Mitochondrial Priming of Chronic Lymphocytic Leukemia Patients Associates Bcl-Xl Dependence with Alvocidib Response

Letters to the Editor 2251 other lymphoma subtypes. The most common gene ontology 2Department of Genetics, University of Groningen, University Medical identified within the list of 88 overlapping genes is the chromatin Center Groningen, Groningen, The Netherlands; modification and transcription gene ontology with 13 genes, 3Centre for Lymphoid Cancer, Department of Experimental Ther- for example, CREBBP, MLL2, MLL3, BCL6, HIST1H1E and EP300. apeutics, The BC Cancer Agency, Vancouver, Mutations of histone-modifying genes are also common in British Columbia, Canada and DLBCL and FL. The common origin of HRS cells and the 4Department of Hematology, University of Groningen, University tumor cells of DLBCL and FL is consistent with the marked Medical Center Groningen, Groningen, The Netherlands overlap in the mutational spectrum. For BL, the overlap is limited E-mail: [email protected] and further genomic studies are awaited to clearly establish the extent of overlap. Despite the significant overlap, the majority of the genes that were mutated in HL remain unique. This indicates REFERENCES that the genetic events occurring during the malignant transfor- 1 Kuppers R, Engert A, Hansmann ML. Hodgkin lymphoma. J Clin Invest 2012; 122: mation in HL are distinct from those in other GC B-cell- 3439–3447. derived NHL. 2 Gunawardana J, Chan F, Telenius A, Woolcock BW, Kridel R, Tan KL et al. Exome- fi In conclusion, we determined the mutational landscape of HL wide association study identi es a TM6SF2 variant that confers susceptibility to cell lines by WES. Several of the mutated genes are in common nonalcoholic fatty liver disease Recurrent somatic mutations of PTPN1 in primary mediastinal B cell lymphoma and Hodgkin lymphoma. Nat Genet 2014; 46: with other GC B-cell-derived lymphoma subtypes. We observed a 352–356. high number of consistently mutated genes, with part of them 3 Morris LG, Ramaswami D, Chan TA. The FAT epidemic: a gene family mapping to regions with copy number gain or loss and part of frequently mutated across multiple human cancer types. Cell Cycle 2013; 12: them showing deregulated expression. This supports their 1011–1012. potential relevance in HL and will help to select candidate genes 4 Yin H, Liang Y, Yan Z, Liu B, Su Q. Mutation spectrum in human colorectal cancers for further analysis. and potential functional relevance. BMC Med Genet 2013; 14:32. 5 Zhang J, Mullighan CG, Harvey RC, Pearson SA, Walters RJ, Nagler B et al. Key pathways are frequently mutated in high-risk childhood acute lymphoblastic CONFLICT OF INTEREST leukemia: a report from the Children’s Oncology Group. Blood 2011; 118: 3080–3087. The authors declare no conflict of interest. 6 Diepstra A, Poppema S, Boot M, Visser L, Nolte IM, Niens M et al. HLA-G protein expression as a potential immune escape mechanism in classical Hodgkin's ACKNOWLEDGEMENTS lymphoma. Tissue Antigens 2008; 71:219–226. 7 Steidl C, Shah SP, Woolcock BW, Rui L, Kawahara M, Farinha P et al. MHC class II We acknowledge the Genome facility of the Genetics Department (Pieter van der transactivator CIITA is a recurrent gene fusion partner in lymphoid cancers. Nature Vlies and others). This work was supported by grants from the Abel Tasman Talent 2011; 471: 377–381. Program, University Medical Center Groningen and the Dutch Cancer Society (KWF: 8 Zhang J, Grubor V, Love CL, Banerjee A, Richards KL, Mieczkowski PA et al. Genetic RUG 2009-4313). heterogeneity of diffuse large B-cell lymphoma. Proc Natl Acad Sci USA 2013; 110: 1398–1403. 9 Morin RD, Mendez-Lago M, Mungall AJ, Goya R, Mungall KL, Corbett RD et al. AUTHOR CONTRIBUTIONS Frequent mutation of histone-modifying genes in non-Hodgkin lymphoma. Nat- YL performed research, collected, interpreted data, performed statistical analyses, ure 2011; 476:298–303. and wrote the manuscript; MT and FRAR performed statistical analyses, analyzed and 10 Morin RD, Mungall K, Pleasance E, Mungall AJ, Goya R, Huff RD et al. Mutational interpreted data; AS performed data and statistical analyses; MN and GvI analyzed and structural analysis of diffuse large B-cell lymphoma using whole-genome mutational spectrum in BL, DLBCL and FL in comparison to HL and contributed to the sequencing. Blood 2013; 122: 1256–1265. writing of the manuscript; LV and JK contributed samples and contributed to writing 11 Pasqualucci L, Trifonov V, Fabbri G, Ma J, Rossi D, Chiarenza A et al. Analysis of the manuscript; FCC, RG and CS performed RNA-seq analysis, interpreted data and of the coding genome of diffuse large B-cell lymphoma. Nat Genet 2011; 43: contributed to writing of the manuscript; AvdB, KK and AD supervised the project, 830–837. designed research, interpreted data and wrote the manuscript. 12 Lohr JG, Stojanov P, Lawrence MS, Auclair D, Chapuy B, Sougnez C et al. Discovery and prioritization of somatic mutations in diffuse large B-cell lymphoma (DLBCL) 109 – Y Liu1, FR Abdul Razak1, M Terpstra2, FC Chan3, A Saber1, by whole-exome sequencing. Proc Natl Acad Sci USA 2012; : 3879 3884. 13 Green MR, Gentles AJ, Nair RV, Irish JM, Kihira S, Liu CL et al. Hierarchy in somatic M Nijland4, G van Imhoff4, L Visser1, R Gascoyne3, C Steidl3, 1 1 2 1 mutations arising during genomic evolution and progression of follicular lym- J Kluiver , A Diepstra , K Kok and A van den Berg 121 – 1 phoma. Blood 2013; : 1604 1611. Department of Pathology & Medical Biology, University of 14 Richter J, Schlesner M, Hoffmann S, Kreuz M, Leich E, Burkhardt B et al. Recurrent Groningen, University Medical Center Groningen, mutation of the ID3 gene in Burkitt lymphoma identified by integrated genome, Groningen, The Netherlands; exome and transcriptome sequencing. Nat Genet 2012; 44: 1316–1320.

