Not for publication or presentation

A G E N D A CIBMTR WORKING COMMITTEE FOR IMMUNOBIOLOGY Orlando, Florida Wednesday, February 22, 2017, 12:15 pm– 4:45 pm

Co-Chair: Michael Verneris, MD, University of Colorado – Denver; Telephone: 303-724-4006; E-mail: [email protected] Co-Chair: Katharina Fleischhauer, MD; Essen University Hospital; Telephone: +49-201-723-4582; E-mail: [email protected] Co-Chair: Katharine Hsu, MD, PhD; Memorial Sloan-Kettering Cancer Center; Telephone: 646-888-2667; E-mail: [email protected] Co-Scientific Dir: Stephanie Lee, MD, MPH, Fred Hutchinson Cancer Research Center Telephone: 206-667-5160; E-mail: [email protected] Co-Scientific Dir: Stephen Spellman, MBS, CIBMTR Immunobiology Research Telephone: 763-406-8334; E-mail: [email protected] Statistical Director: Tao Wang, PhD, CIBMTR Statistical Center Telephone: 414-955-4339; E-mail: [email protected] Statisticians: Michael Haagenson, MS, CIBMTR Statistical Center Telephone: 763-406-8609; E-mail: [email protected]

1. Introduction (M Verneris) 12:15 pm a. Minutes and Overview Plan of Immunobiology Working Committee at 12:20 pm Tandem 2016 (Attachment 1)

2. Published or submitted papers 12:25 pm

a. IB08-06 Rocha V, Ruggeri A, Spellman S, Wang T, Sobecks R, Locatelli F, Askar M, Michel G, Arcese W, Iori AP, Purtill D, Danby R, Sanz GF, Gluckman E, Eapen M. Killer cell immunoglobulin-like receptor ligand matching and outcomes after unrelated cord blood transplantation acute myeloid leukemia. Biol Blood Marrow Transplant. 2016 Jul 1; 22(7):1284-1289. doi:10.1016/j.bbmt.2016.04.007. Epub 2016 Apr 16. PMC4915071.

b. IB08-08 Goyal RK, Lee SJ, Wang T, Trucco M, Haagenson M, Spellman SR, Veneris M, Ferrell RE. Novel HLA-DP region susceptibility loci associated with severe acute GvHD. Bone Marrow Transplant. doi:10.1038/bmt.2016.210. Epub 2016 Sep 5.

c. IB10-01b Gadalla SM, Khincha PP, Katki HA, Giri N, Wong JYY, Spellman S, Yanovski JA, Han JC, De Vivo, Alter BP, Savage SA. The limitations of qPCR telomere length measurement in diagnosing dyskeratosis congenita. Molecular Genetics & Genomic Medicine. 2016 Jul 1; 4(4):475-479. doi:10.1002/mgg3.220. Epub 2016 Mar 20. PMC4947866.

1 Not for publication or presentation

d. IB10-01b Gadalla SM, Wang T, Dagnall C, Haagenson M, Spellman SR, Hicks B, Jones K, Katki HA, Lee SJ, Savage SA. Effect of recipient age and stem cell source on the association between donor telomere length and survival after allogeneic unrelated hematopoietic cell transplantation for severe aplastic anemia. Biol Blood Marrow Transplant. 2016 Dec 1; 22(12):2276-2282. doi:10.1016/j.bbmt.2016.09.012. Epub 2016 Sep 15. PMC5116252.

e. IB10-04 Arora M, Lee SJ, Spellman SR, Weisdorf DJ, Guan W, Haagenson M, Wang T, Horowitz MH, Verneris MR, Fleischhauer K, Hsu K, Thyagarajan B. Validation study failed to confirm an association between genetic variants in the base excision repair pathway and transplant-related mortality and relapse after hematopoietic cell transplantation. Biol Blood Marrow Transplant. 2016 Aug 1; 22(8):1531-1532. doi:10.1016/j.bbmt.2016.04.020. Epub 2016 May 4.

f. IB12-04 Hoff GA, Fischer JC, Hsu K, Cooley S, Miller JS, Wang T, Haagenson M, Spellman S, Lee SJ, Uhrberg M, Venstrom JM, Verneris MR. Recipient HLA-C haplotypes and miRNA 148a/b binding sites have no impact on allogeneic hematopoietic cell transplantation outcomes. Biol Blood Marrow Transplant. doi:10.1016/ j.bbmt.2016.09.028. Epub 2016 Oct 13.

