PDF Full Text

From www.bloodjournal.org by guest on October 10, 2018. For personal use only.

Regular Article

LYMPHOID NEOPLASIA North American ATLL has a distinct mutational and transcriptional proﬁle and responds to epigenetic therapies

Urvi A. Shah,1-3 Elaine Y. Chung,2 Orsi Giricz,3 Kith Pradhan,3 Keisuke Kataoka,4 Shanisha Gordon-Mitchell,3 Tushar D. Bhagat,3 Yun Mai,2 Yongqiang Wei,2,5 Elise Ishida,2 Gaurav S. Choudhary,3 Ancy Joseph,6 Ronald Rice,7 Nadege Gitego,3 Crystall Parrish,3 Matthias Bartenstein,3 Swati Goel,1 Ioannis Mantzaris,1 Aditi Shastri,1,3 Olga Derman,1 Adam Binder,1 Kira Gritsman,1,2 Noah Kornblum,1 Ira Braunschweig,1 Chirag Bhagat,8 Jeff Hall,8 Armin Graber,8 Lee Ratner,6 Yanhua Wang,9 Seishi Ogawa,4 Amit Verma,1,3 B. Hilda Ye,2 and Murali Janakiram1

1Department of Oncology, Monteﬁore Medical Center, Bronx, NY; 2Department of Cell Biology and 3Department of Developmental and Molecular Biology, Albert Einstein College of Medicine, Bronx, NY; 4Department of Pathology and Tumor Biology, Graduate School of Medicine, Kyoto University, Kyoto, Japan; 5Department of Hematology, Nanfang Hospital, Southern Medical University, Guangzhou, China; 6Division of Oncology, Washington University School of Medicine, St Louis, MO; 7Department of Pathology, Phelps Hospital, Northwell Health, Sleepy Hollow, NY; 8Genoptix, Carlsbad, CA; and 9Department of Pathology, Monteﬁore Medical Center, Bronx, NY

KEY POINTS Adult T-cell leukemia lymphoma (ATLL) is a rare T cell neoplasm that is endemic in Japanese, Caribbean, and Latin American populations. Most North American ATLL patients are of l North American ATLL has a distinct genomic Caribbean descent and are characterized by high rates of chemo-refractory disease and landscape with a worse prognosis compared with Japanese ATLL. To determine genomic differences be- high frequency of tween these 2 cohorts, we performed targeted exon sequencing on 30 North American prognostic epigenetic ATLL patients and compared the results with the Japanese ATLL cases. Although the mutations, including EP300 mutations. frequency of TP53 mutations was comparable, the mutation frequency in epigenetic and histone modifying genes (57%) was signiﬁcantly higher, whereas the mutation frequency l ATLL samples with in JAK/STAT and T-cell receptor/NF-kB pathway genes was signiﬁcantly lower. The most mutated EP300 have compromised p53 common type of epigenetic mutation is that affecting EP300 (20%). As a category, epi- function and are genetic mutations were associated with adverse prognosis. Dissimilarities with the selectively sensitive to Japanese cases were also revealed by RNA sequencing analysis of 9 primary patient decitabine treatment. samples. ATLL samples with a mutated EP300 gene have decreased total and acetyl p53 protein and a transcriptional signature reminiscent of p53-mutated cancers. Most importantly, decitabine has highly selective single-agent activity in the EP300-mutated ATLL samples, suggesting that decitabine treatment induces a synthetic lethal phenotype in EP300-mutated ATLL cells. In conclusion, we demonstrate that North American ATLL has a distinct genomic landscape that is characterized by frequent epigenetic mutations that are targetable preclinically with DNA methyltransferase inhibitors. (Blood. 2018;132(14):1507-1518)

Introduction of Caribbean/American ATLL (n 5 53), the median overall survival (OS) was only 6.9 months,3,6 similar to other reported North Adult T-cell leukemia lymphoma (ATLL) is a rare aggressive T-cell American studies.2,7 The outcomes in Japanese cohorts of ATLL neoplasm, which is caused by a retrovirus (human T-cell lym- are more favorable, with median survival times for acute, lymphotropic virus [HTLV]-1), and carries a dismal prognosis. HTLV-1 phomatous, chronic, and smoldering subtypes of 8.3, 10.6, has low sequence variability, enabling its sequence to be used 31.5, and 55.0 months, respectively.8 The percentage of pa- as a molecular tool to follow migrations. The most common tients with aggressive subtypes (acute and lymphomatous) in the subtype of HTLV-1 is the cosmopolitan subtype A, which is American cohorts is ;91%,2,3,7 compared with 78% in the Japa- endemic to Japan, the Caribbean, Central and South America, nese population.8 Thus, despite a younger age at diagnosis, north and west Africa, and parts of the Middle East.1 Conse- NorthAmericanATLLpatientsappeartopresentwithsig- quently, ATLL is diagnosed most frequently in the Japanese and nificantly more aggressive subtypes of the disease than their Caribbean populations2,3 but is likely underreported in Africa, Japanese counterparts. The majority of research on this disease Latin America,4 and the Middle East.5 is from Japan, where it is endemic, and most preclinical models used in published ATLL studies are Japanese-derived cell lines Despite similar viral subtypes, the prognosis of ATLL is worse in and xenografts. It is not known whether North American ATLL is the Caribbean and American populations than in the Japanese genotypically distinct from the Japanese cases and whether such population. In our retrospective analysis of a single-center cohort differences, if present, can be correlated with clinical outcomes.

PATHWAY PI3K/AKT/mTOR signaling B cell receptor signaling Stress response Stress Receptor kinases tyrosine Other Notch signaling Epigenetic and histone modiﬁcation Transcriptional regulation Cell death Jak/Stat signaling Ras signaling TCR signaling FGFR signaling Cell cycle and DNA maintenance Wnt/  -catenin signaling structure Chromatin VHL XPO1 ATLL PATIENTS ATLL AGE GENDER CYTOGENETICS Overall Survival KEAP1 CREBBP PBRM1 TET2 EZH2 MED12 SPEN HIST1H1E IDH1 KRAS HRAS POT1 SETBP1 CDKN2A BRCA1 PALB2 GATA3 TBL1XR1 AR BCL6 TCF3 MYC NOTCH2 CARD11 TNFAIP3 RHOA PLCG2 FGFR3 FGFR1 TSC1 AKT1 MCL1 RIPK1 ERBB3 ALK ERBB2 PDGFRB IGF1R DDR2 RET CD79A SYK SMC3 STAG2 STAT3 PTCH1 SPOP CDH1 FLT1 AXL MAP2K2 MDM4 NTRK1 FAS SUBTYPE TP53 EP300 DNMT3A KMT2A SMARCB1 ASXL1 NRAS BRCA2 PRDM1 GATA2 CEBPA BCOR NKX2-1 KLF2 FAT1 APC DDX3X NOTCH1 FBXW7 TRAF3 FGFR4 FGFR2 AKT2 RICTOR PIK3CD KDR KIT EGFR PDGFRA FLT3 JAK3 STAT5B ITPKB ZYMM3 P2RY8 ZFHX4 MAP3K9 ESR1 ATL 1 83 M 33 ATL 2 59 M 382 ATL 3 30 F 543 ATL 4 66 F 363 ATL 5 74 F 160 ATL 6 53 M88 ATL 7 48 F 225 ATL 8 64 F 215 ATL 9 63 F 739 ATL 10 36 M 52 ATL 11 70 F 451 ATL 12 54 F 23 ATL 13 37 F 17 ATL 14 41 M 771* ATL 15 51 M 312 ATL 16 57 F 2554* ATL 17 36 M 75 ATL 18 67 F 175 ATL 19 65 M 392 ATL 20 74 M 1191* ATL 21 37 F 6345* ATL 22 60 M 126 ATL 23 54 F 503* ATL 24 59 F 176 ATL 25 64 M 73 ATL 26 61 F 102 ATL 27 40 F 351* ATL 28 40 M 322 ATL 29 63 F 192 ATL 30 70 F 153* ATL55T(+) Su9T01 ATL43T(+) ATL43Tb(-) ED40515(+) ED40515(-) ED41214(+) ED41214(-) Japanese

ATL 1-30: North American ATLL Patients ATL55T(+) to ED41214(-) are 8 Japanese derived cell lines Japanese: Mutations also seen in Japanese patients Subtype: Acute Lymphomatous Chronic/Smoldering Cytogenetics: Normal Complex Unknown

