Antigen-Specific T-Lymphocyte Function After Cord Blood Transplantation

Biology of Blood and Marrow Transplantation 12:1335-1342 (2006) 2006 American Society for Blood and Marrow Transplantation 1083-8791/06/1212-0001$32.00/0 doi:10.1016/j.bbmt.2006.08.036 Antigen-Specific T-Lymphocyte Function After Cord Blood Transplantation Geoff Cohen, 1 Shelly L. Carter, 1 Kenneth I. Weinberg, 2 Bernadette Masinsin, 2 Eva Guinan, 3 Joanne Kurtzberg, 4 John E. Wagner, 6 Nancy A. Kernan, 7 Robertson Parkman 2 1 The EMMES Corporation, Rockville, Maryland; 2 Department of Pediatrics, Childrens Hospital Los Angeles, Los Angeles, California; 3 Pediatric Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts; 4 Department of Pediatrics, Duke University Medical Center, Durham, North Carolina; 6 Department of Pediatrics, University of Minnesota, Minneapolis, Minnesota; and 7 Department of Pediatrics, Memorial Sloan-Kettering Cancer Center, New York, New York Correspondence and reprint requests: Robertson Parkman, MD, Division of Research Immunology/Bone Marrow Transplantation, Childrens Hospital Los Angeles, The Saban Research Institute, 4650 Sunset Boulevard, Mail Stop 62, Los Angeles, CA 90027 (e-mail: rparkman@chla.usc.edu). Received June 21, 2006; accepted August 18, 2006 ABSTRACT It has not been possible to determine the singular contribution of naive T lymphocytes to antigen-specific immunity after hematopoietic stem cell transplantation (HSCT), because of the confounding effects of donor-derived antigen-specific T lymphocytes present in most hematopoietic stem cell (HSC) products. Because umbilical cord blood contains only naive T lymphocytes, we longitudinally evaluated the recipients of unrelated cord blood transplantation (UCBT) for the presence of T lymphocytes with specificity for herpesviruses, to determine the contribution of the naive T lymphocytes to antigen-specific immune reconstitution after HSCT. Antigen-specific T lymphocytes were detected early after UCBT (herpes simplex virus on day 29; cytomegalovirus on day 44; varicella zoster virus on day 94). Overall, 66 of 153 UCBT recipients developed antigen-specific T lymphocytes to 1 or more herpesviruses during the evaluation period. The likelihood of developing antigen-specific T lymphocyte function was not associated with immunophenotypic T lymphocyte reconstitution, transplant cell dose, primary disease, or acute and chronic graft-versus-host disease. These results indicate that naive T lymphocytes present in the HSC inoculum can contribute to the generation of antigen-specific T-lymphocyte immunity early after transplantation. 2006 American Society for Blood and Marrow Transplantation KEY WORDS Immune reconstitution Antigen-specific immune function Cord blood transplantation INTRODUCTION Recipients of allogeneic hematopoietic stem cell transplantation (HSCT) are characterized by an immunodeficiency of varying severity and duration that can predispose them to opportunistic infections and possibly neoplastic relapse [1-3]. The T lymphocytes present in the HSC inoculum are composed of both naive and antigen-specific T lymphocytes [4,5]. However, it has not been possible to determine the relative contributions of donor-derived antigen-specific and naive T lymphocytes to antigen-specific immune reconstitution after HSCT. Because umbilical cord blood (UBC) does not contain antigen-specific memory T lymphocytes, unrelated cord blood transplantation (UCBT) represents a unique clinical opportunity to determine the contribution of naive T lymphocytes to posttransplantation antigen-specific immunity without the impact of donorderived antigen-specific T lymphocytes. Consequently, we longitudinally evaluated UCBT recipients for their development of antigen-specific T lymphocytes with specificity for a clinically relevant group of environmental pathogens, the herpesviruses, to determine the contribution of naive T lymphocytes to post-hsct antigen-specific immunity. METHODS Study Population The COBLT (Cord Blood Transplant) study group was a multi-institutional phase II trial of UCBT 1335

