The Journal of

An Anti-IL-12p40 Antibody Down-Regulates Type 1 , , and IL-12/IL-23 in Psoriasis1

Eiko Toichi,*† Gisela Torres,* Thomas S. McCormick,2* Timothy Chang,* Mary Ann Mascelli,‡ Catharine L. Kauffman,§ Nancy Aria,§ Alice B. Gottlieb,¶ Daniel E. Everitt,‡ Bart Frederick,‡ Charles E. Pendley,‡ and Kevin D. Cooper*ʈ

Psoriasis is characterized by activation of T cells with a type 1 profile. IL-12 and IL-23 produced by APCs are essential for inducing Th1 effector cells. Promising clinical results of administration of an Ab specific for the p40 subunit of IL-12 and IL-23 (anti-IL-12p40) have been reported recently. This study evaluated histological changes and mRNA expression of relevant cytokines and chemokines in psoriatic skin lesions following a single administration of anti-IL-12p40, using immunohistochemistry and real-time RT-PCR. Expression levels of type 1 cytokine (IFN-␥) and chemokines (IL-8, IFN-␥-inducible protein-10, and MCP-1) were significantly reduced at 2 wk posttreatment. The rapid decrease of these expression levels preceded clinical response and histologic changes. Interestingly, the level of an anti-inflammatory cytokine, IL-10, was also significantly reduced. Significant reductions in TNF-␣ levels and infiltrating T cells were observed in high responders (improvement in clinical score, >75% at 16 wk), but not in low responders. Of importance, the levels of APC cytokines, IL-12p40 and IL-23p19, were significantly decreased in both responder populations, with larger decreases in high responders. In addition, baseline levels of TNF-␣ significantly correlated with the clinical improvement at 16 wk, suggesting that these levels may predict therapeutic responsiveness to anti- IL-12p40. Thus, in a human Th1-mediated disease, blockade of APC cytokines by anti-IL-12p40 down-regulates expression of type 1 cytokines and chemokines that are downstream of IL-12/IL-23, and also IL-12/IL-23 themselves, with a pattern indicative of coordinated deactivation of APCs and Th1 cells. The Journal of Immunology, 2006, 177: 4917–4926.

soriasis vulgaris is a chronic inflammatory skin disease intralesional cyclosporine A (12) all attest to the ongoing and crit- with a complex pathophysiology. The cellular and molec- ical role of intralesional T cell activation. P ular mechanisms underlying the disease involve interrela- Many studies examining the cytokine profile of T cells derived tionships between hyperplastic epidermal keratinocytes and infil- from psoriatic lesions suggest that the type 1 cytokine-producing T trating leukocytes, including memory T cells, neutrophils, cell subset is a prominent component of the lesions (13–17). Pso- 3 dendritic cells (DCs), and (1–3). The clinical effi- riatic lesions showed elevated mRNA expression for type 1 cyto- by guest on September 30, 2021. Copyright 2006 Pageant Media Ltd. cacy of several interventions targeting T cells, including cyclo- kines (IFN-␥, IL-2, and TNF-␣), compared with lesion-free pso- sporine A (4), (5), CTLA4Ig (6), and Alefa- riatic skin and normal skin, without a significant component of cept (LFA-3 TIP) (7) demonstrates a critical role of T cell type 2 cytokines (IL-4, IL-5, and IL-10) (18). IFN-␥ was shown to activation in , specifically, that of memory/effector cells be an important cytokine involved in the ability of lesional T cell (7). Research using AGR129 mice demonstrated that retained T clones to induce psoriatic uninvolved basal keratinocytes to enter cells in tissue uninvolved by psoriasis could proliferate, leading the cell cycle (13). However, the presence of IFN-␥ does not prove eventually to the development of psoriatic plaques (8). Studies that its role is critical; in experimental autoimmune encephalomy- demonstrating increased activity of lesional APCs (9, 10), their elitis, for example, treatment with anti-IFN-␥ Abs does not result apposition to psoriatic T cells undergoing activation (2), their re- http://classic.jimmunol.org duction during cyclosporine A therapy (11), and the efficacy of in a decrease in clinical severity (19, 20). Further implicating the likely involvement of Th1 cells is the fact that inhibition of TNF-␣, a cytokine that is produced (although not exclusively) by Th1 cells, *Department of Dermatology, Case Western Reserve University, Cleveland, OH results in clinical improvement of psoriasis (21, 22). 44106; †Kyoto University Graduate School of Medicine, Kyoto, Japan; ‡Centocor, Inc., Malvern, PA 19355; §Division of Dermatology, Georgetown University, Wash- IL-12, a heterodimer formed from p40 and p35 subunits, is a key ington, DC 20007; ¶Clinical Research Center, University of Medicine and Dentistry cytokine in Th1 differentiation and proliferation (23–26). Produced of New Jersey, Robert Wood Johnson Medical School, New Brunswick, NJ 08901; ʈ Downloaded from and Veterans Affairs Medical Center, Cleveland, OH 44106 by activated APCs (macrophages and DCs, including human epi- Received for publication July 15, 2005. Accepted for publication June 22, 2006. dermal Langerhans cells), IL-12 strongly promotes differentiation ␥ The costs of publication of this article were defrayed in part by the payment of page of naive T cells into IFN- -producing Th1 cells (25, 27–29). It has charges. This article must therefore be hereby marked advertisement in accordance been reported that both IL-12p40 mRNA and IL-12p70 het- with 18 U.S.C. Section 1734 solely to indicate this fact. erodimer are increased in psoriatic skin lesions (30). IL-23, a re- 1 This work was supported by National Institutes of Health Grants P30AR39750-14 cently discovered member of the IL-12 family, is a disulfide- (to K.D.C. and T.S.M.) and NIH AI41766-09 (to K.D.C.). This study also received funding from Centocor, Inc. bridged complex of a unique p19 subunit and the p40 subunit of 2 Address correspondence and reprint requests to Dr. Thomas S. McCormick, De- IL-12 (26, 31). IL-23 possesses similar as well as distinct proper- partment of Dermatology, Case Western Reserve University and, University Hospi- ties from those of IL-12. IL-23 is particularly efficient at inducing tals of Cleveland, Biomedical Research Building 530, 2109 Adelbert Road, Cleve- land, OH 44106. E-mail address: [email protected] the proliferation of memory Th1 cells (31). IL-23p19 transgenic 3 Abbreviations used in this paper: DC, ; PASI, Psoriasis Area and Se- mice demonstrate multiorgan inflammation, including skin (32). verity Index; IP-10, IFN-␥-inducible protein 10. Keratin 14/p40 transgenic mice that produce IL-23, but not IL-12,

Copyright © 2006 by The American Association of Immunologists, Inc. 0022-1767/06/$02.00 4918 ANTI-IL-12p40 THERAPY IN PSORIASIS

demonstrate skin inflammation (33), and unlike IL-12, the expres- Immunohistochemical staining sion of IL-23 is critical for experimental autoimmune encephalo- Frozen 5-␮m sections from one-half of skin biopsies were fixed in acetone. myelitis (34). Target organ expression of IL-23, consistent with its Sections were pretreated with PBS containing 0.3% hydrogen peroxide to production by microglia and macrophages, is thought to be critical block endogenous peroxidase activity, and then blocked with 10% normal for its role (35), and importantly, IL-23p19 mRNA is increased in goat serum. Sections were incubated with mouse anti-human CD3 Ab or psoriatic skin (36). mouse anti-human CD11c (IgG2; DakoCytomation) for1hatroom tem- perature. After washing, sections were exposed to a goat anti-mouse bio- Early work in murine models indicates that administration of Ab tinylated secondary Ab (DakoCytomation) for 30 min at room temperature. specific for the IL-12p40 subunit (anti-IL-12p40) can prevent and After subsequent washing, sections were exposed to avidin/HRP (Dako- also ameliorate autoimmune-like conditions (37–44). Anti-IL- Cytomation) for 30 min, washed, and developed with diaminobenzidine 12p40 abolished the development of psoriasis in mice, even when followed by copper sulfate. Sections were then incubated with methyl green or hematoxylin for counterstain. A negative control was established administered after transfer of the T cell subset that induced the using a matched isotype primary Ab and developed as described above. psoriasis-like skin disorder, and led to the down-regulation of Frozen sections were also stained using H&E for assessment of epidermal IFN-␥ and TNF-␣ (44). This is an important finding that suggests thickness. Electronic images were acquired using a Zeiss Axiophot micro- that anti-IL-12p40 may be a valuable treatment strategy against scope and an Axiocam HRc-cooled charge-coupled device camera, con- psoriasis, as well as a good tool in determining whether activated trolled by Axiovision software, version 4.4. Epidermal thickness measure- ment and positive cell counts were performed using computer-assisted T cells critical for psoriasis are indeed Th1 cells and IL-12/IL-23 image analysis. Staining and measurements were performed on 12 of 18 dependent. patients, because sufficient quantity of tissues at multiple time points was Anti-IL-12p40 used in our study is a human IgG1 ␬ mAb that only available for these patients. binds to the p40 subunit of human IL-12 and IL-23 and prevents its Quantitative RT-PCR interaction with IL-12R␤1. This first-in-human study of the ad- ministration of anti-IL-12p40 in patients with moderate-to-severe Total RNA was extracted from one-half of a punch biopsy specimen, and psoriasis vulgaris, showed remarkable clinical responses (45). the other half was retained for immunohistochemistry analysis. Total RNA was obtained using an RNeasy mini-kit (Qiagen) following tissue pulver- Safety results showed that anti-IL-12p40 was well tolerated, and ϩ ization in a Mikrodismembrator. RNA was reverse transcribed using 200 U only transient variable decreases in CD4 T cells and of Moloney-murine leukemia virus reverse transcriptase (Invitrogen Life CD16ϩCD56ϩ NK cells were noted in some patients (45). Technologies) for 60 min at 37°C in the presence of 50 mM Tris-HCl (pH The study presented here was intended to evaluate histology and 8.3), 75 mM KCl, 3 mM MgCl2, 10 mM DTT, 100 ng of random hexamer cytokine expression in psoriatic skin, before and after anti- primers (Invitrogen Life Technologies), 0.5 mM dNTPs, and 40 U of re- combinant RNase inhibitor (Fisher). For every reaction set, some RNA IL-12p40 administration, and assess the association with clinical samples were performed without Moloney-murine leukemia virus reverse response. The objectives were 2-fold: first, to evaluate the early transcriptase to provide a negative control in subsequent PCR. cDNA was effect of anti-IL-12p40 administration on cytokine and amplified in the presence of -specific primers and probes (FAM re- expression in an inflammatory pathway; and second, to elucidate porter for specific cytokines and VIC reporter for 18S rRNA) and TaqMan universal master mix (PE Applied Biosystems) in a 96-well microtiter plate the molecular mechanism of psoriasis and reveal the mechanism of format on an ABI PRISM 7700 Sequence Detection System (PE Applied action for anti-IL-12p40 therapy in psoriasis. Biosystems). Each PCR was performed in triplicate, using the following conditions: 2 min at 50°C and 10 min at 95°C followed by 40 cycles of 15 s at 95°C and 1 min at 60°C. Ten-fold serial dilutions of cDNA plasmid for