Supplementary Information accompanies this paper on the Leukemia website (http://www.nature.com/leu) Mitochondrial priming of chronic lymphocytic leukemia patients associates Bcl-xL dependence with alvocidib response

Leukemia (2014) 28, 2251–2254; doi:10.1038/leu.2014.206 EFC6663 (relapsed/refractory patients) were assessed for mitochondrial outer membrane permeabilization following BH3-only Chronic lymphocytic leukemia (CLL) patient outcomes may benefit peptide incubation. Training and test cohort analyses established from targeting therapies to individuals with favorable molecular an association between Bim and Hrk biomarkers and patient profiles. We assessed mitochondrial functionality in apoptosis response that benefitted from inclusion of trisomy 12 status. signaling for CLL patient response to alvocidib (flavopiridol). Separately, tumor lysis syndrome (TLS) occurrence was found to Pretreatment patient specimens from Trials OSU0055 and be associated with Bad priming, which benefitted from inclusion

Accepted article preview online 3 July 2014; advance online publication, 1 August 2014

Bim(0.1) Total Patients Hrk Total Patients 80 80

p = .0027 p = .00046 60 Regression Analysis 60 Regression Analysis

40 40 %Priming %Priming

20 20

0 0

PD SD PR PD SD PR

AUCHrk = 0.73 (p = 0.0002) AUCBIM(0.1) = 0.73 (p = 0.0004) AUCHrk+Trisomy12 = 0.83 (p < 0.0001)

Figure 1. Bim and Hrk BH3 profiling of CLL patients are correlated with alvocidib response. (a) Dot-plot depictions of the combined data set by stratification of response into three categories (PD, SD, PR). Note that increased priming trends are observed for both Bim(0.1) and Hrk from PD to SD and then from SD to PR. (b) Dot-plot and ROC-plot depictions of Bim(0.1) and Hrk display response discrimination (two groups: PD/ SD, PR). (c) Chromosome 12 trisomy multivariate analysis adds to Hrk prediction of CLL patient clinical response to alvocidib. While both Bim (0.1) and Hrk display AUC from ROC-plot depictions of 0.73, Hrk models benefit from inclusion of significant clinical adjustment variable trisomy 12 to yield the increased AUC of 0.83 (Po0.0001).

