Tofacitinib Suppresses Antibody Responses to Protein Therapeutics in Murine Hosts This information is current as of March 4, 2015.

Masanori Onda, Kamran Ghoreschi, Scott Steward-Tharp, Craig Thomas, John J. O'Shea, Ira H. Pastan and David J. FitzGerald

Supplementary Material References Subscriptions Permissions Email Alerts

http://www.jimmunol.org/content/suppl/2014/05/31/jimmunol.140006 3.DCSupplemental.html This article cites 44 articles, 24 of which you can access for free at: http://www.jimmunol.org/content/193/1/48.full#ref-list-1 Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscriptions Submit copyright permission requests at: http://www.aai.org/ji/copyright.html Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/cgi/alerts/etoc

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 9650 Rockville Pike, Bethesda, MD 20814-3994. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606.

Downloaded from http://www.jimmunol.org/ at Univ of Pennsylvania Library on March 4, 2015

J Immunol 2014; 193:48-55; Prepublished online 2 June 2014; doi: 10.4049/jimmunol.1400063 http://www.jimmunol.org/content/193/1/48

The Journal of Immunology

Tofacitinib Suppresses Antibody Responses to Protein Therapeutics in Murine Hosts Masanori Onda,* Kamran Ghoreschi,†,1 Scott Steward-Tharp,†,2 Craig Thomas,‡ John J. O’Shea,† Ira H. Pastan,* and David J. FitzGerald*

R

ecombinant immunotoxins (RITs) have been developed for the treatment of cancer (1). These therapeutic proteins are composed of a 38-kDa truncated form of Pseudomonas exotoxin A (PE38) fused to the variable fragment (Fv) of an Ab. The binding moiety of the Fv targets Ag+ cells that are then killed by the cytotoxic activity of the toxin moiety (2). Three PEbased RITs are currently being evaluated in clinical trials. One of these, moxetumomab pasudotox (HA22), targets CD22 expressed by B cell malignancies. In phase I trials, HA22 produced a high rate of complete remission in patients with drug-resistant hairy cell leukemia and objective responses in acute lymphoblastic leukemia (3, 4). The CD25 targeting RIT anti–Tac(Fv)-PE38 (LMB-2) has shown antitumor activity in patients with hairy cell leukemia and other hematological malignancies (5). Another RIT being tested in phase I studies is the antimesothelin immunotoxin, SS1(dsFv)-PE38 (SS1P). As a monotherapy, SS1P produced only minor responses in patients with mesothelioma (6, 7). However, in *Laboratory of Molecular Biology, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892; †Molecular Immunology and Inflammation Branch, National Institute of Arthritis and Musculoskeletal and Skin Diseases, National Institutes of Health, Bethesda, MD 20892; and ‡ Division of Pre-Clinical Innovation, National Center for Advancing Translational Sciences, National Institutes of Health, Bethesda, MD 20892 1 Current address: Department of Dermatology, University Medical Center, Eberhard Karls University of T€ ubingen, T€ubingen, Germany. 2

Current address: College of Dentistry, University of Iowa, Iowa City, IA.

Received for publication January 23, 2014. Accepted for publication March 8, 2014. This work was supported by the Intramural Research Program of the National Institutes of Health, National Cancer Institute, Center for Cancer Research, and the National Institute of Arthritis and Musculoskeletal and Skin Disease. Address correspondence and reprint requests to Dr. Masanori Onda and Dr. David FitzGerald, Laboratory of Molecular Biology, National Cancer Institute, 37 Convent Drive, Room 5110 (M.O.) or Room 5124 (D.F.), Bethesda, MD 20892-4264. E-mail addresses: [email protected] (M.O.) and [email protected] (D.F.) The online version of this article contains supplemental material. Abbreviations used in this article: ADA, anti-drug Ab; gc, g-chain; Fv, variable fragment; KLH, keyhole limpet hemocyanin; PE38, truncated form of Pseudomonas exotoxin A; PEG, polyethylene glycol; RIT, recombinant immunotoxin; SS1P, SS1 (dsFv)PE38; TI-II, thymus-independent type II; TNP, trinitrophenyl; XSCID, Xlinked SCID. www.jimmunol.org/cgi/doi/10.4049/jimmunol.1400063

preclinical testing, combinations of SS1P with chemotherapy produced more promising results than SS1P alone, suggesting a path forward for clinical trials (8, 9). Importantly, one major factor limiting the efficacy of these RITs is the immunogenicity of PE38, which can lead to Ab responses in treated patients (5, 10). Like other biologicals, protein therapeutics frequently contain immunogenic epitopes with the potential to activate the immune system, including T cells and B cells. This activity can result in the production of anti-drug Abs (ADAs) and the loss of a therapeutic response. In patients with hematological malignancies, the risk for developing ADAs is low. These patients typically present with an immune system impaired by their disease or by chemotherapy, which protects them from ADAs during repeated treatment cycles of RITs. In contrast, in patients with solid tumors, such as mesothelioma or ovarian cancer, the immune system is still functional and the risk for developing ADAs is .75%, as observed after one cycle of treatment with SS1P (7). Several approaches have been proposed to reduce the immunogenicity of protein therapeutics. One way is silencing major B cell epitopes of protein therapeutics by masking them with polyethylene glycol (PEG) or by introducing mutations (11–14). Another approach is to use purine analog–based immune depletion regimens. For example, pentostatin acts synergistically with cyclophosphamide to deplete host lymphoid cells with a minimum effect on myeloid cells. A regimen of pentostatin plus cyclophosphamide abrogated murine host capacity to form anti-RIT Abs (15). Traditional immunosuppressants like azathioprine or methotrexate have been reported to lower the risk for ADAs directed against other biologicals like TNF antagonists (16). However, the use of such general immunosuppressants or chemotherapeutics is often limited by metabolic side effects, such as hepato- or nephrotoxicity. For more specific and precise control of T cell– and B cell–dominated immune responses, novel inhibitors targeting JAK may prove a more powerful tool. JAK3 is an intracellular tyrosine kinase that associates with the common g-chain (gc) of the receptors for IL-2, IL-4, IL-7, IL-9, IL-15, and IL-21 (17). Signal transduction mediated by JAK3 is obligatory for lymphocyte activation, differentiation, and homeostasis.