1336 G. Cohen et al. sponsored by the National Heart, Lung and Blood Institute of the National Institutes of Health. The transplantation protocol was approved by the institutional review board of each participating institution. Pediatric patients (under age 18 years) with both malignant and nonmalignant diseases underwent transplantation after receiving trial-designated preparative regimes. Patients with neoplastic diseases, other than those diagnosed with infant leukemia, were conditioned with total body irradiation (TBI) (9 fractions of 150 cgy) given twice a day on days 8 to 4; cyclophosphamide 60 mg/kg on days 3 and 2; and antithymocyte globulin (ATG), equine, 15 mg/kg twice a day on days 3 to 1, with methylprednisolone (MP) 1 mg/kg given before each dose. Patients diagnosed with infant leukemia received oral busulfan 20-40 mg/m 2 /dose, with dosing based on patient age with pharmakinetic dose adjustment, or intravenous busulfex 0.8-1.0 mg/kg, with dosing based on patient age, for 16 doses on days 8 through 5, and melphalan 45 mg/m 2 rather than TBI. Most patients with nonmalignant disease were prepared with busulfan 1 mg/kg orally given every 6 hours for 16 doses on days 9 to 5; cyclophosphamide 50 mg/kg on day 5 to 2, and ATG and MP on days 3 to 1. On the day of transplantation, patients received 2 doses of intravenous MP (1 mg/kg), with 1 dose given just before the infusion of the UBC unit. Graftversus-host disease (GVHD) prophylaxis comprised intravenous MP, 0.5 mg/kg twice a day on days 1 to 4, followed by 1 mg/kg twice a day on days 5 to 19 or until the first day that the absolute neutrophil count (ANC) reached 500/mm 3, at which time the dose was tapered at a rate of 0.2 mg/kg/week. Cyclosporine was begun on day 3 and was continued to at least day 180, at which point the dose was tapered at a rate of 5% per week of the initial dose if the patient exhibited no evidence of GVHD. Each patient underwent transplantation with only 1 UBC unit. Initial HLA typing was done by low/intermediate molecular typing for HLA-A and HLA-B alleles and high-resolution molecular typing for HLA-DRB1. Initial eligibility criteria required at least a 4 of 6 match, or a3of6match if the match was based on high-resolution molecular typing for HLA-A and -B. Most patients initially tested with low/intermediate molecular typing were retrospectively retyped with high-resolution molecular typing for HLA-A and HLA-B alleles (ie, final HLA typing). For analysis purposes, the final HLA typing was used. Previous infection with herpesviruses (herpes simplex virus [HSV], cytomegalovirus [CMV], and varicella zoster virus [VZV]) was determined by standard pretransplantation serology of the recipients. Serology was performed in the clinical laboratories of the participating transplantation centers according to institutional procedures. All UBC units used for transplantation were negative for CMV IgM. Surveillance for CMV reactivation posttransplantation was done according to institutional policy; CMV antibody negative recipients received CMV-negative or leukocytedepleted blood products. CMV-positive recipients were permitted to receive prophylaxis with ganciclovir after day 100 once their ANC reached 750/mm 3 for 2 days according to institutional policy. HSV-positive recipients could receive prophylactic acyclovir. Pneumocystis carinii and fungal prophylaxes and intravenous immunoglobulin administration were given according to institutional practice. Clinically significant infections and their causative organisms were recorded, as were the causes of death. All data were retrospectively reviewed centrally for accuracy. Assessment of Immune Reconstitution Peripheral blood was collected in preservative-free heparin from UCBT recipients at 1, 3, 6, 9, 12, 18, 24, and 36 months after transplantation. Specimens were transported overnight to the Children s Hospital Los Angeles for central evaluation of both immunophenotype and antigen-specific T-lymphocyte proliferation. Whole blood specimens were analyzed for the percentages and absolute number of CD3-, CD4-, CD8-, CD19-, CD4-, CD45RA-, CD45RO-, and CD56- expressing leukocytes using 3-color immunofluorescence and fluorescent activated cell sorter analysis. Mononuclear cells were isolated on discontinuous gradients (Ficoll 1.077). The interface cells were collected, washed and suspended at 5 10 5 cells/ml in RPMI 1640 medium with 10% heat-activated human A serum, glutamine, and antibodies. Then 1 10 5 cells (0.2 ml) were added to U-bottom microtiter wells, and the optimal concentration of antigen was added. Lysates of HSV-, CMV-, and VZV-infected fibroblasts were obtained from Dr. Myron Levin, University of Colorado, Denver, CO. Tetanus toxoid (Aventis Pasteur, Swiftwater, PA) was used as a specificity control in recipients. Assays were performed in triplicate. Cultures were pulsed with tritiated thymidine (1 Ci/well) on day 6. After 18 hours, cells from the stimulated and control wells were harvested, and the incorporated radioactivity was determined. Increases in counts per minute (CPM) 3000 (meanstimulated cells minus mean-control cells) were considered positive [6]. Statistical Considerations Kaplan-Meier estimates of the cumulative incidence of positive antigen-specific responses were calculated, with recipients who did not develop a positive response censored at the time of their last negative evaluation. Differences between cumulative incidence curves were tested by the log-rank test. Among recip-