of ECOG status and patient age. Taken together, the association of malignancies.7,8 This profiling of Bcl-2 family members serves to Bim and Hrk with CLL patient response is consistent with Bcl-xL elucidate the extent to which pro-apoptotic sensitizers (that is, Puma, dependence as a driver of alvocidib efficacy. Concurrently, Bad’s Noxa, Bad, Hrk) and activators (that is, Bim) may regulate the activity association with TLS, which is indicative of Bcl-2’s dependency, of antiapoptotic proteins (thatis,Mcl-1,Bcl-2,Bcl-xL). Indeed, previous may help identify patients at risk for treatment-related toxicity. investigations have established that alvocidib potentiates pan-BH3 Alvocidib has demonstrated single-agent activity in relapsed/ mimetic activity through upregulation of BH3-only proteins with refractory CLL patients in Phase I and Phase II clinical studies although coordinate downregulation of their antiapoptotic counterparts.9 the activity was originally limited, presumably due to high protein Previous studies assessing alvocidib efficacy in CLL patients ex vivo binding in plasma attenuating the levels of available drug in patients.1,2 and expression levels of Bcl-2 family proteins, including Mcl-1, were However, Byrd and colleagues3–6 recently identified a hybrid intrave- not able to establish correlation with clinical efficacy.10 Based on nous dosing schedule that utilizes a 30-min loading dose (30 mg/m2) these observations, we have investigated the functional context of followed by a 4-h infusion (50 mg/m2). This novel dosing schedule has intrinsic apoptosis BH3-only proteins by mitochondrial priming allowed for improved single-agent activity in relapsed and refractory CLL assessment as a surrogate for cellular response to pro-apoptotic patients in multiple clinical studies, with overall response rates between cues to provide a predictive strategy for CLL patient management.7,11 30 and 53%, including favorable responses in high-risk patients with Patients were randomly segregated into two cohorts that would 17p13.1 deletions.3–6 Response rate limitations have been compounded serve as proof-of-principle (n = 30) and validation sets (n = 32); by treatment-related toxicities including TLS. Patient enrichment cohorts were established merely on the basis of specimen tissue strategies that include molecular and cytogenetic approaches may bank. (Patient information is provided in Supplementary Table 1.) enhance development of alvocidib in CLL by targeting the drug to Percent priming (that is, quantifiable propensity of a given BH3 patients who are most likely to exhibit response and least likely to peptide to induce mitochondrial depolarization relative to an experience adverse events. uncoupling control agent) for each peptide is summarized in Assessment of intrinsic mitochondrial apoptosis pathways by Supplementary Table 2 for patients who exhibited partial response functional profiling has recently shown encouraging results for (PR), stable disease (SD), or progressive disease (PD). In the patient stratification strategies in treatment of hematologic training set, only Bim(0.1) and Hrk elicited significance (P = 0.014