Downloaded from http://www.jimmunol.org/ at Univ of Pennsylvania Library on March 4, 2015

Immunogenicity remains the “Achilles’ heel” of protein-based therapeutics. Anti-drug Abs produced in response to protein therapeutics can severely limit both the safety and efficacy of this expanding class of agent. In this article, we report that monotherapy of mice with tofacitinib (the JAK inhibitor) quells Ab responses to an immunotoxin derived from the bacterial protein Pseudomonas exotoxin A, as well as to the model Ag keyhole limpet hemocyanin. Thousand-fold reductions in IgG1 titers to both Ags were observed 21 d post immunization. In fact, suppression was evident for all IgG isotypes and IgM. A reduction in IgG3 production was also noted with a thymus-independent type II Ag. Mechanistic investigations revealed that tofacitinib treatment led to reduced numbers of CD127+ pro–B cells. Furthermore, we observed fewer germinal center B cells and the impaired formation of germinal centers of mice treated with tofacitinib. Because normal Ig levels were still present during tofacitinib treatment, this agent specifically reduced anti-drug Abs, thus preserving the potential efficacy of biological therapeutics, including those used as cancer therapeutics. The Journal of Immunology, 2014, 193: 48–55.

The Journal of Immunology

Materials and Methods Mice Female BALB/c mice (6–8 wk old) were obtained from the Frederick Cancer Research Facility (Frederick, MD), and JAK3 knockout mice were purchased from The Jackson Laboratory (Bar Harbor, ME). Mice were treated according to an approved animal protocol (LMB-005) and maintained in a specific pathogen–free facility. Experiments were performed according to the National Institutes of Health guidelines for the use of live animals.

Reagents (JAK inhibitor, SS1P, and keyhole limpet hemocyanin) Tofacitinib was synthesized at the National Center for Advancing Translational Sciences using a published protocol (27). Tofacitinib was resuspended at 100 mg/ml in PEG 300 for in vivo studies or in DMSO for in vitro use. SS1P was manufactured in the National Cancer Institute Laboratory of Molecular Biology according to previous methods (28). Hemocyanin from Megathura crenulata keyhole limpet was purchased from Sigma-Aldrich (St. Louis, MO).

Drug treatment and immunizations Mice received tofacitinib in PEG 300 (100 mg/ml) or vehicle alone (PEG 300) by osmotic pump infusion (Alzet Model 2004, 0.25 ml/h, 28 d; Durect, Cupertino, CA). At 4 d prior to immunization, mice were anesthetized and their dorsal surface was shaved. A 1-cm incision was made on the back to create an s.c. pocket and insert the pump. The incision site was closed with wound clips. Mice were injected weekly (i.p.) with SS1P recombinant immunotoxin (RIT; 5 mg per mouse) beginning on day 0; control mice received injections of saline alone. Every week before SS1P or vehicle immunization, ∼50 ml blood was drawn to obtain serum samples. Sera were stored at 280˚C until analyzed.

Thymus-independent type II Ag immunization To characterize responses to a thymus-independent type II (TI-II) Ag, 10 mg trinitrophenyl (TNP)–Ficoll (in PBS) was injected i.p., and TNP-specific Ab titers were determined 5 d later by ELISA. ELISA plates were coated overnight with 1 mg/ml TNP-BSA (50 ml per well). Several dilutions of

both pre- and postimmunization sera were added. Isotypespecific Abs were detected with an HRP-conjugated goat anti-mouse Ig (H+L) Ab (Southern Biotechnology, Birmingham, AL) and a TMB (tetramethylbenzidine) Substrate Kit (Thermo Scientific, Rockford, IL).

Characterization of B cell differentiation and proliferation Splenocyte and bone marrow cell suspensions were prepared from BALB/c mice, and total cells were counted. Cells (1 3 106) were stained with various combinations of the following anti-murine Abs (BD Biosciences, San Jose, CA): CD3, B220, CD43, IgM, Fas, GL-7, CD24, BP-1, CD127, or IgG1 conjugated with FITC, PE, or allophycocyanin (BD Biosciences) and analyzed on a FACSCalibur flow cytometer (BD Biosciences). A total of $10,000 live events were acquired. For assessment of in vitro B cell proliferation CD432 splenocytes were MACS (Miltenyi Biotec, San Diego, CA) purified and labeled with 1 mM CFSE (Life Technologies, Grand Island, NY) according to the manufacturer’s instructions. Labeled cells were activated for 48 h with 25 mg/ml LPS (Escherichia coli 0111: B4; Sigma-Aldrich) and 5 ng/ml IL-4 in the presence of 0, 0.1, 0.3, or 1.0 mM tofacitinib. Following culture, cells were washed, surface stained, and examined by flow cytometry (FACSCalibur; BD Biosciences).

ELISA An ELISA format was used to measure anti-PE38 Abs. Briefly, microtiter plates (MaxiSorp, Sigma-Aldrich) were coated with 100 ng per well of SS1P or keyhole limpet hemocyanin (KLH) in PBS overnight at 4˚C. After adding blocking buffer (25% DMEM, 5% FBS, 25 mM HEPES, 0.5% BSA, and 0.1% azide in PBS), plates were washed with PBS containing 0.05% Tween 20. Then serially diluted sera were transferred to individual wells (50 ml per well). ADAs were detected by rabbit anti-mouse IgG-HRP (Jackson ImmunoResearch Laboratories, West Grove, PA) or goat antimouse Ig isotype-HRP (Southern Biotechnology) and tetramethylbenzidine (TMB Substrate Kit; ThermoScientific). Each plate contained serial dilutions of the anti-PE38 mAb, IP30, as a binding control standard (29). For the measurement of serum Ig isotype studies, ELISA plates were directly coated with goat anti-mouse Ig (H+L) (Southern Biotechnology), before adding the sera samples and secondary Abs-HRP, as described above. Serum Ab concentrations were determined by ELISA using mice isotype control Abs (BD Biosciences).

Tissue staining Spleens were removed after immunization and fixed/paraffin-embedded, and organ sections were stained with H&E (Pathology/Histotechnology Laboratory, SAIC-Frederick, National Cancer Institute–Frederick, MD).

Statistics For comparisons between the experimental groups in the immunogenicity study, the Mann–Whitney nonparametric test was used. The p values , 0.05 were considered statistically significant.