Antigen-Specific Immune Reconstitution 1337 ients who developed positive antigen-specific responses, the times to first positive response were compared using nonparametric tests (Wilcoxon). All calculations were done using SAS, release 8.2 (SAS Institute, Cary, NC). RESULTS Study Population The herpesvirus-specific T-lymphocyte proliferative responses of 153 recipients were determined on 1 or more occasions. Overall, 415 assays of T-lymphocyte proliferation to the herpesviruses were performed. Forty-five recipients were tested once; 36 recipients, twice; 26 recipients, three times, and 46 recipients, 4 or more times. Of the 153 recipients, 66 recipients had a positive proliferative response to 1 or more viruses, whereas 87 recipients had no detectable responses to any virus over the 4 years of evaluation (Table 1). Antigen-Specific T-Lymphocyte Responses HSV. Forty recipients developed a positive proliferative response to HSV, with the first positive response detected on day 29 and the last conversion to a positive response occurring on day 1120. Twenty-five recipients developed a positive proliferative response during the first 12 months after UCBT. Sixty-six recipients had serologic evidence of pretransplantation HSV infection (Table 1). CMV. Thirty-two recipients developed a positive proliferative response to CMV, with the first positive response detected on day 44 and the last conversion to a positive response occurring on day 1506. Thirteen recipients developed a proliferative response during the first 12 months after UCBT. Sixty recipients had pretransplantation serologic evidence of pretransplantation CMV infection. VZV. Fifty recipients developed a positive proliferative response to VZV, with the first positive response detected on day 94 and the last conversion to a positive response occurring on day 1121. Twenty-one recipients developed a positive proliferative response during the first 12 months after UCBT. Eighty recipients were serologically positive for VZV before transplantation. Overall, 28 recipients developed positive proliferative responses to only 1 herpesvirus, 20 recipients did so to 2 viruses, and 18 recipients did so to all 3 viruses, indicating that their proliferative responses were antigen-specific (Table 2). Eighty-seven recipients did not develop detectable antigen-specific T-lymphocyte proliferation to any of the 3 herpesviruses during the evaluation period. Table 1. Characteristics of 153 UCBT Recipients Assessed for Antiherpesvirus T-lymphocyte Proliferative Responses Antigen-Specific Response Negative (n 87) Positive (n 66) Impact of Infection on the Development of Antigen-Specific Responses All (n 153) Patient-related Age, years Median 5.4 7.8 6.3 Range 0.4 17 0.6 17.9 0.4 17.9 Sex, no. (%) Male 48 (55) 46 (70) 94 (62) Female 39 (45) 20 (30) 59 (38) Lansky score < 80 15 (17) 9 (14) 24 > 80 72 (83) 57 (86) 139 Disease-related Primary disease, no. (%) Malignant 79 (91) 58 (88) 137 (90) Nonmalignant 8 (9) 8 (12) 16 (10) Infection-related Pretransplantation HSV status, no. (%) Negative 31 (47) 25 (45) 56 (46) Positive 35 (53) 31 (55) 66 (54) Pretransplantation CMV status, no. (%) Negative 32 (45) 37 (64) 69 (53) Positive 39 (55) 21 (36) 60 (47) Pretransplantation VZV status, no. (%) Negative 20 (34) 11 (21) 31 (28) Positive 38 (66) 42 (79) 80 (72) Pretransplantation CMV/ VZV/HSV status, no. (%) All negative 12 (17) 8 (14) 20 (16) Positive for at least 1 virus 59 (83) 50 (86) 109 (84) Posttransplantation viral infection, no. (%) Noninfected 53 (61) 31 (47) 84 (55) Infected 34 (39) 35 (53) 69 (45) Transplantation-related HLA (final typing), no. (%) < 4/6 59 (68) 34 (52) 93 (61) > 5/6 28 (32) 32 (48) 60 (39) Acute GVHD (reviewed grade), no. (%) < 1 54 (62) 39 (59) 93 (61) > 2 33 (38) 27 (41) 60 (39) Chronic GVHD, no. (%) No 60 (69) 50 (76) 110 (72) Yes 27 (31) 16 (24) 43 (28) Cell dose, cells/kg 10 8 Median 6.5 6.2 6.3 Range 1.4 29.1 1.7 20.4 1.4 29.1 Because the majority (84%) of evaluable recipients had serologic evidence of previous herpesvirus infection, we hypothesized that viral reactivation/infection after UCBT was likely to be the initial antigenic

