. The study was approved by the local bioethics committee (protocol n. 20110009459), and both patients (the parents for patient 1) and controls signed the informed consent.
To characterize circulating immune cells, peripheral blood samples were collected from the 2 patients after HSCT during both GFD and GCD regimen, from patient 1 also before transplantation, and from all control patients on the day of endoscopy. DC were detected at flow cytometry as HLA-DR+/lineage− cells, and distinguished into 2 main subsets: myeloid-DC (mDC, CD11c+) and plasmacytoid-DC (pDC, CD123+), as previously reported (6). Specific antibodies against blood DC antigens (Miltenyi Biotec GmbH; Bergisch Gladbach, Germany), namely BDCA-1 (CD11chigh/CD123low, mDC) and BDCA-2 (CD11c−/CD123high, pDC), were also used (7) In both patients, after HSCT, the absolute number of circulating DC during both GFD and GCD was higher in comparison with the value found in patient 1 before HSCT at time of active CD (Fig. 2A, left panel). A similar increase in both plasmacytoid and monocytoid DC was evident (Fig. 2A). Furthermore, comparable results were found when analyzing BDCA-1 and BDCA-2 subsets (Fig. 2B). A reduction in the number of circulating DC, mainly the plasmacytoid and BDCA-2 subsets (Fig. 2), was found in both untreated and treated CD controls in comparison with dyspeptic patients.
Intracellular staining for the transcription factor forkhead box P3 (FoxP3, Bioscience, San Diego, CA) was carried out on peripheral blood mononuclear cells as previously reported (8), whereas mucosal FoxP3+ T cells were evaluated at immunohistochemistry by using monoclonal antibodies against FoxP3 (Clone 236A/E7, Bioscience) and CD25 (Clone 4C9, Novocastra, Newcastle, UK) on serial sections. The percentage of circulating FoxP3+ T cells observed in patient 1 before HSCT was higher than the values found in both patients after HSCT under either GFD or GCD (Fig. 3A). The percentage of regulatory T cells before HSCT was similar to that of both untreated and treated CD, whereas after HSCT, the percentages were similar to those of dyspeptic patients (Fig. 3A). The same trend was observed at mucosal level (Fig. 3B).
Detection of intracellular cytokines in peripheral blood T cells was performed following manufacturer's instructions (Becton Dickinson, San Jose, CA). As shown in Table 1, we found a low percentage of cytokine-producing T cells in peripheral blood of the 2 patients after transplantation during both the diet regimens with respect to the values found in patient 1 before HSCT and in untreated CD control patients. The cytokine profile was also studied on frozen mucosal samples obtained from patient 1 before and after HSCT, from patient 2 only after HSCT, and from all CD control patients by means of real-time reverse transcription polymerase chain reaction assay (ABIPRISM 7500; Applied Biosystems, Life Technologies, Monza, Italy). In the 2 patients after HSCT, the mucosal profile for all the cytokines investigated showed lower transcript levels as compared with untreated CD controls and patient 1 before HSCT (Fig. 4), with the remarkable exception of transforming growth factor (TGF)-β.
Finally, fresh mucosal samples were also collected to generate gliadin-specific T-cell lines to be tested for antigen reactivity by using proliferation assay (9). A positive T-cell response was defined as a stimulation index (SI) ≥3, calculated by the ratio of mean counts per minute of T cells cultured in the presence of feeder cells plus antigen over the mean counts per minute of T cells alone. The lack of proliferative response upon gliadin stimulation was found when using T cells isolated from duodenal mucosa of both patients after HSCT, during GFD (mean SI 1.5 ± 0.6) or GCD (mean SI 1.2 ± 0.5) regimens. By contrary, a positive response was found when using T cells obtained from mucosal samples of both active and treated CD controls (mean SI 4.1 ± 1.5 and 3.8 ± 1.6, respectively).
In the last decade, HSCT has emerged as an effective treatment to be proposed to those patients with chronic, severe immune-mediated diseases refractory to conventional therapies (10). As regards CD, HSCT (both autologous and allogeneic) has been used in few severely affected patients, namely those with refractory disease or enteropathy-associated T-cell lymphoma, with unconvincing results (11–13). Conversely, 2 patients were reported to be successfully cured of noncomplicated CD after allogeneic HSCT performed for a concomitant hematological disease (3,4); however, duodenal biopsy showing the recovery of normal mucosa after reintroduction of gluten in the diet was performed only in 1 patient, and after a short follow-up period (7 months) (4). According to the criteria of the European Society for Paediatric Gastroenterology and Nutrition whereby most patients with CD have a histological relapse within 2 years after reintroduction of gluten in the diet (14), both of our patients underwent duodenal biopsy 1 and 2 years after the reintroduction of gluten in the diet together with a strict serological monitoring throughout the 5-year follow-up period, which is by far longer than that of the 2 cases previously published (3,4). Considering that our patients had a regenerating immune system, the execution of a duodenal biopsy at 5-year follow-up would have provided further formal evidence of complete histological resolution of CD; however, this investigation was not performed because patients experienced a complete regression of their celiac symptomatology. Moreover, thanks to the possibility to follow an unrestricted diet and to avoid continuous therapy for β-thalassemia major, they recovered an excellent quality of life. Finally, the sustained negativity of the specific CD autoantibody and the absence of histological stigmata despite gluten ingestion, together with the lack of T-cell proliferation upon gliadin stimulation, allow us to speculate that allogeneic HSCT may lead to induction of gluten tolerance. This is an important observation because to date none of the novel therapeutic strategies proposed for uncomplicated CD has been shown to reinduce gluten tolerance.
