Case Report B Cell Function After Haploidentical in Utero Bone Marrow Transplantation in a Patient with Severe Combined Immunodeﬁciency

Bone Marrow Transplantation (2002) 29, 625–628  2002 Nature Publishing Group All rights reserved 0268–3369/02 $25.00 www.nature.com/bmt Case report B cell function after haploidentical in utero bone marrow transplantation in a patient with severe combined immunodeﬁciency

J Bartolome´1, F Porta3, A Lafranchi3, JJ Rodrı´guez-Molina1, E Cela1, A Cantalejo1, E Ferna´ndez-Cruz1,AGo´mez-Pineda2, AG Ugazio3, LD Notarangelo3 and J Gil

1Division of Immunology, Hospital General Universitario ‘Gregorio Maranõń’, Madrid, Spain; 2Division of Haematology, Hospital General Universitario ‘Gregorio Maranõń’, Madrid, Spain; and 3Department of Pediatrics, Spedali Civili, Brescia, Italy

Summary: normal B cell function requires T-B cell co-operation and it is delayed when compared to T cell reconstitution. Full An in utero paternal CD34+ cell transplant was perfor- restoration of B cell function is achieved in 60–70% of B+ med in a T-B+NK+ SCID fetus. We report here the SCID patients only several years after HLA haploidentical results of the 3-year humoral immune reconstitution T cell-depleted BMT.5 The lack of adequate levels of serum study. The methods used were ApoB VNTR typing, flow Igs in those patients who do not achieve B cell reconsti- cytometry, nephelometry, hemagglutination, ELISA, tution requires IVIG treatment. ELISPOT and lymphoproliferative assays. The T cells In this study, we present data of the 3-year follow-up on were of donor origin whereas monocytes, B and NK B cell reconstitution in one SCID patient who was treated cells were of host origin. Peripheral B cell counts and with in utero haploidentical BMT. IgM levels were normal since birth. IVIG therapy was required at 5 months of age until 2 years old. IgA levels у20 mg/dl were detected from month 17 post transplan- Materials and methods tation. Isohemagglutinins were present since month 8 post transplantation, the highest titers (anti-A:1/128, Patient anti-B:1/32) were obtained at month 33 post-transplantation. After immunization with rHBsAg, circulating Immunological analyses which were performed prior to and anti-HBsAg IgG secreting cells and a 7.8-fold increase immediately after in utero BMT are described elsewhere.4 in serum anti-HBsAg Ab were detected. We conclude Briefly, in September 1997, a T-B+NK+ SCID prenatal that split chimerism following in utero haploidentical diagnosis was performed on a male fetus at week 23 of BMT allows complete humoral immune reconstitution gestation. The parents were related and they had a pre- in a T-B+NK+ SCID patient. viously affected T-B+NK+ SCID son (unknown molecular Bone Marrow Transplantation (2002) 29, 625–628. DOI: etiology). Paternal BM CD34+ cells were purified and two 10.1038/sj/bmt/1703410 i.p injections (the second infusion together with ex vivo Keywords: bone marrow; stem cell transplantation; in generated BM stromal cells) were performed in the Spedali utero; immunodeficiency; humoral; reconstitution Civili, Brescia, Italy.4 The donor did not have antibodies to hepatitis B virus (HBV) and he had never been vacci- nated against HBV. The T cell function in the patient developed at month 4 post transplantation and it has Successful treatment for SCID has been accomplished in a remained normal since then. Serum IgG levels progress- very small number of cases by in utero fetal liver trans- ively fell within the first months of life, IVIG therapy (400 plants or BM CD34+ cells transplantation.1 In patients with + mg/kg every 4 weeks, Endobulin IV; Baxter, Vienna, haploidentical (ie paternal) T cell-depleted CD34 BM cell Austria) was started when IgG levels dropped below 200 transplantation, we and others have previously demon- mg/dl (Figure 1). The patient’s growth and development strated lymphoid engraftment and restoration of T cell func- have always been normal. He has presented several upper 2–4 tion early after in utero BMT. However, there is limited respiratory infections, two episodes of gastroenteritis and information on the long-term immune reconstitution, two episodes of otitis media. The patient received prophy- mainly on humoral immune responses in patients with laxis with ketoconazole, acyclovir and co-trimoxazole until 1 SCID and in utero BMT. he was 6 months old. He has completed the following vac- Following post-natal allogeneic BMT, development of cination schedule: HB, DTP, Hib and eIPV immunizations were administered at 3 months of age; HB and eIPV at 4 Correspondence: Dr J Gil, Division of Immunology, Hospital General months of age, DT, Hib and eIPV at 6 months of age, HB Universitario ‘Gregorio Maranõń’, Dr Esquerdo, 46, 28007 Madrid, Spain and meningococcal A+C vaccine when he was 1 year old, Received 30 August 2001; accepted 15 January 2002 and DT and Hib at 19 months of age. He was immunized Humoral immune function after in utero BMT J Bartolome´ et al 626 a IVIG IVIG 62°C and denaturating was carried out for 1 min at 94°C 1250 for a total of 30 extension–denaturation cycles. Electro- 1000 phoresis of the amplification products was performed on 750 2% agarose gel and the amplified alleles were visualized 500 with ethidium bromide. mg/dl 250 0 Lymphocyte immunophenotyping 04812162024283236

