Functional Characterization of BTK C481S Mutation That Confers

ORIGINAL ARTICLE Functional characterization of BTKC481S mutation that confers ibrutinib resistance: exploration of alternative kinase inhibitors

S Cheng1, A Guo2,PLu2,JMa1, M Coleman3 and YL Wang2

The Bruton tyrosine kinase (BTK) inhibitor, ibrutinib, has produced remarkable clinical response in chronic lymphocytic leukemia (CLL) and mantle cell lymphoma. We previously reported the identiﬁcation of BTKC481S mutation in a CLL patient who progressed following 21-month ibrutinib therapy. Initial characterization at structural and biochemical levels revealed that the mutation disrupts the covalent binding of ibrutinib to BTK, reduces its binding afﬁnity and diminishes its ability to inhibit the BTK enzymatic activity. Herein, we further characterized the functional consequences of BTKC481S in terms of molecular signaling, gene expression and cellular behavior in the patient, as well as in lymphoma cells transfected with either the wild-type or the mutant BTK constructs. Further, using an in vitro CLL proliferation model, alternative kinase inhibitors that have the potential to overcome ibrutinib resistance were explored.

Leukemia (2015) 29, 895–900; doi:10.1038/leu.2014.263

INTRODUCTION MATERIALS AND METHODS Bruton tyrosine kinase (BTK) is an essential component of the early Patient, sample collection and cell isolation BCR signaling pathway1 (Supplementary Figure 1). BCR activation The patient was a 49-year-old woman when she was diagnosed with CLL in triggers the formation of a signalosome, which includes protein 2000. In October 2010, after failing multiple therapeutic regimens including γ fludarabine/cyclophosphamide/rutuximab, she was enrolled in a phase I tyrosine kinases LYN, SYK and BTK, as well as phospholipase C 2 5 and phosphoinositide 3-kinase. Within the signalosome, BTK is dose-escalation study of ibrutinib in recurrent B-cell malignancies. She was treated once daily with 560 mg ibrutinib for 21 months and achieved a phosphorylated by LYN and SYK. SYK and BTK subsequently partial response, and then developed disease progression (see Furman et al.9 phosphorylate phospholipase Cγ2, leading to calcium mobiliza- for a detailed case description). A total of four peripheral blood samples tion, activation of ERK, AKT and nuclear factor-κB, which were collected longitudinally before and after the initiation of ibrutinib: subsequently result in cell survival, cell proliferation, differentia- (1) pre-treatment (Pre-Rx); (2) day 1 of month 4, while patient was tion and antibody production.2–4 responding to ibrutinib (Responding); (3) after 20 months on treatment fi Ibrutinib (PCI-32765), an orally administered inhibitor that binds when progressive disease was rst demonstrated (Relapse s1); and (4) 4 weeks later, before dose escalation of ibrutinib to 840 mg (Relapse s2). covalently to cysteine 481 of BTK, has produced remarkable CD5+/CD19+ CLL cells were isolated from these samples as described clinical response in relapsed and refractory chronic lymphocytic previously.10,11 Briefly, cells were isolated from whole blood by negative 5,6 leukemia (CLL) and mantle cell lymphoma. The drug has been selection using RosetteSep Human B Cell Enrichment cocktail (StemCell recently approved by the US Food and Drug Administration for Technologies, Vancouver, BC, Canada) following the manufacturer's the treatment of these diseases. However, it has been reported instructions. The purity of the enriched CLL B cells was 490% by CD5/ fi that in phase I and II studies of CLL, of 246 patients who received CD19 double staining of all four samples. Aliquots of the puri ed cells were immediately cryopreserved as cell suspension in RPMI1640, 20% fetal ibrutinib, 13 patients (5.3%) have progressed following a median 7 bovine serum plus 10% dimethylsulfoxide. Other aliquots of the cells were observation of 14 months. stored at − 80 °C as cell pellets for immunoblot assays or fixed and stored C481S We and others recently reported the identification of BTK for intracellular Ki-67 staining (see below). mutation in several of these patients.8,9 In our patient, the mutation was identified using RNA-Seq and verified using both fi Immunoblotting Sanger sequencing and allele-speci c PCR. We further demon- 12–14 strated that the BTKC481S mutation disrupts the irreversible Immunoblotting was carried out as described previously. The following antibodies were purchased from Cell Signaling Technology covalent binding between the ibrutinib and BTK, and reduces (Beverly, MA, USA): anti-p-BTK (Y223), anti-pERK1/2 (T202/Y204), anti-total the binding affinity of the drug for the enzyme. The result is loss of ERK, anti-p-AKT (S473) and anti-total AKT. Anti-total BTK was obtained from inhibition of BTK enzymatic activity that ultimately causes ibrutinib BD Bioscience (San Jose, CA, USA) and anti-GAPDH from Thermo Scientific resistance in the patient.9 Herein, we further characterize the (Waltham, MA, USA). functional impact of the mutation from the perspectives of cell signaling, gene expression and cellular phenotypes. In addition, RNA-Seq and data analysis attempts were made to identify alternative kinase inhibitors to Total RNA was isolated from the purified CLL cells using QIAamp RNA help overcome ibrutinib resistance. Blood Mini Kit (Qiagen, Valencia, CA, USA), and 1 μg of RNA was used for

