In Kidney Transplant Patients, Alemtuzumab but Not Basiliximab

Home , Alemtuzumab, Everolimus

In Kidney Transplant Patients, Alemtuzumab but Not Basiliximab/Low-Dose Rabbit Anti-Thymocyte Globulin Induces B Cell Depletion and Regeneration, Which This information is current as Associates with a High Incidence of De Novo of September 25, 2021. Donor-Specific Anti-HLA Antibody Development Marta Todeschini, Monica Cortinovis, Norberto Perico, Francesca Poli, Annalisa Innocente, Regiane Aparecida Downloaded from Cavinato, Eliana Gotti, Piero Ruggenenti, Flavio Gaspari, Marina Noris, Giuseppe Remuzzi and Federica Casiraghi J Immunol published online 2 August 2013 http://www.jimmunol.org/content/early/2013/08/01/jimmun ol.1203261 http://www.jimmunol.org/

Supplementary http://www.jimmunol.org/content/suppl/2013/08/02/jimmunol.120326 Material 1.DC1 by guest on September 25, 2021 Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2013 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published August 2, 2013, doi:10.4049/jimmunol.1203261 The Journal of Immunology

In Kidney Transplant Patients, Alemtuzumab but Not Basiliximab/Low-Dose Rabbit Anti-Thymocyte Globulin Induces B Cell Depletion and Regeneration, Which Associates with a High Incidence of De Novo Donor-Speciﬁc Anti-HLA Antibody Development

Marta Todeschini,*,† Monica Cortinovis,*,† Norberto Perico,*,† Francesca Poli,‡ Annalisa Innocente,‡ Regiane Aparecida Cavinato,*,† Eliana Gotti,† Piero Ruggenenti,† Flavio Gaspari,*,† Marina Noris,*,† Giuseppe Remuzzi,*,† and Federica Casiraghi*,† Downloaded from In this single-center matched-cohort study, we evaluated the phenotype of repopulating B cells and its correlation with donor- specific anti-HLA Ab development and long-term graft function in 16 renal transplant recipients and 32 age- and gender- matched controls induced with alemtuzumab or basiliximab (Bas)/low-dose rabbit anti-thymocyte globulin (rATG), respectively. Alemtuzumab, but not Bas/rATG, profoundly depleted peripheral B cells in the first 2 mo posttransplantation. Early posttransplant, naive B cells were significantly depleted, whereas Ag-experienced and memory B cells were partially spared. Transitional B cells

transiently increased 2 mo posttransplant. At month 6 posttransplant, pregerminal center B cells emerged, a process promoted by http://www.jimmunol.org/ increased BAFF serum levels. Thereafter, B cell counts increased progressively, mainly due to expansion of naive B cells. Conversely, Bas/rATG did not modify the B cell phenotype throughout the follow-up period. Alemtuzumab was associated with a higher incidence of de novo DSA compared with Bas/rATG. DSA development was predicted by changes in the B cell compartment and correlated with worse long-term graft function. Thus, alemtuzumab-induced B cell depletion/reconstitution may promote chronic humoral responses against the graft. The Journal of Immunology, 2013, 191: 000–000.

mproved immunosuppression strategies allowed the reduced Induction therapy with alemtuzumab (Campath-1H), a human- incidence and severity of acute rejections after kidney trans- ized monoclonal anti-CD52 Ab inducing near-immediate and pro- I plantation, but they failed to significantly affect the long-term found T and B lymphocyte, NK cell, and monocyte depletion, by guest on September 25, 2021 outcome of the graft (1). Side effects of chronic maintenance allowed successful transplantation in the setting of minimal main- immunosuppression, including increased patient cardiovascular tenance immunosuppression. However, alemtuzumab treatment was morbidity and mortality, and late immunoreactivity against the associated with an increased risk for the development of donor- allograft are the major causes of graft loss in the long-term. Thus, specific anti-HLA Abs (DSA) and with an excess risk for early novel antirejection regimens have been implemented to yield Ab-mediated rejection, particularly in a calcineurin inhibitor–free similar short-term results, limiting the long-term sequelae of chronic immunosuppressive regimen (3–5). The above data converge to immunosuppression and promoting a tolerance-permissive envi- suggest that, in the clinical setting of minimal maintenance immu- ronment (2). nosuppression, alemtuzumab-induced lymphocyte depletion might result in activation of the humoral response toward alloantigens. In contrast, more recent investigations showed that, in kidney trans- *Centro Ricerche Trapianti “Chiara Cucchi de Alessandri e Gilberto Crespi” Istituto plant patients, lymphocyte depletion induced by alemtuzumab was Di Ricovero e Cura a Carattere Scientifico, Istituto di Ricerche Farmacologiche “Mario Negri,” Bergamo 24020, Italy; †Dipartimento di Immunologia e Trapianto, followed by a transient increase in transitional B cells, including Azienda Ospedaliera Papa Giovanni XXIII, Ospedali Riuniti, Istituto Di Ricovero B cells with phenotypic characteristics of regulatory B cells, as well e Cura a Carattere Scientifico, Istituto di Ricerche Farmacologiche “Mario Negri,” as long-term dominance in naive B lymphocytes (6, 7), a B cell Bergamo 24020, Italy; and ‡Fondazione Istituto Di Ricovero e Cura a Carattere Scientifico Ca` Granda, Ospedale Maggiore Policlinico, Milan 20122, Italy repertoire similar to that described to be prominent in the periph- Received for publication November 27, 2012. Accepted for publication July 2, 2013. eral blood of operationally tolerant renal allograft recipients (8). More conventional induction therapy with rabbit anti-thymocyte This work was supported by the Fondazione ART per la Ricerca sui Trapianti (Milan, Italy) and Ricerca Corrente 2011 Cod. No. 580/01 Fondazione Istituto Di Ricovero globulin (rATG) has a limited risk for acute humoral rejection e Cura a Carattere Scientifico Ca` Granda, Ospedale Maggiore Policlinico (Milan, (9), but it has major side effects: acute cytokine release syndrome, Italy). R.A.C. is a recipient of a fellowship from Fondazione ART. increased opportunistic infections, including CMV reactivation, Address correspondence and reprint requests to Dr. Federica Casiraghi, Istituto Di and posttransplant lymphoproliferative disease (10, 11). Induction Ricovero e Cura a Carattere Scientifico, Istituto di Ricerche Farmacologiche “Mario Negri,” Via Gabriele Camozzi 3, Ranica (Bergamo) 24020, Italy. E-mail address: therapy with very low-dose rATG combined with the mAb anti– [email protected] IL-2R basiliximab (Bas), given with the aim of inhibiting the The online version of this article contains supplemental material. activity of T cells surviving the lymphocytolytic effect of rATG, Abbreviations used in this article: Bas, basiliximab; DSA, donor-specific anti-HLA prevented acute graft rejection in kidney transplant patients as Ab; GFR, glomerular filtration rate; MMF, mycophenolate mofetil; rATG, rabbit anti- effectively as did full-dose rATG monotherapy but with remark- thymocyte globulin; ROC, receiver operating characteristic. ably fewer side effects (12, 13) and negligible risk for acute hu- Copyright Ó 2013 by The American Association of Immunologists, Inc. 0022-1767/13/$16.00 moral rejection.