Overall Survival: Number of days from diagnosis to death *: Alive

B p53

TAD DNABD TM

0 100 200 300 393 aa

C p300

TAZ Z TAZ 1 KIX BD HAT Z 2 IBiD

0 400 800 1200 1600 2000 2414 aa

D E 60%

50% 25% 40% 20% 30% 15%

Frequency (%) Frequency 20% 10% Frequency (%) Frequency 5% 10%

0% 0% IDH1 IDH2 TP53 TET2 TET2 JAK3 JAK1 EZH2 SPEN POT1 STAT3 EP300 EP300 RHOA ASXL1 GATA3 PBRM1 MED12 KMT2A PIK3CD CARD11 TNFAIP3 TBL1XR1 CDKN2A NOTCH1 DNMT3A DNMT3A HIST1H1E SMARCB1 All epigenetic Mutated genes Mutated genes North American (n=30) Japanese (n=370) North American (n = 30) Japanese (n = 81)

Figure 1. North American ATLL has a distinct mutational landscape. (A) North American ATLL samples (n 5 30) and Japanese ATLL cell lines (n 5 8) were sequenced by targeted deep next-generation sequencing. Identiﬁed mutations are grouped into various functional categories. Corresponding mutations reported in Japanese ATLL are marked in the last row. (B) Location of point mutations in the p53 protein structure. (C) Location of point mutations in the p300 protein structure. Green, blue, and black circles

1508 blood® 4 OCTOBER 2018 | VOLUME 132, NUMBER 14 SHAH et al From www.bloodjournal.org by guest on October 10, 2018. For personal use only.

Furthermore, understanding the mutational landscape of North genes is determined through bioinformatic analysis and com- American ATLL is critical in an effort to develop new targeted parison with databases (eg, COSMIC and dbSNP). Quality-control therapies for these patients. metrics included a minimum of 200 ng of genomic DNA and average mean sequencing depth of 5003 coverage to give a limit Epidemiology of ATLL in the United States reflects emigration of detection of 5% for SNVs, 10% for insertions/deletions patterns from endemic areas, especially the Caribbean region, and translocation fusions, gene amplifications $ 6 copies, and and is characterized by an increasing incidence in New York City homozygous gene deletions , 0.3 copies. These mutational data and Miami.9,10 Montefiore Medical Center treats a significant were compared with clinical characteristics and patient out- proportion of ATLL patients in the United States as a result of comes to determine the prognostic impact of particular mutations. the large number of Caribbean immigrants in the Bronx, NY.3,6 Here, we present the mutational and transcriptional landscape of Patient records were queried using Clinical Looking Glass soft- Caribbean ATLL and demonstrate that it is characterized by a distinct ware to identify all cases of HTLV positivity and ATLL by searching mutation pattern that targets epigenetic pathways more frequently pathology and laboratory reports of patients who presented than that reported for Japanese ATLL. We also demonstrate that, to Montefiore Medical Center between 2003 and 2017.6 The although ATLL samples with EP300 mutations have compromised diagnosis of ATLL was confirmed based on clinical history, p53 function, they are hypersensitive to DNA methyltransferase pathological findings, and HTLV-1 antibody positivity.6 Cases inhibitors (DNMTIs) and, thus, provide a preclinical rationale for were classified by subtype using Shimoyama criteria.11 The index treating North American ATLL with epigenetic therapies. date was defined as the date on which a diagnosis of ATLL was made. Mortality data included death records within our institution, as well as those reported in Social Security records. For patients Patients, materials, and methods who were discharged to hospice with confirmed documentation Patient samples and cell lines of refractory disease and unknown date of death (n 5 5), the date Specimens were obtained from patients diagnosed with ATLL of discharge to palliative care was used as the date of death. after Institutional Review Board approval by the Albert Einstein College of Medicine, in accordance with the Declaration of Statistical analysis Helsinki. Patient characteristics, including demographics, labo- Data generated by Clinical Looking Glass software and com- ratory parameters, cytogenetics and treatment course, and plemented with individual chart review were transferred to a survival, were collected using retrospective chart review (sup- computer spreadsheet (Microsoft Excel; Microsoft, Redmond, plemental Table 1). To establish long-term cultures, CD4 T cells WA). For the analysis of categorical variables, we reported propor- were sorted from patient peripheral blood mononuclear cells tions and P values calculated using the Pearson x2 test or Fisher fl using uorescence-activated cell sorting (FACS) and expanded exact test, as appropriate. Kaplan-Meier curves were used to fi in Iscove modi ed Dulbecco medium supplemented with 20% compare survival, and statistical significance was examined using human serum and 100 unit/mL interleukin-2 (IL-2). All short-term the Wilcoxon rank-sum test. Statistical analyses were performed and long-term ATLL cultures have been characterized by mul- with Stata v12 (StataCorp, College Station, TX), and a 2-tailed a of tiparameter immunophenotyping with a panel of 17 markers, 0.05 was used to denote significance. including CD3, CD4, CD25, CD44, FoxP3, CTLA-4, ICOS, and – 2 CCR4. The Japanese patient derived ATLL cell lines ATL43Tb , Detailed procedures on in vitro cell cytotoxicity assays, RNA 2 2 Su9T01, ED40515 , and ED41214 were maintained in RPMI sequencing (RNA-seq), bioinformatics analysis, analysis of drug 1640 supplemented with 10% fetal bovine serum, whereas the interaction, mass spectrometry, Western blot, and proviral – 1 1 1 IL-2 dependent cell lines AT55T , ED40515 , ED41214 , and load analysis are described in supplemental Methods, which 1 AT43T were maintained in RPMI 1640 plus 10% fetal bovine are available on the Blood Web site. serum supplemented with IL-2.

Genetic sequencing and clinical characteristics Results Genomic DNA was extracted from peripheral blood, bone marrow, or formalin-fixed paraffin-embedded solid tumor Mutational landscape of North American ATLL samples from 30 ATLL patients and used to identify single nu- patients reveals a high frequency of prognostic cleotide variants (SNVs), insertion/deletions, copy number var- epigenetic gene alterations iations, and translocation fusion genomic alterations in a panel Targeted exon sequencing was performed on samples from of up to 173 reportable genes (patients 1-23) and up to 236 30 North American ATLL patients. They were diagnosed according genes (patients 24-30) (supplemental Table 2). We also se- to the World Health Organization classification12 and sub- quenced 8 cell lines (ATL43T1, ATL43Tb2, Su9T01, ATL55T1, classified using Shimoyama criteria.11 Of these, 43.3% (n 5 13) ED405152, ED405151, ED412142, and ED412141) derived from were acute, 43.3% (n 5 13) were lymphomatous, and 13.4% Japanese ATLL patients. Targeted genomic regions were (n 5 4) were chronic/smoldering ATLL. Baseline demographic sequenced using next-generation sequencing by Genoptix characteristics are shown in supplemental Table 1. For the (https://genoptix.com/test-menu/nexcourse-complete/). The 173 cancer-related genes analyzed in the assay format, 152 mu- presence or absence of genomic alterations within each of the tations in 80 genes were detected, with an average of 5 identified

Figure 1 (continued) depict missense single nucleotide polymorphisms, splice site single nucleotide polymorphisms, and truncating mutations, respectively. (D) Frequency of mutations in North American and Japanese ATLL. (E) Frequency of epigenetic mutations in North American ATLL and whole-exome sequenced Japanese ATLL cases. *P , .05, North American vs Japanese.

MUTATIONAL LANDSCAPE OF NORTH AMERICAN ATLL blood® 4 OCTOBER 2018 | VOLUME 132, NUMBER 14 1509 From www.bloodjournal.org by guest on October 10, 2018. For personal use only.

