Autoimmune enteropathy (AIE) is a disease characterized by intractable, life-threatening diarrhea that typically begins during infancy (1,2). At least 50% of patients have circulating antienterocyte antibodies. Associations with other autoimmune diseases have been described, but most patients do not have an identified abnormality of immune function. Although the pathogenesis of AIE is not entirely clear, activation of T lymphocytes is believed to be a significant mechanism in the generation of the intestinal damage. Without treatment, the diarrhea associated with AIE is unremitting, and the mortality rate has been reported to be as high as 30% (3). To date, the therapeutic options are limited. Responses have been poor or short lived with corticosteroid and azathioprine administration. Remission has been induced and maintained in some patients with chronic use of cyclosporin A and tacrolimus (3,4).
Recently, high-dose cyclophosphamide has been demonstrated to induce durable treatment-free remissions in patients with severe aplastic anemia and other autoimmune disorders, including systemic lupus erythematosus, chronic inflammatory demyelinating polyneuropathy, and paraneoplastic pemphigus (5–7). Given its efficacy for managing other life-threatening autoimmune diseases, this medication was given to an infant with severe AIE.
A 7-month-old female infant was examined for a 1-month history of worsening watery diarrhea that had become blood tinged. The pregnancy, delivery, and neonatal period were normal. The infant had been breast-fed until aged 5 months and was then switched to a standard cow's milk formula. Solid foods were introduced when the infant was aged approximately 4 months. There was no family history of consanguinity, immunodeficiency syndromes, or autoimmune diseases. On examination, the infant was ill appearing and dehydrated. Even while nil per os, her stool output was greater than 40 mL/kg/d with fecal electrolyte concentrations of sodium, potassium, and chloride of 95, 28, and 76 mEq/L, respectively. Examinations of the stool for bacterial and viral pathogens, Clostridium difficile, cryptosporidium, and other parasitic infections were negative. Serum levels of immunoglobulins, vasoactive intestinal peptide and gastrin, and urinary vanillylmandelic acid were normal. In the first 9 days of hospitalization, several attempts were made to feed the infant an amino acid formula (Neocate, SHS, Rockville, MD, U.S.A.), but during each trial, the stool output significantly increased, her weight decreased, and she became hyponatremic and hypoalbuminemic. On hospital day 10, upper and lower gastrointestinal endoscopies were performed. Histopathology of the proximal duodenum (Fig. 1A) showed moderate to marked villous blunting, absence of crypts, and an intense lymphocytic infiltrate in the lamina propria predominantly composed of CD3+ T cells. The surface epithelium was injured with epithelial thinning, cytoplasmic vacuolation, and foci of cytoplasmic lipid hang-up. Despite the intense lamina propria inflammation, there were only a few intraepithelial lymphocytes. The rectosigmoid biopsies (Fig. 1B) also revealed an intense lymphocytic infiltrate within the lamina propria, surface epithelial injury, and architectural distortion, including areas of crypt loss, crypt branching, and patchy cryptitis. Immunoglobulin G antibodies to the enterocyte brush border were present in the serum at a titer of 1:10, suggestive of an AIE (performed in the laboratory of Dr. Pierre Russo, Children's Hospital of Philadelphia, PA, U.S.A.). Intravenous methylprednisolone therapy was initiated at 1.5 mg/kg/d in conjunction with loperamide and intravenous metronidazole.