Age (months) This was performed by direct immunofluorescence and b flow cytometry analysis (Becton Dickinson, Mountain 150 IgA View, CA, USA). Conjugated moAb anti-CD3 120 IgM (FITC)/CD19 (PE)/CD45 (PercP) from Becton Dickinson 90 were used for the evaluation of B lymphocytes. 60 mg/dl 30 Immunoglobulin, isohemagglutinin and specific antibody 0 quantitation 0 48 12 16 20 24 28 32 36 Age (months) Patient serum was incubated with specific Ab to IgG, IgA Figure 1 Serum Igs concentrations during the post-transplant period. (a) and IgM and the resultant immune complexes were meas- Quantification of IgG levels at different time points during follow-up. ured by a nephelometer (Beckman, Brea, CA, USA). Arrows show time interval at which IVIG therapy was administered. At Isohemagglutinin titers were measured by standard method- month 29 an additional IVIG infusion was administered. (b) Quantification ology. Anti-tetanus toxoid Ab titers were assayed by of IgA and IgM levels at different time points during follow-up. ELISA (The Binding Site, Birmingham, UK). For the monitoring of serum-specific Ab responses to hepatitis B by mistake with MMR at 29 months of age, receiving as surface antigen (anti-HBs), pre-immunization and 7, 20 and prophylaxis an IVIG infusion (Figure 1). At present, the 50 days post-immunization serum samples were quantitated patient is 4 years old; he is asymptomatic and with within the same assay by a standard microparticle enzyme excellent health, 90th percentile for height and weight, and immunoassay (Abbott AxSYM System, Weisbaden, without any therapy. Germany). The immunological studies were undertaken with the approval of the Ethics Committee of the Hospital General Detection of Anti-HBsAg Ab secreting cells (HBs AbSC) Universitario Gregorio Maranõń, and written informed con- sent was obtained from the patient’s parents. This was determined by the ELISPOT method described by Bo¨cher et al,7 with slight modifications. Polyvinylidene Isolation of PBMC difluoride-bottom 96-well MultiScreen IP filtration plates (Millipore, Bedford, MA, USA) were coated at 100 ␮l per Cells for chimerism, lymphoproliferative and ELISPOT well with carbonate buffer (pH 9.6) containing 10 ␮g/ml assays were isolated from fresh heparin-anticoagulated of purified recombinant HBsAg (kindly provided by blood by density gradient centrifugation (Lymphoprep, SmithKline Beecham, Rixensart, Belgium) and incubated Nicomed Pharma, Oslo, Norway). Cells for lymphoprolifer- overnight at 4°C. Wells coated with BSA were used as ative assay were obtained 20 days after a recall rHBsAg negative controls. After washing with sterile distilled water, immunization (fourth dose), and those for the ELISPOT plates were blocked with 200 ␮l per well of RPMI 1640 assay 7 and 20 days after this rHBsAg immunization. medium supplemented with 3% BSA (RPMI-BSA) for 1 h at 37°C. Blocking medium was discarded and PBMC were × 5 ␮ Evaluation of engraftment added at 2.5 10 per well in 200 l of RPMI-BSA. After incubation for 6 h at 37°C in a humid atmosphere with 5% Ј Chimerism analysis was performed by typing of the 3 CO2, the plates were washed with PBS containing 0.05% apoB VNTR region.6 The patient’s PBMC were previously Tween 20 (PBS-T) and incubated for 1 h at 37°C with a fractionated into CD3+, CD56+, CD19+ and CD14+ cells by peroxidase-conjugated rabbit anti-human IgG (Dako, using standard immunomagnetic purification methods Glostrup, Denmark) diluted 1:1000 in PBS containing 1% (Miltenyi Biotec, Auburn, CA, USA). Purity of the cellular BSA. The plates were washed with PBS-T and staining was fractions was higher than 90% as assessed by flow cytome- carried out by adding substrate solution containing 3- try. Genomic DNA was extracted from donor PBMC and amino-9-ethyl-carbazole. After 20–30 min the reaction was from host PBMCs and fractions. Amplification of stopped by rinsing water. Granular red spots were enumer- apoB VNTR region was carried out using 5Ј- ated with a dissection microscope and results were ATGGAAACGGAGAAATTATG-3Ј as 5Ј primer and 5Ј- expressed as IgG anti-HBs AbSC per 106 PBMC. CCTTCTCACTTGGCAAATAC-3Ј as 3Ј primer. DNA × 5 5 from 1.5 10 T cells or 10 PBMC and the other cellular Lymphoproliferative assay fractions were used for each reaction. Prior to addition of the Taq polymerase, the reaction mixture was boiled for 7 2 × 105 PBMC/well were seeded in RPMI 1460 medium min. Annealing and extension was carried out for 6 min at containing antibiotics, 2 mm glutamine and 10% pooled