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1203261 2 ALEMTUZUMAB-INDUCED B CELL CHANGES AND DSA

Altogether, the above observations suggest that the risk for Ab- Peripheral B lymphocyte counts and phenotype mediated graft rejection associated with lymphocyte-depleting in- The absolute count of B cell populations was determined using a single- duction therapy in kidneytransplantationvaries greatly, according to platform method (Multitest four-color Abs and Trucount tubes; BD Bio- the specific regimen adopted. We hypothesized that such differ- science, San Jose, CA). PBMC samples were incubated with different ences may depend on the different effects on B cell depletion and fluorochrome-conjugated murine mAbs against human CD3, CD19, CD27, lymphopenia-induced proliferation of B cell subsets. To formally CD38, IgD, CD24, and CD52 Ags (BD Bioscience and BioLegend). The samples were analyzed using a FACSAria flow cytometer (BD Bioscience) test this hypothesis, we conducted a single-center, matched-cohort and FlowJo software. For each marker, blank samples with isotype-matched studyin16 consecutivepatients induced withalemtuzumab andin32 control Abs were analyzed. gender- and age-matched controls induced with Bas/rATG to PBMCs were collected pretransplant at 15–30 d, and at 2, 6, 12, and .24 evaluate the profile of the different circulating B cell subsets in the mo posttransplant. Results obtained from PBMCs collected from 24–48 mo after transplant were pooled and considered .24 month posttrans- early and late posttransplant period, the development of de novo plant. DSA and its impact on long-term graft dysfunction, and the correlations between changes in specific B cell subpopulations and the BAFF and alemtuzumab assays development of de novo DSA. SolubleBAFFwasmeasuredinserumsamplesofkidneytransplantrecipients and healthy volunteers using a commercially available ELISA (R&D Sys- Materials and Methods tems, Minneapolis, MN), following the manufacturer’s instructions. Patients and immunosuppressive therapies Alemtuzumab serum levels also were assessed by ELISA (Supplemental Fig. 1). In this single-center, matched-cohort study, we evaluated the longitudinal profile of B cell subsets, de novo DSA development over a 2-y follow-up, Anti-HLA Ab determination Downloaded from and their impact on long-term graft function in 16 consecutive patients given induction therapy with alemtuzumab and in 32 gender- and age- Anti-HLA Abs in the subjects’ sera collected before surgery and at years 1 matched reference patients given Bas/rATG. The two studies were per- and 2 posttransplant were tested with a bead-based screening assay (19, 20). formed consecutively. For both cohorts, the same inclusion/exclusion Briefly, we used the LABScreen Mixed kit (One Lambda), which simulta- criteria were used: age 18–70 y, non-HLA identical to the donor, nega- neously detects HLA class I and class II Abs with MicroBeads coated with tive cross-match test, and current panel reactive Abs , 10%. All subjects purified Ags. The Single Ag kit (One Lambda) was also used to identify HLA specificities. Results above a cut-off value of 2000 mean fluorescence

were managed by the same transplant physicians, according to the same http://www.jimmunol.org/ standardized protocol of monitoring and nonimmunosuppressive treat- intensity were considered positive. The tests were carried out according to ment. Thus, the two groups of patients differed only with respect to their the manufacturer’s instructions, and the analysis was performed with One immunosuppressive regimens. Protocols were approved by the Ethics Lambda software (HLA Visual Version 2.2). Committee, Azienda Ospedaliera Papa Giovanni XXIII, Bergamo, Italy, Statistics and written informed consent was obtained from all participating patients according to the Declaration of Helsinki. All data were used according to The repeated-measures ANOVA test was used for comparison of variables the standard regulations of the Nord Italia Transplant network for data over time. Categorical values were compared using the Fisher exact test, and registration and use and for the preservation of patients’ anonymity and continuous variables were compared using the Student two-tailed t test. privacy. Individual GFR slopes (GFR changes over time) were calculated by linear- Patients. Study patients received an i.v. infusion of 30 mg alemtuzumab at regression analysis. Computation of the area under the receiver operating the time of surgery (ClinicalTrials.gov identifier: NCT00309270). On day 1, characteristic (ROC) curve for B cell counts and phenotypes was per- by guest on September 25, 2021 they were randomized (1:1) to either low-dose sirolimus or low-dose cy- formed, and the sensitivity and specificity of these parameters for DSA closporine combined with 750 mg twice daily of mycophenolate mofetil positivity were determined. All calculations were made using MedCalc (MMF). The dose of sirolimus was titrated to target trough concentrations 12.3.0.0 (MedCalc Software, Mariakerke, Belgium). The p values , 0.05 of 5–10 ng/ml [by HPLC (14)]. Cyclosporine was initially infused i.v. and were considered statistically significant. was then given per os at doses titrated to target trough blood concentrations of 120–220 ng/ml in the first month postsurgery and 70–120 ng/ml thereafter [by HPLC (15)]. Five hundred milligrams of methylpredniso- Results lone was administered before alemtuzumab infusion and continued for two At the time of transplant, the characteristics of patients (alemtu- additional days posttransplantation (250 and 125 mg, respectively). There- zumab) and reference patients (Bas/rATG) were similar (Table I). after, oral prednisone (75 mg) was administered, which was progressively Characteristics were also similar between patients induced by tapered and discontinued after day 7 postsurgery. alemtuzumab and randomized to cyclosporine or sirolimus main- Sirolimus and cyclosporine trough levels and B cell counts measured through the 24-mo follow-up of the original 16 patients enrolled in this tenance therapy (Supplemental Table I). study, as well as glomerular filtration rate (GFR) values measured through B cell depletion after alemtuzumab or Bas/rATG induction 30-mo follow-up, were reported previously (16, 17) Reference patients. Since the day of transplant, reference patients received Alemtuzumab induced a profound B cell depletion during the first 6 five consecutive rATG (thymoglobulin) infusions at a daily dose of 0.5 mg/ mo posttransplant; subsequently, B cells recovered to pretransplant kg (12, 13). The dose of rATG was reduced or temporary withdrawn when levels and, in the long-term, progressively increased to levels higher WBC or platelet counts decreased to ,2,000/mlor,50,000/ml, respectively. None of the patients experienced lymphopenia or thrombocytope- than those observed in reference patients (Fig. 1). No differences nia, and all received five doses of rATG. Twenty milligrams of Bas was were observed between patients given cyclosporine or sirolimus infused i.v. before transplant and 4 d later. Maintenance immunosuppres- maintenance immunosuppression after alemtuzumab-mediated B sion with low-dose cyclosporine and MMF, as well as steroid treatment, cell depletion, indicating that B cell reconstitution was not influ- was the same as used in alemtuzumab patients. Notably, the study arm with enced by maintenance immunosuppressive drugs. Bas/rATG in- maintenance immunosuppression with low-dose sirolimus/MMF after in- . duction therapy with Bas/rATG was planned in the initial study design. duction therapy did not modify B cell counts during the 24 mo However, the first four renal transplant recipients randomized to the low follow-up period (Fig. 1). sirolimus/MMF maintenance immunosuppressive regimen developed acute allograft rejection. The sirolimus/MMF arm was interrupted, and all pa- Phenotype of repopulating B cells in alemtuzumab and Bas/ tients were given cyclosporine maintenance immunosuppression. rATG cohorts Follow-up and graft outcome The phenotype of repopulating B cells was characterized based on Bm1–Bm5 (21) and IgD-CD27 (22, 23) classifications (Figs. 2A, GFR was measured by plasma clearance of iohexol (18) at 6 and 12 mo posttransplant and at yearly intervals thereafter. At the same time points, 3A) in the alemtuzumab and Bas/rATG kidney transplant cohorts. 24-h urine samples were collected for evaluation of urinary protein ex- In the early posttransplant period, alemtuzumab significantly de- cretion. pleted naive Bm1 and Bm2 cells, as documented by both lower cell The Journal of Immunology 3

Table I. Baseline and clinical characteristics of kidney transplant recipients given induction therapy with either alemtuzumab or Bas/rATG