1338 G. Cohen et al. Table 2. Specificity of Anti-herpesvirus T-lymphocyte Responses Number of Antigens with Positive Responses Antigen (no. of recipients positive) HSV CMV VZV 1 7 5 16 2* 15 9 16 3 18 18 18 Total 40 32 50 *HSV/CMV, 4; HSV/VZV, 11; CMV/VZV, 5. Figure 2. Impact of posttransplantation viral infection on the development of anti-herpes T-lymphocyte responses. Results are the first positive response to the individual virus or the last negative response: (A) HSV; (B) CMV; (C) VZV. A CPM 3000 was considered positive. Figure 1. Time from transplantation to first positive antigen response for 31 patients with no recorded clinical viral infections and 35 patients with viral infections. No significant difference is seen between the 2 samples (P.14; Wilcoxon s test). stimulus for the development of antigen-specific T lymphocytes in most recipients. To determine the impact of clinical infection on developing anti-herpesvirus T-lymphocyte responses, we compared the time of detection of the first positive antigenic-specific response to any virus in recipients who did and did not have clinically detected infections (Figure 1). There was no difference in the timing or the frequency of development of antigen-specific T- lymphocyte responses based on whether clinical infection had or had not occurred (P.14; Wilcoxon s test). Thus, recipients without clinical infection (ie, noninfected) were as likely to develop antigen-specific T-lymphocyte responses as recipients with any detectable infection (ie, infected). The impact of posttransplantation infection on the development of antigen-specific immunity to the individual viruses was also determined (Figure 2). Increases in CPM of 3000 over background ( CPM) were considered positive. There was no association between the time of appearance or the frequency of positivity and clinical infection with any of the 3 herpes viruses. Furthermore, recipients with detectable CMV antigenemia were not more likely to develop positive proliferative responses to CMV than were recipients without antigenemia (data not shown).

Antigen-Specific Immune Reconstitution 1339 Table 3. Causes of Infectious Deaths in UCBT Recipients with and without Antigen-specific T-lymphocyte Responses to Herpesviruses Positive (n 1) Negative (n 11) Viral: Human herpesvirus-6 Viral CMV (2) Adenovirus Epstein-Barr virus Fungal Aspergillus (2) Candida (2) Bacterial Klebsiella Pseudomonas aeruginosa Other Mycobacterium avium intracellulare Impact of a Positive Antigen-Specific T-Lymphocyte Response on Death Due to Infection Although clinical infection did not increase the likelihood of developing antigen-specific proliferation, the presence of an antigen-specific T-lymphocyte response decreased the likelihood of recipients dying from clinical infection. Eleven recipients who did not have detectable antigen-specific T lymphocytes died due to infection, whereas only 1 recipient with a positive response died (P.02). Among these 11 negative recipients, 4 recipients died of viral infection, 4 of fungal infection, 2 of bacterial infection, and 1 due to mycobacterium, whereas the single recipient who had antigen-specific function died of a viral (human herpesvirus-6) infection (Table 3). The variables listed in Table 1 were analyzed for their association with the likelihood of developing an antigen-specific response to 1 or more herpes viruses. No significant associations (P.01) were found for acute GVHD, chronic GVHD, HLA match, recipient age, total cell dose, CD34 cell dose/kg, or pretransplantation remission status. Because viral reactivation after UCBT could be an antigenic stimulus for the development of antigenspecific T lymphocytes, we analyzed the impact of pretransplantation herpesvirus seropositivity on the development of antigen-specific responses after transplantation (Figure 3). Seronegative recipients were as likely as seropositive recipients to develop positive proliferative responses to all 3 herpesviruses. These results indicate that (1) many recipients who are serologically negative for herpesvirus infection before UCBT may have had previous infections, and (2) undetected viral infection/reactivation is a common source of antigenic stimulation after UCBT. Analysis of the kinetics of immunophenotypic reconstitution after UCBT (CD3, CD4, CD8, and CD19) showed no association of any immunophenotype with the development of antigen-specific T-lymphocyte responses (Figure 4). DISCUSSION In the present study, we found that 66 of 153 UCBT recipients (43%) developed T lymphocytes with specificity for 1 or more herpesviruses during the first 4 years after UCBT, and that the first antigenspecific responses could be detected as early as 29 100 days after UCBT, depending on the virus. Because cord blood does not contain antigen-specific memory T lymphocytes, the antigen-specific T lymphocytes detected during the first 12 months after UCBT are most likely derived from the naive T lymphocytes Figure 3. Cumulative incidence of antigen-specific T-lymphocyte proliferation in UCBT recipients who were serologically positive and negative before transplantation. Results are censored at the last negative evaluation. Log-rank test P values: (A) HSV,.59); (B) CMV,.86; (C) VZV,.81.