Bone Marrow Transplantation Humoral immune function after in utero BMT J Bartolome´ et al 627 human AB serum in 96-well U-bottom plates (Costar, Seven days after immunization of the patient with Corning, MD, USA); triplicates with increasing concen- rHBsAg, peripheral IgG anti-HBs AbSC were present, with trations of purified recombinant HBsAg were set from 0.5 a median number of 54/106 PBMC. Children and adult ␮ ° to 5 g/ml. PBMC were incubated at 37 C, 5% CO2 for 5 healthy controls were also tested 7 days after immunization days and on the sixth day of culture 1 ␮Ci of H3Thy with rHBsAg and showed detectable but lower frequencies (Amersham, Bucks, UK) was added. Cultures were har- of anti-HBs AbSC (2 ± 1/106 PBMC). No anti-HBs AbSC vested 18 h later. Radioactivity was measured in a beta were detected 20 days after HBsAg immunization either in counter (MicroBeta TRILUX, Wallac Oy, Turkey, the patient or in the healthy controls (data not shown). The Finland). presence and concentrations of specific serum anti-HBsAg Ab levels were also studied in this patient: serum IgG anti- HBsAg levels pre-immunization and at 7, 20, and 50 days Results post immunization were 13291, 35475, 103920 and 66650 mIU/ml, respectively, showing at 20 days post immuni- Figure 2 shows the chimerism study performed 33 months zation a 7.8-fold increase over the baseline levels. after in utero BMT. There is engraftment of donor T cells The specific T cell lymphoproliferative response of the (along with a small number of autologous T cells), while patient’s PBMC to rHbsAg (1 ␮g/ml) showed an stimu- circulating B cells, NK cells and monocytes are of host lation index (SI) of 4.10 (35282 c.p.m.). The lymphopro- origin. This pattern of engraftment has been stable since liferative response of the adult healthy control included in the first study performed at 9 months post in utero BMT the study showed an SI of 3.03 (18224 c.p.m.). (data not shown). Serum concentrations of IgM have been normal since birth (Figure 1). IgA levels remained р20 mg/dl until 14 Discussion months of age when the levels started to increase reaching levels greater than 50 mg/dl month 20 post in utero BMT. We provide evidence of complete immune reconstitution IgG levels increased from month 5 to month 22 of age, the following in utero transplantation in a T-B+NK+ SCID period in which the patient was receiving IVIG. After IVIG patient. Clinical experience with in utero CD34+ cell trans- withdrawal, the IgG levels remained within the normal plantation is still poor and successful cases have been lim- range (the most recent value at month 35 was 948 mg/dl ited to T-B+NK- SCID affected fetuses, a more common IgG). form of presentation of primary immunodeficiency. Other Serum isohemagglutinin titers were first assessed and approaches using fetal liver transplants have also been found to be positive at month 9 post transplantation (anti- reported in two patients with HLA deficiency and a SCID A: 1/4, anti-B: 1/2), showing the highest values (anti-A: fetus.8 Our data show that the restoration of a fully func- 1/128, anti-B: 1/32) at month 30 of age. Eight months after tional immune system after in utero BMT is not delayed the last IVIG infusion, specific Ab titers to TT were as compared to T cell-depleted, haploidentical post-natal detected (16 ␮g IgG1/dl). BMT performed in SCID patients.5 Circulating absolute B cell numbers remained within nor- Evidence of B cell function in this in utero BMT is dem- mal age-matched levels throughout the follow-up (641, onstrated by IgA levels, production of isohemagglutinins 807, 484 and 480 CD19+ cells/␮l at 7, 15, 27 and 33 months and specific Ab responses evaluated by circulating anti- post in utero BMT, respectively). HBsAg Ab secreting cells and serum anti-HBsAg IgG following HBsAg immunization. Along with the B cell function study, we have detected HBsAg-specific T cell lymphoproliferative responses. This antigen-specific response cannot be related to the transfer of donor memory- specific T cells, since the donor was seronegative for HBV and he has never been immunized with HBsAg. The response could be due to the generation in the recipient of HbsAg-specific T cells which differentiated in his thymus. Other authors have found that NK+ SCID patients who do not receive pretransplant cytoreduction have a poor clinical and immunological outcome.9 However, the T cell engraftment and restoration of immune function in this SCID patient have occurred in the presence of autologous NK cells and without any pretransplant treatment. Approximately 30% of T-B+ SCID patients treated with post-natal HLA non-identical BMT do not develop a normal B cell function. It has been suggested that in T-B+ Figure 2 Chimerism analysis: donor PBMC and different host cellular SCID, lack of Ab responses following haploidentical, T lineages. Amplification products of alleles at the hypervariable region of cell-depleted BMT could be due to intrinsic B cell defects the 3Ј side of the ApoB gene. Gel analysis of DNA from the different 10 cell sources (from left to right): lane 1 contains donor PBMC: lane 2 host and/or to the lack of engraftment of donor B cells. The 11 CD3+ cells; lane 3 host CD14+ cells; lane 4 host CD19+ cells; and lane phenotype T-B+NK+ is rather unusual, the majority of 5 host CD56+ cells. B-positive SCID cases being NK negative. Although the