Alemtuzumab Patients (n = 16) Bas/rATG Reference Patients (n =32) p Value Baseline characteristics Recipient age (y) 48 6 13 44 6 14 0.3252 Recipient gender (%male) 62.5 62.5 1.000 Living donor (%) 12.5 9.4 1.000 HLA-A mismatch 1.57 6 0.51 1.32 6 0.67 0.1473 HLA-B mismatch 1.50 6 0.60 1.30 6 0.57 0.2001 HLA-DR mismatch 1.36 6 0.58 1.08 6 0.49 0.0513 Primary disease (%) Chronic glomerulonephritis 43.8 31.3 0.5242 Polycystic kidney disease 18.8 21.9 1.000 Pyelonephritis 18.8 6.3 0.3164 Other 0 9.4 0.5412 Unknown 18.6 31.1 0.4973 Clinical parameters Cold ischemia time (h) 16 6 3166 3 1.000 Delayed graft function (%) 18.8 18.8 1.000 Acute rejection episodes (%) 18.8 15.6 1.000 Cyclosporine C0 (n =8) (n = 32) At month 6 (ng/ml) 116 6 35 111 6 47 0.7545 Downloaded from At month 12 (ng/ml) 81 6 28 103 6 56 0.2263 At month 48 (ng/ml) 101 6 50 74 6 36 0.0828 Sirolimus C0 (n =8) At month 6 (ng/ml) 7.5 6 2.1 At month 12 (ng/ml) 9.7 6 2.6 At month 48 (ng/ml) 9.3 6 2.7 Proteinuria at month 24 (g/24 h) 0.84 6 0.96 0.25 6 0.27 0.0009 http://www.jimmunol.org/ Proteinuria at month 48 (g/24 h) 1.70 6 2.64 0.23 6 0.16 0.0043

counts (Fig. 2B) and percentages (Fig. 2C) of these B cell subsets at At 6 mo posttransplant, naive Bm1 cells remained significantly 15–30 d and 2 mo posttransplant compared with pretransplant val- depleted in alemtuzumab patients, as reflected by cell counts and ues. Alemtuzumab also significantly depleted Bm29, early Bm5, and percentages that were lower than pretransplant values (Fig. 2B, late Bm5 B cells, as documented by decreases in cell counts at 15– 2C), whereas cell counts and percentages of Bm2, early Bm5, and 30 d (and 2 mo for early and late Bm5 cells) posttransplant (Fig. late Bm5 B cells normalized to pretransplant levels. At 6 mo 2B), although, during the first month posttransplant, the relative posttransplant, the percentages of Bm3+4 cells also had normal- by guest on September 25, 2021 percentages of these B cell subsets did not change (Bm29 B cells) or ized, whereas their counts increased from 2 to 6 mo after trans- increased significantly (early Bm5) (Fig. 2C). A different early plant compared with baseline levels, although not to a statistically posttransplant profile was observed for Bm3+4 cells. Counts of Bm3 significant extent. In contrast, both cell counts and percentages +4 cells (germinal center cells) did not change significantly (Fig. of Bm29 B cells underwent a significant increase compared with 2B), but their relative percentage increased significantly during the pretransplant values; Bm29 B cells represented nearly 20% of the first 2 mo, reaching values that were 10-fold higher than pretrans- peripheral B cells at 6 mo posttransplant (Fig. 2B, 2C). plant levels at 15–30 d posttransplant (Fig. 2C). In the long-term posttransplant, cell counts of Bm1 and late Bm5 cells were significantly higher than pretransplant values, whereas Bm29, Bm3+4, and early Bm5 cell counts returned to pretransplant values (Fig. 2B, 2C). Of note, Bm2 cell counts in the long- term were higher than pretransplant levels and were significantly higher than in the Bas/rATG group. According to IgD/CD27 B cell classification, alemtuzumab significantly depleted all B cell subsets, as documented by IgD/CD27 B cell counts in the first 2 mo posttransplant that were lower than pretransplant numbers (Fig. 3B). The relative percentages of naive and unswitched memory B cells were also reduced in the early posttransplant period (Fig. 3C), whereas the relative percentages of switched memory B cells increased significantly; at 15–30 d posttransplant, they accounted for 70–80% of the residual IgD/ CD27 B cells (Fig. 3C). The percentages of double-negative B cells did not undergo significant changes, with the exception of a transient increase at 2 mo posttransplant (Fig. 3C). At 6 mo FIGURE 1. Kinetics of repopulating B cell counts. B cell counts (cells/ posttransplant, all IgD/CD27 B cell subsets had returned to pre- ml) from baseline (time 0) to .24 mo after transplantation in the peripheral transplant values (Fig. 3B, 3C). In the late posttransplant period, blood of kidney transplant recipients given alemtuzumab induction and either cyclosporine or sirolimus maintenance immunosuppressive therapy cell counts of naive and unswitched B cells increased significantly and in kidney transplant recipients given Bas/rATG induction and cyclo- compared with pretransplant values (Fig. 3B). sporine maintenance immunosuppressive therapy. Data are mean 6 SEM. Changes in both Bm1–Bm5 and IgD/CD27 B cell phenotypes *p , 0.05 versus time 0, xp , 0.05 versus Bas/rATG group at the same induced by alemtuzumab were similar in the cyclosporine and time point. sirolimus groups (Supplemental Table I). 4 ALEMTUZUMAB-INDUCED B CELL CHANGES AND DSA

FIGURE 2. Phenotype of repopulating B cells based on IgD-CD38 expression. (A) Peripheral B cell development and diversification based on Bm1–Bm5 classification. IgD/ CD38 FACS staining provides the “so-called” Bm1–Bm5 classification and can be used to identify multiple subsets, including Bm1 (virgin naive B cells, IgD+CD382), Bm2 (activated naive B cells, IgD+CD38low), Bm29 (pregerminal center cells, IgD+ CD38high), Bm3+4 (centroblasts and centrocytes germinal center cells, IgD2CD38high), and Bm5 (memory B cells, IgD2CD382/low) cells. Bm5 memory cells, which express levels Downloaded from of CD38 ranging from moderately positive to negative, can be divided into early Bm5 (CD38low) and late Bm5 (CD382) cells (21). An infer- ence from this model is that, in peripheral lymphoid organs, naive Bm1

lymphocytes, after Ag encounter, http://www.jimmunol.org/ become Bm2 cells and then develop into germinal center founder Bm29 cells. These cells can differentiate into centroblast Bm3 and centrocyte Bm4 cells (Bm3+4) and, later, into early memory Bm5 cells and late memory Bm5 cells or plasma cells. For the duration of this process, cells representing each subset are released into the blood. Longitudinal proﬁle by guest on September 25, 2021 of Bm1, Bm2, Bm29, Bm3+4, and early and late Bm5 cell counts (cells/ ml) (B) and percentages (on total CD32CD19+ B cells) (C) from baseline to .24 mo after transplant in kidney transplant recipients given either alemtuzumab or Bas/rATG induction therapy. Data are mean 6 SEM. *p , 0.05 versus time 0, ˚p , 0.05 versus 12 and .24 mo posttransplant, xp , 0.05 versus Bas/ rATG group at the same time point.

In reference patients given Bas/rATG, B cell counts did not pression of CD52 Ag was evaluated on Bm1–Bm5 and IgD/CD27 change appreciably (Figs. 2, 3). Specifically, B cell subsets defined B cell subsets before alemtuzumab administration. As shown in based on Bm1–Bm5 and IgD/CD27 were not modified during the Fig. 4A and 4B, ∼12% of Bm3+4 B cells did not express the entire follow-up period (Figs. 2, 3), with the exception of a tran- CD52 Ag, resulting in a significantly lower percentage of CD52+ sient increase in Bm1 and unswitched memory B cell counts Bm3+4 B cells compared with the other Bm1–Bm5 B cell subsets. during the 2 mo after transplant (Figs. 2B, 3B). Because Bm3+4 cells represent 3% of the entire B cell population Representative FACS analysis dot-plots of IgD/CD38 and IgD/ (see gating strategy in Supplemental Fig. 2A), the proportion of CD27 expression at different time points after alemtuzumab in- CD522 Bm3+4 cells in the overall B cell population is ,0.5%. duction are shown in Supplemental Fig. 2. Bm3+4 cells expressed CD52 Ag at a lower intensity compared with the other Bm1–Bm5 B cell subsets (Fig. 4B, 4C). A lower CD52 expression of B cell subsets percentage of CD52+ cells was also observed in early Bm5 B cells, We wondered whether differences in the expression of CD52, although the difference did not reach statistical significance (Fig. the target Ag of alemtuzumab, may account for the differences in 4A). In the IgD/CD27 analysis, a lower percentage of switched B cell composition observed early after transplantation. The ex- memory B cells expressed the CD52 Ag than did all of the other The Journal of Immunology 5

