MEDICALSCIENCES
`
`The Bruton tyrosine kinase inhibitor PCI-32765 blocks
`B-cell activation and is efficacious in models of
`autoimmune disease and B-cell malignancy
`
`Lee A. Honigberga,1, Ashley M. Smitha,1, Mint Sirisawada, Erik Vernera, David Lourya, Betty Changa, Shyr Lib,c,
`Zhengying Panb,d, Douglas H. Thamme, Richard A. Millera,f, and Joseph J. Buggya,2
`
`aPharmacyclics, Sunnyvale, CA 94085-4521; bCelera Genomics, South San Francisco, CA 94080; cExelixis, South San Francisco, CA 94080; dPeking University
`Shenzhen Graduate School, Shenzhen City 518055, China; eColorado State University Animal Cancer Center, Fort Collins, CO 80523; and fStanford University
`Medical Center, Stanford, CA 94305
`
`Edited* by Ronald Levy, Stanford University, Stanford, CA, and approved June 16, 2010 (received for review April 6, 2010)
`
`Activation of the B-cell antigen receptor (BCR) signaling pathway
`contributes to the initiation and maintenance of B-cell malignancies
`and autoimmune diseases. The Bruton tyrosine kinase (Btk) is spe-
`cifically required for BCR signaling as demonstrated by human and
`mouse mutations that disrupt Btk function and prevent B-cell
`maturation at steps that require a functional BCR pathway. Herein
`we describe a selective and irreversible Btk inhibitor, PCI-32765, that
`is currently under clinical development in patients with B-cell non-
`Hodgkin lymphoma. We have used this inhibitor to investigate the
`biologic effects of Btk inhibition on mature B-cell function and the
`progression of B cell-associated diseases in vivo. PCI-32765 blocked
`BCR signaling in human peripheral B cells at concentrations that did
`not affect T cell receptor signaling. In mice with collagen-induced
`arthritis, orally administered PCI-32765 reduced the level of circulat-
`ing autoantibodies and completely suppressed disease. PCI-32765
`also inhibited autoantibody production and the development of
`kidney disease in the MRL-Fas(lpr) lupus model. Occupancy of the Btk
`active site by PCI-32765 was monitored in vitro and in vivo using
`a fluorescent affinity probe for Btk. Active site occupancy of Btk was
`tightly correlated with the blockade of BCR signaling and in vivo
`efficacy. Finally, PCI-32765 induced objective clinical responses in dogs
`with spontaneous B-cell non-Hodgkin lymphoma. These findings
`support Btk inhibition as a therapeutic approach for the treatment
`of human diseases associated with activation of the BCR pathway.
`lymphoma | X-linked agammaglobulinemia
`Bruton tyrosine kinase (Btk) is a Tec family kinase with a well-
`
`defined role in B-cell antigen receptor (BCR) signaling (1, 2).
`Btk is activated by the upstream Src-family kinases Blk, Lyn, and
`Fyn (3, 4), and Btk in turn phosphorylates and activates phos-
`pholipase-Cγ (PLCγ) (5), leading to Ca2+ mobilization and acti-
`vation of NF-κB and MAP kinase pathways. Btk mutations in
`humans cause the inherited disease X-linked agammaglobulin-
`emia, characterized by a lack of peripheral B cells and low levels of
`serum Ig (6). In the mouse, point mutation or deletion of btk causes
`X-linked immunodeficiency (Xid), with approximately 50% fewer
`conventional B2 B cells, absent B1 B cells, and reduced serum Ig
`levels (7, 8). In transgenic mice in which Btk is expressed at ap-
`proximately 25% of WT levels, development of conventional (i.e.,
`B2) B cells is fully restored, but mature B cells are still deficient in
`responding to BCR stimulation. Thus, mature B cells may be
`particularly dependent on Btk for activation (9). Although Btk is
`also expressed in the myeloid lineage and there is some evidence
`that it contributes to other signaling pathways, the primary deficit
`in X-linked agammaglobulinemia is B cell-specific (8). Genetic
`ablation studies in the mouse of many other BCR-pathway kinases
`other than Btk have highlighted complex redundancies as well as
`pleiotropic effects on cell types other than B cells (10, 11); thus, Btk
`is a uniquely attractive kinase target for selective B-cell inhibition.
`Clinical studies using the anti-CD20 antibody rituximab to
`deplete mature B cells have provided evidence for the role of B
`
`cells in the pathogenesis of rheumatoid arthritis (12), systemic
`lupus erythematosus (13), and multiple sclerosis (14). In addi-
`tion, several lines of evidence suggest that the BCR pathway may
`provide a survival signal in tumor cells in non-Hodgkin lym-
`phoma (NHL) (15, 16). In an unbiased screen, Btk was recently
`identified as an essential signaling kinase for survival of a subtype
`of diffuse large B-cell lymphoma (16). Thus, small molecule Btk
`inhibitors may provide therapeutic benefit in the treatment of
`lymphoma and autoimmune diseases.
