7/12/2016

Objectives New Directions in Aplastic Anemia: What's on the Horizon? • To provide information on new concepts emerging in AA in somatic molecular Amy E. DeZern, MD; MHS testing Assistant Professor Hematologic Malignancies Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins • To review updates on markers of Baltimore, MD response to therapy in AA

• To review newer data on emerging therapies in aplastic anemia (AA) 1

Better way to evaluate clonal evolution?

• As survival is improving, we may see more late development of MDS/AML in ≥15% of patients clonal evolution • Extrapolation from MDS field somatic mutational testing TESTING • Goal: to use this testing to ID group of AA pts more likely to get MDS (treatment would be modified)

July 12, 2016 3 Kulasekararaj et al. Blood 2014 4 Yoshizato et al. NEJM 2015

Genes Recurrently Mutated in MDS The Dawn of the Molecular Era of

the Myelodysplastic Syndromes Tyrosine Kinase Pathway Transcription Factors Others • 51% of patients ≥1 mutation (including ~50% with JAK2 KRAS BRAF normal karyotype) RUNX1 GATA2 TP53 NPM1 ETV6 NRAS NOTCH? RTKs MAML? PHF6 • FIVE (in ~30% of pts)  independent PTPN11 CBL WT1 ZSWIM4? BCOR UMODL1? prognostic significance and predicted poor overall Epigenetic Dysregulation Splicing Factors survival ZRSF2 IDH EZH2 U2AF1 – EZH2 ETV6 RUNX1 ASXL1 P53 1 & 2 DNMT3A SF3B1

SETBP1 PRPF40B U2AF2 TET2 UTX ASXL1 SRSF2 PRPF8 SF1 ATRX SF3A1 Bejar et al NEJM Courtesy of Bejar R. 2011

1 7/12/2016

Retrospective from the UK

• From UK: looked at 150 pts with no evidence (by microscope) of MDS Candidate mutations in acquired • Screen for mutations aplastic anemia - correlation with survival – ASXL1, DNMT3A, BCOR, TET2, MPL at least and clonal evolution to myelodysplastic syndrome – 19% AA pts had mutations -courtesy of Dr. Dumitriu – Pts with mutations had a longer disease Yoshizato et al. NEJM 2015 duration (37 vs. 8 months, p<0.04) – AA patients with disease duration of > 6 mos AND mutation40% risk of transformation to MDS (p <0.0002) 7 Kulasekararaj et al. Blood 2014

NIH SAA cohort ACQUIRED MUTATIONS IN SAA – NIH COHORT (n=256)

100% 5% Number mutations “Stable AA” – responders or 7% per SAA patient non-responders to 75% 23% 0 1 2 >2 SAA immunosuppressive therapy

≈ 15% clonal evolution to MDS/AML 50% Targeted gene panel

Transcription Signal DNA Ribosomal Epigenesis Others Receptor factor transduction repair 65% AEBP2 ATM BRAF CDAN1 FANCA FLT3 RPL11 ASXL1 BCOR CSMD1 CDH23 FANCC KIT RPL35a De novo 25% ATRX BCORL1 GNAS DIS3 FANCD1 GCSFR RPL5 DNMT3A CEBPA GPRC5A DKC1 FANCD2 MPL RPS10 EED CTCF JAK1 ELANE FANCE RPS19 MDS/AML EZH1 CUX1 JAK2 ETNK1 FANCG RPS26 EZH2 DAXX JAK3 LAMB4 IDH1 Dido1 KRAS LUC7L2 Telomere Splicing Cohesin IDH2 ETV6 LNK PEG3 NHP2 DDX41 RAD21 0% JARID2 GATA1 MAP3K4 PIGA NOP10 PRPF8 SMC1A KDM6A GATA2 NF1 PRF1 POT1 SF3B1 SMC3 Time RBBP4 IRF1 NRAS SBDS RAP1 SRSF2 STAG2 RBBP7 NCOR2 PTPN11 SETBP1 RTEL1 U2AF1 SUZ12 NPM1 RIT1 UMODL1 TERC U2AF2 Ubiquitin 6 months after IST TET2 PHF6 SRP72 ZSWIM4 TERF1 ZRSR2 BRCC3 Clonal evolution Rb STAT3 TERF2 CBL RUNX1 TERT FBXW7 TP53 TINF2 N = 256 patients (3 institutions) WT1 TPP1 Yoshizato et al. NEJM 2015

ACQUIRED MUTATIONS IN SAA – NIH COHORT (n=256) ACQUIRED MUTATIONS IN SAA – STRONG AGE BIAS

12.0% 8.0% 4.0% 0.0% PIGA PIGA BCOR/BCORL1 BCOR/BCORL1 DNMT3A77% DNMT3A ASXL1 % of patients Splicing ASXL1 cohesin JAKs RUNX1 TP53 TET2 SETBP1 GNAS PRC2 CSMD1 LAMB4 WT1 TERF1/TERT multiple ATRX PHF6 missense ATM nonsense RIT1 CDAN1 frameshift IDH2 CUX1 splicing RBBP4 CBL PRPF8 KRAS MPL NF1 POT1 RAP1A STAT3 PEG3 DIS3 SH2B3