by guest on September 30, 2021. Copyright 2006 Pageant Media Ltd. Materials and Methods each target gene that contained known copy numbers were amplified with Study protocol and patient eligibility the samples during the same PCR. The absolute copy numbers of IL- 12p40, IL-12p35, IL-23p19, IL-8, IL-10, IL-18, RANTES (CCL5), MCP-1 Patients with moderate-to-severe plaque psoriasis, diagnosed at least 6 mo (CCL2), IFN-␥, TNF-␣, and IFN-␥-inducible protein-10 (IP-10; CXCL10) before screening, involving at least 3% body surface area with at least two were calculated against input copy numbers of plasmid standards for each plaques located on either the trunk or extremities, were enrolled in the target gene, and then normalized to the copy numbers of housekeeping first-in-human, phase 1, open-label, single administration, dose-escalating gene 18S rRNA to control the variations in starting RNA amounts. For study of anti-IL-12p40. Details of the study design as well as the safety, CD2 analysis, serial dilutions of a cDNA sample were used as relative pharmacokinetic, and clinical response results are reported elsewhere (45). standards, and relative quantification was made for each sample. Each patient was assigned to one of four dose groups and received a single i.v. infusion of 0.1 mg/kg (n ϭ 4), 0.3 mg/kg (n ϭ 4), 1.0 mg/kg (n ϭ 5), Statistical analysis http://classic.jimmunol.org or 5.0 mg/kg (n ϭ 5) of anti-IL-12p40, administered over a minimum of 120 min. A two-tailed paired t test was used to evaluate the change from baseline in The study was conducted in conformance with the regulations estab- histology and mRNA expression at indicated time points. Pearson corre- lished for the Protection of Human Subjects (21 Code of Federal Regula- lation analysis was used to determine the significance of correlation be- tions, Part 50) and Institutional Review Boards (21 Code of Federal Reg- tween baseline cytokine expression and percent improvement in PASI for ulations, Part 56), in accordance with Good Clinical Practices, and in TNF-␣ analysis. Partial correlation to remove the effect of dose was cal- compliance with local regulations. Before any study-specific procedures culated by first-fit regression models of both variables on dose, and the were performed, all patients provided written informed consent. Pearson correlation of the residuals of the two models was then calculated. Downloaded from Pearson correlation for baseline cytokine and percent change in cytokine Clinical response assessments correlation was also used. Clinical response was evaluated using the Psoriasis Area and Severity In- Results dex (PASI) score (46). The PASI combines scores for the degree of lesional erythema, induration, and desquamation, and the percentage of body sur- Clinical response to anti-IL-12p40 treatment face area affected. PASI scores were assessed at baseline, 72 h, and at 1, 2, Eighteen patients with moderate-to-severe psoriasis vulgaris re- 4, 8, 12, and 16 wk after administration of anti-IL-12p40. ceived a single i.v. administration of 0.1, 0.3, 1.0, or 5.0 mg/kg of anti-IL-12p40 according to dose group assignment. The demo- Skin biopsy graphics, baseline disease characteristics, safety, and pharmacoki- A psoriatic lesion, located on the trunk or extremities, was identified by the netic results, and detailed analysis of clinical response of these investigator to be the target site for biopsy collection. Punch biopsies (6 patients were previously reported (45). A clinical response profile, mm) were obtained from the identified target lesions 24 h before anti-IL- 12p40 administration (baseline) and again at intervals of 48 h and 2 wk as measured by percent improvement from baseline in PASI, for after anti-IL-12p40 administration. The biopsy samples were flash frozen the 18 individual patients in the four dose groups is shown in Fig. in liquid nitrogen and stored at Ϫ80°C before sample processing. 1. Clinical improvement was evaluated at 72 h and at 1, 2, 4, 8, 12, The Journal of Immunology 4919

and G). At 48 h and 2 wk after anti-IL-12p40 administration, a decrease of epidermal acanthosis was not evident, and the contin- ued presence of CD3ϩ cells and CD11cϩ cells in the lesion was observed (Fig. 2, B, C, E, F, H, and I). Measurements of epidermal thickness showed that a decrease at 2 wk was not statistically sig- nificant in either high or low responders (Table I). Total CD3ϩ cells did not decrease in low responders; however, they decreased significantly in high responders at 2 wk ( p Ͻ 0.05; Table I). Al- though the reduction in CD3ϩ cells was not robust (17.3 Ϯ 7.9%), a significant decrease indicates the beginning of improvement in these patients. Epidermal CD3ϩ cells slightly decreased at 2 wk; however, this decrease was not significant in either group (Table I). Total CD11cϩ cells decreased at 2 wk, but the decrease was not statistically significant in both responder groups (Table I). It has been reported that the elevated mRNA level of T cell markers FIGURE 1. Percent improvement in PASI score after anti-IL-12p40 correlates with increased numbers of T cells in psoriatic lesions as treatment. The percent improvement in PASI score from baseline for each measured by histologic analysis (47). CD2 mRNA levels of even patient is plotted over the 16-wk evaluation period. Treatment doses were high responders changed a little; ϩ16.0 Ϯ 10.6% at 48 h and 0.1 mg/kg (n ϭ 4), 0.3 mg/kg (n ϭ 4), 1.0 mg/kg (n ϭ 5), and 5.0 mg/kg Ϫ15.9 Ϯ 6.4% at 2 wk. The changes of CD2 levels in low re- (n ϭ 5). sponders were ϩ2.3 Ϯ 2.4% at 48 h and Ϫ18.4 Ϯ 10.9% at 2 wk. These decreases in the number of CD3ϩ T cells and CD2 mRNA ␥ and 16 wk after anti-IL-12p40 administration. Noticeable clinical levels can be compared with a decrease at 2 wk of IFN- mRNA levels of ϳ60% as discussed later in Fig. 4A. Thus, epidermal improvement across patients was initially detected at 4 wk or ear- ϩ lier, with maximal response observed within 12–16 wk. Overall, thickness and lesional CD11c DCs decreased at 2 wk, but these 12 of 18 patients achieved at least a 75% improvement from their decreases were not statistically significant. Reduction in lesional T baseline PASI score (high responders), which was sustained cells in high responders at 2 wk was statistically significant, but not through the 16-wk evaluation period. Of the remaining 6 patients, sufficient to explain the more dramatic decrease in T cell cytokine 3 had a PASI improvement Ͼ25 and Ͻ75%, and 3 had a PASI mRNA expression. improvement Յ25% (these 6 patients were classified as low re- sponders for this analysis). No patient showed exacerbation during Effects of anti-IL-12p40 treatment on cutaneous cytokine and 2–16 wk compared with baseline. The baseline PASI scores of low chemokine mRNA expression responders (15.7 Ϯ 4.4) did not have significant differences from Expression levels of 11 known or thought to play a role in those of high responders (13.1 Ϯ 1.8). the pathogenesis of psoriasis were analyzed by real-time PCR be- fore and after anti-IL-12p40 administration. Immunohistochemical analysis of lesional T cells and DCs

by guest on September 30, 2021. Copyright 2006 Pageant Media Ltd. First, the effect of anti-IL-12p40 administration on mRNA ex- following anti-IL-12p40 treatment pression of each gene across all patients was assessed. Expression To assess the effect of anti-IL-12p40 administration on lesional T levels of a type 1 cytokine, IFN-␥, across all patients were signif- cells and DCs, frozen sections from punch biopsies were stained icantly reduced in biopsies taken 2 wk after anti-IL-12p40 admin- with anti-CD3 or anti-CD11c Ab. Baseline biopsies showed epi- istration compared with baseline ( p ϭ 0.00001) (Fig. 3A). Indi- dermal acanthosis, a high infiltration of CD3ϩ T cells and CD11cϩ vidually, 17 of the 18 patients had reduced levels of IFN-␥ at this DCs into the lesion including epidermis and dermis (Fig. 2, A, D, time point. Reduction in expression levels of a chemokine, IL-8, http://classic.jimmunol.org