Bone Marrow Transplantation Humoral immune function after in utero BMT J Bartolome´ et al 628 molecular basis of the immunodeficiency in our patient 2 Flake AW, Roncarolo MG, Puck JM et al. Treatment of X- remains unknown, the results suggest that once T cell func- linked severe combined immunodeficiency by in utero trans- tion has been restored the primary defect does not affect B plantation of paternal bone marrow. New Engl J Med 1996; cells. We have not excluded the presence of microchimer- 335: 1806–1810. ism (р3% of donor cells) within the B cell population.12 3 Wengler GS, Lafranchi A, Frusca T et al. In-utero transplantation of parental CD34 haematopoietic progenitor cells in a Therefore, the humoral immune responses demonstrated in patient with X-linked severe combined immunodeficiency the patient could be due to the activation of allogeneic B (SCIDXl). Lancet 1996; 348: 1484–1487. cells. However, it seems unlikely that such a small number 4 Gil J, Porta F, Bartolome´ J et al. Immune reconstitution after of B cells could account for the normal levels of serum Igs. in utero bone marrow transplantation in a fetus with severe We have found a good T–B cell co-operation in this combined immunodeficiency with natural killer cells. Trans- patient demonstrated by the strong specific humoral plant Proc 1999; 31: 2581. responses. Co-operation between donor origin T cells and 5 Haddad E, Landais P, Friedrich W et al. Long-term immune host origin B cells in SCID following post-natal BMT has reconstitution and outcome after HLA-nonidentical T-cell- been recently demonstrated by the presence of autologous depleted bone marrow transplantation for severe combined allotypic determinants and B cell phenotypic subpopula- immunodeficiency: a European retrospective study of 116 10 patients. Blood 1998; 91: 3646–3653. tions. This T–B cell co-operation might be the result of 6 Boerwinkle E, Xiong W, Fourest E et al. Rapid typing of tan- the positive selection of thymocytes derived from the donor + demly repeated hypervariable loci by the polymerase chain CD34 cells by host HLA molecules expressed on the reaction: application to the apolipoprotein B 3´ı hypervariable thymic epithelium. region. Proc Natl Acad Sci 1989; 86: 212–216. Quantification of the concentrations of individual Ig 7Bo¨cher WO, Herzog-Hauff S, Schlaak J et al. Kinetics of classes by conventional serological methods presents sev- hepatitis B surface antigen-specific immune responses in acute eral limitations derived from the rate of synthesis and catab- and chronic hepatitis B or after HBs vaccination: stimulation olism of these molecules. Moreover, transplacentally of the in vitro antibody response by interferon gamma. Hepa- acquired Ig and/or exogenous IVIG, which occurs fre- tology 1999; 29: 238–244. quently in the SCID-BMT setting, does not allow an esti- 8 Touraine JL, Raudrant D, Rebaud A et al. In utero transplantation of stem cells in humans: immunological aspects and mation of the Ig pool contributed by the child’s immune clinical follow-up of patients. Bone Marrow Transplant 1992; system to be made. The single cell functional immunoassay 9 (Suppl. 