Table 1. Subtype, OS, and mutational proﬁle of North American ATLL

Age at ATL no. Sex diagnosis, y Subtype OS, d Mutations Mutations, n Treatment

1 Male 83 A 33 KIT, APC, MYC, TP53, DDX3X, 5 Unknown TSC1

2 Male 59 L 382 NRAS, POT1, CEBPA, KEAP1, AR, 5 Etoposide, cyclophosphamide, NOTCH1, MCL1 cytarabine, cisplatin

3 Female 28 A 543 ALK, BCL6, RIPK1, GATA3, 7 CHOP, cytarabine, denileukin IGF1R, TP53, CEBPA diftitox, IT MTX

4 Female 66 L 363 NOTCH2, DDR2, GATA3 3 EPOCH

5 Female 74 L 160 KDR, FAT1, FGFR2, AKT1, TP53, 7 EPOCH ERBB2, MED12

6 Male 53 A 88 XPO1, TBL1XR1, PDGFRA, KIT, 6 EPOCH (died after cycle 1) FGFR4, EP300, RICTOR

7 Female 48 L 225 KDR, CARD11, NKX2-1 (2) 4 Cyclophosphamide, doxorubicin, etoposide, MTX

8 Female 64 L 215 TNFAIP3, TSC1, BRCA2, 3 Cyclophosphamide, NOTCH1 doxorubicin, etoposide

9 Female 63 L 739 CARD11, NOTCH1, KRAS, POT1 3 EPOCH

10 Male 36 L 52 CEBPA, RHOA, PBRM1, CD79A, 5 Bortezomib, cytarabine, STAG2 doxorubicin, etoposide

11 Female 70 L 451 NRAS, VHL, APC, EGFR, PTCH1, 6 DA EPOCH bortezomib ASXL1, KMT2A raltegravir

12 Female 54 L 23 FAT1, HIST1H1E, PLCG2 3 CHOP died after cycle 1

13 Female 37 A 17 FGFR4, APC, FAT1, SPEN 3 EPOCH

14 Male 41 L 771* EZH2, DDX3X, FAT1, PTCH1 4 EPOCH 1 allogeneic transplant

15 Male 51 L 312 TP53, EP300, PALB2, PTCH1, 5 EPOCH TRAF3

16 Female 57 C 2554* APC, PDGFRB, TET2, DNMT3A 3 IFN-AZT

17 Male 36 A 75 KDR, NOTCH1, TP53, SPOP 3 HyperCVAD, CHOP, EPOCH (1 cycle each)

18 Female 67 A 175 AKT2, CDKN2A, EP300, FAT1, 7 EPOCH 1 raltegravir 1 FBXW7, FLT3, PBRM1, SPEN bortezomib 1 IFN

19 Male 65 A 392 SETBP1, EP300 2 EPOCH

20 Male 74 A 1191* FGFR3, FAT1 2 EPOCH 1 cycle, IFN

21 Female 37 C/S 6345* CDH1, EP300 2 Treated initially as mycosis fungoides, followed by brentuximab

22 Male 60 C/S 126 AKT2, ALK, CD79A, EP300, 6 None, due to poor performance ZMYM3, AR status and comorbidities

23 Female 54 A 503* DDR2, IDH1, TBL1XR1, TP53, 5 ESHAP 1 cycle, followed by ZMYM3 EPOCH

1510 blood® 4 OCTOBER 2018 | VOLUME 132, NUMBER 14 SHAH et al From www.bloodjournal.org by guest on October 10, 2018. For personal use only.

Table 1. (continued)

Age at ATL no. Sex diagnosis, y Subtype OS, d Mutations Mutations, n Treatment

24 Female 59 L 176 FGFR3, FLT1, KRAS, P2RY8, TET2 5 EPOCH 1 raltegravir 1 bortezomib, palliative after cycle

25 Male 64 A 73 AR, JAK3, PDGFRA, SETBP1, 7 EPOCH 1 raltegravir 1 TP53, XPO1, ZFHX4† bortezomib, died after cycle 2

26 Female 61 S 102 AKT1, AXL,† PALB2 2 None

27 Female 40 L 351* BCL6, MDM4† 2 EPOCH 1 raltegravir 1 bortezomib

28 Male 40 A 322 CDH1, HIST1H1E, NOTCH1 3 EPOCH 1 raltegravir 1 bortezomib

29 Female 63 A 192 NOTCH1, APC, APC, GATA2, 10 IFN-AZT, followed by 2 cycles KLF2,† NTRK1, SMARCB1, ESHAP, followed by ICE 1 SPEN, TBL1XR1, TCF3 bortezomib, followed by a clinical trial

30 Female 70 A 153* ALK, ALK, CDKN2A, ERBB3, 11 EPOCH FAS,† FAT1, HRAS, KLF2,† PIK3CD, PIK3CD, TBL1XR1

A, acute; AZT, azidothymidine; CHOP, cyclophosphamide, doxorubicin hydrochloride (hydroxydaunorubicin hydrochloride), vincristine (oncovin), and prednisone; DA EPOCH, dose-adjusted etoposide, prednisone, vincristine, cyclophopshamide, doxorubicin; EPOCH, etoposide, prednisone, vincristine (oncovin), cyclophosphamide, and doxorubicin hydrochloride (hydroxydaunorubicin hydrochloride); ESHAP, etoposide, methylprednisone, cytarabine, cisplatin; ICE, ifosfamide, carboplatin, etoposide; IFN, interferon; IT, intrathecal; L, lymphomatous; MTX, methotrexate. *Alive. †Genes that were identified as part of the extended spectrum of 236 genes in patients ATL24 to ATL30. mutations per patient (Figure 1; Table 1; supplemental Table 2). common to these 2 sequencing panels (173 genes in North American The acute/lymphomatous cases and the chronic/smoldering cases and 88 genes in Japanese) (Figure 1D; supplemental Tables 9-11). had an average of 5.2 and 3.75 identified mutations per sample, respectively. Genes most commonly mutated in North American Although many of the mutations identified in this study have ATLL were TP53 (7/30, 23%), FAT1 (7/30, 23%), EP300 (6/30, 20%), been previously reported for Japanese ATLL, several differences NOTCH1 (6/30, 20%), APC (5/30, 17%), TBL1XR1 (4/30, 13%), and were apparent. Because the large Japanese cohort of 370 cases KDR, ALK,andPTCH1 (3/30, 10% each) (Figure 1A; supplemental was sequenced for some, but not all, epigenetic genes in our Tables 3 and 4). In addition to inactivating mutations in the histone panel, we also compared our findings with their whole-exome acetyltransferase EP300 (Figure 1C-D; supplemental Table 5), data from 81 Japanese cases. More than half of our patients mutations were detected in other epigenetic and histone- (17/30, 57%) had mutations in epigenetic or histone modifier modifying genes (TET2, EZH2, MED12, PBRM1, DNMT3A, KMT2A, genes, most notably inactivating mutations in the histone HIST1H1E, SPEN, IDH1, SMARCB1, and ASXL1), resulting in a acetyltransferase EP300 (Figure 1C; supplemental Table 5); combined mutational frequency of 57% (17/30). mutations were also detected in other epigenetic and histone The cadherin-related tumor suppressor FAT1, mutated in other modifying genes (TET2, EZH2, MED12, PBRM1, DNMT3A, KMT2A, malignancies,13 including T acute lymphoblastic leukemia,14 showed HIST1H1E, SPEN, IDH1, SMARCB1,andASXL1) (Figure 2C). SNVs in 7 of 30 (23%) of our patients. Nevertheless, germline In comparison, these mutations were seen at much lower fre- sequencing revealed that the FAT1 variants in 2 of the samples quencies in the Japanese cohort (18/81, 22.2%) (Figure 1E; sup- (ATL14 and ATL18) were of germline origin. On the other hand, plemental Table 11). Of note, the frequency of EP300 mutations comparison with the germline profiles demonstrated that the in the North American cohort (20%) was about 3 times greater majority of epigenetic and TP53 mutations were somatic in nature than that seen in the Japanese cohort (5.7%; P 5 .01) (Figure 1E). (Figure 1B,D; supplemental Tables 6-8). Integrated molecular analysis of the Japanese patients showed North American ATLL is characterized by alterations that are highly enriched for T-cell receptor–NF-kB significantly more epigenetic mutations and fewer signaling (including JAK/STAT signaling), T-cell trafficking, and JAK/STAT and T-cell receptor/NF-kB pathway other T-cell–related pathways, as well as immunosurveillance mutations compared with Japanese ATLL genes.15 These trends were not seen in the North American ATLL We next compared mutation patterns of North American ATLL with a cohort. For example, mutations in CARD11 were significantly less Japanese cohort of 370 patients who underwent targeted capture common in the North American cohort compared with Japanese sequencing of 88 candidate genes selected from a discovery co- cases (6.6% vs 23.8%). In addition, 21.4%, 2.2%, and 1.4% of hort of 83 patient samples that were studied with whole-genome Japanese patients had mutations in STAT3, JAK3,andJAK1, or whole-exome sequencing.15 There were 17 genes that were respectively, whereas in the North American cohort there was no

MUTATIONAL LANDSCAPE OF NORTH AMERICAN ATLL blood® 4 OCTOBER 2018 | VOLUME 132, NUMBER 14 1511 From www.bloodjournal.org by guest on October 10, 2018. For personal use only.

