Statistics from Altmetric.com
CD40 ligand (CD40L) deficiency, previously known as X linked hyper-IgM syndrome, is a rare primary immune deficiency caused by a mutation in the gene encoding for the CD40L glycoprotein (CD154), expressed on the surface of T lymphocytes.1 2 CD40L expression is critical for the interaction of T lymphocytes with CD40, expressed on both B lymphocytes and monocyte derived cells.3 Lack of interaction prevents immunoglobulin isotype switching from IgM to IgG, IgA, and IgE, as well as failure to activate Kupffer cells and pulmonary macrophages. As a result affected children are at risk of Pneumocystis cariniipneumonia in the first year of life, recurrent bacterial sinopulmonary infections, and persistent Cryptosporidium parvum infections leading to sclerosing cholangitis, cirrhosis, and liver cancer. Other complications include autoimmune disease, neutropenia, and enteroviral meningoencephalitis.4 5 A European survey of CD40L deficient patients showed that 75% develop liver disease and only 20% survive into the third decade of life.6
Prophylactic co-trimoxazole and intravenous immunoglobulin replacement are the mainstays of treatment, together with advice to boil water to prevent cryptosporidium infection. Granulocyte colony stimulating factor (G-CSF) can improve neutropenia. Nevertheless many patients given such optimal treatment still die of liver disease by early adult life. As this is a T cell disorder, at present only haemopoietic stem cell transplantation can cure the genetic defect.
Fasth reported the first allogenic bone marrow transplant in this condition, using an HLA matched brother as donor.7 There have been seven further case reports of successful bone marrow transplantation for X linked hyper-IgM syndrome.8-15However, bone marrow transplantation for immunodeficiency is not without risk, and for other primary immunodeficiencies such as the Wiskott-Aldrich syndrome, older age, ongoing serious infections, lung or liver disease, and lack of availability of a well matched donor are associated with a poor outcome after bone marrow transplantation.
We present the experience of bone marrow transplantation for CD40L deficiency in a single centre, with particular reference to the influence of pretransplant risk factors and donor type.
We studied retrospectively eight patients who received bone marrow transplants between September 1996 and December 2000, with follow up of between 0.5 and 3.5 years. The diagnosis was confirmed by demonstrating absence of CD40L on mononuclear cells separated on Ficoll, stimulated for 4 hours with phorbol myristate acetate 50 ng/ml and ionomycin 1000 ng/ml, and stained with CD3 and CD40L antibodies.16Mutations in the CD40L gene were also demonstrated by single strand conformation polymorphism analysis of individual exons of the CD 40 ligand gene.17
Before transplantation, patients were assessed by chestx ray, computed axial tomography (CT) of the chest, pulmonary function testing, and bronchoalveolar lavage. Other investigations included measurement of serum albumin, alkaline phosphatase, bilirubin, γ-glutamyl transferase, and aspartate aminotransferase, a liver ultrasound scan, endoscopic retrograde cholangiopancreatography, and a liver biopsy with immunohistochemical staining.
Information was obtained about pre-existing infection and pretransplantation immunological and infective data, and details of the transplant procedure were recorded, particularly donor type and HLA match, conditioning regime, and marrow manipulation. The post-transplantation data recorded included description of infective episodes, engraftment details and immune function, complications, and outcome.
All eight boys had defective CD40L expression on activated T lymphocytes and a mutation in the CD40L gene. Six children hadPneumocystis carinii pneumonia (table 1); three required mechanical ventilation with additional extracorporeal membrane oxygenation in case 2. Three had bronchiectasis on CT scan before bone marrow transplantation.
Pretransplantation bronchoalveolar lavage was negative in seven patients. A solitary Pneumocystis cariniicyst was found in case 7, treated with intravenous cotrimoxazole before transplantation. Four patients had documented previous cryptosporidial infection and three had abnormalities on liver biopsy. None had hepatocellular carcinoma or oesophageal varices. There were no cases of cytomegalovirus infection pretransplantation. Five patients required G-CSF before transplantation. None had inflammatory bowel disease. Four had failure to thrive, one requiring total parental nutrition before transplantation for severe liver disease and chronic pancreatitis. All were receiving intravenous immunoglobulin replacement and had normal trough IgG levels.
Age at bone marrow transplantation ranged from 1 to 14 years (median 5.75 years). Standard European bone marrow transplantation working party cytoreductive conditioning protocols for T cell immunodeficiencies were followed. As conditioning for bone marrow transplantation, six patients received busulphan 16 mg/kg and cyclophosphamide 200 mg/kg. Patient 2 also receiving anti-LFA-1 antibody from day −3 to day +10, and anti-CD2 antibody from day −2 to day+1, while patient 7 received fludarabine 30 mg/m2from day −7 to day −3, melphalan 140 mg/m2 on day −2, and antithymocyte globulin 2.5 mg/kg from day −6 to day −2.
Two patients received sibling HLA identical whole marrows, while six received T cell depleted marrow from unrelated donors (table 2). Donor/recipient HLA A, B, C, DR, and DQ typing was performed by molecular DNA methods. Two of the T cell depletion studies were performed with CAMPATH (1G and 1M, respectively) in vitro. The remaining four were performed by CD34+ positive stem cell selection (Miltenyi CliniMACS). All patients received cyclosporin as graft versus host disease prophylaxis. All were isolated under laminar flow and received thrice weekly oral co-trimoxazole (5 mg/kg twice daily) and weekly intravenous immunoglobulins. Patients known to have had cryptosporium infection received additional prophylaxis for this organism in the form of azithromycin and paromomycin; cases 3, 7, and 8 also received bovine serum colostrum.