FIGURE 3. Phenotype of repopulating B cells based on IgD-CD27 expression. (A) Schematic overview of B cell developmental stages based on IgD/CD27 classiﬁcation (22). Na- ive B cells are deﬁned as IgD+CD272. After Ag encounter, unswitched memory B cells (IgD+CD27+) undergo germinal center reaction and Ig class switching. IgD2CD27+ class-switched B cells can either be postgerminal center memory B cells or plasmablasts (both switched memory IgD2 CD27+ B cells). Double-negative 2 2

(IgD CD27 ) B cells are thought to Downloaded from develop outside the germinal center in an extrafollicular reaction and be capable of supporting somatic hypermutation and Ig class switching in a CD40-independent manner (23). Longitudinal proﬁle of naive, unswitched memory, switched mem- http://www.jimmunol.org/ ory and double-negative B cell counts (cells/ml) (B) and percentages (of total CD32CD19+ B cells) (C) from baseline to .24 mo after transplant in kidney transplant recipients given either alemtuzumab or Bas/ rATG induction therapy. Data are mean 6 SEM. *p , 0.05 versus time 0, ˚p , 0.05 versus 12 and .24 mo by guest on September 25, 2021 posttransplant, xp , 0.05 versus Bas/ rATG group at the same time point.

subsets (Fig. 4A), although they expressed CD52 at the same in- counts and percentages returned to pretransplant levels and re- tensity as the other IgD/CD27 subsets (Fig. 4C). mained unchanged until .24 mo. Bas/rATG induction therapy These findings indicate that the increase in the relative per- did not modify cell counts or percentages of transitional B cells centage of Bm3+4 and early Bm5 cells among Bm1–Bm5 B cell (Fig. 5). subsets and of switched memory B cells in the IgD/CD27 pop- Alemtuzumab-induced B cell depletion increases serum BAFF ulation were the result of alemtuzumab sparing of CD522 Bm3+4, levels, which correlates with transitional and Bm29 B cell early Bm5, and switched memory B cells. In addition, a lower expansion expression intensity of the CD52 Ag by CD52+ Bm3+4 B cells conferred upon them resistance to alemtuzumab depletion. The homeostatic expansion of B cells is regulated primarily by the cytokine BAFF; under conditions of B cell lymphopenia induced Alemtuzumab induces a transient increase in transitional by B cell–depleting agents, such as alemtuzumab and rituximab, B cells BAFF availability increased markedly (25, 26). In agreement with Recently, human circulating transitional B cells, identified as these studies, we found very high levels of BAFF in serum samples CD24highCD38high cells (24), were reported to be increased after from alemtuzumab-treated patients at 1 and 2 mo posttransplant alemtuzumab induction in kidney transplant recipients (6, 7). compared with pretransplant values, as well as with posttransplant Transitional B cells were present at a very low frequency (,2%) BAFF values in Bas/rATG-treated reference patients (Fig. 6A). In in the peripheral blood of kidney transplant recipients of the two the alemtuzumab group, BAFF levels were higher at all time points cohorts pretransplant. As shown in Fig. 5, alemtuzumab induced an in patients given maintenance immunosuppressive therapy con- early expansion of transitional B cells, as documented by increased sisting of either cyclosporine or sirolimus (Supplemental Fig. 3). cell counts and percentages during the first 2 mo posttransplant (to To establish whether increased BAFF availability accounted a similar extent in cyclosporine and sirolimus groups; Supplemental for changes in the midterm posttransplant of Bm1–Bm5 and IgD/ Table I); subsequently, beginning at 6 mo posttransplant, both cell CD27 B cells, as well as transitional B cells, we performed linear- 6 ALEMTUZUMAB-INDUCED B CELL CHANGES AND DSA Downloaded from

FIGURE 5. Proﬁle of repopulating B cells with a transitional phenotype.

Longitudinal proﬁle of cell counts (cell/ml) and percentages of transitional http://www.jimmunol.org/ CD24highCD38high B cells in kidney transplant recipients, given either alemtuzumab or Bas/rATG induction therapy, from baseline to .24 mo after transplant. Data are mean 6 SEM. *p , 0.05 versus time 0.

tested by ELISA. As shown in Fig. 7, as early as 1 h after Ab infusion, alemtuzumab serum levels reached 1300 ng/ml and de- clined to ∼1000 ng/ml after 12 h. Seven days after Ab administration, serum levels were reduced further and became negligible (,80 ng/ml) at 1 mo. From this point onward, alemtuzumab was by guest on September 25, 2021 no longer detectable in any of the serum samples, excluding a possible association between circulating levels of the anti-CD52 FIGURE 4. Expression of CD52 Ag on B cell subpopulations from pe- Ab and changes in B cell phenotype in the mid- and long-term ripheral blood taken pretransplant from kidney transplant recipients given period after alemtuzumab administration. alemtuzumab induction therapy. (A) Percentages of CD52+ B cells among the Bm1–Bm5 B cells and switched/unswitched/double-negative/naive B cells in De novo DSA development after alemtuzumab or Bas/rATG the peripheral blood taken pretransplant from kidney transplant recipients induction therapy and long-term graft functional changes B given alemtuzumab. ( ) Overlay graphs of CD52 expression on each Bm1– None of the patients or reference patients had DSA in serum C Bm5 B cell subtype. ( ) Median fluorescence intensity of CD52 expression by samples at the time of transplantation. The incidence of de novo Bm1–Bm5 B cells and by switched/unswitched/double-negative (DN)/naive B DSA at 1 y posttransplantation was significantly higher in alem- cells pretransplantation from kidney transplant recipients given alemtuzumab induction therapy. Data are mean 6 SEM, *p , 0.05 versus all other Bm tuzumab patients than in Bas/rATG-treated reference patients (p= subpopulations, xp , 0.05 versus unswitched, DN, and naive B cells. 0.011, Fig. 8A). In the alemtuzumab group, 50% of the patients positive for DSA had Abs against HLA class I Ags, and 50% had regression analyses of BAFF levels and B cell percentages in the Abs against HLA class II Ags (antidonor DQ, Fig. 8B). DSA combined kidney transplant cohorts (Fig. 6B). Intriguingly, the development was not influenced by maintenance immunosuppres- peak BAFF serum levels at 1 mo posttransplant correlated inversely sive therapy with cyclosporine or sirolimus (Supplemental Table with the percentages of Bm1 B cells at 6 mo, whereas they corre- I). In the Bas/rATG group, 25% of the reference patients positive lated positively and significantly with the percentages of transi- for DSA had Abs against HLA class I, 50% had Abs against HLA tional B cells 2 mo after transplantation and with the percentages of class II, and 25% had Abs against HLA class I and II Ags (Fig. Bm29 cells at 6 mo posttransplant (Fig. 6B), indicating a close re- 8B). At year 2 posttransplant, the frequency of subjects positive lationship between early posttransplant BAFF availability and the for DSA was still significantly greater among patients compared expansion of transitional and Bm29 B cells. BAFF serum levels did with reference patients (p=0.010, Fig. 8A). All individuals who not correlate with the percentages of the other Bm1–Bm5 or IgD/ developed DSA by 1 y posttransplant maintained positivity for CD27 B cells (Fig. 6B). these Abs at 2 y postsurgery. None of the patients or reference patients experienced acute humoral rejection. Circulating levels of alemtuzumab decline rapidly early In alemtuzumab-treated patients, 24-h urinary protein excretion posttransplantation at 2 and 4 y posttransplant was significantly higher (Table I) and the For 13 of the 16 patients given alemtuzumab, serum samples were rate of GFR decline (over 4-y follow-up) was significantly faster available at both early (1 h, 12 h, 7 d) and late (1, 2, 6, 12, 24 mo) (p=0.0284, Fig. 8C) compared with Bas/rATG reference patients. time points posttransplantation, and alemtuzumab levels were The analysis of two cohorts with regard to calcineurin inhibitor The Journal of Immunology 7