FIGURE 2. Histological changes following anti-IL-

Downloaded from 12p40 treatment. Histology and immunohistochemistry of representative patients (high responders) are shown. Biopsies were obtained from the psoriatic skin lesions at baseline (A, D, and G), 48 h (B, E, and H),and2wk(C, F, and I) after administration of anti-IL-12p40. One-half of each biopsy specimen was used for histology and immunohistochemistry. Sections were stained with H&E (A–C), anti-CD3 Ab (D–F), or anti-CD11c Ab (G–I). 4920 ANTI-IL-12p40 THERAPY IN PSORIASIS

Table I. Quantitation of epidermal thickness, CD3-, and CD11c-positive cells

Epidermal Epidermal Total CD3ϩ Total CD11cϩ Thicknessa pb CD3ϩ Cellsc p Cells p Cells p

High responders (n ϭ 9) Baseline 304.9 Ϯ 17.8d 65.5 Ϯ 7.9 161.4 Ϯ 23.0 212.0 Ϯ 28.7 Week 2 290.8 Ϯ 25.4 0.48 48.5 Ϯ 5.6 0.082 122.2 Ϯ 12.3 0.025 178.4 Ϯ 15.5 0.22

Low responders (n ϭ 3) Baseline 362.4 Ϯ 71.7 58.1 Ϯ 18.5 118.3 Ϯ 14.8 236.0 Ϯ 65.2 Week 2 252.0 Ϯ 43.1 0.063 43.9 Ϯ 8.7 0.30 115.7 Ϯ 0.9 0.87 207.0 Ϯ 76.8 0.16

a Values of epidemal thickness (in micrometers). b Paired t test. c Number of indicated cells in epidermis (epidermal) or epidermis and dermis (total) per low power field (ϫ10). d Mean Ϯ SE.

was also significant at 2 wk ( p Ͻ 0.05), with 17 of the 18 patients lower compared with those at 48 h. In contrast to what was seen for having reduced IL-8 levels (Fig. 3B). It is noted that individual IL-12p40 and IL-23p19, the expression levels of IL-12p35 across expression levels of IL-8 at baseline were highly variable across all patients were significantly increased at 2 wk ( p Ͻ 0.01) (Fig. the patient population, with up to a 504-fold difference between 3E). Note that the copy number for IL-12p35 subunit was the patients having the highest and lowest levels. Among other type 1 lowest amplified and thus subject to experimental variation. Inter- cytokines and chemokines, the expression levels of TNF-␣, IP-10, estingly, expression levels of an anti-inflammatory cytokine, IL- and MCP-1 across all patients were also significantly reduced at 2 10, was significantly reduced across all patients at 2 wk ( p ϭ wk ( p Ͻ 0.05, p ϭ 0.001, p ϭ 0.002, respectively; data not 0.00001) (Fig. 3F), with 17 of the 18 patients having reduced shown). Expression levels of another inflammatory cytokine, IL- IL-10 levels. These data indicate that administration of anti-IL- 18, did not show any significant changes at 2 wk (data not shown). 12p40 effectively down-regulates the expression of cytokines and For APC cytokines, expression levels of both IL-12p40 and IL- chemokines associated with activation of Th1 cells and APCs. 23p19 were significantly reduced at 2 wk ( p ϭ 0.007 and p ϭ Next, patients’ mRNA expression levels were examined with 0.0001, respectively) (Fig. 3, C and D). Individually, 16 of the 18 consideration to subsequent clinical responses to the anti-IL-12p40 patients had reduced levels of IL-12p40, and 17 of the 18 patients therapy. Expression levels of IFN-␥ were significantly reduced at had reduced levels of IL-23p19. One patient who exhibited an 2 wk after anti-IL-12p40 administration compared with baseline in increased level of IL-12p40 at 2 wk compared with baseline also both high responders and low responders with 57 and 61% reduc- showed a slight elevation in several other cytokines including IL- tion on average, respectively ( p ϭ 0.001 and p ϭ 0.004, respec- 23p19 at 2 wk. In fact, this patient showed transient increases in tively) (Fig. 4A). In high responders, the levels of TNF-␣ were by guest on September 30, 2021. Copyright 2006 Pageant Media Ltd. the levels of these cytokines at 48 h, so the levels at 2 wk were still significantly reduced at 2 wk ( p Ͻ 0.05), whereas the reduction in http://classic.jimmunol.org

FIGURE 3. Cytokine mRNA expression before and after anti-IL-12p40 treatment. RNA was pre- pared from the one-half of punch biopsies obtained from the psoriatic skin lesions at baseline and 2 wk after anti-IL-12p40 administration. Quantitative RT-

Downloaded from PCR was performed for the indicated genes. Gene expression levels were normalized to 18S rRNA. The expression levels of each cytokine for all the patients (n ϭ 18) at baseline and 2 wk posttreatment are presented. The mean Ϯ SE of all patients is in- ,ءء ;p Ͻ 0.05 ,ء .(dicated by the diamond symbol (᭜ p Ͻ 0.01; statistically significant differences be- tween baseline and 2 wk posttreatment. The Journal of Immunology 4921

FIGURE 4. Changes in the ex- pression levels of type 1 cytokines and chemokines examined following clinical response to anti-IL-12p40 treatment. Data represent the mean Ϯ SE of expression levels of each cyto- kine at baseline, 48 h and 2 wk post- treatment for patients who achieved Ն75% (high responders) and Ͻ75% (low responders) improvement in ,ءء ;p Ͻ 0.05 ,ء .PASI from baseline p Ͻ 0.01; statistically significant dif- ferences between baseline and 2 wk posttreatment.

low responders was not significant (Fig. 4B). At 48 h, there was a Changes in IL-23p19 significantly correlated with changes in IL-8, rapid decrease of IL-8 expression levels only in high responders IL-10, and the other subunit of IL-23, IL-12p40 (all values of p Ͻ (Fig. 4C), although the decrease was not statistically significant. 0.01), in addition to the correlation with IFN-␥. The strongest cor- Eight of 12 high responders showed such large decreases at 48 h. relation observed was between the changes of IL-12p40 and IL-8 The expression levels of chemokines, IP-10 and MCP-1, signifi- (r ϭ 0.93, p Ͻ 0.01). In addition to those discussed above, Table cantly decreased from baseline in both high responders and low II lists the strength of correlation observed among the changes in responders at 2 wk (all values of p Ͻ 0.05) (Fig. 4, D and F). expression of cytokines and chemokines associated with a type 1 by guest on September 30, 2021. Copyright 2006 Pageant Media Ltd. Reductions from baseline in RANTES levels were also observed response. for both responder groups at 2 wk; however, the decreases were not statistically significant (Fig. 4E). Expression levels of IL-18 Baseline cytokine level correlation were almost constant and did not show significant changes in ei- We next examined the potential relationships between baseline ther responder group at 2 wk (Fig. 4G). For APC cytokines, both cytokine mRNA expression levels across all patients. As shown in high responders and low responders exhibited significant decreases Table III, correlation was observed among a number of baseline in the levels of both subunits, IL-12p40 ( p ϭ 0.03 and p ϭ 0.001, cytokine levels. Baseline levels of IFN-␥ and IP-10 demonstrated respectively) (Fig. 5A) and IL-23p19 ( p ϭ 0.002 and p ϭ 0.007, the strongest correlation among all the cytokines examined (r ϭ respectively) (Fig. 5B). The decreases at 2 wk from baseline in 0.80, p Ͻ 0.01). IFN-␥ and IP-10 each also demonstrated statisti- http://classic.jimmunol.org IL-12p40 and IL-23p19 levels were larger in high responders than cally significant correlation with 7 of 10 other cytokines. For ex- in low responders, mainly due to higher baseline levels in high ample, IFN-␥ correlated with IP-10, IL-23p19, TNF-␣, RANTES, responders (Fig. 5, A and B). Expression levels of IL-12p35 and IL-10 ( p Ͻ 0.01) and with less strength of correlation with trended upward at 2 wk in both high responders and low respond- IL-12p40 and MCP-1 ( p Ͻ 0.05). Baseline IL-23p19 levels sig- ers, although neither of them achieved a change from baseline nificantly correlated with baseline levels of the other subunit, IL- level that reached statistical significance (Fig. 5C). IL-10 levels 12p40 ( p Ͻ 0.01). Baseline TNF-␣ mRNA levels had a distinctive Downloaded from were significantly reduced at 2 wk in both high responders and low profile, correlating not only with levels of IFN-␥, but also with responders ( p ϭ 0.0001 and p Ͻ 0.05, respectively) (Fig. 5D). those of RANTES ( p Ͻ 0.01), and less strongly with IL-10 and Thus, the major difference between high responders and low re- IL-23p19 ( p Ͻ 0.05). In contrast, IL-18 and IL-12p35 baseline sponders is a significant reduction in TNF-␣ levels at 2 wk and levels did not correlate significantly with those of any of other greater magnitude of decrease in the levels of IL-12p40 and IL- cytokines examined. 23p19 at 2 wk in high responders. Correlation of TNF-␣ with the efficacy of anti-IL-12p40 Correlation among the changes of cytokine and chemokine treatment expression following anti-IL-12p40 administration A comparison of gene expression levels of high responders vs low Correlation among percent change at 2 wk from baseline values of responders before treatment allows a retrospective analysis of dif- cytokines and chemokines across all patients is presented in Table ferences in the patient population that may explain differential re- II. We found that the extent of reduction in IFN-␥ mRNA expres- sponsiveness to the treatment. The mRNA expression levels of sion strongly correlated with the reductions in mRNA expression TNF-␣ at baseline in high responders were significantly higher of IP-10, IL-23p19, IL-12p40, and IL-8 (all values of p Ͻ 0.01). than those in low responders ( p ϭ 0.02) (Fig. 6A). Across all 4922 ANTI-IL-12p40 THERAPY IN PSORIASIS