A B 1.00 1.00

No Epigenetic mutations No TP53 mutations 0.75 0.75 Epigenetic mutations TP53 mutations

0.50 0.50 Frequency Frequency 0.25 P Val = 0.01 0.25 P Val = 0.07

0.00 0.00 0 500 1000 1500 0 500 1000 1500 Time (days) Time (days)

CD Enrichment plot: P53_DN.V1_UP 0.40 FDR Q = 0.039 0.35 NES = 1.66 ATL13 ATL21 ATL19 ATL29a ATL29d ATL13 ATL18 ATL21 ATL19 ATL30 0.30 Nominal p = 0.0 EP300 WTMUT MUT WT WT WTMUT MUT MUT WT 0.25 0.20 p300 300 kD 0.15 0.10 0.05 TP53 0.00 WT WT WT WT WT WT WT WT WT WT (ES) Enrichment score -0.05 Total p53 53 kD

Acetyl-p53 53 kD 5.0 '1' (positively correlated) Lys382 2.5 0.0 GAPDH 38 kD -2.5

(Signal2Noise) -5.0 '2' (negatively correlated) Ranked list metric Ranked 0 5,000 10,000 15,000 20,000 25,000 Rank in ordered dataset Enrichment proﬁle Hits Ranking metric scores

Figure 2. North American ATLL features a high frequency of epigenetic mutations that are prognostic. (A) OS of North American ATLL patients with epigenetic mutations is worse than those without these mutations. (B) OS of North American ATLL patients with TP53 mutations shows a trend toward worse survival, but it is not statistically signiﬁcant. (C) EP300 mutations are associated with reduced p300 levels, as well as total and acetylated p53 levels, in primary patient samples. (D) Gene set enrichment analysis revealed enrichment of the P53_DN.V1_UP signature in the 2 EP300-mutated samples (ATL21 and ATL18) compared with 2 EP300 WT samples (ATL29 and ATL30).

STAT3 or JAK1 mutation, and only 1 patient showed a JAK3 were still detected at higher frequencies in the North American (3.3%) alteration (Figure 1D; supplemental Tables 9 and 11). cohort relative to the Japanese cohort (supplemental Table 12).

Other mutations detected in both groups at comparable fre- EP300 mutations correlate with adverse prognosis quencies were TP53, NOTCH1, TBL1XR1, TET2, GATA3, POT1, and are associated with reduced p300 and TNFAIP3, DNMT3A, CDKN2A, and JAK3 (Figure 1D). Of note, p53 expression TP53 was mutated at similar frequencies in these 2 patient Next, we evaluated the prognostic significance of the major mu- cohorts. In the Japanese discovery cohort, 13.6% (11/81) of tation groups in North American ATLL. TP53 mutations are gen- patients had TP53 mutations, among which acute, lymphoma- erally accepted as a marker of adverse prognosis in ATLL, and all tous, and chronic cases accounted for 55%, 27%, and 18%, 7 TP53 mutations were seen in acute (n 5 5) or lymphomatous respectively. We also performed targeted exon sequencing disease (n 5 2) and were absent in patients with chronic/smoldering on 8 Japanese-derived cell lines, because they are frequently ATLL (Figure 1B; Table 2; supplemental Table 8). Even though used by ATLL investigators. Six of them had mutations in JAK3, patients with TP53 mutations presented with features of aggressive STAT3,orSTAT5B, consistent with the higher frequency reported for the Japanese cohort (Figure 1A; supplemental Table 14). disease (higher white blood cell [WBC] count and serum calcium) fi Mutations identified in these cell lines, but not detected in (Tables 2 and 3), TP53 mutation was associated with a nonsigni - 5 Japanese or North American patients, include BRCA1, BCOR, cant trend toward worse OS (P .07, log-rank test; Figure 2B). FGFR1, RET, SYK, STAT5B, ITPKB, MAP2K2, MAP3K9,andESR1. Interestingly, as a group, epigenetic mutations did correlate with a significantly worse prognosis (P 5 .01, log-rank test; Because the North American cohort has a higher percentage of ag- Figure 2A). The median survival of patients with and without gressive cases compared with the Japanese cohort, we performed any epigenetic mutations was 176 and 382 days, respectively. a subtype-specific comparison after separating the cases into aggressive (acute/lymphomatous) and indolent (chronic/smoldering) Among the epigenetic alterations detected in our cohort, mu- subtypes. The result showed that, even when the comparison was tations in the EP300 gene are the most common. Therefore, restricted to the aggressive cases, epigenetic and EP300 mutations we examined functional consequences of these mutations in

1512 blood® 4 OCTOBER 2018 | VOLUME 132, NUMBER 14 SHAH et al From www.bloodjournal.org by guest on October 10, 2018. For personal use only.

Table 2. Clinical presentation by ATLL subtype

Patient characteristics Acute (n 5 13) Lymphomatous (n 5 13) Chronic/smoldering (n 5 4)

Age, y 63 59 58

WBC, 109/L 23.1 6.4 12.25

Corrected calcium, mg/dl 13.88 11.52 9.88

Albumin, g/dl 3.6 3.5 3.9

Lactate dehydrogenase, U/L 676 561 280

TP53 mutation, % (n) Mutated 38 (5) 15 (2) 0 (0) Unmutated 62 (8) 85 (11) 100 (4)

Any epigenetic mutation, % (n) Mutated 54 (7) 54 (7) 75 (3) Unmutated 46 (6) 46 (6) 25 (1)

cultured leukemic cells isolated from our patients. As shown in mononuclear cells from healthy controls (n 5 4) (Figure 3; sup- Figure 2C, among the 7 samples examined, the 3 samples with plemental Figure 2). Unsupervised clustering analysis of the amutatedEP300 status (ATL18, ATL19, and ATL21) expressed whole transcriptomes revealed that the ATLL samples had a notably lower levels of p300 protein, as well as total p53 protein, distinct gene-expression profile compared with normal CD4 T compared with the 4 EP300 wild-type (WT) samples (ATL13, cells (Figure 3A). The overexpressed and underexpressed genes ATL29a, ATL29d, and ATL30). Because p53 is a protein substrate in ATLL (Figure 3B) were enriched in many functionally important of p300 lysine acetyltransferase activity,16,17 we also measured pathways (Figure 3C-D), including those associated with cancer, acetyl p53 and found reduced acetyl p53 signals in EP300- apoptosis, and immune cell–related functions (Figure 3D). mutated samples, as predicted. Because acetylation of p53 is indispensable for its in vivo function,18 these results suggest impaired Given the differences in mutational patterns between the North p53 tumor-suppressive activity as a result of EP300 mutation. To American and Japanese ATLL cohorts, we also compared the substantiate this notion, we performed RNA-seq analysis of 2 pairs transcriptional profiles. Unsupervised hierarchical clustering of EP300-mutated and WT samples. Gene set enrichment analysis revealed that all North American samples clustered in a dis- of the differentially expressed genes (supplemental Figure 1) tinct group that is well separated from the Japanese cohort, revealed enrichment of genes typically associated with a mutated suggesting significant differences in transcriptomic features TP53 status in the NCI-60 tumor cell line panel (Figure 2D), con- between these 2 patient populations (Figure 3E). firming attenuated p53 activity in the EP300-mutated samples.

North American ATLL has a distinct transcriptomic Treatment with the DNMTI decitabine leads to proﬁle and differs from Japanese ATLL reduced proliferation and increased apoptosis To determine the transcriptomic characteristics of North American in ATLL ATLL, we performed RNA-seq on primary North American ATLL To evaluate the therapeutic usefulness of targeting such epi- samples (n 5 9), as well as CD4 T cells and peripheral blood genetic marks, we tested the sensitivity of North American and

Table 3. Clinical presentation based on mutational status in acute/lymphomatous ATLL

TP53 (n 5 7) EPI (n 5 11) None (n 5 8) P

Age, y 54 54 63 .51 (between groups)

WBC 19.2 14.9 5.2 .04 (TP53 vs none)

Albumin 3.4 4.1 3.5 .13 (between groups)

Corrected calcium 13.8 11.5 11.8 .02 (TP53 vs none)

Lactate dehydrogenase 2022 579 586 .20 (between groups)

Median OS, d 160 176 382

Progression-free survival, d 98 128 240.5

EPI, any epigenetic mutation.

MUTATIONAL LANDSCAPE OF NORTH AMERICAN ATLL blood® 4 OCTOBER 2018 | VOLUME 132, NUMBER 14 1513 From www.bloodjournal.org by guest on October 10, 2018. For personal use only.