1340 G. Cohen et al. Figure 4. Absolute lymphocyte counts of recipients who did and did not develop antigen-specific T- lymphocyte responses to 1 or more herpes viral antigens: (A) CD3; (B) CD4; (C) CD8; (D) CD19. infused at the time of transplantation. Thus, the naive T lymphocytes contained in the UCB inoculum were capable of differentiating into antigen-specific T lymphocytes with specificity for environmental pathogens within the first 100 days after transplantation. The presence of standard post-ucbt immunosuppression (cyclosporine and steroids) did not interfere with the development of antigen-specific T lymphocytes. The antigen-specific T lymphocytes were capable of normal interleukin (IL)-2 production, because the addition of exogenous IL-2 did not elicit antigen-specific proliferation that was not present without the addition of IL-2 (R. Parkman, unpublished data). However, the presence of antigen-specific proliferation does not indicate the presence of fully functional antigen-specific T lymphocytes. We and others have described defects in -interferon production and cytotoxic T-lymphocyte function in HSCT recipients of bone marrow in whom antigen-specific T-lymphocyte proliferation was detected [7,8]. More than 40 years ago, it was established that newborn infants immunized with Salmonella flagella antigen could produce specific antibody [9]. The presence of intrauterine viral infections can be detected by the presence of specific IgM antibodies at birth [10]. Infants born of mothers infected with Plasmodium falciparum malaria have antigen-specific T lymphocytes detected in their cord blood due to transplacental malaria antigens [11]. Therefore, it is not surprising that the naïve T lymphocytes present in UCB are capable of generating antigen-specific T lymphocytes within months of transplantation. Recipients of T-cell depleted HSCT do not have detectable immunophenotypic T lymphocytes or proliferative responses to mitogen stimulation until 3 months after HSCT [12]. T lymphocytes derived from the newly engrafted donor HSCs can contribute to the generation of antigen-specific T lymphocytes once thymopoiesis is established. Thus, antigen-specific T lymphocytes first detected more than 1 year after transplantation may be derived from either the naive T lymphocytes contained in the UBC innoculum or recent thymic emigrants. However, antigenspecific T lymphocytes detected before 1 year after transplantation are derived predominantly from the infused naive T lymphocytes. Future studies should include T cell excision circles and other analyses of thymopoiesis to assess its impact on the development of antigen-specific immunity. To eliminate the possibility that the antigen-specific T lymphocytes detected after transplantation were derived from antigen-specific T lymphocytes contained in the UBC unit secondary to in utero exposure to herpesvirus antigens, we compared the development of CMV-specific T lymphocytes between recipients who had received UBC units from mothers seropositive (IgG) for CMV and mothers seronegative for CMV. We found no association between maternal CMV antibody status and the development of CMV-specific T lymphocytes (data not shown). Non cord blood HSC products (bone marrow, mobilized peripheral blood cells) contain both naive and mature antigen-specific T lymphocytes. Recipients of mobilized peripheral blood stem cells have antigen-specific T lymphocytes continuously present after transplantation [13]. Consequently, much recent research has focused on the homeostatic expansion of donor-derived antigen-specific T lymphocytes as a source of posttransplantation immunity [14,15]. The present results suggest that the naive T lymphocytes contained in the HSC inoculum may make a greater