creases in epidermal thickness and the number of CD11cϩ DCs were not yet observed. A decrease in the number of T cells in the lesion, although statistically significant in high responders, was not robust (ϳ17%) at this time point (Table I). Therefore, it is unlikely that reduced IFN-␥ mRNA levels are due to decreased infiltration of T cells at this early time point. It has been reported that psoriatic lesions, which improved by other treatments, began to show large and significant decreases in epidermal thickness and lesional T cells and DCs at 4–6 wk after treatment (47–49). Therefore, we presume that all of these markers of histologic changes become more evident at time points later than 2 wk. This finding is con- sistent with our prior observations that induction of the basal ker- atinocyte cell cycle in psoriatic skin is critically dependent upon IFN-␥ (13), and with the report that acanthosis in a murine model of psoriasis is IFN-␥ dependent (44). Thus, cytokine gene expres- sion level is a sensitive marker for anti-IL-12p40 therapy at this early time point. Several chemokines have been reported to be up-regulated in psoriatic skin lesions (50–52). IP-10, produced by keratinocytes with the stimulation of IFN-␥, attracts CXCR3ϩ T cells into the inflamed skin. RANTES induces migration and activation of CCR5ϩ T cells. MCP-1 stimulates migration and activation of and CCR2ϩ T cells. Two weeks after anti-IL-12p40 FIGURE 5. Changes in the expression levels of APC and anti- administration, the expression levels of all three of these chemo- inflammatory cytokines examined following clinical response to anti-IL- kines were reduced, with significant reductions for IP-10 and 12p40 treatment. Data represent the mean Ϯ SE of expression levels of MCP-1 noted (Fig. 4, D–F). Because both CXCR3 and CCR5 are each cytokine at baseline, 48 h, and 2 wk posttreatment for high responders associated with Th1 cells, and CCR2 is associated with both Th1 -p Ͻ 0.01; statistically significant and Th2 cells, reductions in these chemokines lead to less activa ,ءء ;p Ͻ 0.05 ,ء .and low responders differences between baseline and 2 wk posttreatment. tion of Th1 cells and monocytes, which may result ultimately in decreased recruitment of these cells into psoriatic skin lesions. IL-8, produced by keratinocytes, is a main chemoattractant for patients, the baseline mRNA expression levels of TNF-␣ (y) sig- neutrophils into the epidermis. The baseline IL-8 expression levels nificantly correlated with the percent improvement in PASI at 16 we observed were highly variable among patients (Fig. 3B), and ϭ ϭ wk (x) posttreatment (rxy 0.56, p 0.016) (Fig. 6B). This result did not correlate with baseline PASI scores (data not shown). Al- was further confirmed by calculating the partial correlation be- though the reason for such variability in IL-8 expression has not by guest on September 30, 2021. Copyright 2006 Pageant Media Ltd. ϭ tween these two variables with effect of dose removed (rxydose been clarified, it is notable that expression of IL-8 was reduced in 0.65, p ϭ 0.0032). No other cytokine showed a correlation be- almost all of our patients at 2 wk following anti-IL-12p40 therapy tween baseline mRNA levels and PASI improvement. These re- (Fig. 3B). In addition, in high responders, large decreases in IL-8 sults suggest that the baseline TNF-␣ mRNA level may be pre- levels were observed at 48 h (Fig. 4C), which suggests rapid down- dictive of the responsiveness to anti-IL-12p40 treatment. regulation of the level of keratinocyte activation. It has been reported that IL-12 and IL-23 are up-regulated in Discussion psoriatic skin (30, 36). These two cytokines are produced by This study of anti-IL-12p40 therapy in patients with psoriasis dem- APCs. Thus, it is likely that anti-IL-12p40 effectively blocked the onstrated potent and rapid effects on cytokine and chemokine ex- action of intracutaneous IL-12 and IL-23, from lesional APCs on http://classic.jimmunol.org pression in psoriatic skin. IL-12 and IL-23 with their distinct p35 lesional Th1 cells. According to recent reports, CD11cϩ DCs are and p19 subunits and shared IL-12p40 subunit have overlapping greatly increased in psoriatic skin compared with uninvolved skin, yet distinct functions in type 1 response and both cytokines are and the number of CD11cϩ cells exceeds the number of CD3ϩ T targeted by this anti-p40 Ab. Previous reports indicate that anti- cells (53). In patients treated with (anti-CD11a), IL-12p40 ameliorates a psoriasis-like skin disorder in rodent mod- CD11cϩ cells in psoriatic skin significantly decreased at 2 wk els (44). However, this is the first report to demonstrate that anti- posttreatment, probably because efalizumab directly inhibits the Downloaded from IL-12p40 administration to psoriasis patients down-regulates type migration of these cells (53). In our study, as shown in Table I, the 1 cytokines, chemokines and also IL-12/IL-23 themselves in number of CD11cϩ cells in both responder groups decreased, but lesional skin. this decrease was not yet statistically significant at 2 wk posttreat- It has been shown that IFN-␥ mRNA expression is elevated in ment. Therefore, the changes in cytokine mRNA levels relevant to psoriatic skin lesions and the lesions are infiltrated by numerous these cells likely predict initiation of cellular changes. Of partic- IFN-␥-producing T cells (13, 15, 17). In this study, the levels of ular interest is that 2 wk after anti-IL-12p40 administration, the IFN-␥ expression in the lesional skin was reduced in almost all expression levels of IL-12p40 mRNA themselves were decreased patients, as early as 2 wk after anti-IL-12p40 administration (Fig. in almost all patients (Fig. 3C), with larger decreases in high re- 3A). This result directly indicates that anti-IL-12p40 worked in sponders (Fig. 5A). There are several possible explanations for this vivo and suppressed type 1 response. In addition, this effect im- observation. First, IL-12-producing APCs themselves have IL-12 mediately preceded a noticeable clinical improvement. IFN-␥ ex- receptors (26), so neutralizing IL-12p40 could have inhibited a pression was reduced ϳ60%, whereas PASI was only improved positive-feedback loop for IL-12. Second, because IFN-␥ enhances ϳ30% in high responders at 2 wk. Cytokine mRNA change pre- transcription of IL-12p40 and p35 (54), reduced IFN-␥ would re- ceded histologic changes, because statistically significant de- sult in less stimulation for IL-12. Third, a decrease of cell-cell The Journal of Immunology 4923

Table II. Correlation among the changes of biomarkers at 2 wk

IFN-␥ IP-10 IL-23p19 IL-12p40 IL-8 TNF-␣ MCP-1 RANTES IL-10 IL-12p35 IL-18

IFN-␥ 0.83** 0.77** 0.81** 0.82** 0.45 0.47 0.40 0.47 0.47* Ϫ0.10 IP-10 0.83** 0.54* 0.71** 0.68** 0.48* 0.59** 0.29 0.21 0.47* 0.00 IL-23p19 0.77** 0.54* 0.75** 0.79** 0.43 0.29 0.54* 0.62** 0.15 Ϫ0.16 IL-12p40 0.81** 0.71** 0.75** 0.93** 0.50* 0.36 0.28 0.25 0.21 Ϫ0.13 IL-8 0.82** 0.68** 0.79** 0.93** 0.54* 0.37 0.26 0.37 0.12 Ϫ0.17 TNF-␣ 0.45 0.48* 0.43 0.50* 0.54* 0.61** 0.60** 0.30 0.35 Ϫ0.44 MCP-1 0.47 0.59** 0.29 0.36 0.37 0.61** 0.31 0.09 0.48* Ϫ0.15 RANTES 0.40 0.29 0.54* 0.28 0.26 0.60** 0.31 0.49* 0.37 Ϫ0.23 IL-10 0.47 0.21 0.62** 0.25 0.37 0.30 0.09 0.49* 0.07 Ϫ0.09 IL-12p35 0.47* 0.47* 0.15 0.21 0.12 0.35 0.48* 0.37 0.07 Ϫ0.22 IL-18 Ϫ0.10 0.00 Ϫ0.16 Ϫ0.13 Ϫ0.17 Ϫ0.44 Ϫ0.15 Ϫ0.23 Ϫ0.09 Ϫ0.22