`Here we describe a potent irreversibly acting small molecule
`inhibitor of Btk, PCI-32765, that has demonstrated promising
`clinical activity in an ongoing phase I study in patients with B-cell
`NHL. We show that PCI-32765 inhibits BCR signaling down-
`stream of Btk, selectively blocks B-cell activation, and is effica-
`cious in animal models of arthritis, lupus, and B-cell lymphoma.
`
`Results
`PCI-32765 Is a Potent and Selective Inhibitor of Btk. We have pre-
`viously described the synthesis of a series of Btk inhibitors that
`bind covalently to a cysteine residue (Cys-481) in the active site
`leading to potent and irreversible inhibition of Btk enzymatic ac-
`tivity (17). One of these compounds, PCI-32765 (Fig. 1), was se-
`lected for the present studies because of its potency (IC50, 0.5 nM)
`and selectivity for Btk against a screening panel of kinase enzymes
`(Table S1). As only a small subset of kinases is predicted to contain
`a modifiable cysteine residue homologous to Cys-481 in Btk (17),
`we decided to explore the selectivity of PCI-32765 as an irrevers-
`ible kinase inhibitor in cells. To this end, we created a unique cell
`permeable, fluorescently tagged derivative, PCI-33380 by attach-
`ing a Bodipy-FL fluorophore to PCI-32765 via a piperazine linker
`(Fig. 1). In cells transfected with Btk, PCI-33380 bound to Btk
`could be detected by denaturing gel electrophoresis and fluores-
`cent gel scanning. As expected, Btk lacking Cys-481 (C481A) was
`not bound by PCI-33380. In addition PCI-33380 bound to a cata-
`lytically inactive mutant of Btk (K430A), suggesting that binding
`does not require catalytic activity (Fig. 2A). Identification of pro-
`teins capable of binding PCI-33380 was then evaluated in the
`B-cell lymphoma cell line DOHH2, a cell line that endogenously
`expresses the kinases that constitute the BCR signaling pathway,
`along with many other potentially reactive kinase and nonkinase
`proteins. Irreversibly bound proteins in the lysate were resolved by
`denaturing gel electrophoresis. Remarkably, PCI-33380 labeled
`
`Author contributions: L.A.H., D.H.T., R.A.M., and J.J.B. designed research; A.M.S., M.S.,
`E.V., S.L., Z.P., and D.H.T. performed research; L.A.H., D.L., B.C., D.H.T., R.A.M., and J.J.B.
`analyzed data; and L.A.H. and J.J.B. wrote the paper.
`
`The authors declare no conflict of interest.
`
`*This Direct Submission article had a prearranged editor.
`
`Freely available online through the PNAS open access option.
`1Present address: Genentech, South San Francisco, CA 94080.
`2To whom correspondence should be addressed. E-mail: jbuggy@pcyc.com.
`
`This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.
`1073/pnas.1004594107/-/DCSupplemental.
`
`www.pnas.org/cgi/doi/10.1073/pnas.1004594107
`
`PNAS |
`
`July 20, 2010 | vol. 107 | no. 29 | 13075–13080
`
`SAN EX 1015, Page 1
`
`

`

`F
`NB
`
`F
`
`N
`
`O
`
`NH2
`
`N
`
`N
`
`N
`
`N
`
`O
`
`NH
`
`N
`
`N
`
`Linker
`
`Bodipy FL
`
`PCI-29732
`
`Fig. 1. Chemical structure of Btk inhibitors. Chemical struc-
`tures of the irreversible Btk inhibitor PCI-32765, the irre-
`versible Btk inhibitor PCI-33380 (probe), and the reversible
`Btk inhibitor PCI-29732.
`
`O
`
`O
`
`NH2
`
`N
`
`N
`
`N
`
`N
`
`NH2
`
`N
`
`N
`
`N
`
`N
`
`N
`
`O
`
`PCI-32765
`
`N
`
`O
`Core
`PCI-33380
`
`and that only B-cell inhibition is sustained following short dura-
`tion treatment. It is noteworthy that the mean terminal plasma
`half-life of PCI-32765 following oral dosing in mice is 1.7 to 3.1 h
`(Fig. S1). By combining fast irreversible binding to Btk with rapid
`in vivo elimination, PCI-32765 defines a unique approach to im-
`prove selectivity for Btk in vivo relative to reversibly inhibited off-
`target kinases.