Results Ag-specific Ab response To examine the effect of JAK3/JAK1 inhibition on Ab responses, we implanted Alzet osmotic mini-pumps (on day 24) delivering tofacitinib, for a calculated duration of 28 d, or vehicle (PEG solution) to BALB/c mice. On day 0 mice were immunized with SS1P as a protein Ag. Beginning on day 7 after immunization, the serum concentration of ADA (anti-SS1P Abs) was determined weekly. In vehicle-treated mice, ADAs became detectable on day 14, and increasing titers were observed over the following 21 d (Fig. 1A). Of interest, animals that were treated with tofacitinib showed a significantly lower production of ADAs compared with PEG-treated control mice (for 5 wk after initial immunization, p , 0.01, n = 8). Moreover, ADAs became detectable earliest on day 28. A difference of 1000- to 200-fold in titers to SS1P was apparent from days 21 through 35, respectively (Fig. 1A). To demonstrate that the suppressive effect of pharmacological JAK3/ JAK1 inhibition on ADA development was not restricted to SS1P Ag, we next immunized mice with KLH. Mice injected with KLH, compared with SS1P, generated a more rapid Ab response (Fig. 1B). Yet, the administration of tofacitinib reduced anti-KLH

Downloaded from http://www.jimmunol.org/ at Univ of Pennsylvania Library on March 4, 2015

After binding of the IL to its specific type I or II receptor, JAKs will associate with the receptor and activate downstream proteins, STATs (18). Activated STATs control gene expression (19). Although JAKs are typically present in many tissues, JAK3 expression is largely restricted to hematopoietic cells. Thus, JAK3 may be an excellent target for silencing immune responses and reducing ADA production against protein therapeutics without affecting other organs (20). The kinase inhibitor tofacitinib (originally CP-690, 550) was initially reported as a selective JAK3 inhibitor (21–23). However, more recent data indicate that tofacitinib binds JAK3 and JAK1 with roughly equivalent affinity, and JAK2 to a much lesser extent. Therefore, some of the efficacy of tofacitinib in immune responses is likely due to the combined inhibition of JAK3 and JAK1 (24). Tofacitinib has mainly been investigated in a variety of preclinical models of autoimmunity and inflammation. Tofacitinib has demonstrated efficacy and safety in rheumatoid arthritis, thus leading to its approval by the Food and Drug Administration. Clinical trials are ongoing in various arthropathies, psoriasis, and other autoimmune diseases (25, 26). Its exact mode of action on B cell– mediated Ab responses has not been clarified. However, we assumed that a potent inhibitor of the JAK3/JAK1-utilizing cytokines IL-4, IL-7, and IL-21 would have a major impact on B cell responses. Therefore, we investigated the effects of tofacitinib on B cell responses in a mouse model of RIT-induced immunogenicity. Our hypothesis contended that a selective therapeutic regimen, including JAK3/JAK1 inhibition, would abrogate the ADA production and preserve the efficacy of biological protein therapeutics in a cancer setting. We have evaluated the potency of tofacitinib for specifically reducing ADA production and studied the mechanism that controls Ab production in mice receiving RIT.

49

50

TOFACITINIB SUPPRESSES ANTI-DRUG Abs

titers compared with controls (p , 0.05 on day 21, p , 0.01 on day 28, respectively, n = 5). Reductions in titers ranged from 5000- to 250-fold from days 21 through 28, respectively. To confirm the relevance of pharmacological JAK3 inhibition for ADA development, we next used a genetic model of JAK3 deficiency. Wildtype or JAK3 knockout mice were immunized with SS1P, and specific Ab titers were analyzed after 5 wk. As expected, no ADAs were detected in JAK3 knockout mice that were immunized with SS1P (Fig. 1C). From these results we conclude that the JAK3/ JAK1 pathway is important for the generation of relevant titers of Abs to immunized proteins. Next we examined responding titers to SS1P by assaying Agspecific isotypes to determine whether the effect of tofacitinib was general or was restricted to a particular Ab isotype. After four injections, most responding Abs were from the IgG1 subclass, followed by IgG2b (Fig. 2A). No apparent increase in SS1P-specific IgG2a and IgG3 titers was observed when using this immunization protocol and SS1P as Ag. Treatment with tofacitinib completely

FIGURE 2. Tofacitinib inhibits the production of Ag-specific Ig isotypes after repeated immunizations of proteins. (A and B) Mice were treated with tofacitinib (s) or vehicle alone (d) and immunized with SS1P (A, n = 8 per group) or KLH (B, n = 5 per group), as described in the legend of Fig. 1. The mice received either five injections of SS1P (A) or four injections of KLH (B). At day 28 (A) or on day 21 (B), blood was drawn from each mouse, and SS1P- or KLHspecific Ig isotype levels were evaluated by Agspecific ELISA. Data are expressed as the mean 6 SD. *p , 0.05, **p , 0.01, ***p , 0.001.

blocked the anti-SS1P responses by IgG1 and IgG2b subclasses (p , 0.001, p , 0.05, respectively; n = 8; Fig. 2A). The IgM titer to SS1P was reduced by tofacitinib treatment (p , 0.05; Fig. 2A). To show that tofacitinib activity was not restricted to a particular thymus-dependent Ag, we used a similar immunization protocol with the model Ag KLH. Mice treated with or without tofacitinib were immunized weekly with KLH. After three injections, sera were analyzed for the presence of KLH-specific Igs. This time all subclasses of IgG, including IgG2a and IgG3, showed increased titers to KLH (Fig. 2B). Similar to SS1P, the titers of all subclasses of responding IgGs were reduced in the tofacitinib-treated mice (p , 0.01; n = 5, Fig. 2B), confirming that inhibition is not restricted to SS1P but apparently is general for thymus-dependent Ags. These results suggest that tofacitinib treatment effectively interferes with Ag-specific immune responses mediated by T and B cell interaction. Typically, TI-II Ags stimulate the production of IgG3 Abs (30). Therefore, we studied the effects of JAK3/JAK1 inhibition

Downloaded from http://www.jimmunol.org/ at Univ of Pennsylvania Library on March 4, 2015