Figure 2. Bad peptide BH3 profiling correlates with TLS in CLL patients following treatment with alvocidib. (a) Dot-plot depictions indicate that higher Bad BH3 profiling readout values are significantly associated with the presence of TLS versus those patients who did not experience TLS. (b) The Bad AUC from ROC-plot analysis was 0.75; this improved to 0.85 when combined with clinical adjustment variables age and ECOG status. and P = 0.0098, respectively) between biomarker and PR, SD and BH3 profiling biomarkers were analyzed relative to progression- PD patients (regression analyses). These two markers validated free survival (PFS) and overall survival (OS); no significant (Bim P = 0.0051; Hrk P = 0.015) in the second set. When the two association between biomarker and outcome was observed sets were combined (n = 62), regression indicated that Bim(0.1) (Supplementary Figure 3) although there was a trend between and Hrk were both significant (Bim(0.1) P = 0.0027 and Hrk Bim(0.1) and OS (P = 0.03), and between Hrk and PFS (P = 0.08), P = 0.00046, respectively; P-value significance at o0.01 for suggesting that an appropriately powered study may demonstrate Bonferroni correction as described in Methods) (Figure 1a). more significant correlation between these parameters. Although not considered in the core group of patients initially As TLS remains a concern in patients prescribed certain identified and examined in this study, a subset of patients from therapies, including alvocidib, we sought to determine if BH3 OSU trial 0491 for which there were sufficient available vials was profiling could be used to identify patients at risk for experiencing considered as a secondary validation set. Here, Hrk priming in PD/ toxicity. Supplementary Table 3 summarizes the association of SD patients displayed a mean of 6.4%, and PR patients a mean of BH3 profiling biomarkers with TLS status for 48 patients for whom 22.2% (n = 13; 4 PD/SD, 9 PR). Although not statistically powered, toxicity data were available. Here, both higher Bad and Puma(10) the trend towards Hrk priming correlation with clinical response priming scores were significantly associated with TLS (Figure 2a) was consistent (Supplementary Figure 2). by Wilcoxon (P = 0.0083 and P = 0.0080, respectively) and logistic Figure 1b indicates dot plot depictions of Bim(0.1) and Hrk regression (P = 0.012 and P = 0.012, respectively). None of the relative to response when the profiling data are further analyzed other BH3 profiling biomarkers was significant (P40.05). The AUC as PD/SD versus PR. To test the ability of either Bim(0.1) or Hrk to statistic indicated that each marker held a significant ability to serve as a predictive biomarker of response, we employed area predict TLS status; Puma(10) AUC = 0.75 (P = 0.00079) and Bad under the curve (AUC) analysis to determine the sensitivity and AUC = 0.75 (P = 0.00072). Puma(10) and Bad when added together specificity of this biomarker; Bim AUC = 0.73 (CI (0.60–0.85); did not improve the predictive capacity of either marker P = 0.00037) and Hrk AUC = 0.73 (CI (0.61–0.86); P = 0.00022) independently. (significant P-valueo0.01 employing Bonferroni correction). Bim Patient age (P = 0.034, Wilcoxon; P = 0.067, logistic regression) (0.1) displays a P of 0.04, while Hrk indicates a P of 0.0039 and baseline ECOG status (P = 0.021, Wilcoxon; P = 0.031, logistic (significant P-valueo0.01). Adjusted analyses of Bim(0.1) and Hrk regression) showed significant association or borderline associa- priming in which we accounted for trisomy 12 status as a second tion with toxicity (Supplementary Table 3). Multivariate analyses, covariate improved the AUC to 0.83 (CI (0.71–0.95); Po0.0001) with inclusion of both patient age and ECOG status to Bad BH3 (Figure 1c). In contrast to Hrk, trisomy 12 status did not improve profiling, resulted in increase in AUC from 0.75 to 0.85, and the model when added with Bim(0.1). improved P (from 0.0007 to o0.0001) (Figure 2b). Trisomy 12 is a frequent event in CLL, occurring in ~ 20% of Mitochondrial priming readout values from Hrk and Bim(0.1), patients12 generally carrying an intermediate prognosis.12,13 The both with a capacity to bind to and modulate the activity of frequency of trisomy 12-positive patients in the EFC6663 study is antiapoptotic Bcl-xL, are the markers statistically associated with in line with this (8 of 62 (13%)). Although small in number, all eight response, indicating this clinical end point is driven by Bcl-xL of the patients on study experienced clinical benefit (7 PR, 1 SD); dependency. Patients likely to exhibit TLS are predicted by Bad whether this extends to a larger patient population needs to be priming; the indicative toxicity may be driven by its antiapoptotic explored. Approximately 35–50% of trisomy 12-positive patients binding partner Bcl-2. One plausible mechanism to explain the also have a mutation in NOTCH1,14 conferring adverse outcomes observed phenomenon is that clinical response and TLS are in some series.15 This patient stratification strategy could benefit occurring via the action of alvocidib on two clonally related but from layering of BH3 profiling to stratify patients who are most biologically distinct cell subpopulations, one driven by Bcl-xL and likely to benefit from alvocidib in future studies. the other by Bcl-2. Interestingly, both subpopulations would by