1Department of Pathology and Laboratory Medicine, Weill Cornell Medical College, New York, NY, USA; 2Department of Pathology, University of Chicago, Chicago, IL, USA and 3Department of Medicine, Weill Cornell Medical College, New York, NY, USA. Correspondence: Dr YL Wang, Department of Pathology, University of Chicago, MC1089, N316, 5841South Maryland Avenue, Chicago, IL 60637, USA. E-mail: [email protected] Received 8 April 2014; revised 27 August 2014; accepted 29 August 2014; accepted article preview online 5 September 2014; advance online publication, 10 October 2014 Functional characterization of BTKC481S mutation S Cheng et al 896 poly-A RNA selection with oligo (dT) beads. The enriched mRNA was used variety of drugs. After 10 days of culture, the CLL cells were harvested and for the construction of cDNA libraries. RNA sequencing was carried out in BrdU incorporation was determined with FITC BrdU Flow Kit following the an Illumina HiSeq 2000 instrument (Illumina, San Diego, CA, USA). Paired- manufacturer’s instructions (BD). The stained CLL cells were counted in an end sequencing was conducted with each sample running in a separate LSR2 cytometer and data were analyzed with FlowJo software (Tree Star, sequencing lane. More than 100 millions reads were yielded with an Ashland, OR, USA). average of 88x coverage per sample. The reads were aligned to the HG19 genome assembly using TopHat package with Bowtie2 aligner. Reads not mapped to the genome or siRNA-mediated BTK knockdown fl potential PCR duplicates were excluded with samtools. Cuf inks and The small interfering RNA (siRNA) against human BTK was customized and Cuffmerge were used to quantify levels of transcript expression. Reads per purchased from Thermo Scientific. The sense strand sequence of BTK kilobase per million mapped reads were computed: the number of reads siRNA was: 5′-GUAUGAGUAUGACUUUGAAUU-3′, and the antisense mapped to each transcript sequence was normalized by the template sequence was: 5′-UUCAAAGUCAUACUCAUAGUU-3′. A non-targeting siRNA length in kbs and divided by the number of reads mapping to the whole transcriptome. Hierarchical cluster analysis for BCR signature genes was with scrambled sequence was included as the negative control. siRNA transfection were performed with the Amaxa Nucleofector II device performed using Cluster 3.0 software, and heat map was generated using 7 TreeView. The 27-gene BCR signature was previously established by (Amaxa), and 1 × 10 CLL cells were resuspended in Amaxa Solution V Herishanu et al.15 containing 1 μM nontarget control siRNA or siBTK (Life Technologies, Carlsbad, CA, USA). Nucleofection was performed using the Program U-015. The knockdown efficiency was assessed by western blots and Cell transfection shown in Supplementary Figure 2. After transfection, cells were plated in a HBL1 cells were seeded overnight and transfected with a pCMV6 vector 24-well plate for 8 days in 1 ml of RPMI/10% fetal calf serum, CpG C481S that expresses either WT BTK or BTK mutant using the Amaxa oligodeoxynucleotides (dsp30), CD40L and BrdU before cells were Nucleofection technology according to the manufacturer's protocols collected for the BrdU incorporation assay. (Amaxa, Cologne, Germany; Program U-13). The cells were incubated for 12 h, and ibrutinib or vehicle was then added. Cells were continuously cultured for 12 h before cell harvest. Numbers of viable HBL1 cells were Statistical analysis counted with propidium iodide exclusion following periods of time One-way analysis of variance was applied to determine the differences in indicated in Figure 1d. the levels of the 27-gene BCR signature15 among four samples.