A B C 350 Differentially Expressed Genes 60 RNA-Seq 300 3882 2098

50 GLB1L2 10 PTPRN2 8 250 PTGER3 6 CCR8

HNRNPA1P21 4

AIRE 2 CDKN2B-AS1 40 0 200 CDKN2A Height PRKY

TRIM2

TRBV6-5 VLDLR 150 30 SPARC SDPR PEG10 PPBP

KIR3DL2 ATL18

ATL23 ZNF503 100 ATL6 APBB2 ATL13 20 GNAI1 BEND4 (-log(P value)) CTTN ATL29 ATL30 ATL19 ATL25 ATL22 • CDKN1A COL6A4P1 GPA33 MLF1 10 SH3BGRL2 PBX1

Signi ﬁ cance of difference IGHD

BANK1 MS4A1

TMEM178B SCNBA 0 IGHA1 KCNK5

MUC16 RFX8 CTRLCD4_1 CTRLCD4_2 CTRLCD4_4 CTRLCD4_3 CYP24A1 -8 -6 -4 -2 024 GRAMD1C ATP9A NFASC

ARNT2

POF1B Diff in mean expression PARM1 BCAS1

TNFSF11

NRXN3 IL1R2

CACNA1D FST

ARHGAP6 Overexpressed in Underexpressed in RAB25 POU4F1

FOXB1

ATLL ATLL TP73 FBN1

PTPLA

CX3CR1

MYO1E

FCRL1

HMCN1

MIOX

CRMP1

LTBP1 D CD70 P WWTR1 -Log ( val) 0 2 4 6 810121416 GTSF1 P2RY1

PDLIMY

SLC1A1

FCRL5

BOLL

DIAPH3

ESYT3 Cancer FCGR3B TPT1P12

HNRNPA1P34 DPY19L2 Organismal Injury PKP2 GATA6

ENPP3 Gastrointestinal Disease GNG11 TCL6

PCDHB9 Hematologic Disease SGCE PTH2R

ZYG11A Immunological Disease NAT8L CELF2-AS2

DISC1FP1

NRP2

MSRB3

CACNA2D3 Cell death and survival RAPGEF5 CA2

IGF2BP2 Cell Cyle NFE2 GMPR

TMCC2 Cellular Development TMOD1 KANK2

ALAS2 Cellular Growth and Proliferation TRIM58 CA1

HEMGN Cellular Movement SNCA HBG2

HBQ1

SMIM1

BAMBI

DYRK3 Lymphoid Tissue Development MDS2 POU3F1

MDGA1 Hematological System Development C10orf11 TAS1R1

TSHR Tissue Morphology PCDHGA10 PCDHGA6 MME Immune Cell Trafﬁcking PCDHGB4 ZG16B KIAA1217 Connective Tissue Development RN7SL443P CEACAM8 C9orf139 MYO5B CREG2 STARD13 FAM227A DUSP19 INSL6 TDRD9 PLS3 SH3RF2 GREB1 WNT5A E TMEM56 IFI27 Unsupervised Clustering KRT7 ARMC2-AS1 RHOXF1 FAM83F RNA-seq NBL1 FAM153C HLF Japanese ATLL GSTM3 RBM11 MYO16 CDS1 North American ATLL ETNK2 TRAV8-3 80 TRAV8-2 TRAV20-1

SPRY1 ADTRP DCHS1 SEMA6B ACVR1C SYDE2 KANK1 RNVU1-13 DTX1 TCEA3 IL4I1 DPP4 RGS16 PTPN13 GPRAP1 SCML1 20 CD7 RASGRF2 MAN1C1 DOCK3 CASS4 BEX4 DTX3 PI16 LTK MACAM PERP AKR1E2 MRC2 TUFT1 COL1A1 FAM153A AKAP7 SUSD4 0 RBMS3 GSTT1 HLA-DRB1 CD6 NELL2 DUSP8 ZBTB10 SMAD7 TNS1 OSBP2

American_ATL013 American_ATL030 American_ATL019 American_ATL025 American_ATL029 American_ATL022 American_ATL006 American_ATL018 American_ATL023 Japanese_ATL067 Japanese_ATL064 Japanese_ATL068 Japanese_ATL073 Japanese_ATL074 Japanese_ATL041 Japanese_ATL038 Japanese_ATL040 Japanese_ATL079 Japanese_ATL080 Japanese_ATL072 Japanese_ATL023 Japanese_ATL034 Japanese_ATL019 Japanese_ATL033 Japanese_ATL024 Japanese_ATL012 Japanese_ATL086 Japanese_ATL017 Japanese_ATL002 Japanese_ATL007 Japanese_ATL039 Japanese_ATL035 Japanese_ATL059 Japanese_ATL087 Japanese_ATL089 Japanese_ATL053 Japanese_ATL021 Japanese_ATL004 Japanese_ATL030 Japanese_ATL032 Japanese_ATL020 Japanese_ATL008 Japanese_ATL025 Japanese_ATL027 Japanese_ATL028 Japanese_ATL036 Japanese_ATL022 Japanese_ATL016 Japanese_ATL029 Japanese_ATL010 Japanese_ATL081 Japanese_ATL083 Japanese_ATL077 Japanese_ATL076 Japanese_ATL084 Japanese_ATL078 Japanese_ATL026 Japanese_ATL037 Japanese_ATL011 Japanese_ATL075 Japanese_ATL046 Japanese_ATL050 Japanese_ATL049 Japanese_ATL056 Japanese_ATL009 Japanese_ATL014 SPTB KLF8 PPFIBP1 SEMA3G DCAF12 ANK1 CD38 IL12RB2 ALPK2 ADCY1 ZNF365 MTOF PLD1 ADAM28 ARMC2 TCF4 ZNF462 CADM1 TIAM2 IGHM PDE8B GPR125 EVC FCRL3 LRRC32 CACNA1H NEFL ZDHHC11B ATL6 ATL13 ATL23 ATL19 ATL18 ATL29 ATL30 ATL25 CTRLCD4_1 CTRLCD4_2 CTRLCD4_3 CTRLCD4_4 chronic ATL22 chronic

CTRL NA ATLL

Figure 3.

1514 blood® 4 OCTOBER 2018 | VOLUME 132, NUMBER 14 SHAH et al From www.bloodjournal.org by guest on October 10, 2018. For personal use only.