© 2014 Macmillan Publishers Limited Leukemia (2014) 2241 – 2272 Letters to the Editor 2254 definition comprise B-cell populations as the BH3 profiling assay REFERENCES here is performed on a patient specimen after purification and 1 Flinn IW, Byrd JC, Bartlett N, Kipps T, Gribben J, Thomas D et al. Flavopiridol elimination of non-B-cells. Additional work will be required to administered as a 24-hour continuous infusion in chronic lymphocytic leukemia further characterize the presence and composition of these lacks clinical activity. Leuk Res 2005; 29:1253–1257. distinct subpopulations. Priming by Noxa, a specific pro- 2 Byrd JC, Peterson BL, Gabrilove J, Odenike OM, Grever MR, Rai K et al. Treatment apoptotic modulator of antiapoptotic Mcl-1, is not associated of relapsed chronic lymphocytic leukemia by 72-hour continuous infusion or 1-hour bolus infusion of flavopiridol: results from Cancer and Leukemia Group B with either alvocidib response or TLS, consistent with Mcl-1 not 11 – being a driver mechanism. study 19805. Clin Cancer Res 2005; : 4176 4181. fi 3 Lin TS, Ruppert AS, Johnson AJ, Fischer B, Heerema NA, Andritsos LA et al. Phase II Conceivably, patients could be strati ed by cytogenetics and study of flavopiridol in relapsed chronic lymphocytic leukemia demonstrating high BH3 profiling to increase the overall response rate from 35 to 50% response rates in genetically high-risk disease. JClinOncol2009; 27:6012–6018. as observed in previous clinical studies. In the current patient set, 4 Byrd JC, Lin TS, Dalton JT, Wu D, Phelps MA, Fischer B et al. Flavopiridol admi- 23 of the 62 analyzed patients achieved a PR to alvocidib (37.1%.) nistered using a pharmacologically derived schedule is associated with marked In our analyses, the combination of Hrk and trisomy 12 yielded an clinical efficacy in refractory, genetically high-risk chronic lymphocytic leukemia. AUC of 0.83. By closer examination of the ROC curves, 95.6% of Blood 2007; 109:399–404. responder patients were identified (sensitivity) concurrently to 5 Stephens DM, Ruppert AS, Blum K, Jones J, Flynn JM, Johnson AJ et al. Flavopiridol discrimination of 66.7% of likely PD/SD patients (specificity). If the treatment of patients aged 70 or older with refractory or relapsedchronic predictive value of such an applied biomarker is to triage the likely lymphocytic leukemia is a feasible and active therapeutic approach. Haematologica 2012; 97:423–427. non-responder patients (here the triaging of 26 PD/SD patients, 6 Phelps MA, Lin TS, Johnson AJ, Hurh E, Rozewski DM, Farley KL et al. Clinical response while also mis-identifying 1 likely PR patient), then the response and pharmacokinetics from a phase 1 study of an active dosing schedule of flavo- rate increases from 37.1 to 62.9% (an overall 69.5% improvement piridol in relapsed chronic lymphocytic leukemia. Blood 2009; 113:2637–2645. of response rate). Similarly, TLS incidence could be decreased from 7 Ni Chonghaile T, Sarosiek KA, Vo TT, Ryan JA, Tammareddi A, Moore Vdel G et al. 13% observed in the EFC6663 study to a minimal number moving Pretreatment mitochondrial priming correlates with clinical response to cytotoxic forward. While larger cohorts will be required to fully define the chemotherapy. Science 2011; 334: 1129–1133. extent of medical utility for such diagnostics, these early 8 Pierceall WE, Kornblau SM, Carlson NE, Huang X, Blake N, Lena R et al. BH3 profiling discriminates response to Cytarabine-based treatment of acute observations may hold promise for alvocidib impact in CLL 12 – patient options and improved outcomes. myeloid leukemia. Mol Cancer Ther 2013; : 2940 2949. 9 Chen S, Dai Y, Pei X-Y, Myers J, Wang L, Kramer LB et al. CDK inhibitors up-regulate BH3-only proteins to sensitize human myeloma cells to BH3 mimetic therapies. Cancer Res 2012; 72: 4225–4237. CONFLICT OF INTEREST 10 Woyach JA, Lozanski G, Ruppert AS, Lozanski A, Blum KA, Jone JA et al. Outcome of patients with relapsed or refractory chronic lymphocytic leukemia treated with WEP,CD,RJL,NB,ME,andMHCareemployeesofEutropics,Inc.SLWandDJBare flavopiridol: impact of genetic features. Leukemia 2012; 26: 1442–1444. employees of Tolero Pharmaceuticals, Inc. JCB and MRG are co-inventors of a patent 11 Byrd JC, Shinn C, Waselenko JK, Fuchs EJ, Lehman TA, Nguyen PL et al. pending for the use of alvocidib. The remaining authors declare no conflict of interest. Flavopiridol induces apoptosis in chronic lymphocytic leukemia cells via activa- tion of caspase-3 without evidence of bcl-2 modulation or dependence on 1 2 1 1 1 WE Pierceall , SL Warner , RJ Lena , C Doykan , N Blake , functional p53. Blood 1998; 92: 3804–3816. 1 2,3 2 1 4 M Elashoff , DV Hoff , DJ Bearss , MH Cardone , L Andritsos , 12 Döhner H, Stilgenbauer S, Benner A, Leupolt E, Krober A, Bullinger L et al. JC Byrd4, MC Lanasa5, MR Grever4 and AJ Johnson4 Genomic aberrations and survival in chronic lymphocytic leukemia. New Engl J 1Eutropics Inc., Cambridge, MA, USA; Med 2000; 343: 1910–1916. 2Tolero Pharmaceuticals, Inc., Lehi, UT, USA; 13 AbdelSalam M, El Sissy A, Samra MA, Ibrahim S, El-Markaby D, Gadallah F. The 3 impact of trisomy 12, retinoblastoma gene and P53 in prognosis of B-cell chronic Translational Genomics Research Institute, Scottsdale, AZ, USA; 13 – 4Department of Internal Medicine, Division of Hematology, lymphocytic leukemia. Hematology 2008; :147 153. 14 Balatti V, Bottoni A, Palamarchuk A, Alder H, Rassenti LZ, Kipps TJ et al. The Ohio State University Comprehensive Cancer Center, Columbus, NOTCH1 mutations in CLL associated with trisomy 12. Blood 2012; 119: 329–331. OH, USA and 15 Willander K, Dutta RK, Ungerback J, Gunnarsson R, Juliusson G, Fredrikson M et al. 5 Department of Medicine, Duke Cancer Institute, Durham, NC, USA NOTCH1 mutations influence survival in chronic lymphocytic leukemia patients. E-mail: [email protected] or [email protected] BMC Cancer 2013; 13: 274.