FIGURE 1. Tofacitinib inhibits specific IgG1 Ab responses to immunized proteins in mice. (A and B) BALB/c mice received tofacitinib (white bars) or vehicle alone (black bars) by osmotic pump infusion and were immunized i.p. every week with 5 mg of SS1P on days 0, 7, 14, 21, and 28 (n = 8 per group) (A) or with 5 mg of KLH on days 0, 7, 14, and 21 (n = 5 per group) (B). Before each immunization, sera were collected and tested for anti-SS1P (A) or antiKLH (B) Abs by ELISA. Each titration curve was fitted in a four-parameter logistic curve, and the serum dilutions showing 10% of the highest absorbance value of a positive control were taken as the specific Ab titers. Error bars indicate SD. (C) JAK3 wild-type mice or JAK3 knockout mice were immunized with SS1P (5 mg) weekly. After 5 wk, sera were taken and anti-SS1P titers were measured by ELISA. Anti-SS1P Ab was not detected in JAK3 knockdown mice by ELISA. Data are expressed as the mean 6 SD. *p , 0.05, **p , 0.01. n.d., not detected in this assay.

The Journal of Immunology

51

on IgG3 responses to the TI-II Ag TNP-Ficoll. Mice received TNPFicoll i.p., and 5 d later TNP-specific IgM and IgG3 Ab levels were determined. Although the IgM response in mice treated with tofacitinib was similar to that in control mice, the TNP-specific IgG3 responses were reduced by ∼3-fold in tofacitinib-treated mice (Fig. 3). Thus, tofacitinib inhibits T cell–independent Ab production of the IgG3 subclass. Taken together, the results with SS1P, KLH, and TNP immunization suggest that tofacitinib affects B cell differentiation and/or inhibits B cell proliferation. To address these issues, we decided to study the overall effects of tofacitinib treatment on the lymphoid cells of the immune system. Specifically, we analyzed the proportion of B and T cell subsets in spleen and bone marrow and measured serum Ig content. B cell numbers

B cell development To investigate whether JAK inhibition by tofacitinib treatment directly affects early B cell development, we isolated and analyzed

FIGURE 3. Effects of tofacitinib on responses to TI-II Ag. Mice were either pretreated with tofacitinib (O and ▴) or vehicle control (N and n) and immunized with TNP-Ficoll. After 5 d, TNP-specific IgM and IgG3 responses were measured by ELISA. Symbols represent average values of five mice. The N and O indicate Ab levels in the preimmune serum, whereas n and ▴ represent titers from TNP-Ficoll–immunized mice. Starting at 1:100, 3-fold dilutions were made and analyzed for reactivity to TNP. Data are expressed as the mean 6 SD. **p , 0.01.

FIGURE 4. Effect of tofacitinib on the number of splenocytes. Spleens were harvested from KLH-immunized mice with (■)/without (d) tofacitinib treatment. The cell numbers are calculated from the results of flow cytometry. **p , 0.01.

bone marrow for the presence of cells expressing B220, CD24, and CD43. As defined (34), prepro–B cells lack or minimally express CD24, whereas pro–B cells typically express CD24. As shown in Fig. 5A, a small decrease in B220low CD43+ CD24+ B lymphocytes (pro–B cells) and a similar percentage of B220low CD43+ CD242 B cells (prepro–B cells) were observed in the tofacitinib treatment group, indicating that the partial developmental block mediated by JAK3/JAK1 inhibition appeared after the prepro–B cell stage. Bone marrow cells were also analyzed for the expression of CD127 (IL-7R) and B220. As depicted in Fig. 5B, the majority of cells expressing IL-7R were positive for B220, indicative of a pro– B cell and pre–B cell population. Tofacitinib treatment strongly reduced the percentage of CD127+ cells. This feature of reduced CD127+ cells resembles the defect in B cell development reported in JAK3-deficient mice (35, 36). Our results suggested that the suppression within the B cell population in tofacitinibtreated mice was likely caused by impaired expression of IL-7R and IL-7R signaling. During B cell development, IL-7Ra is expressed by common lymphoid precursors, which can generate both T and B cells. This induction of IL-7Ra expression might be in response to stimulation by the FLT3 ligand, which has been shown to induce IL-7Ra expression in vitro, and expression of both FLT3 and IL7Ra occurs at the same developmental stage (37). Tofacitinib may suppress the FLT3 ligand expression by an unknown mechanism. Presumably, JAK3/JAK1 inhibition prevents further differentiation of B cells from the prepro–B cell stage through intervention with the IL-7R/STAT5 signaling pathway (38, 39). B cell proliferation We next performed functional studies on splenic B cells by culturing them in the presence or absence of tofacitinib. The effects of JAK3/JAK1 inhibition on B cell proliferation were investigated by labeling the cells with CFSE and stimulation with LPS and IL-4, a cytokine that, like IL-7, signals through JAK3 and JAK1. After 48 h the cells were studied by flow cytometry and proliferation was

Downloaded from http://www.jimmunol.org/ at Univ of Pennsylvania Library on March 4, 2015

In agreement with previous reports, tofacitinib-treated mice thrived and showed no signs of toxicity (31). Of interest, when immunized with KLH, mice treated with tofacitinib showed a 3- to 4-fold reduction in total splenic T cells (p = 0.008) and B cells (p = 0.004), compared with sham-treated controls (Fig. 4). However, the relative proportions of B and T cells were minimally affected (Supplemental Fig. 1A). Because JAK3 knockout mice have been reported to have some development and signaling defects within B and T cell lineages (32, 33), we decided to test whether treatment with tofacitinib produced a similar phenotype. Although no reduction in the percentage of splenic B220+ CD19+ IgMlow CD432 cells (mature B cell phenotype; Supplemental Fig. 1B) was observed, we detected a reduction in the percentage of splenic Fas+ GL-7+ germinal center B cells in KLH-immunized mice treated with tofacitinib. This resulted in a significant reduction of total germinal center B cells upon treatment with tofacitinib (Fig. 4, Supplemental Fig. 1C, p = 0.01).