Japanese ATLL samples to a DNMTI (decitabine). Among the This is the first North American cohort studied using a next- 6 Japanese-derived cell lines tested, the cell line with mutated generation sequencing approach. The JAK/STAT signaling path- EP300 (ATL43Tb2) was highly sensitive (50% inhibitory con- way is constitutively active in ATLL and has been documented in , m 19,20 21 centration [IC50] 1 M), showing a pronounced decrease in vitro and in vivo studies. IL-2 is an autocrine growth factor in viability after 5 days, whereas IC50 was not reached in the for activated T cells. In vitro, HTLV-1–infected cultures become IL-2 5 resistant samples with a WT EP300 status (Figure 4A,C). independent with time, and demonstrate constitutive activation of the JAK/STAT signaling pathway.19,20 As such, it is not surprising that Because there are no established cell lines for North American ATLL, 23.8% of the Japanese cases carry activating mutations in STAT3, we derived short- and long-term CD4 cultures from circulating JAK3,orJAK1. However, these mutations were rarely detected in ATLL cells of our patients and tested their decitabine sensitivity. North American ATLL (Figure 1D). This may be accounted for by Six patient-derived ATLL cultures with known EP300 mutational and the lower numbers of chronic/smoldering cases in our cohort, be- protein status were treated with decitabine in vitro. Two samples cause STAT3 mutations are preferentially detected among in- with EP300 mutations (ATL18 and ATL21) were found to be sensitive dolent cases in the Japanese cohort.22 The gene encoding to decitabine treatment, whereas 4 EP300 WT samples were resistant to decitabine treatment (Figure 4B-C). Annexin V/propidium CARD11, a cytoplasmic scaffolding protein required for T-cell – – k iodide–based staining confirmed that decitabine triggered apo- receptor and B-cell receptor mediated activation of the NF- B 23 ptosis in the sensitive samples (Figure 4F). signaling pathway, was also mutated much more frequently in Japanese patients compared with North American cases (Figure Because the majority of North American ATLL cases present 1D). It is possible that these differences in mutational profile re- in the aggressive stages2,3,7 and eventually require cytotoxic flect different pathogenic mechanisms, including the relative im- chemotherapy, we also tested the efficacy of combining deci- portance of dysregulated epigenetic vs cell signaling programs. tabine and doxorubicin, a major component of the combination chemotherapy for lymphoid malignancies. In Japanese and We identified a high number of epigenetic mutations in our North American ATLL samples, this combination synergistically patients. Epigenetic mutations in TET2, DNMT3A, and IDH have reduced total cell viability (Figure 4D-E). Decitabine treat- been well described in T-cell lymphomas.24,25 Mutations in TET2 ment led to on-target decreases in total cytosine methylation, as (10/31, 32%) and MLL3 have also been described in ATLL.26 measured by sensitive mass spectrometry analysis (Figure 4G). EP300, a histone acetyltransferase, had a significantly higher rate RNA-seq analysis of 4 pairs of decitabine-treated North Amer- of mutation in North American ATLL patients. This gene has ican ATLL cells (ATL18, ATL21 ATL29, and ATL30) demonstrated been characterized as a tumor suppressor and plays an impor- 286 probes (red dots) that passed both thresholds (absolute tant role in cell proliferation and differentiation via transcriptional . , log 2-fold change 1.0 and adjusted P value .05) and regulation by chromatin remodeling.27 Mutations in EP300 have were upregulated following decitabine treatment (Figure 4H). been identified in diffuse large B cell lymphoma,28,29 epithelial Among the decitabine-induced genes was CDKN1A, a p53 cancers,27 and ATLL.15 In our patient cohort, 6 (20%) of the cases target gene that encodes the cell cycle inhibitor p21/WAF. Other carry inactivating mutations in EP300, the functional loss of which re-expressed genes mapped to important pathways, such as has been shown to be associated with a gain of repressive his- cellular development and growth, cell death and survival, cell tone marks and gene silencing.30 Additionally, 2 (6.7%) of our cycle, cancer, organismal injury and abnormalities, tissue development, and cell morphology (supplemental Table 13). cases harbor inactivating mutations in TET2. Such mutations are associated with increased DNA methylation and aberrant gene silencing in myeloid leukemias.31 Recently, it was also reported Discussion that p300-dependent acetylation stabilizes TET2,which,inturn, 32 North American patients have a more aggressive clinical course inhibits cancer-associated CpG island hypermethylation. If – – and present at a younger age compared with Japanese this p300 Tet2 CpG island methylation connection functions patients.3,8 The differences in aggressiveness, despite the same in ATLL as well, TET2 inactivation and reduced p300 activity HTLV-1 serotype, may be accounted for by different genetic al- (due to p300 mutation/underexpression) are expected to impair terations. Our study examines the mutational landscape and this regulatory pathway, leading to the CpG island hyper- transcriptomics of North American ATLL. We report that, com- methylation phenotype, a hallmark of ATLL.15 pared with Japanese ATLL patients, North American ATLL is characterized by a higher frequency of prognostic epigenetic TP53 mutations are known to be associated with worse prog- mutations and fewer T-cell receptor/NF-kB and JAK/STAT mu- noses; however, the prognostic impact of epigenetic mutations tations.15 We also characterized functional consequences of EP300 has not been described before for ATLL. In our cohort of mutations, the most common epigenetic alteration identified in 30 cases, mutations in TP53 and EP300 were mutually exclusive our patient cohort. Our results suggest that ATLLs with a mutated with 1 exception. Although TP53 mutations alone did not reach EP300 gene have compromised p53 activity, which could account, statistical significance due to cohort size (Figure 2B), cases carry- at least in part, for their therapy-resistant characteristics, and ing mutations in any of the epigenetic modifier genes had are hypersensitive to DNMTIs, such as decitabine. adverse survival relative to those with WT genes (Figure 2A).

Figure 3. Distinct transcriptomic features are seen in North American ATLL. (A) Unsupervised clustering analysis of RNA-seq proﬁles of healthy CD4 controls and ATLL samples. (B) Volcano plot shows aberrantly expressed genes in North American ATLL compared with healthy controls. (C) Supervised clustering analysis reveals genes that are aberrantly expressed in North American ATLL. (D) Functional pathways enriched in differentially expressed genes. (E) Unsupervised clustering based on RNA-seq proﬁles shows transcriptomic differences between Japanese (blue, n 5 57) and North American (orange, n 5 9) ATLL samples.

MUTATIONAL LANDSCAPE OF NORTH AMERICAN ATLL blood® 4 OCTOBER 2018 | VOLUME 132, NUMBER 14 1515 From www.bloodjournal.org by guest on October 10, 2018. For personal use only.

A B C North American ATLL Cell lines (o) & Decitabine Short term cultures () Sample TP53 EP300 Japanese ATLL: Cell Lines IC50 (uM) 110 110 ATL21 WT Mut 0.1 100 100 ATL18 WT Mut 0.2 90 90 ATL13 WT WT 1.2 80 80 ATL29 WT WT 3 70 70 ATL25 Mut WT 3.8 N Am Patient

60 60 derived cultures ATL30 WT WT NR 50 50 ATL43Tb(-) Mut Mut 0.5 40 40 AT43T(+) Mut WT NR 30 30 Relative Survival (%) Relative Relative survival (%) Relative Su9T01 Mut WT NR 20 20

Lines ED40515(-) WT WT NR 10 10 ED41214(-) WT WT NR 0 0 Japanese Cell AT55T(+) WT WT NR 0110 0.1 1 10 Log scale decitabine (M) Log scale decitabine (M) ATL43Tb(-) ED40515(-) ED41214(-) ATL13 ATL18 ATL21 Su9T01(-) AT43T+ AT55T+ ATL25 ATL29 ATL30 D E F ATL43Tb(-) 100 ATL18 120 140 80 120 100 100 80 60

80 60 40 60

40 (%) fraction Cell 40 20

Relative survival (%) Relative 20 Relative survival (%) Relative 20 0 0 0 0 500 1,000 1,500 2,000 2,500 3,000 0 500 1,000 1,500 2,000 2,500 3,000 DMSO DMSO Dox (nM) Dox (nM) DMSO DMSO Dox DeclC10+Dox (0.05 uM) Dox Dox + Dec lC10 (0.05 uM) decitabine decitabine decitabine decitabine DeclC20+Dox (0.10 uM) Dox + Dec lC20 (0.1 uM) Viable AV+ Viable AV+ ATL18 ATL43Tb(-) G H 60 12 286 • CDKN1A %5mdc Control %5mdc Treated 50 8 7 40 6 * 30 5 * * * * 4 * 20

3 (-log10(p value)) %5mdc by LC-MS 2 Signi ﬁ cance of difference 10 1 0 0 -4-2 02 4 Su9T01 ATL19 ATL18 Diff in mean expression ATL43Tb(-) ED40515(-) ED41214(-)

Underexpressed Re-expressed after Decitabine Tx after Decitabine Tx

Figure 4. Decitabine has cytotoxic activity in ATLL. Results from cell-viability assays following 5 days of decitabine treatment in Japanese (A) and North American (B) ATLL samples. (C) Summary of decitabine IC50 doses for all samples tested with EP300 and TP53 mutation status. (D-E) Doxorubicin and decitabine combination showed synergistic efﬁcacy in 1 ATLL cell line and 1 North American ATLL culture. (F) Apoptosis measured by Annexin V/propidium iodide staining in decitabine-treated samples. Results shown are representative of 2 (ATL43Tb2) and 3 (ATL18) independent experiments. (G) Reduction in total 5-methylcytosine content after 48 hours of decitabine treatment at 0.6 mM was quantiﬁed by mass spectrometry. *P , .05. (H) Volcano plot. RNA-seq analysis revealed many re-expressed genes in 4 primary samples (ATL18, ATL21, ATL29, ATL30) following decitabine treatment.

Our observation of reduced p300 and acetylated p53 in EP300- mutations is our RNA-seq analysis, which shows enrichment of mutated primary samples (Figure 2C) is consistent with a pub- a mutated p53 transcriptional signature in the EP300-mutated lished study in diffuse large B-cell lymphoma, in which mutated samples (Figure 2D). EP300/CREBBP was found to cause defective p53 acetylation,33 as well as with the fact that acetylated p53 is more active as a Prior reports have shown that belinostat, a histone deacetylase transcription activator that can upregulate its own expression.17 inhibitor, along with azidothymidine, can trigger apoptosis of Other evidence supporting the biological signiﬁcance of EP300 ATLL cell lines derived from North American patients.34 Previous

1516 blood® 4 OCTOBER 2018 | VOLUME 132, NUMBER 14 SHAH et al From www.bloodjournal.org by guest on October 10, 2018. For personal use only.