Pearson correlation coefficients among changes of cytokine expression levels at 2 wk (relative percentage change from baseline, n ϭ 18). **, p Ͻ 0.01; *, p Ͻ 0.05.

interaction (such as CD40 expressed on APC with CD40L on ac- decreased, and this was true for both high responders and low tivated T cells) caused by reduced activation of infiltrating T cells responders (Fig. 5D). These data fit with the inconclusive IL-10 could result in decreased production of IL-12. Of additional im- clinical trials and suggest that the improvement of psoriasis by portance is that the expression levels of IL-23p19 were also re- anti-IL-12p40 is not mediated by either an absolute or a relative duced after anti-IL-12p40 administration (Fig. 3D), with larger increase of the anti-inflammatory cytokine IL-10. Given the ob- decrease in high responders (Fig. 5B). Although the regulation of served reduction of IL-10, there is no evidence that anti-IL-12p40 the production of this novel cytokine has not been as well studied administration induces a shift from a type 1 to a type 2 immune as that of IL-12, similar reasons may account for this observation. response. Although Th2 cells can produce IL-10, untreated psori- The coordinate reduction of IL-23p19 and p40 represents a pro- atic skin lesion has very few Th2 cells. Although keratinocytes can found effect on IL-23 production, which may in turn reduce the produce IL-10, they are less secretory for IL-10 than skin macro- generation of pathogenic memory/effector T cells. It is tempting to phages (64). Therefore, it is likely that IL-10 reduction posttreat- speculate that pathogenic T cell reduction could result in disease ment is indicative of reduced activation, as would be whose expression is clinically subthreshold. In an environment suggested by the concomitant reduction in IL-12p40, IL-23p19, where functional regulatory T cells in psoriatic lesions are dys- MCP-1, and TNF-␣. One proposed IL-10-mediated anti- functional and/or numerically insufficient (55), a treatment target- inflammatory mechanism is inhibition of IL-12 production via ing the pathogenic memory/effector T cells may result in a rebal- blocked IL-12 gene transcription (65, 66). A possible explanation ancing of the immune system such that regulatory T cells could for IL-10 reduction is that suppressed production of IL-12 and effectively restrain the remaining pathogenic memory/effector T other cytokines (including macrophage activators such as IFN-␥) cell population. Additionally, the regulation of IL-17, which has by anti-IL-12p40 eliminates the positive-feedback signals for by guest on September 30, 2021. Copyright 2006 Pageant Media Ltd. been reported to play a role in psoriasis, may also be critical to the IL-10 usually induced for cessation of the inflammatory response. outcome of anti-IL-12p40 therapy, given the proposed role of IL-18, a member of the IL-1 family that is constitutively ex- IL-23 to promote IL-17 production (56). Interesting concepts have pressed in the skin, acts in synergy with IL-12 to induce IFN-␥ been raised in recent reports suggesting a potential key role for (67). Of note, IL-18 can induce a type 2 response in the absence myeloid DC subpopulations (53) and plasmacytoid DCs (57). of IL-12; however, our patients did not demonstrate the increase Whether anti-IL-12p40 Ab therapy has an effect on these DC sub- in IL-10 that would be expected if a type 2 response were gener- populations remains to be investigated. However, as pointed out in ated. IL-18 is overexpressed in psoriatic lesional skin, suggesting the commentaries to Kauffman et al. (45), this therapy is the first that it plays a role in the pathogenesis of psoriasis (68). The ob- to selectively target human cytokines that are produced exclusively servation that no reduction in IL-18 expression at 2 wk after anti- http://classic.jimmunol.org by activated DCs (58) and anti-IL-12p40 may be targeting the IL-12p40 administration even in patients demonstrating rapid re- “next master cytokine(s) in the pathogenesis and therapy of pso- sponses in other parameters (Fig. 4G) suggests that anti-IL-12p40 riasis” (59). therapy does not affect the expression levels of IL-18 at least at an IL-10 is an anti-inflammatory cytokine that suppresses produc- early time point. tion of inflammatory cytokines and activation of DCs and mono- In inflammation, cytokines and chemokines function in a cas- cytes/macrophages. IL-10 is also a type 2 cytokine that suppresses cade of reactions. IFN-␥ and TNF-␣ stimulate the production of Downloaded from type 1 response. Therefore, antipsoriatic effects by an administra- chemokines by various cell types (52, 69, 70). Therefore, it is tion of recombinant human IL-10 have been studied. It has been possible that reductions of IFN-␥ and TNF-␣ by anti-IL-12p40 reported that IL-10 administration could enhance IL-10 mRNA could lead to a reduction in chemokines, resulting in decreased expression in the skin (60); however, two of seven patients re- activation and migration of T cells and other leukocytes, followed lapsed before study termination after 4 mo (61). In the most com- in turn by a further decrease in IFN-␥ and TNF-␣. Indeed, strong prehensive study, only a modest improvement (mean 35.6% im- correlation among the extent of reduction of several cytokines and provement in PASI) peaking at 8 wk and rebound flares afterward chemokines were observed at 2 wk following anti-IL-12p40 ad- have been reported (62). Expression levels of IL-10 in the psoriatic ministration (Table II). lesion and its change after therapies varied among reports. Con- TNF-␣, a type 1 and inflammatory cytokine, is increased in stant low expression of IL-10 after treatment with recombinant psoriatic skin lesions (71). TNF-␣ is produced by various types of human IL-11 (type 2 cytokines) (47), but also high baseline levels cells, including Th1 cells, mast cells, macrophages, DCs, and ker- of IL-10 and its reduction during PUVA therapy have been re- atinocytes. TNF-␣ increases DC migration, promotes T cell acti- ported (63). Interestingly, after anti-IL-12p40 administration, the vation and migration, increases keratinocyte proliferation, and expression of IL-10 was not increased but was in fact significantly leads to increased levels of NF-␬B in almost all cell types. It has 4924 ANTI-IL-12p40 THERAPY IN PSORIASIS

Table III. Correlation among biomarkers at baseline values

IFN-␥ IP-10 IL-23p19 IL-12p40 IL-8 TNF-␣ MCP-1 RANTES IL-10 IL-12p35 IL-18

IFN-␥ 0.80** 0.61** 0.52* 0.42 0.69** 0.56* 0.78** 0.76** 0.31 Ϫ0.06 IP-10 0.80** 0.63** 0.56* 0.57* 0.38 0.68** 0.65** 0.62** 0.30 Ϫ0.18 IL-23p19 0.61** 0.63** 0.64** 0.64** 0.51* 0.41 0.47 0.32 Ϫ0.04 0.05 IL-12p40 0.52* 0.56* 0.64** 0.62** 0.45 0.53* 0.12 0.42 Ϫ0.07 0.02 IL-8 0.42 0.57* 0.64** 0.62** 0.26 0.48* 0.27 0.30 Ϫ0.14 Ϫ0.07 TNF-␣ 0.69** 0.38 0.51* 0.45 0.26 0.35 0.61** 0.58* 0.13 0.07 MCP-1 0.56* 0.68** 0.41 0.53* 0.48* 0.35 0.27 0.59** 0.07 0.03 RANTES 0.78** 0.65** 0.47 0.12 0.27 0.61** 0.27 0.57* 0.41 0.03 IL-10 0.76** 0.62** 0.32 0.42 0.30 0.58* 0.59** 0.57* 0.22 Ϫ0.05 IL-12p35 0.31 0.30 Ϫ0.04 Ϫ0.07 Ϫ0.14 0.13 0.07 0.41 0.22 0.27 IL-18 Ϫ0.06 Ϫ0.18 0.05 0.02 Ϫ0.07 0.07 0.03 0.03 Ϫ0.05 0.27

Pearson correlation coefficients among cytokine expression levels at baseline (n ϭ 18). **, p Ͻ 0.01; *, p Ͻ 0.05.

been reported that TNF-␣ mRNA levels were higher in psoriatic examine the number of positive responders to treatment within our lesions compared with nonlesional skin and they decreased in re- subjects’ test-positive group. If we consider that TNF-␣ copy num- sponders after treatment (47). Because the origin of TNF-␣ is not ber of 100 or greater is the test-positive group, then 11 of 18 only Th1 cells, the production of TNF-␣ is not necessarily IL-12/ patients meet this criterion. Of the 11 patients, 9 patients achieved IL-23 dependent. An important finding is that blockade of IL- Ն75% PASI improvement (82% of the test-positives were also 12p40 down-regulated TNF-␣ expression in high responders (Fig. PASI improved). Of the 7 patients whose TNF-␣ copy number is 4B), which is consistent with the recent clinical results that anti- Ͻ100, 3 patients achieved Ն75% PASI improvement (only 43% of TNF-␣ therapy is effective in the treatment of psoriasis (21, 22). the test-negatives were PASI improved). As previously reported, The significance of baseline TNF-␣ expression levels in psori- high baseline PASI scores could explain poor improvement of asis patients is unknown. Baseline TNF-␣ levels do not seem to some low responders, but not all (45). Our results suggest that the reflect the severity of psoriasis, because they did not correlate with baseline TNF-␣ level might be a predictor for therapeutic re- baseline PASI scores in our study (data not shown). With a retro- sponsiveness to anti-IL-12p40 in psoriasis, potentially identi- spective analysis, we found that, across all patients, baseline fying a patient subset. It has been reported that some psoriatic TNF-␣ levels correlated significantly with the percent improve- patients have TNF-␣ 238A allele and that their transcription and ment in PASI at 16 wk (Fig. 6B). In each of the two lower dosage production of TNF-␣ is low (72). Alternatively, the baseline groups, the higher the patient baseline TNF-␣ level, the higher the TNF-␣ level in each patient might reflect an immunologically observed percent PASI improvement at 16 wk (data not shown). In responsive state at that time point. It is also possible that the the two higher dose groups, this correlation was not absolute; this TNF-␣ mRNA levels are not reflective of protein levels in the may be because the high degree of improvement obscured the tissue. It is possible that TNF-␣ protein levels have a close

by guest on September 30, 2021. Copyright 2006 Pageant Media Ltd. correlation (indeed, only one patient in the higher dose groups relationship to the degree of inflammation (e.g., PASI score), failed to achieve Ն75% PASI improvement). It is interesting to and mRNA levels are indicative of a likely responder to the http://classic.jimmunol.org Downloaded from