Antigen-Specific Immune Reconstitution 1341 contribution to antigen-specific immunity early after HSCT than had previously been realized. An antigen-specific T lymphocyte response requires both a competent immune system and a stimulating antigen. In the present study, we chose to not immunize UCBT recipients with a neoantigen, but rather to follow the development of antigen-specific T-lymphocyte immunity to the herpesviruses secondary to viral reactivation or a de novo infection [16]. Because the time of viral reactivation/infection is unknown for most recipients, the lack of an antigenspecific T-lymphocyte response at any particular time may be due to a lack of antigeneic stimulation rather than to immunoincompetence. Most recipients responded to only 1 or 2 herpesviruses and not to tetanus toxoid (TT), demonstrating the specificity and the lack of breadth of their proliferative responses. Future clinical studies of UCBT recipients should include scheduled immunizations starting at the time of transplantation with an antigen (ie, TT) to determine the kinetics of immune reconstitution. Sixteen UCBT recipients developed proliferative responses to TT 1 or more years after UCBT after routine posttransplantation reimmunization. A pilot study was undertaken to evaluate the response of UCBT recipients to TT immunization early (90 days) after transplantation. Ten recipients were immunized, 1 of whom developed an antigen-specific proliferative response. Because antigen-specific T lymphocytes are present in non cord blood HSC products, recall antigens like TT or herpesvirus antigens cannot be used to evaluate the de novo development of posttransplantation antigenspecific T-lymphocyte function unless the donor is negative. Non cord blood transplantation recipients should be immunized at the time of transplantation with a neoantigen, such as keyhole limpet hemocyanin (KLH), to determine whether antigen-specific T-lymphocyte immunity can develop in the immediate posttransplantation period [17]. The present study demonstrates that pretransplantation serology underestimates the frequency of previous herpesvirus infection. In fact, the earliest response to HSV after UCBT (day 29) was in a recipient who was serologically negative for HSV. Because many of the algorithms used for the peritransplantation administration of prophylactic anti-herpesvirus drugs are based on serologic assessments, some recipients who might benefit from peritransplantation prophylaxis may not be treated because of false-negative serologic results [18,19]. The in vitro assessment of anti-herpes T-lymphocyte proliferation in serologically negative leukemic patients has demonstrated a 10% incidence of positive T-lymphocyte proliferative response (R. Parkman, unpublished data). Presumably, the lack of detectable pretransplantation antibody was due to the immunosuppressive effect of previous antileukemic therapy. Because cord blood contains only naive T lymphocytes, concerns have been raised that UCBT recipients might be at increased risk of opportunistic infections, particularly with viral and fungal organisms, to which antigen-specific T lymphocytes might be present in other HSC sources. The potential functional immaturity of cord blood T lymphocytes also might contribute to the inability of UCBT recipients to respond fully to opportunistic infections [20,21]. The present data demonstrate that the naive T lymphocytes present in UCB are capable of generating T lymphocytes with specificity for herpesviruses within the first 100 days after transplantation, which may contribute to a decreased incidence of infectious deaths after transplantation. In addition, UCBT recipients who develop antigen-specific immunity have a decreased likelihood of acute leukemic relapse (P.0003) and improved relapse-free survival (P.0001) [22]. It will be difficult in non cord blood transplantation settings to determine the singular contribution of the naive T lymphocytes contained in the HSC inoculum to posttransplantation immune reconstitution. However, after UCBT, naive T lymphocytes can develop into antigen-specific T lymphocytes early after transplantation and may contribute to both a decreased likelihood of death due to infection and improved overall survival. ACKNOWLEDGMENTS This work was supported by the National Heart, Lung and Blood Institute through contracts N01-HB- 67135 (R.P., K.I.W., B.M.), N01-HB-67132 (G.C., S.L.C., N.A.K.), N01-HB-67139 (J.E.W.), N01-HB- 67138 (J.K.), and N01-HB-67113 (E.G.) and grant P01-CA100265 (R.P.). We are indebted to Ms. Manuela Alvarez-Wilson and Ms. Angela Norman for their assistance in the preparation of this manuscript. REFERENCES 1. Noel DR, Witherspoon RP, Storb R, et al. Does graft-versushost disease influence the tempo of immunologic recovery after allogeneic human marrow transplantation? An observation on 56 long-term survivors. Blood. 