`
`PCI-33380 (µM)
`
`1 2 4 8
`
`0.5
`0.25
`0.125
`0.062
`0.031
`.016
`
`00
`
`B
`
`NT
`
`Y551A
`
`C481A
`
`K430A
`
`WT
`
`A
`
`6.46
`7.60
`
`7.64
`
`7.63
`
`6.46
`
`4.20
`
`1.97
`
`1.06
`
`0.49
`.17
`
`0 0
`
`Band intensity (x 105)
`
`DOHH2
`Jurkat
`+ - + + - + :PCI-32765
`- + + -
`+ + :Probe
`
`Btk
`
`D
`
`Sup (Btk)
`
`Sup (mock)
`
`Btk IP
`
`Mock
`
`Lysate
`
`Btk
`
`PCI-32765 (nM)
`
`200
`
`100
`
`50
`
`25
`
`12.5
`
`6.25
`
`3.12
`
`1.56
`
`0.78
`
`.39
`
`0 0
`
`Btk
`
`C
`
`E
`
`Btk
`
`Fig. 2. Selective irreversible targeting of Btk. (A–E) (Upper) Fluorescent gel
`scans of lysates from cells that were incubated with the affinity probe PCI-
`33380. Arrows indicate the predominant band labeled by the probe (approx-
`imately 76 kDa, the expected MW of Btk) and this band was confirmed to align
`with Btk as detected by Western blot of the same gel (Lower). (A) 293H
`cells transfected with WT Btk, K430A, C481A, or Y551A Btk mutants. Non-
`transfected (NT) 293H cells were included as a negative control. (B) DOHH2
`cells incubated with increasing concentrations of PCI-33380 for 1 h. Densi-
`tometry values for the band corresponding to Btk are shown below the gel. (C)
`Btk immunoprecipitated (IP) from affinity probe labeled DOHH2 lysates. Total
`lysate (10 μg), and mock IP or Btk IP (from 50 μg of total lysate) were analyzed.
`A fluorescent gel scan of the immunodepleted supernatants (Sup) is also
`shown. (D) Affinity probe labeling of DOHH2 or Jurkat cells with and without
`1 μM PCI-32765 pretreatment. (E) DOHH2 cells incubated with increasing
`concentrations of PCI-32765 for 1 h before labeling with affinity probe. By
`densitometry, the IC50 for active site occupancy by PCI-32765 is 4 nM.
`
`a single predominant band; Western blotting of the gel indicated
`that the band labeled by PCI-33380 was likely to be Btk (Fig. 2B).
`To confirm that Btk was indeed the predominant band labeled by
`PCI-33380 in cells, Btk was immunoprecipitated from PCI-33380–
`labeled DOHH2 lysates. Btk immunoprecipitation captured PCI-
`33380–labeled protein and depleted the corresponding band from
`the lysate (Fig. 2C). Next, PCI-33380 labeling in DOHH2 cells was
`compared with PCI-33380 labeling in Jurkat, a T cell line that
`expresses several Tec family members but does not express Btk. As
`expected, the predominant labeled band (i.e., Btk) was found to be
`present in DOHH2, but not in Jurkat (Fig. 2D). Finally, pre-
`treatment of cells with PCI-32765 completely prevented labeling
`of this single predominant band (Fig. 2 D and E). Taken together,
`these data indicate that both PCI-33380 and PCI-32765 are re-
`markably selective for irreversible binding to Btk in the context of
`a complex proteome.
`
`PCI-32765 Selectively Inhibits B-cell Signaling and Activation. PCI-
`32765 was examined in a series of cellular signaling assays. First, in
`DOHH2, a cell line in which the BCR pathway can be activated by
`stimulation with anti-IgG, we confirmed that PCI-32765 inhibits
`autophosphorylation of Btk (IC50, 11 nM), phosphorylation of
`Btk’s physiological substrate PLCγ (IC50, 29 nM), and phos-
`phorylation of a further downstream kinase, ERK (IC50, 13 nM;
`Fig. 3A). In this experiment, PCI-32765 was washed out before
`stimulation, so the results are consistent with irreversible in-
`hibition of these phosphorylation events. As expected, phos-
`phorylation of Syk, which functions upstream or in parallel to Btk,
`was not affected. To model therapeutic applications, we evaluated
`PCI-32765 in primary cultures of human peripheral B cells. To
`distinguish reversible from irreversible effects of PCI-32765 we
`directly compared continuous versus pulse exposure. We also
`included a structurally similar but reversibly acting Btk inhibitor
`PCI-29732 (17, 18) (IC50, 0.3 nM; Fig. 1) as a control. In human
`CD20+ B cells stimulated at the BCR, both PCI-32765 and PCI-
`29732 blocked the transcriptional up-regulation of a panel of
`B-cell activation genes that occurs within 6 h of stimulation (Fig.