52

TOFACITINIB SUPPRESSES ANTI-DRUG Abs inhibition correlated with a reduction of total serum Ig levels, we followed Ig isotypes in mice immunized with SS1P (Fig. 7) or KLH (Supplemental Fig. 2). As shown in Supplemental Fig. 2, serum IgG1 and IgG2a levels were suppressed in mice treated with tofacitinib compared with sham-treated control mice during the treatment period. In contrast, the serum levels of other Ig isotypes were not affected by tofacitinib treatment. This finding indicates that in vivo significant numbers of Ig-producing plasma cells were still induced in mice treated with tofacitinib, but that Ag-specific IgG1 and IgG2a production was diminished (Fig. 2). As shown in Supplemental Fig. 2, when KLH was used as an immunogenic Ag, we had similar findings as observed with SS1P immunization (Fig. 7). IgG1 and IgG2a levels were suppressed with tofacitinib treatment, and the serum levels of other Ig isotypes were generally not affected. Failure of germinal center formation by tofacitinib

Tofacitinib did not prevent the specific cytotoxicity of immunotoxin on cancer cells evaluated by CFSE dilution. Tofacitinib had no negative effects on B cell proliferation by LPS/IL-4 activation at concentrations ranging from 0.1 to 1.0 mM (Fig. 6A). In addition, we studied the IL-4–dependent expression of IgG1 by proliferating B cells. Even though B cell proliferation in vitro by LPS was not negatively affected by tofacitinib, the IL-4–dependent expression of IgG1 was suppressed in a dose-dependent manner (Fig. 6B, 6C, p , 0.01). Serum Ig levels To investigate whether the decrease in splenic B cell numbers and the block in B cell development by pharmacological JAK3/JAK1

FIGURE 6. Tofacitinib blocks IgG1 expression by B cells. Isolated B cells were labeled with 1 mM CFSE and stimulated in vitro with LPS and IL-4 in the presence of indicated doses of tofacitinib. (A and B) B cell proliferation was determined by CFSE dilution. (C) IgG1 expression was determined by Ab staining and flow cytometry. n = 5, ***p , 0.001.

To study whether tofacitinib affected the specific cytotoxicity of immunotoxin on a cancer cell line, we measured the cytotoxic activity of 1 ng/ml of immunotoxin, HB21(Fv)-PE40 (HB21 is an Ab that targets human transferrin receptor), on several cancer cell lines (DLD1 cells are a colorectal adenocarcinoma, KB3-1 cells are derived from cervical carcinoma, and Raji cells are a Burkitt lymphoma) with different amounts of tofacitinib (0–60 mM). As most growing cells express transferrin receptors, we can estimate the activity of this immunotoxin on a variety of cancer cell types. There was no difference in the cytotoxic effect of 1 ng/ml HB21 when tofacitinib-treated cells were compared with nontreated cells

Downloaded from http://www.jimmunol.org/ at Univ of Pennsylvania Library on March 4, 2015

FIGURE 5. Altered B cell profile in bone marrow cells after tofacitinib administration. Bone marrow cells were harvested from KLH-immunized mice with/without tofacitinib. (A) Cells were stained with anti-CD43, antiCD24, and anti-B220 Abs. (B) Cells were stained with anti-CD127 and anti-B220 Abs. Expression was analyzed by flow cytometry. Numbers next to the gates in the dot plots indicate percentage of gated cells.

To investigate whether JAK3/JAK1 inhibition affects germinal center formation, a requirement for efficient Ab responses, we prepared histological sections from spleens of mice on day 17 after immunization with KLH and treatment with tofacitinib or control PEG. The tissue sections were stained with H&E. In sham-treated mice immunized for KLH, we could readily detect germinal centers in the spleen on day 17 after immunizations. In contrast, in mice treated with tofacitinib we could not detect germinal center formation on day 17 after immunization with KLH (Supplemental Fig. 3). Thus, our data suggest that mechanisms affecting germinal center formation, B cell development, and Ig class switching are potentially responsible for the suppression of specific Ab responses by pharmacological JAK3/JAK1 inhibition.

The Journal of Immunology

53

Discussion

(Supplemental Fig. 4). Thus, tofacitinib, by inhibiting JAK3/JAK1, could improve the therapeutic index of immunotoxins by decreasing ADA response without blocking antitumor effects.

Downloaded from http://www.jimmunol.org/ at Univ of Pennsylvania Library on March 4, 2015

FIGURE 7. Effect of tofacitinib on serum Ig isotype levels. Mice were treated as described in the legend of Fig. 1A. Mice were immunized with SS1P, 5 mg per mouse i.p., every week with/without tofacitinib treatment. Blood was drawn from each mouse, and serum Ig isotype levels were determined by ELISA. IgG1 and IgG2a levels were lower in the tofacitinibtreated mice than in the nontreated mice. Symbols show serum Ig levels of each mouse. Bars indicate averages and SDs. Statistical significances were calculated versus control groups. *p , 0.05, **p , 0.01. n.a., not analyzed in this group.