sporine (22.9456 6 2.8574 ml/min/1.73 m2/y versus 20.2112 6 3.0810 ml/min/1.73 m2/y, respectively; p = 0.0354). Combined analysis of the two cohorts based on the presence or absence of DSA showed that the rate of GFR decline over 4 y posttransplant was signiﬁcantly faster in subjects who had developed de novo DSA 1 y posttransplant compared with DSA- negative individuals (Fig. 8D). Among subjects without DSA at 1 y, the GFR slope was similar following alemtuzumab and Bas/ rATG induction (p=0.0974). All of the other speciﬁed characteristics of subjects, with and without de novo DSA at 1 y posttransplant, were comparable, with the exception of a trend toward worse proteinuria at 48 mo posttransplant in DSA-positive patients compared with subjects who did not develop de novo DSA (Table II). Relationship between B cell depletion/reconstitution and de novo DSA development We investigated whether early posttransplant changes in the B cell

compartment could predict DSA development in the combined Downloaded from cohorts of kidney transplant recipients. ROC curve analyses were used to evaluate the ability of B cell counts during the ﬁrst month posttransplantation (i.e., B cell nadir) to predict de novo DSA development at 1 y posttransplantation. B cell counts , 4/ml at 15– 30 d posttransplant signiﬁcantly predicted de novo DSA devel-

opment (Fig. 9A). To exclude the possibility that these results http://www.jimmunol.org/ were biased by the more profound depletion induced by alemtuzumab on B cells compared with Bas/rATG, ROC curve analysis was repeated using only patients treated with alemtuzumab. Even in this cohort alone we confirmed that low B cell counts in the first month posttransplant (,2/ml) was associated significantly with de novo DSA development at 1 y (p=0.0249). Considering ROC curve analysis with B cell subsets, switched memory B cell percentages . 55.9% during the first month after

FIGURE 6. Posttransplant BAFF serum levels and correlation with B cell surgery were associated significantly with DSA at 1 y posttrans- by guest on September 25, 2021 A subsets in kidney transplant recipients. ( ) Serum BAFF levels measured by plant (Fig. 9B). In turn, naive B cell percentages , 21.9% in the ELISA in kidney transplant recipients receiving alemtuzumab and in a sub- first month posttransplant predicted de novo DSA development at group of 12 reference patients given Bas/rATG as induction therapies during a 1-y follow-up posttransplantation. Data are mean 6 SEM. (B) Relationship 1 y (Fig. 9C). between serum BAFF concentrations at 1 mo and percentages of B cell With regard to changes in the B cell compartment in the late subsets. *p , 0.05 versus healthy controls and pretransplant values, ˚p , posttransplant period (i.e., the significant increase in naive B cells), + 2 0.001 versus Bas/rATG-treated group at the same time point posttransplant. we found that percentages of naive Bm2 and IgD CD27 cells in the long-term were higher in patients who had developed DSA treatment showed that the rate of GFR decline during 4 y of (Fig. 10), and both correlated with the degree of early B cell 2 2 follow-up was significantly faster in the eight alemtuzumab-treated lymphopenia at 15–30 d (r = 0.6908, p=0.0011; r = 0.7624, patients given cyclosporine than in the gender- and age-matched p=0.0001; respectively). reference patients in the Bas/rATG group who received cyclo- Discussion In this study, we showed that the phenotype of B cells repopulating the peripheral blood in kidney transplant recipients after alemtuzumab, but not Bas/rATG, induction underwent significant variations that can be summarized in three phases: in the early posttransplant period, the few remaining B cells were composed primarily by Ag-experienced Bm3+4 and switched memory B cells. The second phase (midterm posttransplant, 2–6 mo) was characterized by a transient increase in transitional B cells, followed by the emergence of Bm29 B cells, processes that are mediated by the increase in serum levels of BAFF. The third phase in the long-term (.24 mo posttransplant) was characterized by a progressive increase in circulating B cell numbers, primarily due to the expansion of B cells with a naive phenotype. Both early and late changes in the B cell compartment were associated with an FIGURE 7. Alemtuzumab serum levels in alemtuzumab-treated kidney increased incidence of de novo DSA at 1 y posttransplantation. De transplant recipients. Serum concentrations of alemtuzumab measured by novo DSA development at 1 y posttransplant was coupled with ELISA in 13 kidney transplant recipients at 1 and 12 h and at 7, 30, 60, worse long-term graft function, consistent with the notion that 360, and 720 d after transplantation. Data are mean 6 SD. a substantial period of time is required before damage to the 8 ALEMTUZUMAB-INDUCED B CELL CHANGES AND DSA

FIGURE 8. De novo DSA development in kidney transplant recipients given induction therapy with either alemtuzumab or Bas/rATG and long-term graft function. (A) Percentages of subjects positive for de novo DSA in alemtuzumab and Bas/rATG groups at 1 and 2 y posttransplantation. (B) HLA speciﬁcity and titer (mean ﬂuorescence intensity) of de novo DSA in kidney transplant recipients given either alemtuzumab or Bas/rATG induction therapy. Rate of GFR decline at 4 y posttransplantation in subjects given induction therapy with alemtuzumab or Bas/rATG (C) or in subjects with or D without de novo DSA at 1 y ( ). Data are Downloaded from mean 6 SD. *p , 0.05 versus reference patients treated with Bas/rATG, ˚p , 0.05 versus patients without DSA. http://www.jimmunol.org/

kidney induced by DSA becomes clinically apparent (27), even- trast, we provide evidence that early changes in the phenotype of tually leading to poor long-term graft outcome (27–30). circulating B cells in alemtuzumab-treated patients correlated with The alloantibody response in alemtuzumab-treated patients was the development of DSA posttransplantation. A previous report in attributed to a population of depletion-resistant effector/memory patients with multiple sclerosis showed that even multiple doses of T cells that homeostatically expanded in the first month after alemtuzumab failed to significantly decrease CD27+ memory B cell transplantation and could provide B cell help (31). However, this percentages 1 mo after administration (32), confirming our findings by guest on September 25, 2021 memory T cell subset homeostatically expanded in rATG-treated of a partial resistance of memory B cells to alemtuzumab-mediated patients as well (31), but it was not associated with an increased risk depletion. Higher percentages of switched memory IgD2CD27+ for acute humoral rejection. Thus, the memory T cell hypothesis B cells predicted the late development of DSA, indicating a causa- does not fully explain the higher DSA development that we found tive link between this B cell subset and DSA production. Of note, by in kidney transplant recipients given alemtuzumab compared with transferring purified B cell subsets in immunized SCID/SCID mice, reference patients receiving Bas/rATG induction therapy. In con- human switched memory CD27+ B cells were able to produce IgG

Table II. Characteristics and baseline and immunological parameters of the subjects according to de novo DSA development at 1 y posttransplant

No DSA (n = 36) DSA (n = 12) p Value Characteristics Recipient age (y) 46 6 14 46 6 12 0.8290 Recipient gender (%male) 68 50 0.2243 Living donor (%) 11 8.3 1.000 HLA-A mismatch 1.40 6 0.63 1.44 6 0.63 0.8597 HLA-B mismatch 1.33 6 0.61 1.50 6 0.52 0.3121 HLA-DR mismatch 1.14 6 0.56 1.31 6 0.48 0.2783 Clinical and immunological parameters Cold ischemia time (h) 16 6 3176 3 0.6878 Delayed graft function (%) 19 17 1.000 Acute rejection episodes (%) 17 17 1.000 Cyclosporine C0 at month 6 (ng/ml) 114 6 47 105 6 36 0.5509 Cyclosporine C0 at month 12 (ng/ml) 106 6 55 74 6 25 0.0622 Cyclosporine C0 at month 48 (ng/ml) 77 6 35 90 6 52 0.3295 Sirolimus C0 at month 6 (ng/ml) 7.1 6 0.9 8.3 6 3.5 0.3944 Sirolimus C0 at month 12 (ng/ml) 10.7 6 2.5 8.3 6 2.2 0.1482 Sirolimus C0 at month 48 (ng/ml) 10.1 6 2.7 7.4 6 1.8 0.2560 Proteinuria at month 48 (g/24 h) 0.51 6 0.65 1.51 6 3.16 0.0595 T cell counts before transplant (cells/ml) 1049 6 378 1246 6 260 0.1842 T cell counts during month 1 (cells/ml) 335 6 409 200 6 353 0.2485 T cell counts at month 12 (cells/ml) 997 6 497 715 6 585 0.1246 The Journal of Immunology 9

FIGURE 9. B cell compartment with regard to positivity for DSA 1 y posttransplant. (A) According to the ROC curve analysis, B cell counts ,4cells/mlduringthefirst month posttransplant predicted positivity for de novo DSA at 1 y posttransplant. (B)ROC curve analysis shows that percentages of switched memory IgD2CD27+ B cells . 55.9% during the first month posttransplantation predicted de novo DSA development at 1 y posttransplant. (C)ROC curve analysis shows that a naive IgD+ CD272 B cell percentage , 21.9% in the first month posttransplant predicted de novo DSA development at 1 y posttransplant.