FIGURE 6. Baseline TNF-␣ mRNA expression levels are involved in the efficacy of anti-IL-12p40 treatment. A, Baseline TNF-␣ expression levels in .p Ͻ 0.05; statistically significant differences between two patient groups ,ء .high responders and those in low responders are presented. Bars, Mean values B, Correlation between the expression levels of TNF-␣ mRNA at baseline and the percent improvement in PASI at 16 wk posttreatment. r ϭ 0.56; p ϭ 0.016. The Journal of Immunology 4925

therapy. Posttranscriptional regulation and posttranslational References processing are important in TNF-␣ protein production (73, 74) 1. Valdimarsson, H., B. S. Baker, I. Jonsdottir, and L. Fry. 1986. Psoriasis: a disease and mRNA levels may also reflect increased half-life of mRNA. of abnormal keratinocyte proliferation induced by T lymphocytes. Immunol. To- day 7: 256–259. Because our study has only a limited number of patients, and 2. Morganroth, G. S., L. S. Chan, G. D. Weinstein, J. J. Voorhees, and TNF-␣ mRNA and protein processing is quite complicated, a K. D. Cooper. 1991. Proliferating cells in psoriatic dermis are comprised primar- ϩ more highly powered study specifically designed to address ily of T cells, endothelial cells, and factor XIIIa perivascular dendritic cells. ␣ J. Invest. Dermatol. 96: 333–340. TNF- levels and clinical improvement will be necessary to 3. Wrone-Smith, T., and B. J. Nickoloff. 1996. Dermal injection of immunocytes validate these observations. induces psoriasis. J. Clin. Invest. 98: 1878–1887. Among new antipsoriatic reagents, anti-IL-12p40 therapy 4. Brooks, D. B. 1986. Clearance of psoriasis with low dose cyclosporine. Br. Med. J. 293: 1098–1099. achieved high efficacy according to the data we currently have (7, 5. Gottlieb, S. L., P. Gilleaudeau, R. Johnson, L. Estes, T. G. Woodworth, 21, 22, 45, 47, 48). Our findings in this study have revealed the A. B. Gottlieb, and J. G. Krueger. 1995. Response of psoriasis to a lymphocyte- mechanism of action for anti-IL-12p40 therapy. Steps in a cuta- selective toxin (DAB389IL-2) suggests a primary immune, but not keratinocyte, pathogenic basis. Nat. Med. 1: 442–447. neous type 1 response consist of APC activation, T cell activation, 6. Abrams, J. R., M. G. Lebwohl, C. A. Guzzo, B. V. Jegasothy, M. T. Goldfarb, differentiation and proliferation of Th1 cells, migration of acti- B. S. Goffe, A. Menter, N. J. Lowe, G. Krueger, M. J. Brown, et al. 1999. vated memory Th1 cells to skin, and subsequent production of CTLA4Ig-mediated blockade of T-cell costimulation in patients with psoriasis vulgaris. J. Clin. Invest. 103: 1243–1252. inflammatory cytokines and chemokines. APC cytokines, IL-12/ 7. Ellis, C. N., and G. G. Krueger. 2001. Treatment of chronic plaque psoriasis by IL-23, are fairly “upstream” in the inflammatory cascade. Our selective targeting of memory effector T lymphocytes. N. Engl. J. Med. 345: 248–255. study demonstrated that blockade of IL-12p40 down-regulated 8. Boyman, O., H. P. Hefti, C. Conrad, B. J. Nickoloff, M. Suter, and F. O. Nestle. type 1 cytokines and chemokines that are downstream of IL-12/ 2004. Spontaneous development of psoriasis in a new animal model shows an IL-23. If IL-12/IL-23 were continuously produced by activated essential role for resident T cells and -␣. J. Exp. Med. 199: 731–736. APCs, the effect of neutralizing secreted IL-12/IL-23 would not 9. Baadsgaard, O., A. K. Gupta, R. S. Taylor, C. N. Ellis, J. J. Voorhees, and last. However, we also demonstrated that the blockade of IL-12p40 K. D. Cooper. 1989. Psoriatic epidermal cells demonstrate increased numbers and down-regulated mRNA for IL-12/IL-23 themselves, with larger function of non-Langerhans antigen-presenting cells. J. Invest. Dermatol. 92: 190–195. decreases in high responders, suggesting a single administration of 10. Nestle, F. O., L. A. Turka, and B. J. Nickoloff. 1994. Characterization of dermal anti-IL-12p40 leads to a long-lasting effect, likely breaking the dendritic cells in psoriasis: autostimulation of T lymphocytes and induction of APC-Th1 cell self-sustaining cycle proposed in psoriasis. These Th1 type cytokines. J. Clin. Invest. 94: 202–209. 11. Cooper, K. D., O. Baadsgaard, C. N. Ellis, E. Duell, and J. J. Voorhees. 1990. findings also suggest that activation of APCs, more upstream of Mechanisms of cyclosporine A inhibition of antigen-presenting activity in unin- targeted cytokines in the cascade, may be suppressed by this ther- volved and lesional psoriatic epidermis. J. Invest. Dermatol. 94: 649–656. apy. It is desirable to selectively target the type 1 response rather 12. Ho, V. C., C. E. M. Griffiths, C. N. Ellis, A. K. Gupta, C. C. McCuaig, B. J. Nickoloff, K. D. Cooper, T. A. Hamilton, and J. J. Voorhees. 1990. Intrale- than global immune suppression for the treatment of Th1-mediated sional cyclosporine in the treatment of psoriasis: a clinical, immunologic, and diseases. It has been recently reported that anti-IL-12p40 therapy pharmacokinetic study. J. Am. Acad. Dermatol. 22: 94–100. 13. Bata-Csorgo, Z., C. Hammerberg, J. J. Voorhees, and K. D. Cooper. 1995. Ki- was also effective in patients with active Crohn’s disease and in- netics and regulation of human keratinocyte stem cell growth in short-term pri- duced remission (75). This therapy may also be beneficial to pa- mary ex vivo culture: cooperative growth factors from psoriatic lesional T lym- tients with other Th1-mediated inflammatory diseases. phocytes stimulate proliferation among psoriatic uninvolved, but not normal, stem keratinocytes. J. Clin. Invest. 95: 317–327. In conclusion, this study indicates that the administration of anti- 14. Barker, J. N., G. D. Karabin, T. J. Stoof, V. J. Sarma, V. M. Dixit, and by guest on September 30, 2021. Copyright 2006 Pageant Media Ltd. IL-12p40 rapidly down-regulates mRNA expression of type 1 cy- B. J. Nickoloff. 1991. Detection of -␥ mRNA in psoriatic epidermis by tokines and chemokines that are downstream of APC cytokines, polymerase chain reaction. J. Dermatol. Sci. 2: 106–111. 15. Szabo, S. K., C. Hammerberg, Y. Yoshida, Z. Bata-Csorgo, and K. D. Cooper. and APC cytokines themselves in psoriatic skin lesions. Interest- 1998. Identification and quantitation of interferon-␥ producing T cells in psoriatic ingly, an anti-inflammatory cytokine, IL-10, is also down-regu- lesions: localization to both CD4ϩ and CD8ϩ subsets. J. Invest. Dermatol. 111: lated. High responders, but not low responders, have significant 1072–1078. 16. Schlaak, J. F., M. Buslau, W. Jochum, E. Hermann, M. Girndt, H. Gallati, K.-H. reductions in TNF-␣ levels and infiltrating T cells, and larger de- Buschenfelde, and B. Fleischer. 1994. T cells involved in psoriasis vulgaris be- creases in the levels of IL-12p40 and IL-23p19. A retrospective long to the Th1 subset. J. Invest. Dermatol. 102: 145–149. analysis suggests that baseline expression level of TNF-␣ may be 17. Austin, L. M., M. Ozawa, T. Kikuchi, I. B. Walters, and J. G. Krueger. 1999. The majority of epidermal T cells in psoriasis vulgaris lesions can produce type 1 a predictor of patient therapeutic responsiveness to anti-IL-12p40 cytokines, interferon-␥, -2, and tumor necrosis factor-␣, defining TC1 http://classic.jimmunol.org therapy. It is proposed that a cyclic interaction between APCs and (cytotoxic T lymphocyte) and TH1 effector populations: a type 1 differentiation bias is also measured in circulating blood T cells in psoriatic patients. J. Invest. Th1 cells mediated by IL-12/IL-23 and type 1 cytokines results in Dermatol. 113: 752–759. the maintenance of chronic inflammation in psoriasis. The obser- 18. Uyemura, K., M. Yamamura, D. F. Fivenson, R. L. Modlin, and B. J. Nickoloff. vations from this study strongly suggest that blockade of IL-12p40 1993. The cytokine network in lesional and lesion-free psoriatic skin is charac- terized by a T-helper type 1 cell-mediated response. J. Invest. Dermatol. 101: is an effective treatment strategy for psoriasis acting through con- 701–705. trol of type 1 cytokine cascade and IL-12/IL-23 production. This 19. Billiau, A., H. Heremans, F. Vandekerckhove, R. Dijkmans, H. Sobis, Downloaded from study also provides further evidence for the hypothesis that APC- E. Meulepas, and H. Carton. 1988. Enhancement of experimental allergic en- cephalomyelitis in mice by antibodies against IFN-␥. J. Immunol. 140: Th1 cell interaction is indeed a critical element of psoriasis. 1506–1510. 20. Lublin, F. D., R. L. Knobler, B. Kalman, M. Goldhaber, J. Marini, M. Perrault, C. D’Imperio, J. Joseph, S. S. Alkan, and R. Korngold. 1993. Monoclonal anti- Acknowledgments ␥-interferon antibodies enhance experimental allergic encephalomyelitis. Auto- We acknowledge technical assistance from A. Jovanovski, writing support immunity 16: 267–274. from N. Sobieck and S. Hobson of Precision Research, Inc., and statistical 21. Mease, P. J., B. S. Goffe, J. Metz, A. VanderStoep, B. Finck, and D. J. Burge. 2000. in the treatment of psoriatic and psoriasis: a randomised support from B. Zhou of Centocor, Inc., and Syamal K. Dey of Case West- trial. Lancet 356: 385–390. ern Reserve University. 22. Chaudhari, U., P. Romano, L. D. Mulcahy, L. T. Dooley, D. G. Baker, and A. B. Gottlieb. 2001. Efficacy and safety of infliximab monotherapy for plaque- type psoriasis: a randomised trial. Lancet 357: 1842–1847. Disclosures 23. Gately, M. K., L. M. Renzetti, J. Magram, A. S. Stern, L. Adorini, U. Gubler, and M. A. Mascelli, D. E. Everitt, B. Frederick, and C. E. Pendley, are em- D. H. Presky. 1998. The interleukin-12/interleukin-12- system: role in normal and pathologic immune responses. Annu. Rev. Immunol. 16: 495–521. ployed by Centocor, Inc., and T. S. McCormick, C. L. Kauffman, 24. Trinchieri, G. 1995. Interleukin-12: A proinflammatory cytokine with immuno- A. B. Gottlieb, and K. D. Cooper have received research support from regulatory functions that bridge innate resistance and antigen-specific adaptive Centocor. immunity. Annu. Rev. Immunol. 13: 251–276. 4926 ANTI-IL-12p40 THERAPY IN PSORIASIS