We have analyzed the effects of tofacitinib on the murine immune response to two thymus-dependent Ags and one thymusindependent Ag. Tofacitinib reduced Ag-specific Ig responses to each of these Ags, sometimes by as much as 5000-fold. Further, tofacitinib reduced the number of B and T cells, leading to a small reduction in total serum IgG levels. Tofacitinib treatment regimens were generally well tolerated with no apparent side effects. Specifically in mice, in which the production of ADA against SS1P was induced by weekly immunizations, tofacitinib significantly reduced ADA both in time to response and in overall titer. Although protein therapeutics constitute a promising class of drugs, their efficacy can be reduced considerably when they elicit an Ab response. Certain ADAs work by neutralization, whereas others reduce drug efficiency by hastening clearance or preventing access to target tissue. In this study, we have used an in vivo model to show that tofacitinib-mediated inhibition of JAK3/JAK1 reduced immunogenicity even in the face of repetitive immunotoxin dosing. These results establish a proof-of-principle for combining immunotoxin treatments with immune-modifying tofacitinib treatment. It is noteworthy that the therapeutic effect in the current study was achieved with continuous infusion. It is likely that tofacitinib treatment, producing very low anti-immunotoxin responses, relates to reduced immunogenicity of Pseudomonas exotoxin. These observations have broad applicability not only for protein therapeutics per se but also for gene therapy strategies in which wild-type or “corrected” proteins are expressed from engineered vectors. Further, for immunotoxins specifically, our results will likely inform the next generation of immunotoxin clinical trials. Traditional immunosuppressants like methotrexate or azathioprine have also been shown to reduce ADA during concomitant treatment with biologicals. However, those drugs can induce severe immunosuppression and accompanying toxicities. As tofacitinib suppresses Ig production more specifically than other immunosuppressants, clinical tolerance by cancer patients will be easier and safer. Our experiments also point to the importance of maintenance tofacitinib treatment to prevent immunogenicity during prolonged immunotoxin dosing. Indeed, we observed that the continuous infusion of tofacitinib was required for complete protection against ADA formation during five weekly challenges with immunotoxin. In the data from healthy volunteers, tofacitinib is rapidly absorbed after oral administration, with peak plasma concentrations of tofacitinib attained within 0.5–1.0 h (40). The oral bioavailability of the tofacitinib after a single oral 10-mg dose was 74% (40). For clinical application, it is possible that the continuous infusion regimen used in our mice experiments could be replaced by a daily oral administration regimen. It should be noted that our experiments were carried out in a tumor-free model. Further evaluation of combination between immunotoxin and tofacitinib treatment may be warranted in experiments that incorporate tumor cells, such as xenograft models incorporating a mesothelin-expressing tumor cell line (41). Yet, additional treatment with tofacitinib had no impact on the cytotoxicity of an immunotoxin directed to DLD1, KB3-1, and Raji cells in vitro (Supplemental Fig. 4). As the constitutive activation of JAKs and STATs is a common feature of a variety of cancers, JAK inhibitors are being tested in malignancies. For example, the constitutive activation of JAK1 has been recently reported in mesothelioma cell lines and in primary malignant mesothelioma (42). Thus, inhibition of JAKs may even be beneficial in mesothelioma treatment as well as blocking ADA production. We have identified that the percentage of CD127 (IL-7R)+ pro– B cells in tofacitinib-treated mice was greatly reduced. IL-7 is an essential pre–B cell growth factor in mice (43). Prepro–B cells

54

Acknowledgments We thank Gal Ben-Josef for conducting immunotoxin assays in the absence and presence of tofacitinib.

Disclosures The U.S. National Institutes of Health holds patents related to immunotoxins and Janus family kinases, and J.J.O. and the National Institutes of Health

have a Collaborative Research Agreement and Development Award with Pfizer. K.G. has been a consultant or investigator for Pfizer.

References 1. Pastan, I., R. Hassan, D. J. Fitzgerald, and R. J. Kreitman. 2006. Immunotoxin therapy of cancer. Nat. Rev. Cancer 6: 559–565. 2. Chaudhary, V. K., C. Queen, R. P. Junghans, T. A. Waldmann, D. J. FitzGerald, and I. Pastan. 1989. A recombinant immunotoxin consisting of two antibody variable domains fused to Pseudomonas exotoxin. Nature 339: 394–397. 3. Kreitman, R. J., W. H. Wilson, K. Bergeron, M. Raggio, M. Stetler-Stevenson, D. J. FitzGerald, and I. Pastan. 2001. Efficacy of the anti-CD22 recombinant immunotoxin BL22 in chemotherapy-resistant hairy-cell leukemia. N. Engl. J. Med. 345: 241–247. 4. Wayne, A. S., R. J. Kreitman, H. W. Findley, G. Lew, C. Delbrook, S. M. Steinberg, M. Stetler-Stevenson, D. J. Fitzgerald, and I. Pastan. 2010. AntiCD22 immunotoxin RFB4(dsFv)-PE38 (BL22) for CD22-positive hematologic malignancies of childhood: preclinical studies and phase I clinical trial. Clin. Cancer Res. 16: 1894–1903. 5. Kreitman, R. J., W. H. Wilson, J. D. White, M. Stetler-Stevenson, E. S. Jaffe, S. Giardina, T. A. Waldmann, and I. Pastan. 2000. Phase I trial of recombinant immunotoxin anti-Tac(Fv)-PE38 (LMB-2) in patients with hematologic malignancies. J. Clin. Oncol. 18: 1622–1636. 6. Kreitman, R. J., R. Hassan, D. J. Fitzgerald, and I. Pastan. 2009. Phase I trial of continuous infusion anti-mesothelin recombinant immunotoxin SS1P. Clin. Cancer Res. 15: 5274–5279. 7. Hassan, R., S. Bullock, A. Premkumar, R. J. Kreitman, H. Kindler, M. C. Willingham, and I. Pastan. 2007. Phase I study of SS1P, a recombinant anti-mesothelin immunotoxin given as a bolus I.V. infusion to patients with mesothelinexpressing mesothelioma, ovarian, and pancreatic cancers. Clin. Cancer Res. 13: 5144–5149. 8. Zhang, Y. J., L. M. Xiang, R. Hassan, and I. Pastan. 2007. Immunotoxin and Taxol synergy results from a decrease in shed mesothelin levels in the extracellular space of tumors. Proc. Natl. Acad. Sci. USA 104: 17099–17104. 9. Zhang, Y., L. Xiang, R. Hassan, C. H. Paik, J. A. Carrasquillo, B. S. Jang, N. Le, M. Ho, and I. Pastan. 2006. Synergistic antitumor activity of taxol and immunotoxin SS1P in tumor-bearing mice. Clin. Cancer Res. 12: 4695–4701. 10. Kreitman, R. J., M. Stetler-Stevenson, I. Margulies, P. Noel, D. J. Fitzgerald, W. H. Wilson, and I. Pastan. 2009. Phase II trial of recombinant immunotoxin RFB4(dsFv)-PE38 (BL22) in patients with hairy cell leukemia. J. Clin. Oncol. 27: 2983–2990. 11. Tsutsumi, Y., M. Onda, S. Nagata, B. Lee, R. J. Kreitman, and I. Pastan. 2000. Site-specific chemical modification with polyethylene glycol of recombinant immunotoxin anti-Tac(Fv)-PE38 (LMB-2) improves antitumor activity and reduces animal toxicity and immunogenicity. Proc. Natl. Acad. Sci. USA 97: 8548–8553. 12. Onda, M., R. Beers, L. Xiang, S. Nagata, Q. C. Wang, and I. Pastan. 2008. An immunotoxin with greatly reduced immunogenicity by identification and removal of B cell epitopes. Proc. Natl. Acad. Sci. USA 105: 11311–11316. 13. Onda, M., R. Beers, L. Xiang, B. Lee, J. E. Weldon, R. J. Kreitman, and I. Pastan. 2011. Recombinant immunotoxin against B-cell malignancies with no immunogenicity in mice by removal of B-cell epitopes. Proc. Natl. Acad. Sci. USA 108: 5742–5747. 14. Liu, W., M. Onda, B. Lee, R. J. Kreitman, R. Hassan, L. Xiang, and I. Pastan. 2012. Recombinant immunotoxin engineered for low immunogenicity and antigenicity by identifying and silencing human B-cell epitopes. Proc. Natl. Acad. Sci. USA 109: 11782–11787. 15. Mossoba, M. E., M. Onda, J. Taylor, P. R. Massey, S. Treadwell, E. Sharon, R. Hassan, I. Pastan, and D. H. Fowler. 2011. Pentostatin plus cyclophosphamide safely and effectively prevents immunotoxin immunogenicity in murine hosts. Clin. Cancer Res. 17: 3697–3705. 16. Garceˆs, S., J. Demengeot, and E. Benito-Garcia. 2013. The immunogenicity of anti-TNF therapy in immune-mediated inflammatory diseases: a systematic review of the literature with a meta-analysis. Ann. Rheum. Dis. 72: 1947–1955. 17. Ghoreschi, K., A. Laurence, and J. J. O’Shea. 2009. Janus kinases in immune cell signaling. Immunol. Rev. 228: 273–287. 18. Levy, D. E., and J. E. Darnell, Jr. 2002. Stats: transcriptional control and biological impact. Nat. Rev. Mol. Cell Biol. 3: 651–662. 19. Shuai, K., and B. Liu. 2003. Regulation of JAK-STAT signalling in the immune system. Nat. Rev. Immunol. 3: 900–911. 20. Borie, D. C., J. J. O’Shea, and P. S. Changelian. 2004. JAK3 inhibition, a viable new modality of immunosuppression for solid organ transplants. Trends Mol. Med. 10: 532–541. 21. Pesu, M., A. Laurence, N. Kishore, S. H. Zwillich, G. Chan, and J. J. O’Shea. 2008. Therapeutic targeting of Janus kinases. Immunol. Rev. 223: 132–142. 22. Changelian, P. S., M. E. Flanagan, D. J. Ball, C. R. Kent, K. S. Magnuson, W. H. Martin, B. J. Rizzuti, P. S. Sawyer, B. D. Perry, W. H. Brissette, et al. 2003. Prevention of organ allograft rejection by a specific Janus kinase 3 inhibitor. Science 302: 875–878. 23. Ghoreschi, K., A. Laurence, and J. J. O’Shea. 2009. Selectivity and therapeutic inhibition of kinases: to be or not to be? Nat. Immunol. 10: 356–360. 24. Ghoreschi, K., M. I. Jesson, X. Li, J. L. Lee, S. Ghosh, J. W. Alsup, J. D. Warner, M. Tanaka, S. M. Steward-Tharp, M. Gadina, et al. 2011. Modulation of innate and adaptive immune responses by tofacitinib (CP-690,550). J. Immunol. 186: 4234–4243.