2 7/12/2016

ACQUIRED MUTATIONS DO NOT CORRELATE WITH CYTOPENIAS, ACQUIRED MUTATIONS DO NOT CORRELATE WITH PRESENCE OF PNH CLONE, OR RESPONSE TO IST REFRACTORY DISEASE

Odds ratio (95% CI) p value Odds ratio (95% CI) p value ARC<25k/uL 1.09 (0.59, 2.07) 0.885 Refractory to 2 rounds of IST 1.45 (0.49, 4.31) 0.498 Refractory to 2nd IST given for ANC<200/uL 1.46 (0.79, 2.68) 0.185 0.85 (0.22, 3.15) 0.805 relapse PNH+ 1.46 (0.80, 2.64) 0.202 HSCT (for relapse/refractory/clonal 0.51 (0.23, 1.10) 0.08 Response to IST 1.42 (0.78, 2.62) 0.256 evolution)

MUTATIONS IN BCOR/BCOL1 CORRELATE WITH BETTER ACQUIRED MUTATIONS CORRELATE WITH PROGRESSION TO MDS RESPONSE TO to IST

1.0 Odds ratio (95% CI) p value 0.8 BCOR/BCORL1 2.73 (1.05, 7.11) 0.038 0.6

DNMT3A 0.89 (0.31, 2.55) 0.835 OS 0.4

ASXL1 0.68 (0.25, 1.83) 0.447 0.2 0.0

None of these mutations correlate with clonal evolution/refractory disease

ACQUIRED MUTATIONS CORRELATE WITH OVERALL SURVIVAL CONCLUSIONS

• Somatic mutations in 35% of cases 1.0

0.8 • PIGA, DNMT3A, ASXL1, and BCOR/BCORL1 were most frequently mutated genes 0.6

OS 0.4 • Presence and number of mutations correlated with age only

0.2 • BCOR/BCORL1 mutated clones associated with a better chance of 0.0 responding to IST

• Response to IST, clonal evolution and overall survival are associated with acquired mutations

• Mutations remain independent predictors of clinical outcomes in multivariate analysis

3 7/12/2016

Telomeres

• Telomeres: regions of repetitive nucleotides at the ends of that are there to protect the chromosomes from damage and breakdown. • Telomere length testing very helpful in inherited AA DKC (very short) • Reports suggest that telomeres are shorter (not very short) in up to one-third of patients with acquired SAA • At NIH, telomere lengths measured in the white blood cells of 183 patients treated with IST • Shorter telomeres not found to predict who would have improved blood counts at 6 months after IST • Shorter lengths may be associated with late effects such as relapse or clonal evolution to MDS

July 12, 2016 Scheinberg et al. JAMA, 304 (2010), 1358-64. DeZern and Armanios (unpublished)

PNH clones and telomeres predict response to IST in Pediatric AA • Prospective study of 113 children (Ages 0-16) • All had PNH clones and telomeres done in CLIA certified way before IST • Response assessed based on PNH clone presence and telomere length – If PNH clone + and telomeres “long”- ~70% response – If PNH clone neg and telomeres “short”- ~19% response TREATMENTS • Suggestion for upfront HSCT if in this group

Narita et al. Haematolgica 2015 21 July 12, 2016 22

Severe Aplastic Anemia 2015 Treatment Paradigm Alternative Donor transplants in Diagnosis of Severe Treatment based upon age and Aplastic Anemia donor availability SAA

Age < 40 years with Age > 40 years OR HLA matched sibling No HLA matched sib • Reserve for relapsed or refractory SAA – After failing ATG

Bone Marrow Immunosuppressive • Should be done in a “specialist center with Transplantation Therapy major experience in hematopoietic SCT procedures” Response No Response Rate ~70% ~30% • Perform in setting of clinic trial “designed ~40% Relapse or Clinical Clonal Evolution TRANSPLANT specifically to address the prevention of follow up graft rejection and GVHD”

Ciceri et al BMT 2013

4 7/12/2016

Unrelated Donor or Alternative donor New Trial BMT

• The sooner, the better after relapse/ refractoriness noted

• Bone marrow still the best source

• Similar conditioning regimens

• Often in setting of clinical trial

• Unrelated donor (NMDP- “MUDs”)

• Haplo-identical donor (half match)