Those cell populations were endowed with the capacity to influence peripheral tolerance, and we found that although the number of peripheral blood DC and the percentage of both circulating and mucosal regulatory T cells before HSCT were comparable with those of patients with CD, after HSCT these values were similar to those of non-CD controls independently of the absence or presence of gluten in the diet. The persistence over time of these values provides support to the hypothesis that a restoration of the right “balance” between tolerance and immunity may ensue after allogeneic HSCT (15). Moreover, the robust decrease of the percentage of circulating and mucosal regulatory T cells, and T-cell unresponsiveness to gliadin after HSCT, suggests that the induction of gluten tolerance is likely because of the elimination of pathogenic T-cell clones and resetting of the immune system. The full recovery may have been enhanced by the fact that both cases were likely in full remission on GFD at the time of HSCT. Similarly, analysis of the cytokine profile demonstrated that although before HSCT the peripheral blood cytokine profile of patient 1 was comparable with that of CD control patients, the post-HSCT profile mimicked that of dyspeptic controls, thus reinforcing the concept of a reconstitution of immune homeostasis. The only mucosal cytokine whose expression after HSCT remained high was TGF-β. This finding could be because of an expansion of TGF-β–producing “TH3” regulatory cells, whereas the lack of an upregulation of interleukin-10 leads to hypothesize the absence of a role for interleukin-10–producing regulatory T cells. TGF-β is consistently secreted also by pericryptal myofibroblasts and plays a central role in orchestrating enterocyte differentiation along the crypt-villous axis (16). The high levels of TGF-β found after HSCT, therefore, could be because of tissue remodeling. In this regard, we cannot rule out the possibility that HSCT may have also provided mechanisms of gut mucosa repair and regeneration (17).
In conclusion, allogeneic HSCT may induce gluten tolerance; in view of this observation, this treatment may be considered in complicated patients with CD in whom the morbidity and mortality risks associated with the procedure are lower than the potential benefits deriving from the eradication of pathogenic T-cell clones and the cure of the underlying disease. Moreover, the immunological study performed in our patients casts some light on mechanisms governing immune tolerance and gut mucosa healing useful for new targeted therapies of patients with CD.
The authors thank Dr Federico Biagi and Dr Paola Bianchi for anti-endomysium antibody detection, Dr Marco Danova and Dr Bianca Rovati for characterization of circulating dendritic cells and cytokine-producing T cells, and Prof Gianbattista Parigi, who carried out the endoscopy of the pediatric cases. The authors also thank Mrs Susan West for the English-language revision.
1. Di Sabatino A, Corazza GR. Coeliac disease. Lancet
2. Bargetzi MJ, Schönenberger A, Tichelli A, et al. Celiac disease transmitted by allogeneic non-T cell-depleted bone marrow transplantation. Bone Marrow Transplant
3. Kline RM, Neudorf SM, Baron HI. Correction of celiac disease after allogeneic hematopoietic stem cell transplantation for acute myelogenous leukemia. Pediatrics
4. Hoekstra JH, Groot-Loonen JJ, van der Weij A, et al. Successful treatment of celiac disease by allogeneic haematopoietic stem cell transplantation. J Pediatr Gastroenterol Nutr
5. Corazza GR, Frazzoni M, Dixon MF, et al. Quantitative assessment of the mucosal architecture of jejunal biopsy specimens: a comparison between linear measurement, stereology, and computer aided microscopy. J Clin Pathol
6. MacDonald KPA, Munster DJ, Clark GJ, et al. Characterization of human blood dendritic cell subsets. Blood
7. Dzionek A, Fuchs A, Schmidt P, et al. BDCA-2, BDCA-3, and BDCA-4: three markers for distinct subsets of dendritic cells in human peripheral blood. J Immunol
8. Maccario R, Podestà M, Moretta A, et al. Interaction of human mesenchymal stem cells with cells involved in alloantigen-specific immune response favors the differentiation of CD34+ T cell subsets expressing a regulatory/suppressive phenotype. Haematologica
9. Gianfrani C, Levings MK, Sartirana C, et al. Gliadin-specific type 1 regulatory T cells from the intestinal mucosa of treated celiac patients inhibit pathogenic T cells. J Immunol
10. Farge D, Labopin M, Tyndall A, et al. Autologous hematopoietic stem cell transplantation for autoimmune diseases: an observational study on 12 years’ experience from the European Group for Blood and Marrow Transplantation Working Party on Autoimmune Diseases. Haematologica
11. Al-Toma A, Visser O, van Roessel HM, et al. Autologous hematopoietic stem cell transplantation in refractory celiac with aberrant T-cells. Blood
12. Al-Toma A, Verbeek WH, Visser OJ, et al. Disappointing outcome of autologous stem cell transplantation for enteropathy-associated T-cell lymphoma. Dig Liver Dis
13. Tack GJ, Wondergem MJ, Al-Toma A, et al. Auto-SCT in refractory celiac disease type II patients unresponsive to cladribine therapy. Bone Marrow Transplant
14. McNeish AS, Harms HK, Rey J, et al. The diagnosis of celiac disease. A commentary on the current practices of members of the European Society for Paediatric Gastroenterology and Nutrition (ESPGAN). Arch Dis Child
15. Roncarolo MG, Gregori S, Lucarelli B, et al. Clinical tolerance in allogeneic hematopoietic stem cell transplantation. Immunol Rev
16. Andoh A, Bamba S, Brittan M, et al. Role of intestinal subepithelial myofibroblasts in inflammation and regenerative response in the gut. Pharmacol Ther
17. Okamoto R, Yajima T, Yamazaki M, et al. Damaged epithelia regenerated by bone marrow-derived cells in the human gastrointestinal tract. Nat Med
Keywords:Copyright 2013 by ESPGHAN and NASPGHAN
celiac disease; gluten; hematopoietic stem cell transplantation; tolerance