`3B). A 1-h pulse exposure of PCI-32765 followed by washout was
`sufficient to prevent up-regulation of these genes, indicating that
`PCI-32765 acts as an irreversible inhibitor of the BCR pathway. In
`contrast, pulse exposure to the reversible inhibitor PCI-29732 did
`not result in BCR inhibition. As further confirmation that PCI-
`32765 inhibits BCR signaling, we found that continuous exposure
`to 10 nM PCI-32765 for 18 h completely prevented up-regulation
`of the B-cell activation marker CD69 (Fig. 3C). A 1-h pulse ex-
`posure to 10 nM PCI-32765 resulted in a similar level of CD69
`inhibition in B cells, consistent with the effect being primarily
`caused by an irreversible inhibition of Btk. By using the fluo-
`rescently tagged derivative PCI-33380, we found that 10 nM of
`PCI-32765 was sufficient to fully occupy the active site of Btk in
`primary B cells in culture. Thus, the concentration of PCI-32765
`required to covalently bind Btk was well correlated with the con-
`centration required to inhibit B-cell activation (Fig. 3D). In pri-
`mary cultures of T cells (which do not express Btk), up-regulation
`of CD69 induced by T cell receptor stimulus was blocked only by
`continuous exposure to PCI-32765 at 10 μM (Fig. 3C). These
`results indicate that PCI-32765 is more than 1,000-fold selective
`for inhibition of antigen receptor signaling in B cells over T cells,
`
`13076 | www.pnas.org/cgi/doi/10.1073/pnas.1004594107
`
`Honigberg et al.
`
`SAN EX 1015, Page 2
`
`

`

`A
`
`C
`
`B
`
`D
`
`MEDICALSCIENCES
`
`Inhibition of B-cell receptor signaling. (A) Concentration-dependent inhibition of BCR stimulation-induced phosphorylation events in DOHH2 cells.
`Fig. 3.
`Cells were exposed to PCI-32765 and then drug was washed out before stimulation with anti-IgG. Blots were probed with the indicated antibodies. (B)
`Comparison of the reversible Btk inhibitor PCI-29732 to the irreversible Btk inhibitor PCI-32765 on the transcriptional response of six genes induced by BCR
`stimulation. Purified human peripheral B cells were treated with vehicle or 1 μM inhibitor for 1 h and then stimulated for 6 h with anti-IgM. In the washout
`condition, inhibitor-containing cell media was replaced with fresh media before addition of anti-IgM. Gene expression levels (mean ± SD) were measured by
`TaqMan RT-PCR and normalized to unstimulated cells. Btk expression is not affected by BCR stimulation or drug treatment. (C) Concentration-dependent
`inhibition of antigen receptor stimulation induced cell surface expression of the lymphocyte activation marker CD69 (mean ± SD). Purified human B or T cells
`were treated with PCI-32765 for 1 h and then stimulated for 18 h. In the washout condition, inhibitor-containing cell media was replaced with fresh media
`before stimulation. (D) Concentration-dependent covalent binding of PCI-32765 to Btk in purified human B cells as measured by the ability of Btk to bind to
`the fluorescent affinity probe PCI-33380. Total Btk levels are measured by Western blot (Bottom).
`
`Btk Inhibition by PCI-32765 Is Efficacious in Mouse Models of
`Autoimmune Disease. To test the effects of PCI-32765 in vivo, we
`focused initially on two autoimmune models in which Btk or B-
`cell function has been previously implicated. xid mice have been
`shown to be resistant to the induction of collagen-induced ar-
`thritis and to partially suppress disease in the MRL-Fas(lpr) lupus
`model (19, 20). To evaluate the consequences of inhibiting Btk
`activity during the establishment of arthritis, arthritic DBA/1 mice
`were assigned to treatment groups when their disease had par-
`tially progressed as measured by a mean clinical arthritis score
`between 1.0 and 1.5. PCI-32765 was administered orally for 11
`consecutive d at dosages of 3.125, 12.5, or 50 mg/kg per day and
`clinical arthritis scores, reflecting paw swelling and joint in-
`flammation, were measured daily. As shown in Fig. 4A, marked
`inhibition of clinical arthritis scores was seen in mice treated at all
`
`doses. A partial and a nearly complete elimination of clinical signs
`of disease occurred after 9 to 11 d of treatment at dosages of 3.125
`and 12.5 mg/kg per day, respectively. Consistent with in vivo in-
`hibition of B-cell activation, there was a significant reduction in
`the production of anticollagen autoantibodies, and a modest re-
`duction of total IgG levels (Fig. 4 B and C). A oral single dose of
`PCI-32765 at 3.125 mg/kg per day resulted in partial Btk occu-
`pancy in splenocytes, and the maximally efficacious dose (12.5 mg/
`kg per day) was sufficient to fully occupy Btk for 12 h (Fig. 4D).