Abs against pneumococcal surface protein A, a T cell–dependent distinction of the more immature B cells (CD272) from pre- Ag (33). Similarly, the CD27+ B cell subset isolated from the pe- germinal center cells (CD27+) (36). Findings from our cohort of ripheral blood of sensitized patients was able to produce alloanti- alemtuzumab-treated patients that transitional B cells (peaking at 2 Downloaded from bodies in vitro (34). In tetramer-binding studies, a higher frequency mo) and Bm29 B cells (peaking at 6 mo) had a completely different of HLA tetramer–positive B cells among the CD27+ B cells was regeneration profile and that, on average, 60% of Bm29 B cells were found in sensitized subjects (35); these cells were able to produce positive for CD27 expression (data not shown), suggest that the alloantibodies when stimulated in vitro. These and previous findings Bm29 B cell subtype represented activated B cells committed to led us to hypothesize that, in the presence of alloantigens, spared germinal center maturation rather than a population of immature/ switched memory B cells could be rapidly converted to plasmablasts transitional B cells. and to Ab-secreting cells, leading to DSA development. The homeostatic expansion of B cells is regulated primarily by http://www.jimmunol.org/ From the second month posttransplant, B cells started repopu- the cytokine BAFF. Serum BAFF levels increased after B cell de- lating the peripheral blood, reaching basal levels at 6 mo post- pletion induced by alemtuzumab, probably as a result of reduced transplant. The first B cell subtype that transiently dominated the receptor-mediated clearance. Similar to previous studies (25, 32), we B cell compartment during the early B cell expansion was transi- found abnormally high levels of BAFF in serum from patients given tional B cells, followed by Bm29 B cells. Consistently, a transient alemtuzumab induction during the nadir of B cell lymphopenia. increase in transitional B cells, as well as Bm29 B cells, was re- In addition to controlling B cell proliferation, BAFF is the lim- ported recently in alemtuzumab-treated kidney transplant recipi- iting factor for which immature B cell clones compete for their

ents 6 mo posttransplantation (6) and in patients with multiple survival and selection. Under lymphoreplete conditions in which by guest on September 25, 2021 sclerosis (32). The peripheral population of Bm29 cells could in- BAFF is limiting, immature B cells with lower BCR avidity and clude immature/transitional B cells in addition to pregerminal specificity are eliminated at the transitional stage (37, 38). How- center B cells. The expression of CD27 on Bm29 cells allows the ever, under lymphopenic conditions in which BAFF is in excess, selective stringency is relaxed, and all clones, including those with autoreactive specificities, may be allowed to mature (39–41). In- terestingly, BAFF-dependent rescue of self-reactive B cells from deletion also was proposed as the mechanism responsible for autoimmune disease development in patients receiving alemtuzumab (42). Intriguingly, in our kidney transplant recipients, BAFF levels correlated positively with the percentages of transitional B cells and Bm29 B cells, as reported in patients with Sjo¨gren’s syndrome (43) and chronic graft-versus-host disease (44), suggesting that, in vivo, these B cell subsets depend on BAFF availability. Moreover, in vitro BAFF promoted the conversion of Bm1 cells to Bm2/Bm29 (43). In vitro evidence demonstrates that BAFF promotes the survival of activated B cells by lowering BCR-activation thresholds and by inducing the upregulation of antiapoptotic protein expression (41, 45). This finding, together with an increased incidence of DSA at 1 y posttransplantation in our alemtuzumab-treated patients, led us to formulate the following hypothesis: transitional B cells emerging after B cell lymphopenia are constantly exposed to kidney graft Ags, and those B cells bearing a BCR specific for the donor alloantigens receive sufficient BCR signaling to upregulate BAFF receptor expression. In a BAFF-enriched environment, these cells are allowed to survive and to enter the germinal center for T cell FIGURE 10. Profile of repopulating naive B cells according to de novo DSA development. Percentages of naive IgD+CD272 B cells (A)andBm2B help–mediated maturation and differentiation into donor-specific cells (B) from baseline to .24 mo after transplant in the combined cohort of Ab-producing cells. kidney transplant recipients according to de novo DSA development at 1 y. Changes in peripheral B cell subsets after alemtuzumab nor- Data are mean 6 SEM. *p , 0.05, versus no DSA, ˚p , 0.01, versus no DSA. malized 1 y posttransplant; thereafter, the B cell compartment be- 10 ALEMTUZUMAB-INDUCED B CELL CHANGES AND DSA came dominated by naive B cells, the B cell subtype that started 3. Knechtle, S. J., J. D. Pirsch, J. H Fechner, Jr., B. N. Becker, A. Friedl, R. B. Colvin, L. K. Lebeck, L. T. Chin, Y. T. Becker, J. S. Odorico, et al. 2003. repopulating the periphery as late as 6 mo after transplantation. The Campath-1H induction plus rapamycin monotherapy for renal transplantation: delayed expansion of naive B cells was reported after alemtuzumab results of a pilot study. Am. J. Transplant. 3: 722–730. induction in both kidney transplant recipients (6) and in patients with 4. Hill, P., E. Gagliardini, P. Ruggenenti, and G. Remuzzi. 2005. Severe early acute humoral rejection resulting in allograft loss in a renal transplant recipient with multiple sclerosis (32), as well as in patients with rheumatoid ar- Campath-1H induction therapy. Nephrol. Dial. Transplant. 20: 1741–1744. thritis after rituximab treatment (46) and after myeloablation in 5. Mao, Q., P. I. Terasaki, J. Cai, K. Briley, P. Catrou, C. Haisch, and L. Rebellato. the setting of hematopoietic stem cell transplantation (44, 47). We 2007. Extremely high association between appearance of HLA antibodies and failure of kidney grafts in a five-year longitudinal study. Am. J. Transplant. 7: speculate that the late naive B cell expansion in the peripheral blood 864–871. of alemtuzumab-treated patients is the result of time needed for 6. Heidt, S., J. Hester, S. Shankar, P. J. Friend, and K. J. Wood. 2012. B cell repo- pulation after alemtuzumab induction-transient increase in transitional B cells and both new B cell generation (48) and for expression of the BAFF long-term dominance of naı¨ve B cells. Am. J. Transplant. 12: 1784–1792. receptors BR3 and TACI on their cell surface. Indeed, during the 7. Cherukuri, A., A. D. Salama, C. Carter, N. Smalle, R. McCurtin, E. W. Hewitt, maturation process, newly forming B cells exiting the bone marrow M. Hernandez-Fuentes, B. Clark, and R. J. Baker. 2012. An analysis of lymphocyte phenotype after steroid avoidance with either alemtuzumab or progress through the transitional stage before becoming naive B basiliximab induction in renal transplantation. Am. J. Transplant. 12: 919–931. cells. During this maturation course, the cell surface expression 8. Newell, K. A., A. Asare, A. D. Kirk, T. D. Gisler, K. Bourcier, M. Suthanthiran, of BR3 and TACI increases progressively, reaching the highest level W. J. Burlingham, W. H. Marks, I. Sanz, R. I. Lechler, et al; Immune Tolerance Network ST507 Study Group. 2010. Identification of a B cell signature associ- at the final stage of mature naive B cells (49, 50). Although sig- ated with renal transplant tolerance in humans. J. Clin. Invest. 120: 1836–1847. naling through the BR3 receptor is crucial for the survival of all 9. Bestard, O., J. M. Cruzado, M. Mestre, A. Calde´s, J. Bas, M. Carrera, J. Torras, I. Rama, F. Moreso, D. Seroń, and J. M. Grinyo´. 2007. Achieving donor-specific preimmune B cell subsets from the transitional stage to the naive hyporesponsiveness is associated with FOXP3+ regulatory T cell recruitment in stage, TACI signals negatively regulate naive B cell survival (51). human renal allograft infiltrates. J. Immunol. 179: 4901–4909. Downloaded from This suggests that the complex interplay of intracellular signals 10. Lebranchu, Y., F. Bridoux, M. Bu¨chler, Y. Le Meur, I. Etienne, O. Toupance, B. Hurault de Ligny, G. Touchard, B. Moulin, P. Le Pogamp, et al. 2002. activated by positive (BR3) and negative (TACI) BAFF receptors on Immunoprophylaxis with basiliximab compared with antithymocyte globulin naive B cells can eventually contribute to delaying and prolonging in renal transplant patients receiving MMF-containing triple therapy. Am. J. their response to the increased circulating BAFF levels that we Transplant. 2: 48–56. 11. Sollinger, H., B. Kaplan, M. D. Pescovitz, B. Philosophe, A. Roza, K. Brayman, documented after alemtuzumab-induced B cell lymphopenia. and K. Somberg. 2001. Basiliximab versus antithymocyte globulin for preven-