1978;51:1087-1105. 2. Ochs L, Snu XO, Miller J, et al. Late infections after allogeneic bone marrow transplantations: comparison of incidence in related and unrelated donor transplant recipients. Blood. 1995;86: 3979-86. 3. Parkman R, Weinberg KI. Immunological reconstitution following bone marrow transplantation. Immunol Rev. 1997;157: 73-78. 4. Mackall CL, Gress RE. Pathways of T-cell regeneration in mice and humans: implications for bone marrow transplantation and immunotherapy. Immunol Rev. 1997;157:61-72. 5. Storek J, Dawson MA, Maloney DG. Correlation between the numbers of naive T cells infused with blood stem cell allografts

1342 G. Cohen et al. and the counts of naive T cells after transplantation. Biol Blood Marrow Transplant. 2003;9:781-784. 6. Petersen JM, Weinberg KI, Annett G, et al. Correlation of antigen-specific T-lymphocyte function by recombinant cytokines in children infected with human immunodeficiency virus type 1. J Pediatr. 1992;121:565-568. 7. Levin MJ, Parkman R, Oxman MN, et al. Proliferative and interferon responses by peripheral blood mononuculear cells after bone marrow transplantation in humans. Infect Immun. 1978;20:678-684. 8. Reusser R, Riddell SR, Meyers JD, et al. Cytotoxic T-lymphocyte response to cytomegalovirus after human allogeneic bone marrow transplantation: pattern of recovery and correlation with cytomegalovirus infection and disease. Blood. 1991;78: 1373-1380. 9. Smith RT, Eitzman DV, Catlin ME, et al. The development of the immune response: characterization of the response of the human infant and adult to immunization with salmonella vaccines. Pediatrics. 1964;33:163-183. 10. Fung JC, Tilton RC. TORCH serologies and specific IgM antibody determination in acquired and congenital infections. Ann Clin Lab Sci. 1985;15:204-211. 11. Brustoski K, Moller U, Kramer M, et al. U. IFN-gamma and IL-10 mediate parasite-specific immune responses of cord blood cells induced by pregnancy-associated Plasmodium falciparum malaria. J Immunol. 2005;174:1738-1745. 12. Small TN, Papadopoulos EB, Boulad F, et al. Comparison of immune reconstitution after unrelated and related T-cell depleted bone marrow transplantation: effect of patient age and donor leukocyte infusions. Blood. 1999;93:467-480. 13. Ottinger HK, Beelen DW, Scheulen B, et al. Improved immune reconstitution after allotransplantation of peripheral blood stem cells instead of bone marrow. Blood. 1996;88: 2775-2779. 14. Lenz DC, Kurz SK, Lemmens E, et al. IL-7 regulates basal homeostatic proliferation of antiviral CD4 T cell memory. Proc Natl Acad Sci USA. 2004;101:9357-9362. 15. Giver CR, Li JM, Hossain MS. Reconstructing immunity after allogeneic transplantation. Immunol Res. 2004;29:269-282. 16. Winston DJ, Huang ES, Miller MJ, et al. Molecular epidemiology of cytomegalovirus infections associated with bone marrow transplantation. Ann Intern Med. 1985;102:16-20. 17. Reichardt VL, Okada CY, Liso A, et al. Idiotype vaccination using dendritic cells after autologous peripheral blood stem cell transplantation for multiple myeloma: a feasibility study. Blood. 1999;93:2411-2419. 18. Guidelines for preventing opportunistic infections among hematopoietic stem cell transplant recipients. MMWR Recomm Rep. 2000;49(RR-10):1-125. 19. Ljungman P, Reusser P, de la Camara R, et al. Management of CMV infections: recommendations from the infectious diseases working party of the EBMT. Bone Marrow Transplant. 2004;33: 1075-1081. 20. Harris DT, Schumacher MJ, Locascio J. Phenotypic and functional immaturity of human umbilical cord blood T lymphocytes. Proc Natl Acad Sci USA. 1992;89:10006-10010. 21. Nonoyama S, Penix LA, Edwards CP, et al. Diminished expression of CD40 ligand by activated neonatal T cells. J Clin Invest. 1995;95:66-75. 22. Parkman R, Cohen G, Carter SL, et al. Successful immune reconstitution decreases leukemic relapse and improves survival in recipients of unrelated cord blood transplantation. Biol Blood Marrow Transplant. 2006, in press.