`Thus, the level of Btk inhibition in vivo by PCI-32765 is well
`correlated with the degree of efficacy in the collagen-induced
`arthritis model.
`We next tested PCI-32765 in the MRL-Fas(lpr) lupus model, in
`which a mutation in the Fas receptor leads to survival of autor-
`eactive cells, production of autoantibodies and progressive glo-
`
`Honigberg et al.
`
`PNAS |
`
`July 20, 2010 | vol. 107 | no. 29 | 13077
`
`SAN EX 1015, Page 3
`
`

`

`A
`
`A
`
`Vessels
`Interstitial Nephritis
`Glomerulonephritis
`
`∗∗
`
`50 mg/kg
`12.5 mg/kg
`3.125 mg/kg
`Vehicle
`
`012345678
`
`D
`
`Scores, (0-5 each)
`
`Mean histopathology
`
`50 mg/kg
`12.5 mg/kg
`3.125 mg/kg
`Vehicle
`
`C
`
`40
`
`30
`
`01
`Anti-dsDNA IgG(mg/ml)
`
`20
`
`0
`
`∗∗∗
`
`∗∗∗
`
`∗∗∗
`
`50 mg/kg
`12.5 mg/kg
`3.125 mg/kg
`Vehicle
`
`B
`
`70
`60
`50
`40
`30
`20
`10
`0
`
`Serum BUN (mg/dl)
`
`PCI-32765
`
`PCI-32765
`
`PCI-32765
`
`Inhibition of Btk reduces renal disease and autoantibody production
`Fig. 5.
`in MRL-Fas(lpr) mice. (A) Eight-week-old MRL-Fas(lpr) mice (n = 12) were
`randomized and treated orally with PCI-32765 or vehicle once daily for 12
`wk at different concentrations as indicated. Range of urine protein con-
`centration is calculated as a proteinurea score. Proteinurea scores from PCI-
`32765 treatments were significantly lower than vehicle group (**P < 0.01 for
`3.125 mg/kg and ***P < 0.001 for 12.5 and 50 mg/kg treatments; repeated-
`measure ANOVA). (B) Serum BUN in PCI-32765–treated and vehicle-treated
`mice (***P < 0.001). (C) dsDNA-specific IgG in PCI-32765 treated and vehicle
`treated mice. (D) Histopathology scores at week 20 from the animals shown
`in A. Scores are mean ± SEM (n = 12; *P < 0.05).
`
`drug-induced changes in peripheral blood Btk expression levels. To
`date, eight dogs have been treated. We have observed three partial
`responses per Response Evaluation Criteria In Solid Tumors
`(RECIST), including one dog in which measurable tumor burden
`was reduced 77%, and three instances of stable disease (Table 1).
`
`Discussion
`We have described a selective and irreversible Btk inhibitor and
`its efficacy in models of autoimmune disease and spontaneous
`B-cell lymphoma. The use of irreversible inhibitors has previously
`been shown to be a viable method to achieve potent and selective
`inhibition of kinase enzymes (22, 23). We previously reported the
`discovery and characterization of a series of Btk-selective irre-
`versible inhibitors that bind covalently to a noncatalytic Cys (Cys-
`481) residue in Btk (17). Structural alignments revealed that only
`10 kinases have a Cys at this position (17), representing a signifi-
`cant selectivity filter. Indeed, a fluorescently tagged derivative of
`PCI-32765, PCI-33380, was shown to bind predominantly to
`a single protein, Btk, in B-cell lysates. Cellular selectivity of PCI-
`32765 was also demonstrated by the observation that PCI-32765
`inhibits antigen receptor signaling in B cells but not in T cells.
`We used the fluorescently tagged affinity probe PCI-33380 as
`a tool to directly monitor the degree of Btk occupancy by PCI-
`32765 in cells or target tissues. Validation of this probe in dogs
`with spontaneous NHL and the agreement between Btk occu-
`pancy in PBMC and tumor biopsies supports the current use of
`PCI-33380 as a pharmacodynamic marker in PBMCs in ongoing
`human clinical trials. This approach to direct measurement of
`kinase occupancy in a clinical setting enables the determination of
`a minimum dosage level that leads to full inhibition of Btk in
`specific populations of patients with cancer.