Increases in both total peripheral B cell numbers and naive B cell tion of acute renal allograft rejection. Transplantation 72: 1915–1919. http://www.jimmunol.org/ 12. Ruggenenti, P., I. Codreanu, P. Cravedi, A. Perna, E. Gotti, and G. Remuzzi. percentages were described recently as the main immunologic 2006. Basiliximab combined with low-dose rabbit anti-human thymocyte glob- features of transplant recipients who developed spontaneous tol- ulin: a possible further step toward effective and minimally toxic T cell-targeted erance to kidney allografts (8). Actually, in our study, we found therapy in kidney transplantation. Clin. J. Am. Soc. Nephrol. 1: 546–554. 13. Gennarini, A., P. Cravedi, M. Marasa`, A. Perna, G. Rota, M. Bontempelli, that the percentages of naive B cells in the long-term were higher S. Sandrini, G. Remuzzi, and P. Ruggenenti. 2012. Perioperative Minimal In- in subjects who had previously developed DSA, and they corre- duction Therapy: A Further Step toward More Effective Immunosuppression in lated with the degree of early B cell lymphopenia at 15–30 d, Transplantation. J. Transplant. 2012: 426042. 14. Cattaneo, D., S. Merlini, M. Pellegrino, F. Carrara, S. Zenoni, S. Murgia, questioning the relevance of the long-term enhanced naive B cells S. Baldelli, F. Gaspari, G. Remuzzi, and N. Perico. 2004. Therapeutic drug as a biomarker of the state of tolerance, at least in patients given monitoring of sirolimus: effect of concomitant immunosuppressive therapy and

optimization of drug dosing. Am. J. Transplant. 4: 1345–1351. by guest on September 25, 2021 alemtuzumab induction therapy. 15. Baldelli, S., S. Zenoni, S. Merlini, N. Perico, and D. Cattaneo. 2006. Simulta- In conclusion, we document that B cells re-emerging after neous determination of everolimus and cyclosporine concentrations by HPLC alemtuzumab-induced lymphopenia showed a dysregulated phe- with ultraviolet detection. Clin. Chim. Acta 364: 354–358. 16. Noris, M., F. Casiraghi, M. Todeschini, P. Cravedi, D. Cugini, G. Monteferrante, notype both early and late posttransplant; this phenomenon did not S. Aiello, L. Cassis, E. Gotti, F. Gaspari, et al. 2007. Regulatory T cells and occur in Bas/rATG-treated reference patients. Induction therapy T cell depletion: role of immunosuppressive drugs. J. Am. Soc. Nephrol. 18: with alemtuzumab was associated with a higher incidence of de novo 1007–1018. 17. Ruggenenti, P., N. Perico, E. Gotti, P. Cravedi, V. D’Agati, E. Gagliardini, DSA development and inferior graft function, and changes in B cell M. Abbate, F. Gaspari, D. Cattaneo, M. Noris, et al. 2007. Sirolimus versus phenotype correlated with de novo DSA development. Thus, in cyclosporine therapy increases circulating regulatory T cells, but does not protect renal transplant patients given alemtuzumab induction from chronic allograft kidney transplant recipients, B cell depletion and regeneration injury. Transplantation 84: 956–964. promote chronic humoral immune responses against graft alloan- 18. Gaspari, F., N. Perico, P. Ruggenenti, L. Mosconi, C. S. Amuchastegui, E. Guerini, tigens that contribute to DSA production and, eventually, long-term E. Daina, and G. Remuzzi. 1995. Plasma clearance of nonradioactive iohexol as a measure of glomerular filtration rate. J. Am. Soc. Nephrol. 6: 257–263. graft dysfunction. These data could provide useful insights for 19. Terasaki, P. I., and J. D. McClelland. 1964. Microdroplet Assay of Human Serum improving the choice of induction therapy that limits the risk for Cytotoxins. Nature 204: 998–1000. B cell–induced late graft loss, at least in kidney transplantation. 20. Pei, R., J. H. Lee, N. J. Shih, M. Chen, and P. I. Terasaki. 2003. Single human leukocyte antigen flow cytometry beads for accurate identification of human leukocyte antigen antibody specificities. Transplantation 75: 43–49. Acknowledgments 21. Liu, Y. J., F. Malisan, O. de Bouteiller, C. Guret, S. Lebecque, J. Banchereau, We thank Dr. Mario Scalamogna for suggestions and helpful discussion; F. C. Mills, E. E. Max, and H. Martinez-Valdez. 1996. Within germinal centers, isotype switching of immunoglobulin genes occurs after the onset of somatic Dr. Mario Bontempelli for the lymphocyte subset profiling in the peripheral mutation. Immunity 4: 241–250. blood; Paolo Cravedi, Maddalena Marasa`, and Valentina Portalupi for 22. Klein, U., K. Rajewsky, and R. Ku¨ppers. 1998. Human immunoglobulin (Ig)M monitoring patients before and after kidney transplantation; and Giuseppe +IgD+ peripheral blood B cells expressing the CD27 cell surface antigen carry Monteferrante and Daniela Cugini for contributing to peripheral blood somatically mutated variable region genes: CD27 as a general marker for somatically mutated (memory) B cells. J. Exp. Med. 188: 1679–1689. mononuclear cell isolation. 23. Wei, C., J. Anolik, A. Cappione, B. Zheng, A. Pugh-Bernard, J. Brooks, E. H. Lee, E. C. Milner, and I. Sanz. 2007. A new population of cells lacking Disclosures expression of CD27 represents a notable component of the B cell memory com- The authors have no financial conflicts of interest. partment in systemic lupus erythematosus. J. Immunol. 178: 6624–6633. 24. Blair, P. A., L. Y. Norenã, F. Flores-Borja, D. J. Rawlings, D. A. Isenberg, M. R. Ehrenstein, and C. Mauri. 2010. CD19(+)CD24(hi)CD38(hi) B cells ex- hibit regulatory capacity in healthy individuals but are functionally impaired in References systemic Lupus Erythematosus patients. Immunity 32: 129–140. 1. Lodhi, S. A., K. E. Lamb, and H. U. Meier-Kriesche. 2011. Solid organ allograft 25. Bloom, D., Z. Chang, K. Pauly, J. Kwun, J. Fechner, C. Hayes, M. Samaniego, survival improvement in the United States: the long-term does not mirror the and S. Knechtle. 2009. BAFF is increased in renal transplant patients following dramatic short-term success. Am. J. Transplant. 11: 1226–1235. treatment with alemtuzumab. Am. J. Transplant. 9: 1835–1845. 2. Salama, A. D., G. Remuzzi, W. E. Harmon, and M. H. Sayegh. 2001. Challenges 26. Cambridge, G., W. Stohl, M. J. Leandro, T. S. Migone, D. M. Hilbert, and to achieving clinical transplantation tolerance. J. Clin. Invest. 108: 943–948. J. C. Edwards. 2006. Circulating levels of B lymphocyte stimulator in patients The Journal of Immunology 11