Downloaded from http://www.jimmunol.org/ at Univ of Pennsylvania Library on March 4, 2015

were not diminished in IL-7R–deficient mice, but the pro–B cell population and pre–B cell population were reduced compared with these populations in control mice (38). Our findings, which highlight a partial developmental block at the pro–B cell stage, suggest that the reduced number of B cells in tofacitinib-treated mice was likely caused by lower expression of the IL-7R. However, this suppression was weak, and we can detect serum Igs in the tofacitinib-treated mice to some extent. This finding suggested that immune impairment by tofacitinib is limited and that a remarkable number of T and B cells were still present in tofacitinib-treated mice. Tofacitinibtreated mice also showed about half the amount of serum IgG1 and IgG2a, but identical amounts of other serum Igs (Fig. 7, Supplemental Fig. 2). Ag-specific IgG1 responses, however, were remarkably diminished in the tofacitinib-treated mice. This observation can be explained by the inhibition of IL-4 signaling, which is important for IgG1 response (44). In addition, during tofacitinib treatment, all mice appeared to retain full health without obvious side effects caused by immunosuppression. In animal models of organ engraftment, tofacitinib treatment prolonged allograft survival in both heart and kidney transplant models. Recently, the results of clinical trials of tofacitinib to treat patients with refractory rheumatoid arthritis have led to drug approval in the United States. Treatment with tofacitinib resulted in significant dose-dependent improvements in patients’ responses. After 6-wk treatments, disease activity was significantly reduced in the 5 mg, 15 mg, and 30 mg twice daily treatment groups (25). The incidences of adverse events in the tofacitinib 5 mg twice daily and placebo groups were similar. This kind of progress together with low toxicity suggests that it may be possible to develop a safe and effective treatment regimen for combining tofacitinib and immunotoxin therapy. We have shown that tofacitinib treatments successfully reduced the immunogenicity of an immunotoxin and other immunogenic agents in a mouse model. Remarkably, this result was achieved with a compound that does not induce neutropenia. However, several additional considerations are important when translating these findings from mice to humans. It is still unclear whether IL-7 is an essential pre–B cell growth factor in humans, as it is in mice. X– linked SCID (XSCID) is an inherited disease in humans, and this disease results from mutation in the common cytokine receptor gc. As JAK3 is primarily associated with gc, and as no other signaling molecules have been identified that associate with gc, IL-7 signaling should be blocked in XSCID patients. However, XSCID patients have normal or even increased B cell numbers. This finding is quite different from the findings in gc-deficient mice. The role of JAK3 in B cell homeostasis may differ in mice and humans; however, the effect of tofacitinib on function may very well be the same. Several considerations will need to be addressed when considering tofacitinib-immunotoxin treatments in cancer patients. Clinical trials should seek to determine the depth of B and T cell depletion required to abrogate host immunity to PE-based immunotoxins. Timing and dosing of tofacitinib will need to be planned to limit potential infectious complications. Successful translation of this work into the clinic holds promise as a new addition to the existing targeted therapy of cancer using protein therapeutics.