`B-cell activation is a complex process involving Btk-dependent
`signal transduction initiated by stimulation of the BCR. BCR
`stimulation recruits Btk to the cell membrane via interactions
`
`50 mg/kg
`
`12.5 mg/kg
`3.125 mg/kg
`Dex
`
`Vehicle
`
`Naive
`
`3500
`3000
`2500
`2000
`1500
`1000
`500
`0
`
`C
`
`Total IgG(µg/ml)
`
`∗∗∗
`
`∗∗∗
`
`∗∗
`
`50 mg/kg
`
`12.5 mg/kg
`3.125 mg/kg
`Dex
`
`Vehicle
`
`Naive
`
`250
`
`200
`
`150
`
`100
`
`50
`
`0
`
`Anti-Collagen IgG(µg/ml)
`
`B
`
`Veh
`
`3.125
`
`12.5
`
`Veh
`
`3 12 24 3 12 24 3 12 24
`
`D
`
`Time (hr):
`Probe
`
`Btk
`
`Fig. 4. Btk inhibition by PCI-32765 inhibits collagen-induced arthritis in
`mice. (A) Mean clinical arthritis scores ± SEM (n = 5) from daily oral treat-
`ment for 11 d with different doses of PCI-32765 or dexamethasone as in-
`dicated. Disease control mice received vehicle only. Mean clinical scores from
`dexamethasone or PCI-32765 treatments at 3.125, 12.5, and 50 mg/kg were
`significantly different when compared with vehicle treatment (***P < 0.001,
`repeated-measures ANOVA). (B and C) Antibody levels from experiment
`shown in A. Data are mean ± SEM (**P < 0.01 and ***P < 0.001, ANOVA). (D)
`PCI-32765 occupancy of splenocyte Btk following a single oral dose in DBA/1
`mice. Binding of PCI-32765 to Btk prevents binding of PCI-33380, detected by
`fluorescent gel scanning of PCI-33380-labeled lysates (Upper). Subsequent
`Western blotting confirms the presence of Btk (Lower).
`
`merulonephritis. Eight-week-old MRL-Fas(lpr) mice were treated
`for 12 wk with daily oral doses of PCI-32765. Treatment with PCI-
`32765 reduced proteinuria, a measure of glomerular dysfunction,
`and reduced blood urea nitrogen (BUN), a general measure of
`renal impairment (Fig. 5 A and B). Consistent with inhibition of
`B-cell activation, serum anti-dsDNA levels were reduced at all
`dose levels compared with vehicle-treated controls (Fig. 5C).
`Histopathological evaluation of the kidneys revealed a significant
`reduction in interstitial nephritis,
`reduced perivascular
`in-
`flammation, and a nonsignificant trend toward reduction of glo-
`merulonephritis at dosages of 12.5 and 50 mg/kg per day (Fig. 5D).
`
`Btk inhibition by PCI-32765 Leads to Objective Clinical Responses in
`Spontaneous Canine B-cell Lymphomas. To determine if blocking
`BCR signaling by inhibiting Btk would affect the progression of
`lymphoma, we initiated a trial of PCI-32765 in naturally occurring
`B-cell NHL in companion dogs. Canine NHL shares many char-
`acteristics with human NHL, including diagnostic classifications
`and response to CHOP-based chemotherapy (cyclophosphamide,
`doxorubicin, vincristine, and prednisone/prednisolone) (21). In the
`study, both treatment-naive and relapsed dogs were enrolled and
`PCI-32765 was dosed orally once per day using the capsule for-
`mulation prepared for human clinical trials. Inhibition of Btk was
`monitored in vivo by labeling peripheral blood mononuclear cell
`(PBMC) and tumor lysates ex vivo with PCI-33380 and labeled Btk
`was visualized by fluorescent gel scanning. In five dogs in which
`tissue samples were analyzed, a single administration of PCI-32765
`at dosage levels ranging from 2.5 to 20 mg/kg per day was sufficient
`to fully occupy Btk in peripheral blood and tumor tissue for 24 h
`(Fig. 6). Total Btk levels varied significantly across samples, which
`may reflect heterogeneity in biopsy sampling as well as potential
`
`13078 | www.pnas.org/cgi/doi/10.1073/pnas.1004594107
`
`Honigberg et al.
`
`SAN EX 1015, Page 4
`
`

`

`MEDICALSCIENCES
`
`in myeloid lineage cells such as mast cells and macrophages may
`also have some contribution to efficacy in these disease models,
`although in both models efficacy was associated with decreases in
`circulating autoantibody levels, consistent with a mechanism of
`action involving inhibition of B-cell activation in vivo.
`Chronic activation of the BCR pathway has been demonstrated
`in some B-cell lymphomas, and this activation has been shown to
`be required for tumor cell survival (16, 25). Traditional human
`xenograft studies have limited relevance to the study of BCR
`dependent lymphoma growth in vivo as, in these systems, tumor
`growth is not thought to be driven by BCR signaling. To evaluate
`a general antitumor effect of PCI-32765 in a true B-cell lym-
`phoma, we studied dogs with spontaneously occurring disease. In
`these dogs, treatment with PCI-32765 at well-tolerated doses led
`to complete active site occupancy of Btk in both peripheral blood
`and on tumor biopsies. Disease stabilization and objective clinical
`responses were observed in dogs diagnosed with rapidly pro-
`gressive disease. Although it is not possible to assess the signifi-
`cance of disease stabilization in this study without a randomized
`controlled trial, the objective responses argue that PCI-32765 is
`an active drug in this setting, supporting a role for BTK in the
`maintenance of tumor growth.