CLL proliferation assays To measure Ki-67 positivity, the fixed and cryopreserved cells were thawed RESULTS at room temperature, washed with 1x PBS once and permeabilized with C481S 50% methanol. The cells were then stained with PE-Cy5.5-conjugated anti- Functional consequences of the BTK mutation on BCR Ki-67 (BD). The Ki-67 gate was set based on the study of 14 pre-treatment molecular signaling samples collected from patients who participated in the trial. Ki-67 9 Using the serial samples collected from the patient, we analyzed positivity in these 14 patients ranged from 0.33 to 14%. This range was the consequences of the BTKC481S mutation on BCR molecular made consistent with that of Ki-67 positivity (0.1–13.8%) reported previously in the peripheral bloods of 95 CLL patients.16 signaling. As BTK autophosophorylation at Y223 results in kinase For bromodeoxyuridine (BrdU) incorporation assay, 4 × 104 NK-tert, a activation, the level of p-BTK (Y223) was measured by immuno- human bone marrow-derived stroma cell line, was seeded for 6 h in blotting as a surrogate marker for the BTK enzymatic activity. The RPMI1640 (10% fetal bovine serum), 5–10 million CLL cells were then immunoblot analysis demonstrated that p-BTK was decreased added and cocultured in the presence of 10 μM BrdU with or without a relative to the baseline with response to ibrutinib (Figure 1, Responding vs Pre-Rx), and restored to the baseline level during Before relapse Relapsed disease progression (Figure 1, Relapse s1 and s2). Changes in p-ERK displayed a similar trend as p-BTK, decreasing during the Pre-Rx Responding s1 s2 clinical response, and subsequently rising to the pre-treatment p-BTK level at disease progression. In contrast, p-AKT changed little (Y223) during the clinical course. Overall, changes in the levels of p-BTK and p-ERK correlated with each other and with the activity of patient’s clinical disease. The results suggest that the mutation t-BTK allows reactivation of BCR signaling in the patient still receiving ibrutinib. p-ERK1/2 (T202/Y204) Functional consequences of the BTKC481S mutation on BCR- t-ERK controlled gene expression To corroborate this finding, we analyzed the RNA-Seq data for gene expression using an established 27-gene BCR expression p-AKT signature.15 The expression of this cluster of genes in the BCR (S473) signature was downregulated in the Responding sample relative to the pre-treatment baseline as shown in Figures 2a and b. t-AKT However, the signature was rebounded in both Relapse s1 and Relapse s2 samples. Notably, the trend of changes in gene expression was similar to the trend of changes in p-BTK and p-ERK. GAPDH Collectively, the data from the previous mutational analyses, signal transduction and gene expression profiling strongly suggest Figure 1. C481S that (1) BCR signaling activity over the treatment course correlates Functional consequences of the BTK mutation on BCR ’ molecular signaling. Immunoblots of BTK phosphorylation and well with the status of the patient s disease; and (2) C481S downstream BCR signaling activities in the four serial samples. mutation confers resistance to ibrutinib leading to increased BCR GAPDH was included as the loading control. signaling at patient’s relapse.