When the infant's clinical status did not improve, she was transferred to The Johns Hopkins Children's Center at 8 months of age. Within a few days of admission, a central intravenous catheter was placed for the administration of parenteral nutrition. Cultures and antigen detection tests of stool failed to identify an infectious etiology. Additional immunologic analyses, performed while the patient was receiving corticosteroids, revealed a low lymphocyte count of 1,687/mm3 with a normal distribution of lymphocyte subsets (61% CD3 T cells, normal range, 13–35%; 43% CD4 T cells, normal range, 33–58%; 15% CD8 T cells, normal range, 13–26%; 33% CD19 B cells, normal range, 13–35%; and 5% CD16/CD56 NK cells, normal range, 2–13%). Antinuclear antibody and antismooth muscle antibodies were detected at titers of 1:80 and 1:20. Antithyroglobulin, antimicrosomal, antiislet cell, antimitochondrial, and antiliver and kidney antibodies were negative. During the next 7 months, the patient received intravenous methylprednisolone in doses up to 2 mg/kg/d almost continuously along with multiple other therapeutic interventions (Fig. 2). Intravenous cyclosporin A was administered on two separate occasions for approximately 1.75 and 2.75 months, with a mean blood level of 239 ng/mL (range, 42–630 ng/mL). Clinical response was suboptimal, enabling the infant's enteral feedings to reach only 20% of her estimated caloric requirement. Tacrolimus was administered in doses as high as 0.37 mg/kg/d. Oral absorption proved to be erratic, although the mean blood level during treatment was 7.8 ng/mL (range, 2.7–16.7 ng/mL). When the 2-month administration of tacrolimus failed to improve the patient's intestinal absorption, it was replaced with 6-mercaptopurine administered in conjunction with cyclosporin A and methylprednisolone. Ten weeks later, the infant developed clinical and biochemical pancreatitis presumed to be induced by the 6-mercaptopurine, and the drug was discontinued. During these treatment courses, diarrhea persisted, and parenteral nutrition was required to maintain the infant's nutritional status. A percutaneous endoscopic gastrostomy was placed when the infant was aged 10 months; however, intermittent trials of continuous enteral feeding with an amino acid formula (EleCare, Ross, Columbus, OH, U.S.A.) were complicated by increased stool volume, dehydration, and acidosis. Endoscopic biopsies obtained at 10, 11, and 13 months of age revealed persistent histopathologic features of severe AIE. The infant's course was additionally complicated by multiple episodes of central venous line catheter sepsis, including an episode of methicillin-resistant Staphylococcus aureus sepsis associated with an acute worsening of her diarrhea, sudden hyponatremia, and seizures. She subsequently developed deep vein thromboses in the iliac and femoral veins requiring long-term anticoagulation.
When aged 14 months, the infant had been hospitalized for more than 6 months since her initial examination. Given her poor response to other immunosuppressive therapies, consideration was given to treatment with high-dose cyclophosphamide. After obtaining informed consent from the parents, the infant was treated with intravenous cyclophosphamide at 50 mg/kg/d for 4 consecutive days. All other immunosuppressive medications were discontinued. Prophylaxis against hemorrhagic cystitis included intravenous 2-mercaptoethanesulfonate (MESNA; 10 mg/kg/d) administered 30 minutes before and 3, 6, and 8 hours after cyclophosphamide administration. Granulocyte colony-stimulating factor (5 μg/kg/d) was begun 6 days after the last cyclophosphamide dose and continued until the absolute neutrophil count reached 1,000/mm3. During the immediate postcyclophosphamide period, the infant developed neutropenia and a Klebsiella pneumoniae urinary tract infection. She required a single packed erythrocyte cell transfusion. She did not require platelet transfusion and was neutropenic (absolute neutrophil count <500/mm3) for only 12 days.
The infant's clinical status dramatically improved without the need for additional immunosuppressive therapy. She required intermittent hospitalizations for venous catheter-associated bacteremia during the first few months after therapy but had no other significant complications. Six months after cyclophosphamide administration, duodenal biopsies showed marked improvement with a return of villous and crypt architecture and a marked decrease in epithelial injury and vacuolation (Fig. 1C). There continued to be mild villous blunting and a moderate increase in lymphocytic and plasma cell inflammation within the lamina propria. The histopathology of the colon biopsies revealed restitution of normal crypt architecture and the expected amount of lymphocytic infiltrate in the mucosa (Fig. 1D). Seven months after cyclophosphamide therapy, when the child was aged 21 months, parenteral alimentation was discontinued and the central venous catheter was removed. When aged 27 months, she was consuming an age-appropriate diet without complications, and her gastrostomy tube was removed at 33 months. Aged 3 years, the child remains well without evidence of disease recurrence. Her height and weight are at approximately the 50% percentile for age, and her development is within normal limits.
We have presented the case of an infant with severe AIE who remains free of active disease, without receiving any immunosuppressive agents, after a single treatment course of high-dose intravenous cyclophosphamide. In addition to remission of symptoms, intestinal histopathology also improved. To our knowledge, this is the first reported case of AIE successfully managed with this regimen.
The term AIE was proposed nearly two decades ago for patients who had 1) intractable diarrhea with severe small intestine enteropathy not responding to dietary restriction; 2) circulating intestinal autoantibodies or other autoimmune disease; and 3) absence of a primary immune deficiency (1,2). Although patients typically are examined during the first 6 months of life, AIE has been described in adults (8). Two recent European studies have shown AIE to be the most commonly identified cause of intractable diarrhea in children (9,10). The prognosis of patients with AIE remains poor, and the relative impact of this disorder on health system resources is high because of the need for long-term parenteral nutrition.