TOFACITINIB SUPPRESSES ANTI-DRUG Abs

The Journal of Immunology

34. Nagasawa, T. 2006. Microenvironmental niches in the bone marrow required for B-cell development. Nat. Rev. Immunol. 6: 107–116. 35. Erlandsson, L., S. Licence, F. Gaspal, S. Bell, P. Lane, A. E. Corcoran, and I. L. Ma˚rtensson. 2004. Impaired B-1 and B-2 B cell development and atypical splenic B cell structures in IL-7 receptor-deficient mice. Eur. J. Immunol. 34: 3595–3603. 36. Kikuchi, K., A. Y. Lai, C. L. Hsu, and M. Kondo. 2005. IL-7 receptor signaling is necessary for stage transition in adult B cell development through up-regulation of EBF. J. Exp. Med. 201: 1197–1203. 37. Borge, O. J., J. Adolfsson, A. Ma˚rtensson, I. L. Ma˚rtensson, and S. E. Jacobsen. 1999. Lymphoid-restricted development from multipotent candidate murine stem cells: distinct and complimentary functions of the c-kit and flt3-ligands. Blood 94: 3781–3790. 38. Peschon, J. J., P. J. Morrissey, K. H. Grabstein, F. J. Ramsdell, E. Maraskovsky, B. C. Gliniak, L. S. Park, S. F. Ziegler, D. E. Williams, C. B. Ware, et al. 1994. Early lymphocyte expansion is severely impaired in interleukin 7 receptordeficient mice. J. Exp. Med. 180: 1955–1960. 39. Mazzucchelli, R., and S. K. Durum. 2007. Interleukin-7 receptor expression: intelligent design. Nat. Rev. Immunol. 7: 144–154. 40. Scott, L. J. 2013. Tofacitinib: a review of its use in adult patients with rheumatoid arthritis. Drugs 73: 857–874. 41. Weldon, J. E., L. Xiang, J. Zhang, R. Beers, D. A. Walker, M. Onda, R. Hassan, and I. Pastan. 2013. A recombinant immunotoxin against the tumor-associated antigen mesothelin reengineered for high activity, low off-target toxicity, and reduced antigenicity. Mol. Cancer Ther. 12: 48–57. 42. Menges, C. W., Y. Chen, B. T. Mossman, J. Chernoff, A. T. Yeung, and J. R. Testa. 2010. A phosphotyrosine proteomic screen identifies multiple tyrosine kinase signaling pathways aberrantly activated in malignant mesothelioma. Genes Cancer 1: 493–505. 43. Leonard, W. J., E. W. Shores, and P. E. Love. 1995. Role of the common cytokine receptor gamma chain in cytokine signaling and lymphoid development. Immunol. Rev. 148: 97–114. 44. Xia, J. Q., D. A. Rickaby, K. J. Kelly, H. Choi, C. A. Dawson, and V. P. Kurup. 1999. Immune response and airway reactivity in wild and IL-4 knockout mice exposed to latex allergens. Int. Arch. Allergy Immunol. 118: 23–29.

Downloaded from http://www.jimmunol.org/ at Univ of Pennsylvania Library on March 4, 2015

25. van Vollenhoven, R. F., R. Fleischmann, S. Cohen, E. B. Lee, J. A. Garcı´a Meijide, S. Wagner, S. Forejtova, S. H. Zwillich, D. Gruben, T. Koncz, et al; ORAL Standard Investigators. 2012. Tofacitinib or adalimumab versus placebo in rheumatoid arthritis. N. Engl. J. Med. 367: 508–519. 26. Papp, K. A., A. Menter, B. Strober, R. G. Langley, M. Buonanno, R. Wolk, P. Gupta, S. Krishnaswami, H. Tan, and J. A. Harness. 2012. Efficacy and safety of tofacitinib, an oral Janus kinase inhibitor, in the treatment of psoriasis: a Phase 2b randomized placebo-controlled dose-ranging study. Br. J. Dermatol. 167: 668–677. 27. Jiang, J. K., K. Ghoreschi, F. Deflorian, Z. Chen, M. Perreira, M. Pesu, J. Smith, D. T. Nguyen, E. H. Liu, W. Leister, et al. 2008. Examining the chirality, conformation and selective kinase inhibition of 3-((3R,4R)-4-methyl-3-(methyl(7Hpyrrolo[2,3-d]pyrimidin-4-yl)amino)piperidin-1-yl)-3-oxopropanenitrile (CP690,550). J. Med. Chem. 51: 8012–8018. 28. Onda, M., S. Nagata, Y. Tsutsumi, J. J. Vincent, Q. Wang, R. J. Kreitman, B. Lee, and I. Pastan. 2001. Lowering the isoelectric point of the Fv portion of recombinant immunotoxins leads to decreased nonspecific animal toxicity without affecting antitumor activity. Cancer Res. 61: 5070–5077. 29. Onda, M., M. Willingham, S. Nagata, T. K. Bera, R. Beers, M. Ho, R. Hassan, R. J. Kreitman, and I. Pastan. 2005. New monoclonal antibodies to mesothelin useful for immunohistochemistry, fluorescence-activated cell sorting, Western blotting, and ELISA. Clin. Cancer Res. 11: 5840–5846. 30. Slack, J., G. P. Der-Balian, M. Nahm, and J. M. Davie. 1980. Subclass restriction of murine antibodies. II. The IgG plaque-forming cell response to thymusindependent type 1 and type 2 antigens in normal mice and mice expressing an X-linked immunodeficiency. J. Exp. Med. 151: 853–862. 31. Kudlacz, E., B. Perry, P. Sawyer, M. Conklyn, S. McCurdy, W. Brissette, M. Flanagan And, and P. Changelian. 2004. The novel JAK-3 inhibitor CP690550 is a potent immunosuppressive agent in various murine models. Am. J. Transplant. 4: 51–57. 32. Nosaka, T., J. M. van Deursen, R. A. Tripp, W. E. Thierfelder, B. A. Witthuhn, A. P. McMickle, P. C. Doherty, G. C. Grosveld, and J. N. Ihle. 1995. Defective lymphoid development in mice lacking Jak3. Science 270: 800–802. 33. Thomis, D. C., C. B. Gurniak, E. Tivol, A. H. Sharpe, and L. J. Berg. 1995. Defects in B lymphocyte maturation and T lymphocyte activation in mice lacking Jak3. Science 270: 794–797.

55