studies have shown that ATLL has a polycomb repressive Xenotransplantation Facility, which is supported by a grant from the complex 2–mediated trimethylation at histone H3Lys27 in about New York State Department of Health (NYSTEM Program) for the shared facility (C029154). The authors also acknowledge the Einstein Epi- half of the genes in ATLL cells; these global alterations involved genomics Core for assistance with RNA-seq analysis. The authors thank fi ATLL-speci c gene-expression changes that included several Xiaoxin Ren for technical assistance with isolating CD4 cells from control tumor suppressors, transcription factors, epigenetic modifiers, samples, as well as our ATLL patients and their families without whose microRNAs, and developmental genes.35 Combined with these support and cooperation this work would not have been possible. previous reports, our findings strengthen the notion that the dysregulated epigenetic program plays an important role in ATLL This work was supported by Leukemia & Lymphoma Society Translational Research grant 6471-15 (B.H.Y.) and AIDS Malignancy Consortium grant pathogenesis, as well as the therapeutic response. UM1CA121947 (M.J.). E.Y.C. is supported by the Harry Eagle Scholarship from the Department of Cell Biology, Albert Einstein College of Medi- Aberrant DNA hypermethylation is associated with inappropri- cine. U.A.S. is supported by the hematology/oncology fellowship pro- ate transcriptional silencing of many important genes, including gram, Montefiore Medical Center and Jacobi Medical Center. tumor suppressors; thus, it leads to unregulated cell growth and development of malignancies.36 It has been implicated in the pathogenesis of ATLL,15 as well as many other malignancies.37 Authorship Hypermethylation of numerous genes has been described in the Contribution: U.A.S. designed the study, collected and analyzed data, performed the experiments, designed the figures, and wrote the man- pathogenesis of ATLL. PDLIM2 (a potent suppressor of Tax),38 uscript. E.Y.C., S.G.-M., O.G., T.D.B., Y.M., Y. Wei, E.I., G.S.C., N.G., C.P., 39,40 fi CDKN2A (a cell cycle regulatory gene), C2H2 zinc- nger and M.B. performed experiments. K.P. performed biostatistical and genes and MHC class I genes,15 BMP-6 (a regulator of cell RNA-seq analysis. K.K. and S.O. collaborated and shared the Japanese growth),41 SHP1, HCAD,andDAPK,42 and APC43 have been targeted gene and RNA-seq data. R.R. and Y. Wang diagnosed found to be hypermethylated in ATLL, with increasing incidence and collected patient samples. S.G., I.M., A.S., O.D., A.B., K.G., N.K., I.B., A.V., M.J., and U.A.S. diagnosed and treated ATLL patients and as the disease progresses. Because most chemotherapeutic collected samples. C.B., A.G., and J.H. assisted in gene sequencing agents depend on the same apoptosis and differentiation for ATLL patients. A.J. and L.R. performed proviral load assays. A.V., pathways of cells for their action, hypermethylation and in- B.H.Y., and M.J. designed the study, analyzed data, designed the fi activation of tumor suppressors are often associated with gures, and wrote the manuscript. marked chemoresistance, resulting in failure of chemotherapy.36 Conflict-of-interest disclosure: The authors declare no competing fi- nancial interests. Chemosensitization with DNMTIs has been successfully dem- 36 onstrated in patients with diffuse large B cell lymphoma, ORCID profiles: U.A.S., 0000-0001-8419-1091; T.D.B., 0000-0002-4527- ovarian cancer,44 and breast cancer.45 It was noted that deme- 5505; A.J., 0000-0001-7371-1696; M.B., 0000-0002-0908-770X; B.H.Y., thylation, followed by reprogramming at cancer-critical loci, is 0000-0002-2281-422X; M.J., 0000-0001-9323-8007. key to chemosensitization. In this regard, we also observed a significant synergy between doxorubicin and decitabine in Correspondence: Amit Verma, Albert Einstein College of Medicine, 1300 Morris Park Ave, Chanin Building, Room 302B, Bronx, NY 10461; North American ATLL cells. Of note, decitabine used as single e-mail: [email protected]; B. Hilda Ye, Albert Einstein Col- agent caused significant cytotoxicity, as well as demethylation, lege of Medicine, 1300 Morris Park Ave, Chanin Building, Room 302C, which is infrequently seen in acute myeloid leukemia. Most Bronx, NY 10461; e-mail: [email protected]; Murali Janakiram, importantly, the close correlation between EP300 mutation status 111 E 210 ST, Hoffheimer 100, Bronx, NY, 10467; e-mail: mjanakir@ montefiore.org. and decitabine sensitivity suggests that decitabine treatment induced a synthetic lethal phenotype in EP300-mutated cases. Taken together, our findings show that North American ATLL has a distinct set of genetic alterations that may cause the Footnotes Submitted 1 January 2018; accepted 25 July 2018. Prepublished online chemorefractoriness seen in these patients. This study also pro- as Blood First Edition paper, 13 August 2018; DOI 10.1182/blood-2018- vides a preclinical rationale for the use of epigenetic therapies in 01-824607. the treatment of North American ATLL with epigenetic alterations. Part of the exon sequencing data was presented in poster form at the 58th annual meeting of the American Society of Hematology, San Diego, Acknowledgments CA, 3 December 2016. Some of the data were presented in poster form at the 59th annual meeting of the American Society of Hematology, The authors thank Thomas Waldmann and Michael Petrus (National In- Atlanta, GA, 9 December 2017. stitutes of Health/National Cancer Institute, Bethesda, MD), Naomichi Arima (Kagoshima University, Kagoshima, Japan) for the gift of the Japanese ATLL cell lines, and Naomichi Arima for valuable advice on The online version of this article contains a data supplement. establishing long-term ATLL cultures. Flow cytometry analyses were performed at the Albert Einstein College of Medicine Flow Cytometry The publication costs of this article were defrayed in part by page facility, which is supported by National Institutes of Health National charge payment. Therefore, and solely to indicate this fact, this article is Cancer Institute Cancer Center Support grant P30CA013330. FACS hereby marked “advertisement” in accordance with 18 USC section sorting was performed at the Einstein Human Stem Cell FACS and 1734.

REFERENCES outcome of HTLV-1 related adult T-cell is a distinct clinical entity with dismal response leukemia-lymphoma in a prevalent area in to conventional chemotherapy. Oncotarget. 1. Gessain A, Cassar O. Epidemiological aspects the United States. Blood. 2017;130(suppl 1): 2016;7(32):51981-51990. and world distribution of HTLV-1 infection. 373. Front Microbiol. 2012;3:388. 4. Oliveira PD, de Carvalho RF, Bittencourt AL. 2. Malpica Castillo LEPA, Reis IM, Gotuzzo E, 3. Zell M, Assal A, Derman O, et al. Adult T-cell Adult T-cell leukemia/lymphoma in South et al. Epidemiology, clinical features, and leukemia/lymphoma in the Caribbean cohort and Central America and the Caribbean:

MUTATIONAL LANDSCAPE OF NORTH AMERICAN ATLL blood® 4 OCTOBER 2018 | VOLUME 132, NUMBER 14 1517 From www.bloodjournal.org by guest on October 10, 2018. For personal use only.