Before relapse Relapse Pre-Rx Responding s1 s2 BCR signature *** CCL4 DUSP2 5 LPL PAICS PDGFA GFI1 * EGR3 GNPDA1 SLAMF7 OAS3 *** * *** NAB2 TXNRD1 LILRA4 DDX21 NR4A3 NFKBIE 0 EGR1 RGS10 CTLA4 CD83

Ly9 BCR Signature Score IRF4 KLF10 CASP3 -5 DDIT3 Pre-Rx Responding Relapse s1 Relapse s2

-3 -2 -1 0 1 2 3 Figure 2. Functional consequences of the BTKC481S mutation on BCR-controlled gene expression. (a) A graphical heat map showing the expression proﬁles of a 27-gene BCR signature across all 4 serial samples. The RPKM-normalized expression counts were represented by colors using the color reference range shown below. (b) The RPKM-normalized expression counts were plotted for each individual gene in the cluster of the BCR signature. Differences among the four specimens were analyzed by ANOVA one-way analysis of variance. *Po0.05, ***Po0.001.

Before Relapse CLL+ Stroma + ibrutinib Pre-Rx Responding Iso 0 nM 250nM 500nM

0.05% 4.5% 0.53% 0.03% 20% 0.0% Pre-Rx eoeRlpeRelapse Before Relapse Responding 0.0% Relapse 19% 0.13% 0.14% Ki67 s1 s2

5.3% 8.0% BrdU 0.36% 35% 14% 15% s1

CD19

0.53% 36% 13% 8.6% s2

7-AAD Figure 3. Functional consequences of the BTKC481S mutation on cell proliferation. (a) Longitudinal changes in Ki67+ CLL cells over patient clinical course. (b) In vitro stroma-induced proliferation was eliminated by ibrutinib in Pre-Rx and Responding samples, but not in relapsed samples. Concentrations of ibrutinib were indicated on the top of the each column. Percentage of BrdU+ cells are indicated in each plot. Iso, isotype control.

Functional consequences of the BTKC481S mutation on cell In the current study, the BrdU+ proliferative CLL population proliferation increased from 19–20% in the Pre-Rx and Responding samples to We next investigated the cellular consequences of the BCR 35–36% in the relapsed samples (Figure 3b, second column, rows reactivation. Previously, we demonstrated that ibrutinib directly 1–4vsfirst column, rows 1–4). Treatment of the patient’s CLL cells inhibits in vivo CLL proliferation. The number of Ki-67+ cells with ibrutinib (250 or 500 nM) completely blocked BrdU incorpora- decreased over the ibrutinib treatment course in a cohort of nine tion in the Pre-Rx and Responding samples (Figure 3b, top two patients.9 Analysis of serial samples from the current patient also rows, 3rd and 4th column vs 2nd). In contrast, a significant fraction demonstrated a reduction in the percentage of Ki-67+ CLL cells of proliferative CLL cells remained at 13–14% in both relapsed from baseline 4.5 to 0.53% in the Responding sample. Ki-67+ cells samples (Figure 3b, bottom two rows, 3rd and 4th column vs 2nd). then increased to 5.3% in Relapse s1, and subsequently to 8.0% in Thus, CLL in vitro proliferative response to ibrutinib agrees well Relapse s2 (Figure 3a). with the patient’s clinical sensitivity and resistance to ibrutinib To corroborate this finding, we also determined BrdU positivity, treatment. which reflects DNA synthetic activity. In a previously characterized coculture system, culture of CLL cells with NKTert, a human bone BTK has a direct role in promoting CLL proliferation marrow-derived stromal cell line, induces the appearance of a To assess whether BTK has a direct role in cell proliferation, we BrdU+ proliferative CLL population.9 reduced the level of BTK in Relapsed s2 using BTK-specific siRNA