25. Trinchieri, G. 1998. Interleukin-12: a cytokine at the interface of inflammation 50. Raychaudhuri, S. P., W. Y. Jiang, E. M. Farber, T. J. Schall, M. R. Ruff, and and immunity. Adv. Immunol. 70: 83–243. C. B. Pert. 1999. Upregulation of RANTES in psoriatic keratinocytes: a possible 26. Trinchieri, G. 2003. Interleukin-12 and the regulation of innate resistance and pathogenic mechanism for psoriasis. Acta. Derm. Venereol. 79: 9–11. adaptive immunity. Nat. Rev. Immunol. 3: 133–146. 51. Gillitzer, R., R. Berger, V. Mielke, C. Muller, K. Wolff, and G. Stingl. 1991. 27. Hsieh, C. S., S. E. Macatonia, C. S. Tripp, S. F. Wolf, A. O’Garra, and Upper keratinocytes of psoriatic skin lesions express high levels of NAP-1/IL-8 K. M. Murphy. 1993. Development of TH1 CD4ϩ T cells through IL-12 pro- mRNA in situ. J. Invest. Dermatol. 97: 73–79. duced by Listeria-induced macrophages. Science 260: 547–549. 52. Giustizieri, M. L., F. Mascia, A. Frezzolini, O. De Pita, L. M. Chinni, 28. Manetti, R., P. Parronchi, M. G. Giudizi, M.-P. Piccinni, E. Maggi, G. Trinchieri, A. Giannetti, G. Girolomoni, and S. Pastore. 2001. Keratinocytes from patients and S. Romagnani. 1993. stimulatory factor () with atopic dermatitis and psoriasis show a distinct chemokine production profile induces T helper Type 1 (Th1)-specific immune responses and inhibits the de- in response to T cell-derived cytokines. J. Allergy Clin. Immunol. 107: 871–877. velopment of IL-4-producing Th cells. J. Exp. Med. 177: 1199–1204. 53. Lowes, M. A., F. Chamian, M. V. Abello, J. Fuentes-Duculan, S. L. Lin, 29. Kang, K., M. Kubin, K. D. Cooper, S. R. Lessin, G. Trinchieri, and A. H. Rook. R. Nussbaum, I. Novitskaya, H. Carbonaro, I. Cardinale, T. Kikuchi, et al. 2005. 1996. IL-12 synthesis by human Langerhans Cells. J. Immunol. 156: 1402–1407. Increase in TNF-␣ and inducible nitric oxide synthase-expressing dendritic cells 30. Yawalkar, N., S. Karlen, R. Hunger, C. U. Brand, and L. R. Braathen. 1998. in psoriasis and reduction with efalizumab (anti-CD11a). Proc. Natl. Acad. Sci. Expression of interleukin-12 is increased in psoriatic skin. J. Invest. Dermatol. USA 102: 19057–19062. 111: 1053–1057. 54. Ma, X. J., J. M. Chow, G. Gri, G. Carra, F. Gerosa, S. E. Wolf, R. Dzialo, and 31. Oppmann, B., R. Lesley, B. Blom, J. C. Timans, Y. Xu, B. Hunte, F. Vega, G. Trincheri. 1996. The interleukin 12 p40 gene promoter is primed by interfer- N. Yu, J. Wang, K. Singh, et al. 2000. Novel p19 protein engages IL-12p40 to on-␥ in monocytic cells. J. Exp. Med. 183: 147. form a cytokine, IL-23, with biological activities similar as well as distinct from 55. Sugiyama, H., R. Gyulai, E. Toichi, E. Garaczi, S. Shimada, S. R. Stevens, IL-12. Immunity 13: 715–725. T. S. McCormick, and K. D. Cooper. 2005. Dysfunctional blood and target tissue 32. Wiekowski, M. T., M. W. Leach, E. W. Evans, L. Sullivan, S. C. Chen, CD4ϩCD25high regulatory T cells in psoriasis: mechanism underlying unre- G. Vassileva, J. F. Bazan, D. M. Gorman, R. A. Kastelein, S. Narula, and strained pathogenic effector T cell proliferation. J. Immunol. 174: 164–173. S. A. Lira. 2001. Ubiquitous transgenic expression of the IL-23 subunit p19 56. Aggarwal, S., N. Ghilardi, M. H. Xie, F. J. de Sauvage, and A. L. Gurney. 2003. induces multiorgan inflammation, runting, infertility, and premature death. J. Im- Interleukin-23 promotes a distinct CD4 T cell activation state characterized by the munol. 166: 7563–7570. production of interleukin-17. J. Biol. Chem. 278: 1910–1914. 33. Kopp, T., P. Lenz, C. Bello-Fernandez, R. A. Kastelein, T. S. Kupper, and 57. Nestle, F. O., C. Conrad, A. Tun-Kyi, B. Homey, M. Gombert, O. Boyman, G. Stingl. 2003. IL-23 production by cosecretion of endogenous p19 and trans- G. Burg, Y. J. Liu, and M. Gilliet. 2005. Plasmacytoid predendritic cells initiate genic p40 in keratin 14/p40 transgenic mice: evidence for enhanced cutaneous psoriasis through interferon-␣ production. J. Exp. Med. 202: 135–143. immunity. J. Immunol. 170: 5438–5444. 58. Blauvelt, A. 2004. In this issue—full court press on psoriasis. J. Invest. Dermatol. 34. Cua, D. J., J. Sherlock, Y. Chen, C. A. Murphy, B. Joyce, B. Seymour, L. Lucian, 123: vii–viii. W. To, S. Kwan, T. Churakova, et al. 2003. Interleukin-23 rather than interleu- 59. Nestle, F. O., and C. Conrad. 2004. The IL-12 family member p40 chain as a kin-12 is the critical cytokine for autoimmune inflammation of the brain. Nature master switch and novel therapeutic target in psoriasis. J. Invest. Dermatol. 123: 421: 744–748. xiv–xv. 35. Li, J., B. Gran, G. X. Zhang, E. S. Ventura, I. Siglienti, A. Rostami, and 60. Asadullah, K., W. Sterry, K. Stephanek, D. Jasulaitis, M. Leupold, H. Audring, M. Kamoun. 2003. Differential expression and regulation of IL-23 and IL-12 H. D. Volk, and W. D. Docke. 1998. IL-10 is a key cytokine in psoriasis: proof subunits and receptors in adult mouse microglia. J. Neurol. Sci. 215: 95–103. of principle by IL-10 therapy: a new therapeutic approach. J. Clin. Invest. 101: 36. Lee, E., W. L. Trepicchio, J. L. Oestreicher, D. Pittman, F. Wang, F. Chamian, 783–794. M. Dhodapkar, and J. G. Krueger. 2004. Increased expression of interleukin 23 61. Friedrich, M., W. D. Docke, A. Klein, S. Philipp, H. D. Volk, W. Sterry, and J. Exp. Med. p19 and p40 in lesional skin of patients with psoriasis vulgaris. 199: K. Asadullah. 2002. Immunomodulation by interleukin-10 therapy decreases the 125–130. incidence of relapse and prolongs the relapse-free interval in psoriasis. J. Invest. 37. Leonard, J. P., K. E. Waldburger, and S. J. Goldman. 1995. Prevention of ex- Dermatol. 118: 672–677. perimental autoimmune encephalomyelitis by antibodies against interleukin 12. 62. Kimball, A. B., T. Kawamura, K. Tejura, C. Boss, A. R. Hancox, J. C. Vogel, J. Exp. Med. 181: 381–386. S. M. Steinberg, M. L. Turner, and A. Blauvelt. 2002. Clinical and immunologic 38. Simpson, S. J., S. Shah, M. Comiskey, Y. P. de Jong, B. Wang, E. Mizoguchi, assessment of patients with psoriasis in a randomized, double-blind, placebo- A. K. Bhan, and C. Terhorst. 1998. T cell-mediated pathology in two models of controlled trial using recombinant human . Arch. Dermatol. 138: experimental colitis depends predominantly on the interleukin 12/signal trans- 1341–1346. ducer and activator of transcription (Stat)-4 pathway, but is not conditional on 63. Olaniran, A. K., B. S. Baker, D. G. Paige, J. J. Garioch, A. V. Powles, and L. Fry. interferon-␥ expression by T cells. J. Exp. Med. 187: 1225–1234.