Histopathology of the small intestine in patients with AIE typically reveals severe villous atrophy and an inflammatory infiltrate of the lamina propria. In contrast to other etiologies for villous blunting, such as celiac disease, there is typically a milder degree of intraepithelial lymphocytosis. Electron microscopy does not reveal specific ultrastructural abnormalities (11). As in our case, the inflammatory process can be extensive, involving the small bowel and colonic mucosa. The pathophysiology of AIE is incompletely understood, although there is little doubt that the intestinal lesions are generated by an immune mechanism and that activated T lymphocytes play a role in the pathogenesis of the villous atrophy (12,13). Antienterocyte antibodies are not found in all patients, and their significance is unclear. Their generation may be the result of bowel injury rather than a primary event.
The morbidity and mortality of severe AIE remain high because of its limited responsiveness to corticosteroids, azathioprine, cyclosporin A, and tacrolimus (14). One teenaged boy with total villous atrophy, selective immunoglobulin A deficiency, and antiepithelial cell antibody had resolution of his intestinal symptoms after being treated with oral low-dose cyclophosphamide (15). However, subsequently reported attempts to manage AIE with cyclophosphamide in doses up to 3 mg/kg/d have not been successful (16,17). The total dose of cyclophosphamide given to our patient (200 mg/kg) is approximately 20 times greater than those previously used to manage AIE.
Cyclophosphamide is an alkylating agent related to nitrogen mustard and is a potent immunosuppressive agent used in bone marrow transplantation (BMT) conditioning regimens. The possibility that high-dose cyclophosphamide could be therapeutic for autoimmune diseases was first suggested in the mid-1970s, when the treatment of patients with severe aplastic anemia using high-dose cyclophosphamide conditioning and allogeneic BMT became standard (6). Several reports appeared of autologous hematopoietic reconstitution after allogeneic BMT, suggesting that high-dose cyclophosphamide alone was capable of managing the disease (18). Additionally, a case was reported of a patient with aplastic anemia successfully treated with high-dose cyclophosphamide without BMT (19).
High-dose cyclophosphamide is immunoablative, but not myeloablative, because its metabolites can be inactivated by aldehyde dehydrogenase, an enzyme that is highly expressed in hematopoietic stem cells but not in lymphoid cells. Thus, T lymphocytes, B lymphocytes, and natural killer cells are rapidly eliminated by high doses of cyclophosphamide (20). The transient bone marrow suppression that develops after this therapy may provide a period that is free of memory T-cell influence during which new lymphocyte progenitors may mature without recruitment to autoimmunity (21).
In treating patients with AIE, the side effects of high-dose cyclophosphamide need to be weighed against the need for prolonged exposure to immunosuppressive drugs and hyperalimentation. Parenteral nutrition given via central venous catheters is associated with a high risk of infection and cholestasis that can lead to end-stage liver disease. Corticosteroids can cause suppression of the hypothalamic-pituitary-adrenal axis, and their use has been associated with an array of serious adverse effects, including hypertension, osteopenia, and diabetes. Cyclosporin A and tacrolimus can cause nephrotoxicity, hypertension, gingival hyperplasia, neuropathy, hirsutism, and coarsening of facial features. In addition, there is a clear risk of infection and lymphoproliferative disease with long-term use of these agents (22). High-dose cyclophosphamide can cause transient nausea, vomiting, reversible alopecia, and neutropenia. Hemorrhagic cystitis can rarely occur, and the risk is minimized by adequate hydration and prophylaxis with MESNA. Granulocyte colony-stimulating factor may shorten the duration of deep aplasia and thereby limit risk of infection. With a single treatment course, as was given in this case, there is no risk of prolonged myelosuppression, and the risk of developing a secondary malignancy, such as leukemia or lymphoma, or impaired fertility is considered minimal. It should be noted that no relapse or lymphoproliferative disease has been seen in a series of 10 patients with aplastic anemia treated with this regimen during a follow-up period of more than 10 years (23).
The risks and benefits of high-dose cyclophosphamide were carefully analyzed for this infant. The pervasive extent of the inflammatory process in her gastrointestinal tract, resistance to standard immunosuppressive therapies, and her poor prognosis led us to consider this regimen as a reasonable alternative. The treatment was associated with self-limited and manageable acute side effects without significant complications. Although a fortuitous spontaneous recovery cannot be excluded in this case, it is unlikely in light of the severity of the infant's AIE and the lack of response to multiple previous therapies. The clinical and histopathologic response of disease activity, the ability to discontinue all immunosuppressive drugs, and the lack of long-term complications more than 23 months after therapy suggest that high-dose cyclophosphamide may be a promising treatment for some patients with AIE.
1. Mirakian R, Richardson A, Milla PJ, et al. Protracted diarrhoea of infancy: evidence in support of an autoimmune variant. BMJ 1986; 293:1132–6.