© 2015 Macmillan Publishers Limited Leukemia (2015) 895 – 900 Functional characterization of BTKC481S mutation S Cheng et al 898 (siBTK). Figure 4a shows that BrdU incorporation was reduced the presence of the drug. In the absence of the drug, cells carry from 9.7 to 4.6% comparing control siRNA-transfected cells with either the WT construct or the mutant construct grew at the same siBTK-transfected cells. Next, we performed the reverse experi- rate. This is not surprising as the C481S change only affects ment by transiently transfecting an ibrutinib-sensitive lymphoma ibrutinib binding and does not confer BTK with additional cell line, HBL1, with either WT or mutant BTKC481S constructs and functions. Notably, the growth rate of drug-treated mutant cells tested their effects on cell growth in the presence or absence of was not the same as that of untreated mutant cells, suggesting ibrutinib. As shown in Figure 4b, growth of the WT cells was that the mutant cells maintain certain sensitivity to the drug. This inhibited by ibrutinib, whereas mutant cells continued to grow in observation may be explained by our structural finding showing that the binding of the drug was not abolished but significantly 9 Non-target diminished (by 500-fold). In this in vitro system, ibrutinib was Isotype siRNA siBTK constantly present in the cell culture, the equilibrium between bound and unbound drug may keep BTK inhibited to a higher 0.1% 9.7% 4.6% degree than in patient’s plasma where the drug is actively metabolized. Nonetheless, these results provide more definitive functional assessment, indicating that BTK has a direct role in cell

BrdU proliferation and the mutation causes ibrutinib resistance with regard to cell proliferation. 7-AAD Sensitivity of ibrutinib-resistant cells to other kinase inhibitors 30 As resistance to ibrutinib results in reactivation of the BCR pathway, we investigated whether other inhibitors of the BCR or wt alternative pathways can suppress CLL proliferation and override wt BTK + 400nM ibrutinib 25 the resistance. Using the in vitro CLL proliferation model, cells from mt Relapsed s1 were treated with dasatinib (mainly targets LYN and BTK), PRT062070 (Cerdulatinib, an SYK/JAK dual inhibitor), mt BTK + 400nM ibrutinib 17 20 PRT060318 (a highly speciﬁc SYK inhibitor ) and tofacitinib (a JAK inhibitor), as well as ibrutinib (Figure 5a). The concentrations for ibrutinib, dasatinib and tofacitinib were chosen based on 5,18,19 15 human pharmacokinetic and pharmacodynamic studies. A ﬁxed dose of 250 nM were chosen for the two SYK inhibitors to match that of ibrutinib where pharmacokinetic/pharmacodynamic 10 data are not available. Figure 5b demonstrates that ibrutinib brought proliferation down to 37.3% relative to no drug control,

Fold increase in cell number whereas tofacitinib decreased the proliferation only to 67.7%. 5 In contrast, dasatinib and the two inhibitors with activity against SYK (PRT062070 and PRT060318) led to a complete block of proliferation. Structural modeling suggests that dasatinib ﬁts into 0 the ATP binding pocket of both WT and mutant BTK 02448 72 96 120 144 168 (Supplementary Figure 3) and it indeed binds and inhibits the Culture duration (hours) kinase.20 Dasatinib, as a poorly selective drug, may also affect the Figure 4. BTK has a direct role in promoting CLL proliferation. cells through its actions on other molecules including LYN. (a) BTK knockdown reduced the BrdU incorporation. Treatment We also tested idelalisib (CAL-101 or GS-1101), a phosphoinosi- conditions are indicated on top of each dot plots. (b) BTK-mutated tide 3-kinaseδ inhibitor that is highly effective against CLL in 21,22 cells continue to grow in the presence of ibrutinib (400 nM). phase 1 and phase 3 clinical trials, for its potential to Cell numbers were counted in a ﬂow cytometer. overcome ibrutinib resistance. A pharmacokinetic study in human