All patients had evidence of engraftment after transplantation. Full or partial chimaerism was demonstrated in all seven patients who were tested (table 2). All patients had evidence of infection during the transplant period (table 2). Three had ongoing cryptosporidial infections. Adenovirus type 5 was found in stool from patient 2 from day −1, treated with oral and intravenous immunoglobulin and intravenous ribavarin. Human herpesvirus 6 was also detected in peripheral blood by polymerase chain reaction on day +55, and treated with foscarnet.
Four patients are alive and well, with normal CD40L expression and no graft versus host disease. Two remain on intravenous immunoglobulin because they received their transplant less than a year ago. IgM levels are normal. Two patients off intravenous immunoglobulin have made adequate seroresponses to vaccination against tetanus,Haemophilus influenzae type B, and pneumococcus.
Four patients died. The unrelated donor for patient 5 was declared unfit on the day of harvest. A 9/10 substitution unrelated donor was found and harvested within 48 hours. The patient developed biopsy confirmed grade III acute skin and gut graft versus host disease on day 7, treated with high dose methylprednisolone for three days followed by antithymocyte globulin (rabbit) 5 mg/kg/day from day 16 to day 23. Adenovirus was detected in stool by ELISA from day 10 and treated with oral and high dose intravenous immunoglobulin infusion (12 g/week), ribivarin, and ganciclovir. By day 68 lung, liver, and gut disease was evident with adenovirus isolated from bronchoalveolar lavage, liver, and lower gastrointestinal tract. The child died of multisystem organ failure and adenovirus hepatitis on day 82.
Patient 6 had severe liver disease before his bone marrow transplant, with diarrhoea, chronic cryptosporidial infection, sclerosing cholangitis, cirrhosis, and chronic pancreatitis; long term total parental nutrition was required before the transplant, with G-CSF for chronic neutropenia. He was treated with paromomycin, bovine serum colostrum, and ursodeoxycholic acid in addition to azithromycin. He died on day 34 from fulminant liver failure. Veno-occlusive disease, cryptosporidium infestation of bile ducts, biliary fibrosis, and graft versus host disease were demonstrated at necropsy examination.
Patient 7 had cryptosporidial ascending cholangitis with portal inflammation on liver biopsy. He developed severe cryptosporidial enteritis with massive diarrhoea despite treatment with azithromycin, paromomycin, bovine serum colostrum, and nitazoxamide. Capillary leak syndrome and pneumonitis led to multiorgan failure and death on day +87.
Patient 8 had extensive bilateral bronchiectasis and severe liver disease before transplantation. Veno-occlusive disease requiring recombinant tissue plasminogen activator on day +9 and heparin infusion on day +11 complicated the post-transplant course, followed by severe progressive liver graft versus host disease. He died from haemorrhage following a liver biopsy while being assessed for orthoptic liver transplantation.
The long term outlook for CD40L deficiency remains poor despite full supportive treatment with intravenous immunoglobulin replacement therapy, co-trimoxazole prophylaxis forPneumocystis carinii infection, and G-CSF for neutropenia, together with prompt antibacterial treatment and nutritional support. Our results confirm that bone marrow transplantation can be curative and it had a successful outcome with T cell depleted mismatched unrelated donor bone marrow transplantation in four cases. However, bone marrow transplantation is risky and four patients died. Perhaps not surprisingly these were older patients with more advanced liver disease. Furthermore their deaths were all due to liver disease and occurred either because cryptosporidial infection progressed rapidly following pretransplantation cytotoxic conditioning therapy or because ongoing hepatic inflammation activated donor T lymphocytes, setting off severe liver graft versus host disease. Previous Pneumocystis carinii pneumonia did not appear to be associated with increased risk of death. Pre-existing liver disease appeared to be a greater risk factor than suboptimal donor matching, as the four survivors all received marrow from unrelated donors, whereas two of the four who died received marrow from HLA identical siblings. The number of patients in this series is small and caution is needed before drawing conclusions. Nevertheless these results suggest that, as in other primary immunodeficiency diseases like the Wiskott-Aldrich syndrome, bone marrow transplantation is more likely to be successful if performed in early childhood before significant liver damage has occurred.
For patients who already have significant liver disease, planning the best treatment programme remains a challenge, particularly as cryptosporidial infection cannot be eradicated despite the use of agents such as paromomycin, azithromycin, and bovine serum colostrum. Whether longer periods of anticryptosporidial treatment before bone marrow transplantation, with more drugs delivered intravenously or through a nasojejunal tube, would slow the rate of progression of cryptosporidial infection is unclear. Less damaging pretransplantation cytotoxic conditioning therapy might lessen the progression of cryptosporidial infection and liver inflammation, so reducing deaths from cryptosporidium and graft versus host disease. Regimens using fludarabine and melphalan hold promise18 19 and a recent report describes success with this method in children with CD40L deficiency and sclerosing cholangitis,20 but it should be noted that in our series, patient 8 was given such conditioning and still developed severe liver disease. Finally, the role of combined liver and bone marrow transplantation needs to be explored further, following the recent report of a successful case.13 As liver disease appears to recur rapidly in patients given a liver transplant without a bone marrow transplant it is likely that the two procedures should be undertaken within a short space of time.
Multicentre studies are needed for accurate assessment of pretransplantation risk factors, in particular detecting and quantifying cryptosporidial infection with new methods such as poymerase chain reaction, and assessing any consequent liver damage.21 It may be that less toxic conditioning regimens will make bone marrow transplantation less risky for older patients with significant damage. On the basis of our results so far we would advocate bone marrow transplantation at an early age before significant damage has occurred, when both HLA identical sibling and unrelated donor bone marrow transplantation can be successful.
We are very grateful to Dr Alison Jones for performing CD40L mutation analyses, Dr Gavin Spickett and Dawn Barge for performing flow cytometric analysis of CD40L expression, and Dr Steve Hodges for assessment of liver disease.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.