• Umbilical cord donor

July 12, 2016 26

Hopkins Haplo BMT for Acquired and Primary Objective inherited SAA Day of Degree of engraftm Donor Age/ Pre-BMT Follow- HLA ent chimerism aGVHD cGHVD Response Sex59 Therapy Up (mos) match (ANC at day 100 >1000) 35M ATG/ CsA 5/10 30 100 none none CR 50 Assess overall survival in 2 cohorts 52M ATG/ CsA 5/10 24 100 none none CR 27 (unrelated cord blood and haplo- 45F ATG/ CsA 5/10 27 100 Gr 2 skin Gr 2 skin CR 22 27F ATG 5/10 24 100 none none CR 15 identical) at 1 year post-hematopoietic 33M HiCY 5/10 16 100 none none CR 17 stem cell transplantation in patients with 17M HiCY 5/10 17 100 none none CR 9 severe aplastic anemia 54M ATG/ CsA 5/10 15 100 none None CR 9 ATG/ 26F CsA/TPO 5/10 17 100 none None CR 6 ATG/ 100 (Day 59 CsA/TPO 5/10 24 60 none None CR 3 Hypothesis—both cohorts >75% 1yr OS 21 Steroids 5/10 18 100 none none CR 33 38 Danazol 9/10 17 100 none Gr 2 skin CR 20 27 28

Mini-BMT for Refractory SAA Adult URD Treatment schema

• Survival now >75% • Similar conditioning regimens- may add TBI

Bacigalupo and Marsh BMT 2013

5 7/12/2016

Kaplan-Meier survival rate estimates (and 95% Umbilicord Transplants confidence intervals) for recipients of an HLA-matched sibling or unrelated donor HCT for SAA registered with CIBMTR from 1990 through 2011, by donor type, age group, and graft source • Survival is still lower ~40% HLA-Matched Sibling Donor Unrelated Donor Age < 40y Age ≥ 40y Age < 40y Age ≥ 40y Survival rate (95% CI), Graft Source: BM and PB 1 year 84 (82 – 85) 68 (64 – 72) -- -- 3 years 80 (79 – 82) 61 (57 – 65) -- -- • Conditing regimens more varied 5 years 80 (78 – 81) 57 (52 – 62) -- -- 10 years 77 (75 – 79) 50 (45 – 56) -- -- Survival rate (95% CI), Graft Source: CB – CY/Flu/ TBI 1 year 79 (56 – 94) -- 52 (44 – 59) Not estimated* 3 years 72 (48 – 90) -- 44 (36 – 51) Not estimated* – Melphalan/Flu/ TBI 5 years 72 (48 – 90) -- 40 (32 – 49) Not estimated* * CIBMTR requires a sample size of at least 20 for KM survival-rate estimations Yamamoto et al Blood 2011 Peffault et all BMT 2013 • Japanese study Yoshimi et al BBMT 2008 – 12 adult patients, 11/12 engrafted, survival The data presented here are preliminary and were obtained from the Statistical Center of the Center for International Blood and Marrow Transplant Research. The analysis has not been reviewed or approved by the Advisory or Scientific Committee of the CIBMTR.’ 10/12 median 36 mos

URD in children 2nd HSCT in AA

• 44 children in UK (40 s/p IST) • Retrospective of 162 European pts • Fludarabine, CY, Alemtuzumab – 1998-2009: 1°or 2°in 14% of pts with 1 txp for AA conditioning • Used same donor 81% of time • Overall survival and failure free survival • Changed from BM to PBSC in 56% was 95% – Excluded 2nd txp using cord blood • Graft failure still occurred 26% of 2nd txp • Follow up 3.5 years (median) 60% OS

Samarasinghe et al BJH 2012 Cesaro et al BJH 2015 34

ASH 2014: Transplants Androgens: Danazol

• Abstract 256 Dufour et al • Suggestion in literature that androgens may slow • Similar Outcome of Upfront Unrelated and Matched telomere shortening in some pts Sibling Donor Hematopoietic Stem Cell Transplantation in • Phase I/II at NIH in 27 pts (20 with mod AA) to take Idiopathic Aplastic Anaemia danazol daily for 2 years of Childhood and Adolescence • Increased telomere lengths and hematologic • 29 SAA children had unrelated improvement seen donor HSCT without prior IST in UK • No significant liver toxicity • Compares each with 3 matched • At publication 9 pts stopped drug prior to 2 years controls from the database of the SAAWP of the EBMT with MSD HSCT

July 12, 2016 35 Dumitriu et al. Blood 36 2014

6 7/12/2016

Iron Overload Iron Chelation

• Iron burden increased in • Deferasirox (Exjade oral) or patients with AA at Desferoxamine (IV) presentation and over – To discuss with doctor as time with transfusions meds have liver and – Follow ferritin (lab kidney side effects and measure of body’s iron can lower stores) • Recent study of 53 AA pts • Iron overload in other showed deferasirox marrow failure (MDS) improves ferritin and liver studied and problematic iron overload when taken – Deposition in liver daily and heart

Cheong et al Transfusion Jin et al Int J Hem 2015 37 38 2015

THANK YOU!

July 12, 2016 39

7