250nM 125nM No drug Ibrutinib Dasatinib 100 100

80 67.7

60 250nM 250nM 250nM

BrdU PRT062070 PRT060318 Tofacitinib 37.3 40

20 Normalized Proliferate Rate (%) Normalized Proliferate 0.16 0.06 0.06 0 No 250nM 125nM 250nM 250nM 250nM 7-AAD drug Ibrutinib Dasatinib PRT PRT Tofacitinib 062070 060318 Figure 5. Sensitivity of ibrutinib-resistant cells to other small-molecule inhibitors. (a) Effects of other kinase inhibitors on CLL proliferation. Concentrations used in the experiment were indicated. Untreated cells (+NK) served as the control. The ibrutinib-resistant Relapsed s1 were treated with: Dasatinib, a multikinase inhibitor; PRT060318, an SYK inhibitor; PRT062070, an SYK/JAK dual inhibitor; and tofacitinib, a JAK inhibitor. Left panel, percentages of BrdU+ CLL cells were measured. (b) Proliferation rates (%) were normalized to no drug control (100%).

Leukemia (2015) 895 – 900 © 2015 Macmillan Publishers Limited Functional characterization of BTKC481S mutation S Cheng et al 899 showed that the in vivo maximal achievable concentration is investigation provided mechanistic insights highlighting not only 23 ~6μM with a once-daily dose of 150 mg. We thus tested a range BTK but also the key role of BCR signaling and cell proliferation in of idelalisib concentrations below this level. Supplementary Figure 4 the pathogenesis of CLL. As clinical trials of ibrutinib in many other demonstrated a dose-dependent reduction in CLL proliferation types of non-Hodgkin lymphoma are ongoing, cases of resistance with increasing concentrations of idelalisib. At 5 μM of the drug, have emerged.24 Knowledge of the molecular and cellular the population of proliferative CLL cells dropped to 1.9%. These mechanisms would allow early detection and time for the results suggest that idelalisib may serve as an alternative therapy development of strategies to overcome the resistance. in the setting of ibrutinib resistance at clinically achievable doses. Taken together, our results suggest that BTK-mutated CLL cells may still be sensitive to other agents that targeted the B-cell CONFLICT OF INTEREST receptor signaling pathway including dasatinib, SYK inhibitors and The authors declare no conflict of interest. idelalisib. ACKNOWLEDGEMENTS DISCUSSION The study was supported by funding from the Leukemia and Lymphoma Society (LLS In the current study, we investigated the functional consequences 6364-13) to YLW. of BTKC481S mutation from the following aspects: (1) BCR signaling, (2) gene expression and (3) cellular proliferative responses. Notably, BCR signaling, gene expression profiles and in vivo REFERENCES Ki-67 measurement exhibited very similar up and down changes 1 Rickert RC. New insights into pre-BCR and BCR signalling with relevance to B cell across all four samples. Further, the baseline cell proliferative malignancies. Nat Rev Immunol 2013; 13: 578–591. capacity and their response to ibrutinib also correlate well with 2 Dal Porto JM, Gauld SB, Merrell KT, Mills D, Pugh-Bernard AE, Cambier J. B cell 41 – patient’s clinical sensitivity and resistance to the treatment. antigen receptor signaling 101. Mol Immunol 2004; :599 613. 3 Kurosaki T, Shinohara H, Baba Y. B cell signaling and fate decision. Annu Rev Importantly, BTK transfection or knockdown experiments Immunol 2009; 28:21–55. demonstrated a direct role of BTK in CLL cell proliferation. 4 Harwood NE, Batista FD. Early events in B cell activation. Annu Rev Immunol 2010; In attempts to overcome ibrutinib resistance in this patient, 28: 185–210. several drugs were tested using our in vitro CLL proliferation 5 Advani RH, Buggy JJ, Sharman JP, Smith SM, Boyd TE, Grant B et al. Bruton model. The CLL cells were sensitive to inhibition by dasatinib and tyrosine kinase inhibitor ibrutinib (PCI-32765) has significant activity in two SYK inhibitors and a phosphoinositide 3-kinaseδ inhibitor, patients with relapsed/refractory B-cell malignancies. J Clin Oncol 2013; 31: whereas a JAK inhibitor had poor activity (Figure 5). The patient 88–94. was subsequently treated with dasatinib at 100 mg two times a 6 Byrd JC, Furman RR, Coutre SE, Flinn IW, Burger JA, Blum KA et al. Targeting BTK day. Her disease stabilized with a small reduction in palpable with ibrutinib in relapsed chronic lymphocytic leukemia. N Engl J Med 2013; 369: – lymphadenopathy and an improvement in hemoglobin. The 32 42. 