systematic search and review. Int J STD AIDS. 19. Migone TS, Lin JX, Cereseto A, et al. of acetyltransferase genes in B-cell lymphoma. 2017;28(3):217-228. Constitutively activated Jak-STAT pathway in Nature. 2011;471(7337):189-195. T cells transformed with HTLV-I. Science. 5. Bitar N, Hajj HE, Houmani Z, et al. Adult T-cell 1995;269(5220):79-81. 34. Toomey NL RJ. Histone deacetylase inhibitors leukemia/lymphoma in the Middle East: abolish HBZ protein expression and induce first report of two cases from Lebanon. 20. Xu X, Kang SH, Heidenreich O, Okerholm M, cell death in HTLV-1 related adult T-cell leu- Transfusion. 2009;49(9):1859-1864. O’Shea JJ, Nerenberg MI. Constitutive acti- kemia-lymphoma. Blood. 2017;130 (suppl 1): vation of different Jak tyrosine kinases in hu- 4012. 6. Licata MJ, Janakiram M, Tan S, et al. man T cell leukemia virus type 1 (HTLV-1) tax Diagnostic challenges of adult T-cell protein or virus-transformed cells. J Clin 35. Fujikawa D, Nakagawa S, Hori M, et al. – leukemia/lymphoma in North America a Invest. 1995;96(3):1548-1555. Polycomb-dependent epigenetic landscape clinical, histological, and immunophenotypic in adult T-cell leukemia. Blood. 2016;127(14): fl correlation with a work ow proposal. Leuk 21. Takemoto S, Mulloy JC, Cereseto A, et al. 1790-1802. Lymphoma. 2018;59(5):1188-1194. Proliferation of adult T cell leukemia/ lymphoma cells is associated with the con- 36. Clozel T, Yang S, Elstrom RL, et al. 7. Phillips AA, Shapira I, Willim RD, et al. A critical stitutive activation of JAK/STAT proteins. Proc Mechanism-based epigenetic chemo- analysis of prognostic factors in North Amer- Natl Acad Sci USA. 1997;94(25):13897-13902. sensitization therapy of diffuse large B-cell ican patients with human T-cell lymphotropic lymphoma. Cancer Discov. 2013;3(9): virus type-1-associated adult T-cell leukemia/ 22. Kataoka K, Iwanaga M, Yasunaga JI, et al. 1002-1019. lymphoma: a multicenter clinicopathologic Prognostic relevance of integrated genetic experience and new prognostic score. Cancer. profiling in adult T-cell leukemia/lymphoma. 37. van Doorn R, Zoutman WH, Dijkman R, et al. 2010;116(14):3438-3446. Blood. 2018;131(2):215-225. Epigenetic profiling of cutaneous T-cell lymphoma: promoter hypermethylation of 8. Katsuya H, Ishitsuka K, Utsunomiya A, et al; 23. Jiang C, Lin X. Regulation of NF-kBbythe multiple tumor suppressor genes including ATL-Prognostic Index Project. Treatment and CARD proteins. Immunol Rev. 2012;246(1): BCL7a, PTPRG, and p73. J Clin Oncol. 2005; survival among 1594 patients with ATL. Blood. 141-153. 23(17):3886-3896. 2015;126(24):2570-2577. 24. Couronne´ L, Bastard C, Bernard OA. TET2 and 38. Yan P, Fu J, Qu Z, et al. PDLIM2 suppresses 9. Adams SV, Newcomb PA, Shustov AR. Racial DNMT3A mutations in human T-cell lym- patterns of peripheral T-cell lymphoma in- phoma. N Engl J Med. 2012;366(1):95-96. human T-cell leukemia virus type I Tax- mediated tumorigenesis by targeting Tax into cidence and survival in the United States. J Clin Oncol. 2016;34(9):963-971. 25. Palomero T, Couronne´ L, Khiabanian H, et al. the nuclear matrix for proteasomal degrada- Recurrent mutations in epigenetic regulators, tion. Blood. 2009;113(18):4370-4380. 10. Chihara D, Ito H, Katanoda K, et al. Increase in RHOA and FYN kinase in peripheral T cell incidence of adult T-cell leukemia/lymphoma lymphomas. Nat Genet. 2014;46(2):166-170. 39. Nosaka K, Maeda M, Tamiya S, Sakai T, in non-endemic areas of Japan and the United Mitsuya H, Matsuoka M. Increasing methyl- States. Cancer Sci. 2012;103(10):1857-1860. 26. Yeh CH, Bai XT, Moles R, et al. Mutation of ation of the CDKN2A gene is associated with epigenetic regulators TET2 and MLL3 in pa- the progression of adult T-cell leukemia. 11. Shimoyama M. Diagnostic criteria and classi- tients with HTLV-I-induced acute adult T-cell Cancer Res. 2000;60(4):1043-1048. fication of clinical subtypes of adult T-cell leukemia [published correction appears leukaemia-lymphoma. A report from the in Mol Cancer. 2016;15:20]. Mol Cancer. 40. Endo M, Kobayashi C, Setsu N, et al. Lymphoma Study Group (1984-87). Br J 2016;15(1):15. Prognostic significance of p14ARF, Haematol. 1991;79(3):428-437. p15INK4b, and p16INK4a inactivation in 27. Gayther SA, Batley SJ, Linger L, et al. malignant peripheral nerve sheath tumors. 12. Swerdlow SH, Campo E, Pileri SA, et al. The Mutations truncating the EP300 acetylase Clin Cancer Res. 2011;17(11):3771-3782. 2016 revision of the World Health Organiza- in human cancers. Nat Genet. 2000;24(3): tion classification of lymphoid neoplasms. 300-303. 41. Taniguchi A, Nemoto Y, Yokoyama A, et al. Blood. 2016;127(20):2375-2390. Promoter methylation of the bone morpho- 28. Andersen CL, Asmar F, Klausen T, Hasselbalch genetic protein-6 gene in association with 13. Morris LG, Kaufman AM, Gong Y, et al. H, Grønbæk K. Somatic mutations of the adult T-cell leukemia. Int J Cancer. 2008; Recurrent somatic mutation of FAT1 in mul- CREBBP and EP300 genes affect response to 123(8):1824-1831. tiple human cancers leads to aberrant Wnt histone deacetylase inhibition in malignant activation. Nat Genet. 2013;45(3):253-261. DLBCL clones. Leuk Res Rep. 2012;2(1):1-3. 42. Sato H, Oka T, Shinnou Y, et al. Multi-step aberrant CpG island hyper-methylation is as- 14. Neumann M, Seehawer M, Schlee C, et al. 29. Morin RD, Mendez-Lago M, Mungall AJ, et al. sociated with the progression of adult T-cell FAT1 expression and mutations in adult acute Frequent mutation of histone-modifying leukemia/lymphoma. Am J Pathol. 2010; lymphoblastic leukemia. Blood Cancer J. genes in non-Hodgkin lymphoma. Nature. 176(1):402-415. 2014;4(6):e224. 2011;476(7360):298-303. 43. Yang Y, Takeuchi S, Tsukasaki K, et al. 15. Kataoka K, Nagata Y, Kitanaka A, et al. 30. Lunning MA, Green MR. Mutation of chro- Methylation analysis of the adenomatous Integrated molecular analysis of adult T cell matin modifiers; an emerging hallmark of polyposis coli (APC) gene in adult T-cell leu- leukemia/lymphoma. Nat Genet. 2015;47(11): germinal center B-cell lymphomas. Blood kemia/lymphoma. Leuk Res. 2005;29(1): 1304-1315. Cancer J. 2015;5(10):e361. 47-51. 16. Dutta R, Tiu B, Sakamoto KM. CBP/p300 31. Ko M, Huang Y, Jankowska AM, et al. Impaired acetyltransferase activity in hematologic ma- hydroxylation of 5-methylcytosine in myeloid 44. Suh DH, Kim MK, Kim HS, Chung HH, Song YS. Epigenetic therapies as a promising lignancies. Mol Genet Metab. 2016;119(1-2): cancers with mutant TET2. Nature. 2010; strategy for overcoming chemoresistance in 37-43. 468(7325):839-843. epithelial ovarian cancer. J Cancer Prev. 2013; 17. Grossman SR. p300/CBP/p53 interaction 32. Zhang YW, Wang Z, Xie W, et al. Acetylation 18(3):227-234. and regulation of the p53 response. Eur enhances TET2 function in protecting against J Biochem. 2001;268(10):2773-2778. abnormal DNA methylation during oxidative 45. Gordian E, Ramachandran K, Singal R. stress. Mol Cell. 2017;65(2):323-335. Methylation mediated silencing of TMS1 in 18. Tang Y, Zhao W, Chen Y, Zhao Y, Gu W. breast cancer and its potential contribution to Acetylation is indispensable for p53 activation. 33. Pasqualucci L, Dominguez-Sola D, docetaxel cytotoxicity. Anticancer Res. 2009; Cell. 2008;133(4):612-626. Chiarenza A, et al. Inactivating mutations 29(8):3207-3210.

1518 blood® 4 OCTOBER 2018 | VOLUME 132, NUMBER 14 SHAH et al From www.bloodjournal.org by guest on October 10, 2018. For personal use only.

2018 132: 1507-1518 doi:10.1182/blood-2018-01-824607 originally published online August 13, 2018

North American ATLL has a distinct mutational and transcriptional profile and responds to epigenetic therapies

Urvi A. Shah, Elaine Y. Chung, Orsi Giricz, Kith Pradhan, Keisuke Kataoka, Shanisha Gordon-Mitchell, Tushar D. Bhagat, Yun Mai, Yongqiang Wei, Elise Ishida, Gaurav S. Choudhary, Ancy Joseph, Ronald Rice, Nadege Gitego, Crystall Parrish, Matthias Bartenstein, Swati Goel, Ioannis Mantzaris, Aditi Shastri, Olga Derman, Adam Binder, Kira Gritsman, Noah Kornblum, Ira Braunschweig, Chirag Bhagat, Jeff Hall, Armin Graber, Lee Ratner, Yanhua Wang, Seishi Ogawa, Amit Verma, B. Hilda Ye and Murali Janakiram

Updated information and services can be found at: http://www.bloodjournal.org/content/132/14/1507.full.html Articles on similar topics can be found in the following Blood collections Lymphoid Neoplasia (2893 articles)

Information about reproducing this article in parts or in its entirety may be found online at: http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests

Information about ordering reprints may be found online at: http://www.bloodjournal.org/site/misc/rights.xhtml#reprints

Information about subscriptions and ASH membership may be found online at: http://www.bloodjournal.org/site/subscriptions/index.xhtml

Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036. Copyright 2011 by The American Society of Hematology; all rights reserved.