by guest on September 30, 2021. Copyright 2006 Pageant Media Ltd. 1996. Cytokine expression in psoriatic skin lesions during PUVA therapy. Arch. 39. Davidson, N. J., S. A. Hudak, R. E. Lesley, S. Menon, M. W. Leach, and Dermatol. Res. 288: 421–425. D. M. Rennick. 1998. IL-12, but not IFN-␥, plays a major role in sustaining the 64. Kang, K., C. Hammerberg, L. Meunier, and K. D. Cooper. 1994. CD11bϩ mac- chronic phase of colitis in IL-10-deficient mice. J. Immunol. 161: 3143–3149. rophages that infiltrate human epidermis after in vivo ultraviolet exposure po- 40. Neurath, M. F., I. Fuss, B. L. Kelsall, E. Stuber, and W. Strober. 1995. Antibodies tently produce IL-10 and represent the major secretory source of epidermal IL-10 to interleukin 12 abrogate established experimental colitis in mice. J. Exp. Med. protein. J. Immunol. 153: 5256–5264. 182: 1281–1290. 41. Malfait, A. M., D. M. Butler, D. H. Presky, R. N. Maini, F. M. Brennan, and 65. D’Andrea, A., M. Aste-Amezaga, N. M. Valiante, X. Ma, M. Kubin, and G. Trinchieri. 1993. Interleukin 10 (IL-10) inhibits human lymphocyte interferon M. Feldmann. 1998. Blockade of IL-12 during the induction of collagen-induced ␥ arthritis (CIA) markedly attenuates the severity of the arthritis. Clin. Exp. Immu- -production by suppressing natural killer cell stimulatory factor/IL-12 synthesis nol. 111: 377–383. in accessory cells. J. Exp. Med. 178: 1041–1048. 42. Matthys, P., K. Vermeire, T. Mitera, H. Heremans, S. Huang, and A. Billiau. 66. Aste-Amezaga, M., X. Ma, A. Sartori, and G. Trinchieri. 1998. Molecular mech- 1998. Anti-IL-12 antibody prevents the development and progression of collag- anisms of the induction of IL-12 and its inhibition by IL-10. J. Immunol. 160: http://classic.jimmunol.org en-induced arthritis in IFN-␥ receptor-deficient mice. Eur. J. Immunol. 28: 5936–5944. 2143–2151. 67. Nakanishi, K., T. Yoshimoto, H. Tsutsui, and H. Okamura. 2001. Interleukin-18 43. Yokoi, H., K. Kato, T. Kezuka, J. Sakai, M. Usui, H. Yagita, and K. Okumura. regulates both Th1 and Th2 responses. Annu. Rev. Immunol. 19: 423–474. 1997. Prevention of experimental autoimmune uveoretinitis by monoclonal an- 68. Ohta, Y., Y. Hamada, and K. Katsuoka. 2001. Expression of IL-18 in psoriasis. tibody to interleukin-12. Eur. J. Immunol. 27: 641–646. Arch. Dermatol. Res. 293: 334–342. 44. Hong, K., A. Chu, B. R. Ludviksson, E. L. Berg, and R. O. Ehrhardt. 1999. IL-12, 69. Li, J., G. W. Ireland, P. M. Farthing, and M. H. Thornhill. 1996. Epidermal and independently of IFN-␥, plays a crucial role in the pathogenesis of a murine oral keratinocytes are induced to produce RANTES and IL-8 by cytokine stim- psoriasis-like skin disorder. J. Immunol. 162: 7480–7491. ulation. J. Invest. Dermatol. 106: 661–666. 45. Kauffman, C. L., N. Arai, E. Toichi, T. S. McCormick, K. D. Cooper, 70. Boorsma, D. M., J. Flier, S. Sampat, C. Ottevanger, P. de Haan, L. Hooft, Downloaded from A. B. Gottlieb, D. E. Everitt, B. Frederick, Y. Zhu, M. A. Graham, et al. 2004. R. Willemze, C. P. Tensen, and T. J. Stoof. 1998. Chemokine IP-10 expression A phase I study evaluating the safety, pharmacokinetics, and clinical response of in cultured human keratinocytes. Arch. Dermatol. Res. 290: 335–341. a human IL-12p40 antibody in subjects with plaque psoriasis. J. Invest. Dermatol. 71. Krueger, G., and K. Callis. 2004. Potential of tumor necrosis factor inhibitors in 123: 1037–1044. psoriasis and . Arch. Dermatol. 140: 218–225. 46. Fredriksson, T., and U. Pettersson. 1978. Severe psoriasis—oral therapy with a 72. Kaluza, W., E. Reuss, S. Grossmann, R. Hug, R. E. Schopf, P. R. Galle, new retinoid. Dermatologica 157: 238–244. E. Maerker-Hermann, and T. Hoehler. 2000. Different transcriptional activity and 47. Trepicchio, W. L., M. Ozawa, I. B. Walters, T. Kikuchi, P. Gilleaudeau, in vitro TNF-␣ production in psoriasis patients carrying the TNF-␣ 238A pro- J. L. Bliss, U. Schwertschlag, A. J. Dorner, and J. G. Krueger. 1999. Interleu- moter polymorphism. J. Invest. Dermatol. 114: 1180–1183. kin-11 therapy selectively downregulates type I cytokine proinflammatory path- 73. Johansen, C., A. T. Funding, K. Otkjaer, K. Kragballe, U. B. Jensen, M. Madsen, ways in psoriasis lesions. J. Clin. Invest. 104: 1527–1537. L. Binderup, T. Skak-Nielsen, M. S. Fjording, and L. Iversen. 2006. Protein 48. Ghoreschi, K., P. Thomas, S. Breit, M. Dugas, R. Mailhammer, W. van Eden, expression of TNF-␣ in psoriatic skin is regulated at a posttranscriptional level by R. van der Zee, T. Biedermann, J. Prinz, M. Mack, et al. 2003. Interleukin-4 MAPK-activated protein kinase 2. J. Immunol. 176: 1431–1438. therapy of psoriasis induces Th2 responses and improves human autoimmune 74. Pradines-Figueres, A., and C. R. Raetz. 1992. Processing and secretion of tumor disease. Nat. Med. 9: 40–46. necrosis factor ␣ in endotoxin-treated Mono Mac 6 cells are dependent on phor- 49. Chamian, F., M. A. Lowes, S. L. Lin, E. Lee, T. Kikuchi, P. Gilleaudeau, bol myristate acetate. J. Biol. Chem. 267: 23261–23268. M. Sullivan-Whalen, I. Cardinale, A. Khatcherian, I. Novitskaya, et al. 2005. 75. Mannon, P. J., I. J. Fuss, L. Mayer, C. O. Elson, W. J. Sandborn, D. Present, reduces infiltrating T cells, activated dendritic cells, and inflammatory B. Dolin, N. Goodman, C. Groden, R. L. Hornung, et al. 2004. Anti-interleu- genes in psoriasis vulgaris. Proc. Natl. Acad. Sci. USA 102: 2075–2080. kin-12 antibody for active Crohn’s disease. N. Engl. J. Med. 351: 2069–2079.