7 Chang BY, Furman RR, Zapatka M, Barrientos JC, Li D, Steggerda S et al. patient ultimately succumbed to sepsis. Use of tumor genomic profiling to reveal mechanisms of resistance to the BTK It is interesting that both SYK inhibitors completely suppressed inhibitor ibrutinib in chronic lymphocytic leukemia (CLL). ASCO Meet Abstr 2013; ’ CLL proliferation. Besides BTK, SYK s substrates include at least 31: 7014. BLNK, phospholipase Cγ2 and BCAP in the BCR pathway 8 Woyach JA, Furman RR, Liu TM, Ozer HG, Zapatka M, Ruppert AS et al. Resistance (Supplementary Figure 1) that may enable SYK-targeted inhibitors mechanisms for the Bruton's tyrosine kinase inhibitor ibrutinib. N Engl J Med 2014; to have an antitumor effect when BTK-targeted molecule fails. It is 370: 2286–2294. also very interesting to note that idelalisib may work in the setting 9 Furman RR, Cheng S, Lu P, Setty M, Perez AR, Guo A et al. Ibrutinib of ibrutinib resistance. Given its success in advanced clinical trials, resistance in chronic lymphocytic leukemia. N Engl J Med 2014; 370: the drug was approved in July for treating relapsed CLL/SLL 2352–2354. patients by the US Food and Drug Administration and may be 10 Cheng S, Ma J, Guo A, Lu P, Leonard JP, Coleman M et al. BTK inhibition targets 21,22 in vivo CLL proliferation through its effects on B-cell receptor signaling activity. approved in Europe soon. We further suggest that it may be Leukemia 2014; 28:649–657. used to rescue patients with BTK mutations. Altogether, our 11 Song Z, Lu P, Furman RR, Leonard JP, Martin P, Tyrell L et al. Activities of SYK and observations suggest that therapeutic intervention by other PLCgamma2 predict apoptotic response of CLL cells to SRC tyrosine kinase kinase inhibitors may shut down BCR signaling and inhibit CLL inhibitor dasatinib. Clin Cancer Res 2010; 16: 587–599. C481S proliferation in patients who have acquired the BTK mutation. 12 Yang C, Lu P, Lee FY, Chadburn A, Barrientos JC, Leonard JP et al. Tyrosine Combining with our previous findings, the results from the kinase inhibition in diffuse large B-cell lymphoma: molecular basis for current study are consistent with the following sequence of events antitumor activity and drug resistance of dasatinib. Leukemia 2008; 22: in our patient: (1) the patient acquired BTKC481S mutation during 1755–1766. the course of ibrutinib treatment; (2) C481S disrupted BTK- 13 Jo SH, Yang C, Miao Q, Marzec M, Wasik MA, Lu P et al. Peroxisome proliferator- ibrutinib covalent binding; (3) decreased binding and loss of activated receptor gamma promotes lymphocyte survival through its actions on cellular metabolic activities. J Immunol 2006; 177: 3737–3745. inhibition led to increased BCR signaling that resulted in increased 14 Lu P, Yang C, Guasparri I, Harrington W, Wang YL, Cesarman E. Early events of gene expression; (4) increased gene expression subsequently led B-cell receptor signaling are not essential for the proliferation and viability of to increased cell proliferation; and (5) accelerated cell proliferation AIDS-related lymphoma. Leukemia 2009; 23: 807–810. led to increased cellularity that eventually manifested in the 15 Herishanu Y, Perez-Galan P, Liu D, Biancotto A, Pittaluga S, Vire B et al. The lymph patient as rising lymphocyte count, progressive lymphadenopathy node microenvironment promotes B-cell receptor signaling, NF-kappaB activa- as parts of disease progression. Lastly and importantly, the tion, and tumor proliferation in chronic lymphocytic leukemia. Blood 2011; 117: resistance may be overcome by other BCR inhibitors. 563–574. It is noteworthy that cysteine to serine change in BTK is one of 16 Damle RN, Temburni S, Calissano C, Yancopoulos S, Banapour T, Sison C et al. several mechanisms cells used to escape ibrutinib’s antitumor CD38 expression labels an activated subset within chronic lymphocytic leukemia 110 – effects. Mutations in other genes, such as phospholipase Cγ2, SF3B1 clones enriched in proliferating B cells. Blood 2007; : 3352 3359. 17 Cheng S, Coffey G, Zhang XH, Shaknovich R, Song Z, Lu P et al. SYK inhibition and and MLL2 were observed in CLL patients who developed 118 8,24 response prediction in diffuse large B-cell lymphoma. Blood 2011; : resistance to ibrutinib. However, it remains unclear whether 6342–6352. these mutations cause or concur with ibrutinib resistance. 18 Christopher LJ, Cui D, Wu C, Luo R, Manning JA, Bonacorsi SJ et al. Metabolism Functional characterization described herein demonstrated that and disposition of dasatinib after oral administration to humans. Drug Metab C481S BTK is responsible for the drug resistance. Further, the Dispos 2008; 36: 1357–1364.