2. Walker-Smith JA, Unsworth DJ, Hutchins P, et al. Autoantibodies against gut epithelium in a child with small-intestinal enteropathy. Lancet 1982; 1:566–7.
3. Seidman EG, Lacaille F, Russo P, et al. Successful treatment of autoimmune enteropathy with cyclosporine. J Pediatr 1990; 117:929–32.
4. Bousvaros A, Leichtner AM, Book L, et al. Treatment of pediatric autoimmune enteropathy with tacrolimus (FK506). Gastroenterology 1996; 111:237–43.
5. Brodsky RA, Petri M, Smith BD, et al. Immunoablative high-dose cyclophosphamide without stem-cell rescue for refractory, severe autoimmune disease. Ann Intern Med 1998; 129:1031–5.
6. Brodsky RA. High-dose cyclophosphamide for aplastic anemia and autoimmunity. Curr Opin Oncol 2002; 14:143–6.
7. Nousari HC, Brodsky RA, Jones RJ, et al. Immunoablative high-dose cyclophosphamide without stem cell rescue in paraneoplastic pemphigus: report of a case and review of this new therapy for severe autoimmune disease. J Am Acad Dermatol 1999; 40:750–4.
8. Corazza GR, Biagi F, Volta U, et al. Autoimmune enteropathy and villous atrophy in adults. Lancet 1997; 350:106–9.
9. Catassi C, Fabiani E, Spagnuolo MI, et al. Working Group of the Italian society of Pediatric Gastroenterology and Hepatology (SIGEP). Severe and protracted diarrhea: results of the 3-year SIGEP multicenter survey. J Pediatr Gastroenterol Nutr 1999; 29:63–8.
10. Goulet OJ, Brousse N, Canioni D, et al. Syndrome of intractable diarrhea with persistent villous atrophy in early childhood: a clinicopathologic survey of 47 cases. J Pediatr Gastroenterol Nutr 1998; 26:151–61.
11. Cutz E, Sherman PM, Davidson GP. Enteropathies associated with protracted diarrhea of infancy: clinicopathologic features, cellular and molecular mechanisms. Pediatr Pathol Lab Med 1997; 17:335–67.
12. Murch SH, Fertleman CR, Rodrigues C, et al. Autoimmune enteropathy with distinct mucosal features in T-cell activation deficiency: the contribution of T cells to the mucosal lesion. J Pediatr Gastroenterol Nutr 1999; 28:393–9.
13. Hill SM, Milla PJ, Bottazzo GF, et al. Autoimmune enteropathy and colitis: is there evidence of a generalized autoimmune gut disorder? Gut 1991; 32:36–42.
14. Russo PA, Brochu P, Seidman EG, et al. Autoimmune enteropathy. Pediatr Dev Pathol 1999; 2:65–71.
15. McCarthy DM, Katz SI, Gazze L, et al. Selective IgA deficiency associated with total villous atrophy of the small intestine and an organ-specific-anti-epithelial cell antibody. J Immunol 1978; 120:932–8.
16. Savilahti E, Pelkonen P, Holmberg C, et al. Fatal unresponsive villous atrophy of the jejunum, connective tissue disease and diabetes in a girl with intestinal epithelial cell antibody. Acta Paediatr Scand 1985; 74:472–6.
17. Colletti RB, Guillot AP, Rosen S, et al. Autoimmune enteropathy and nephropathy with circulating anti-epithelial cell antibodies. J Pediatr Gastroenterol Nutr 1991; 118:858–64.
18. Speck B, Cornu P, Jeannet M, et al. Autologous marrow recovery following allogeneic marrow transplantation in a patient with severe aplastic anemia. Exp Hematol 1976; 4:131–7.
19. Baran DT, Griner PF, Klemperer MR. Recovery from aplastic anemia after treatment with cyclophosphamide. N Engl J Med 1976; 295:1522–3.
20. Brodsky RA, Fuller AK, Ratner LE, et al. Elimination of alloantibodies by immunoablative high-dose cyclophosphamide. Transplantation 2001; 71:482–4.
21. Traynor AE, Schroeder J, Rosa RM, et al. Treatment of severe systemic lupus erythematosus with high-dose chemotherapy and haemopoietic stem-cell transplantation: a phase I study. Lancet 2000; 356:701–7.
22. McDiarmid SV, Wallace P, Vargas J, et al. The treatment of intractable rejection with tacrolimus (FK506) in pediatric liver transplant recipients. J Pediatr Gastroenterol Nutr 1995; 20:291–9.
23. Brodsky RA, Sensenbrenner LL, Smith D, et al. Durable treatment-free remission after high-dose cyclophosphamide therapy for previously untreated severe aplastic anemia. Ann Intern Med 2001; 135:477–83.