Supplementary Information accompanies this paper on the Leukemia website (http://www.nature.com/leu) Immunoglobulin light-chain amyloidosis shares genetic susceptibility with multiple myeloma

Leukemia (2014) 28, 2254–2256; doi:10.1038/leu.2014.208 myeloma (MM) pass with a relative risk of 25.3,4 AL amyloidosis, MGUS and MM are hypothesized to be the same disease entity at Amyloid light-chain (AL) amyloidosis is a rare clonal plasma cell the cellular level, with AL amyloidosis just being a clonal plasma disorder with an estimated incidence of 3 per million.1 The disease cell disorder with an ‘unlucky protein’.5 Nevertheless, the genetic is characterized by deposition of amyloid fibers derived from etiology of AL amyloidosis is poorly understood. According immunoglobulin light chains systematically in many organs.2 The to a 15-year follow-up study of MGUS patients the relative risk symptoms arise in the organs where amyloids accumulate, for AL amyloidosis was 8.4.3 Two recent genome-wide association including the heart, kidney, liver and peripheral nerves; heart studies (GWAS) of MM have identified seven single-nucleotide failure is the critical life-threatening condition that may ensue in a polymorphisms (SNPs) that are associated with disease risk, few months after diagnosis.2 Monoclonal gammopathy of including loci mapping to 2p23.3 (rs6746082), 3p22.1 undetermined significance (MGUS) is an asymptomatic pre- (rs1052501), 3q26.2 (rs10936599), 6p21.33 (rs2285803), 7p15.3 malignant clonal condition, through which all cases of multiple (rs4487645), 17p11.2 (rs4273077) and 22q13.1 (rs877529).6,7 These

Accepted article preview online 3 July 2014; advance online publication, 25 July 2014