© 2015 Macmillan Publishers Limited Leukemia (2015) 895 – 900 Functional characterization of BTKC481S mutation S Cheng et al 900 19 Dowty ME, Jesson MI, Ghosh S, Lee J, Meyer DM, Krishnaswami S et al. Preclinical 22 Brown JR, Byrd JC, Coutre SE, Benson DM, Flinn IW, Wagner-Johnston ND et al. to clinical translation of tofacitinib, a Janus kinase inhibitor, in rheumatoid Idelalisib, an inhibitor of phosphatidylinositol 3-kinase p110delta, for relapsed/ arthritis. J Pharmacol Exp Ther 2014; 348:165–173. refractory chronic lymphocytic leukemia. Blood 2014; 123: 3390–3397. 20 Hantschel O, Rix U, Schmidt U, Burckstummer T, Kneidinger M, Schutze G et al. 23 Jin F, Robeson M, Zhou H, Moyer C, Ramanathan S. Pharmacokinetics and safety The Btk tyrosine kinase is a major target of the Bcr-Abl inhibitor dasatinib. of idelalisib, a novel PI3Kδ inhibitor, in Japanese and Caucasian subjects. Proc Natl Acad Sci USA 2007; 104: 13283–13288. ASH Annu Meet Abstr 2013; abstract 5575. 21 Furman RR, Sharman JP, Coutre SE, Cheson BD, Pagel JM, Hillmen P et al. Idelalisib 24 Burger JA, Landau D, Hoellenriegel J, Sougnez C, Schlesner M, Ishaque N et al. and rituximab in relapsed chronic lymphocytic leukemia. N Engl J Med 2014; 370: Clonal evolution in patients with chronic lymphocytic leukemia (CLL) developing 997–1007. resistance to BTK inhibition. ASH Annu Meet Abstr 2013; abstract 866.

Supplementary Information accompanies this paper on the Leukemia website (http://www.nature.com/leu)