`PCI-32765 is currently undergoing human clinical development
`in patients with B-cell malignancies and has shown promising
`clinical activity. Here we have shown that this compound is a po-
`tent, selective, and irreversible Btk inhibitor that blocks signaling
`downstream of the BCR in mature human B cells, inhibits disease
`progression in a collagen-induced arthritis model, and can treat
`the progressive lupus-like autoimmunity of the MRL-Fas(lpr)
`model. The degree of active site occupancy of Btk by PCI-32765
`was tightly correlated with the efficacy response, both in vitro and
`in vivo, indicating that the probe assay will have utility as a tool for
`assessing pharmacodynamics in human clinical trials. Potent and
`highly selective Btk inhibitors such as PCI-32765 represent
`a unique therapeutic approach for the treatment of diseases
`driven by inappropriate activation of the BCR pathway.
`
`Materials and Methods
`B and T Cells. CD20+ B and CD3+ T cells were purified by negative selection
`(RosetteSep, >90% purity) from buffy coat PBMCs and viably frozen in 10%
`DMSO. Cells were thawed at 37 °C and maintained in growth media (RPMI
`media containing 10% FCS). B cells were stimulated with goat antihuman IgM
`F(ab′)2 (10 μg/mL; Invitrogen) and T cells were stimulated with anti-CD3/CD28
`coated beads (Dynabeads) at a 1:1 bead/cell ratio. Cells were stained with PE-
`CD69 (BD Biosciences) and analyzed by flow cytometry, gating on viable
`lymphocytes. PCI-32765 at concentrations lower than 10 μM did not decrease
`B- or T-cell viability during the course of the experiment, although PCI-32765
`did block the modest survival benefit of anti-IgM stimulation in B cells. For
`washout experiments, cells were rinsed three times in 10 volumes of growth
`media, a protocol that was confirmed to completely wash away inhibition of
`BCR signaling by PCI-29732, a reversible Btk inhibitor.
`
`Real-Time RT-PCR. B cells were stimulated as described earlier and mRNA
`prepared by RNeasy 96 (Qiagen). Real-time RT-PCR was performed using
`standard cycling conditions (ABI 7300) and predesigned assay reagents for the
`
`Fig. 6. Orally-dosed PCI-32765 leads to sustained occupancy of Btk in dogs
`with lymphoma. PBMCs and biopsy specimens from affected lymph nodes
`(LN) were collected from dogs (Table 1) treated with PCI-32765 (oral capsule
`formulation). Tissue samples were then treated with PCI-33380 to determine
`Btk occupancy by PCI-32765. Shown are predose (P), 4 h, 24 h, and predose
`d 7 occupancy data for all available week 1 samples in the study. Arrow
`marked “P” indicates fluorescent probe (PCI-33380) signal, arrow marked
`“Btk” indicates Btk protein level by Western blot. Full occupancy was ach-
`ieved in all dogs. Western blots indicate that some LN biopsies did not
`contain Btk, presumably as a result of sample collection heterogeneity.
`
`between the N-terminal PH domain and cell membrane phos-
`phoinositides and membrane-associated Btk is then phosphory-
`lated at Tyr-551 in the activation loop by Src family kinases (3).
`Subsequent Btk autophosphorylation at Tyr-223 stabilizes the
`active conformation and fully activates Btk kinase activity (18).
`Activated Btk phosphorylates PLCγ (5), initiating calcium mo-
`bilization and generating diacylglycerol as secondary signals,
`eventually leading to NF-κB transcriptional activation and am-
`plification of BCR stimulation. PCI-32765 was shown to block
`signaling downstream of Btk, and completely and irreversibly
`inhibit B-cell activation as assayed by either gene expression or
`the cell surface marker CD69.