with rheumatoid arthritis following rituximab treatment: relationships with 39. Parsons, R. F., K. Vivek, R. R. Redfield, III, T. S. Migone, M. P. Cancro, A. Naji, B cell depletion, circulating antibodies, and clinical relapse. Arthritis Rheum. 54: and H. Noorchashm. 2010. B-lymphocyte homeostasis and BLyS-directed im- 723–732. munotherapy in transplantation. Transplant. Rev. (Orlando) 24: 207–221. 27. Terasaki, P. I., M. Ozawa, and R. Castro. 2007. Four-year follow-up of a pro- 40. Mackay, F., S. A. Woodcock, P. Lawton, C. Ambrose, M. Baetscher, spective trial of HLA and MICA antibodies on kidney graft survival. Am. J. P. Schneider, J. Tschopp, and J. L. Browning. 1999. Mice transgenic for BAFF Transplant. 7: 408–415. develop lymphocytic disorders along with autoimmune manifestations. J. Exp. 28. Lee, P. C., L. Zhu, P. I. Terasaki, and M. J. Everly. 2009. HLA-specific antibodies Med. 190: 1697–1710. developed in the first year posttransplant are predictive of chronic rejection and 41. Thien, M., T. G. Phan, S. Gardam, M. Amesbury, A. Basten, F. Mackay, and renal graft loss. Transplantation 88: 568–574. R. Brink. 2004. Excess BAFF rescues self-reactive B cells from peripheral de- 29. Wiebe, C., I. W. Gibson, T. D. Blydt-Hansen, M. Karpinski, J. Ho, L. J. Storsley, letion and allows them to enter forbidden follicular and marginal zone niches. A. Goldberg, P. E. Birk, D. N. Rush, and P. W. Nickerson. 2012. Evolution and Immunity 20: 785–798. clinical pathologic correlations of de novo donor-specific HLA antibody post 42. Cossburn, M., A. A. Pace, J. Jones, R. Ali, G. Ingram, K. Baker, C. Hirst, J. Zajicek, kidney transplant. Am. J. Transplant. 12: 1157–1167. N. Scolding, M. Boggild, et al. 2011. Autoimmune disease after alemtuzumab 30. Gaston, R. S., J. M. Cecka, B. L. Kasiske, A. M. Fieberg, R. Leduc, F. C. Cosio, treatment for multiple sclerosis in a multicenter cohort. Neurology 77: 573–579. S. Gourishankar, J. Grande, P. Halloran, L. Hunsicker, et al. 2010. Evidence for 43. d’Arbonneau, F., J. O. Pers, V. Devauchelle, Y. Pennec, A. Saraux, and antibody-mediated injury as a major determinant of late kidney allograft failure. P. Youinou. 2006. BAFF-induced changes in B cell antigen receptor-containing Transplantation 90: 68–74. lipid rafts in Sjo¨gren’s syndrome. Arthritis Rheum. 54: 115–126. 31. Pearl, J. P., J. Parris, D. A. Hale, S. C. Hoffmann, W. B. Bernstein, K. L. McCoy, 44. Sarantopoulos, S., K. E. Stevenson, H. T. Kim, C. S. Cutler, N. S. Bhuiya, S. J. Swanson, R. B. Mannon, M. Roederer, and A. D. Kirk. 2005. Immuno- M. Schowalter, V. T. Ho, E. P. Alyea, J. Koreth, B. R. Blazar, et al. 2009. Altered competent T-cells with a memory-like phenotype are the dominant cell type B-cell homeostasis and excess BAFF in human chronic graft-versus-host disease. following antibody-mediated T-cell depletion. Am. J. Transplant. 5: 465–474. Blood 113: 3865–3874. 32. Thompson, S. A., J. L. Jones, A. L. Cox, D. A. Compston, and A. J. Coles. 2010. 45. Saito, Y., Y. Miyagawa, K. Onda, H. Nakajima, B. Sato, Y. Horiuchi, H. Okita, B-cell reconstitution and BAFF after alemtuzumab (Campath-1H) treatment of Y. U. Katagiri, M. Saito, T. Shimizu, et al. 2008. B-cell-activating factor inhibits multiple sclerosis. J. Clin. Immunol. 30: 99–105. CD20-mediated and B-cell receptor-mediated apoptosis in human B cells. Im- 33. Moens, L., M. Wuyts, I. Meyts, K. De Boeck, and X. Bossuyt. 2008. Human munology 125: 570–590. memory B lymphocyte subsets fulfill distinct roles in the anti-polysaccharide and 46. Roll, P., A. Palanichamy, C. Kneitz, T. Dorner, and H. P. Tony. 2006. Regen- Downloaded from anti-protein immune response. J. Immunol. 181: 5306–5312. eration of B cell subsets after transient B cell depletion using anti-CD20 anti- 34. Han, M., J. A. Rogers, B. Lavingia, and P. Stastny. 2009. Peripheral blood B cells bodies in rheumatoid arthritis. Arthritis Rheum. 54: 2377–2386. producing donor-specific HLA antibodies in vitro. Hum. Immunol. 70: 29–34. 47. Storek, J., M. Geddes, F. Khan, B. Huard, C. Helg, Y. Chalandon, J. Passweg, 35. Zachary, A. A., D. Kopchaliiska, R. A. Montgomery, and M. S. Leffell. 2007. and E. Roosnek. 2008. Reconstitution of the immune system after hematopoietic HLA-specific B cells: I. A method for their detection, quantification, and iso- stem cell transplantation in humans. Semin. Immunopathol. 30: 425–437. lation using HLA tetramers. Transplantation 83: 982–988. 48. Fo¨rster, I., and K. Rajewsky. 1990. The bulk of the peripheral B-cell pool in mice 36. Sims, G. P., R. Ettinger, Y. Shirota, C. H. Yarboro, G. G. Illei, and P. E. Lipsky. is stable and not rapidly renewed from the bone marrow. Proc. Natl. Acad. Sci.

2005. Identiﬁcation and characterization of circulating human transitional USA 87: 4781–4784. http://www.jimmunol.org/ B cells. Blood 105: 4390–4398. 49. Cancro, M. P., and S. H. Smith. 2003. Peripheral B cell selection and homeo- 37. Smith, S. H., and M. P. Cancro. 2003. Cutting edge: B cell receptor signals stasis. Immunol. Res. 27: 141–148. regulate BLyS receptor levels in mature B cells and their immediate progenitors. 50. Cancro, M. P. 2004. Peripheral B-cell maturation: the intersection of selection J. Immunol. 170: 5820–5823. and homeostasis. Immunol. Rev. 197: 89–101. 38. Lentz, V. M., M. P. Cancro, F. E. Nashold, and C. E. Hayes. 1996. Bcmd governs 51. Treml, J. F., Y. Hao, J. E. Stadanlick, and M. P. Cancro. 2009. The BLyS family: recruitment of new B cells into the stable peripheral B cell pool in the A/WySnJ toward a molecular understanding of B cell homeostasis. Cell Biochem. Biophys. mouse. J. Immunol. 157: 598–606. 53: 1–16. by guest on September 25, 2021