1): 121–126. ELISPOT provides accurate information on Ig production 9O’Reilly RJ, Friedrich W, Small TN. Hematopoietic cell trans- at a cellular level by detecting Ab secreting cells.13,14 The plantation for immunodeficiency disorders. In: Thomas ED, use of ELISPOT has enabled us to show active and specific Blume KG, Forman SJ (eds). Hematopoietic Cell Transplan- Ab production by B cells of the BMT recipient, and may tation, 2nd edn. Blackwell Science: Malden, MA, 1999, pp be a clinically useful method for the study of the humoral 1154–1172. immune responses in patients with BMT. 10 Haddad E, Le Deist F, Aucouturier P et al. Long-term chimer- The data support in utero BMT as an alternative thera- ism and B-cell function after bone marrow transplantation in peutic procedure to selective abortion or to post-natal BMT patients with severe combined immunodeficiency with B cells: a single-center study of 22 patients. Blood 1999; 94: 2923– in SCID disorders following prenatal diagnosis. More cases 2930. of in utero BMT in other fetuses with SCID are necessary 11 Buckley RH, Schiff RI, Schiff SE et al. Human severe com- in order to confirm the efficacy of the procedure, in parti- bined immunodeficiency: genetic, phenotypic and functional cular in SCID patients bearing the T-B+NK+ phenotype. diversity in one hundred eight infants. J Pediatr 1997; 130: 378–387. 12 O’Reilly J, Meyer B, Stoner M et al. Very early analysis of Acknowledgements graft establishment after allogeneic bone marrow transplantation using the polymerase chain reaction. Br J Haematol The authors thank Dr W Bo¨cher (Johannes Gutenberg University, 1993; 85: 169–172. Mainz, Germany) for kind support in the ELISPOT technique, 13 Lee FK, Nahmias AJ, Spira T et al. Enumeration of human Consuelo Pedras and Ana Hernańdez (laboratory technicians), Dr peripheral blood lymphocytes secreting immunoglobulins of MA Llaurado´ for efficient administrative work regarding the major classes and subclasses in healthy children and adults. J BMT, and the patient’s parents for their collaboration in this Clin Immunol 1991; 11: 213–218. investigation. 14 Stoll BJ, Lee FK, Hale E et al. Immunoglobulin secretion by the normal and the infected newborn infant. J Pediatr 1993; 122: 780–786. References

1 Flake AW, Zanjani ED. In utero hematopoietic stem cell transplantation: ontogenic opportunities and biologic barriers. Blood 1999; 94: 2179–2191.

Bone Marrow Transplantation