`As an orally bioavailable, selective, and irreversible inhibitor,
`PCI-32765 is an appropriate tool to examine the effects of Btk
`inhibition in animal models of disease. The xid mutation in mice
`has been shown to suppress or partially suppresses disease in
`models of autoimmune disease (19, 20, 24), but these genetic
`experiments cannot distinguish Btk’s developmental role from its
`role in mature B lymphocyte function in a disease caused by sus-
`tained generation of autoimmunity. PCI-32765 treatment led to
`a dose-dependent decrease in disease manifestation in both the
`collagen-induced arthritis and MRL-Fas(lpr) lupus models. The
`level of Btk inhibition in vivo by PCI-32765 was shown to correlate
`with the degree of efficacy in the collagen-induced arthritis model,
`consistent with Btk as the relevant target in vivo; however, a role
`for other cysteine-containing kinases that may also be inhibited by
`PCI-32765 (e.g., Blk, Bmx) cannot be ruled out. Inhibition of Btk
`
`Table 1. Study summary of the effect of Btk inhibitor PCI-32765 in naturally occurring canine lymphomas
`
`Dog
`
`Stage
`
`Histology
`
`A1
`A2
`A3
`A4
`B1
`B2
`B3
`B4
`
`IIIa
`Va
`IIIa
`IIIa
`IIIa
`IIIa
`IIIa
`Va
`
`NA
`NA
`Follicular large cell
`Diffuse immunoblastic
`Diffuse immunoblastic
`Follicular large cell
`Diffuse immunoblastic
`Follicular large cell
`
`PARR
`(monoclonal
`BCR)
`
`Previous
`Treatment
`
`—
`+
`—
`+
`+
`+
`+
`+
`
`—
`COP
`CHOP
`COP
`—
`
`—
`
`—
`
`—
`
`Dose,
`mg/kg
`
`20
`20
`20
`20
`2.5/5.0/7.5
`2.5/5.0
`2.5
`2.5/5.0
`
`Outcome
`(RECIST)
`
`Progression
`free
`interval, d
`
`Decrease
`in tumor
`sums, %
`
`SD
`PD
`SD
`PR
`SD
`PR
`PD
`PR
`
`28
`0
`14
`35
`21
`70
`0
`63
`
`10
`—
`
`—
`77
`—
`31
`—
`62
`
`CR, complete response; NA, not applicable; PARR, PCR of antigen receptor rearrangement; PD, progressive disease; PR, partial response; SD, stable disease.
`
`Honigberg et al.
`
`PNAS |
`
`July 20, 2010 | vol. 107 | no. 29 | 13079
`
`SAN EX 1015, Page 5
`
`

`

`indicated genes (Applied Biosystems). Expression levels were normalized (in
`arbitrary units) to total RNA amount as determined by Ribogreen (Invitrogen).
`
`Phospho-Blots. DOHH2 cells were preincubated with compound for 1 h,
`washed three times in 10 volumes of PBS solution, and then stimulated with
`anti-IgG F(ab′)2 (30 μg/mL; Invitrogen) for 2 min. Cells were lysed in LDS
`sample buffer containing sample reducing agent (Invitrogen) and analyzed
`using phospho-specific antibodies Syk pY352 (cat. no. 2701; Cell Signaling),
`Btk pY223 (EP420Y; Epitomics), PLCγ1 pY783 (cat. no. 2821; Cell Signaling),
`and ERK1/2 pT202/pY204 (197G2; Cell Signaling); chemiluminescent de-
`tection (Pierce Dura) was then performed. Blots were then stripped and
`reprobed with antibodies Syk (cat. no. 2712; Cell Signaling), Btk (53/Btk; BD
`Biosciences), PLCγ1 (cat. no. 2822; Cell Signaling), and ERK1/2 (3A7; Cell
`Signaling) to detect total protein levels.
`
`PCI-33380 Labeling. For PCI-33380 labeling of human B cells, 16 cells were
`treated with PCI-33380 at 2 μM for 1 h, washed, lysed in LDS sample buffer
`containing sample reducing agent (Invitrogen), and analyzed by SDS/PAGE and
`fluorescent gel scanning using a Typhoon scanner (Ex, 532 nm; Em, 555 nm). The
`gel was then blotted and total Btk levels detected by standard Western blot. Btk
`occupancy in cellular assays was assessed by preincubation with PCI-32765 for
`1 h before labeling with PCI-33380. For analysis of Btk occupancy in vivo fol-
`lowing oral dosing of PCI-32765, spleens were processed to splenocytes fol-
`lowed by 5 min incubation in red blood cell lysing buffer (Sigma-Aldrich). Cells
`were then PCI-33380–labeled and lysates analyzed by fluorescent gel scanning
`as described earlier. For immunoprecipitation of Btk, DOHH2 cells were lysed
`and labeled with PCI-33380 (2 μM) before immunoprecipitation. Lysates were
`then incubated at room temperature for 1 h with Btk antibody (53/Btk; BD
`Biosciences) or mouse IgG2a isotype control antibody. Immune complexes were
`then incubated for 1 h at 4 °C with immobilized Protein A (Thermo Fisher Sci-
`entific). Immune complexes were washed with immunoprecipitation buffer
`(20 mM Tris-HCl, pH 7.4, 150 mM NaCl) four times and eluted with LDS sample
`buffer and sample reducing agent (Invitrogen) and analyzed by Western blot as
`described earlier.
`
`Arthritis and Lupus Models. Male DBA/1 mice were immunized with type II
`collagen plus Freund adjuvant and bo