g. IB12-05/RT10-01 Kornblit B, Wang T, Lee SJ, Spellman SR, Zhu X, Fleischhauer K, Müller C, Johansen JS, Vindelov L, Garred P. YKL-40 in allogeneic hematopoietic cell transplantation after AML and myelodysplastic syndrome. Bone Marrow Transplant. doi:10.1038/bmt.2016.192. Epub 2016 Jul 18.

h. IB12-06 Bachanova V, Weisdorf DJ, Wang T, Marsh SGE, Trachtenberg E, Haagenson MD, Spellman SR, Ladner M, Guethlein LA, Parham P, Miller JS, Cooley SA. Donor KIR B genotype improves progression-free survival of non-Hodgkin lymphoma patients receiving unrelated donor transplantation. Biol Blood Marrow Transplant. 2016 Sep 1; 22(9):1602-1607. doi:10.1016/j.bbmt.2016.05.016. Epub 2016 May 21. PMC4981536.

i. IB13-02 Marino SR, Lee SM, Binkowski TA, Wang T, Haagenson M, Wang H-L, Maiers M, Spellman S, van Besien K, Lee SJ, Karrison T, Artz A. Identification of high-risk amino-acid substitutions in hematopoietic cell transplantation: a challenging task. Bone Marrow Transplant. 2016 Oct 1; 51(10):1342-1349. doi:10.1038/bmt.2016.142. Epub 2016 May 23. PMC5052106.

j. IB13-03 Lazaryan A, Wang T, Spellman SR, Wang H-L, Pidala J, Nishihori T, Askar M, Olsson R, Oudshoorn M, Abdel-Azim H, Yong A, Gandhi M, Dandoy C, Savani B, Hale G, Page K, Bitan M, Reshef R, Drobyski W, Marsh SGE, Schultz K, Müller CR, Fernandez-Viña M, Verneris MR, Horowitz MM, Arora M, Weisdorf DJ, Lee SJ. Human leukocyte antigen supertype matching after myeloablative hematopoietic cell transplantation with 7/8 matched unrelated donor allografts: A report from the Center for International Blood and Marrow Transplant Research. Haematologica. 2016 Oct 1; 101(10):1267-1274. doi:10.3324/haematol.2016.143271. Epub 2016 May 31. PMC5046657.

2 Not for publication or presentation

k. IB13-05 Askar M, Sobecks R, Wang T, Haagenson M, Majhail N, Madbouly A, Thomas D, Zhang A, Fleischhauer K, Hsu K, Verneris M, Lee SJ, Spellman SR and Fernández-Viña M. MHC Class I Chain-Related Gene A (MICA) Donor-Recipient Mismatches and MICA-129 Polymorphism in Unrelated Donor Hematopoietic Cell Transplants (HCT) For Hematological Malignancies: A CIBMTR Study. Biol Blood Marrow Transplant. 2016 Dec 14. pii: S1083-8791(16)30521-3. doi:10.1016/j.bbmt.2016.11.021.

l. IB14-03b Lindsley RC, Saber W, Mar B, Medd, Wang T, Haagenson M, Grauman P, Zhu Z, Spellman S, Lee SJ, Verneris M, Hsu K, Fleischhauer K, Cutler C, Antin JH, Neuberg D, Ebert BL. Prognostic Mutations in Myelodysplastic Syndrome After Stem Cell Transplantation. In press. New Engl J Med.

m. IB10-01c Gadalla SM, Ballew BJ, Haagenson M, Spellman S, Hicks B, Alter BP, Zhu B, Zhou W, Yeager M, Wang T, Fleischhauer K, Hsu K, Verneris M, Freedman N, Lee SJ, Savage SA. Germline Mutations in Marrow Failure Predisposition Genes in Patients Receiving Unrelated Donor Hematopoietic Cell Transplant for Severe Aplastic Anemia. Submitted. J Clin Invest.

n. IB12-02B Fleischhauer K, Ahn KW, Wang H-L, Zito L, Crivello P, Müller C, Verneris M, Shaw BE, Pidala J, Oudshorn M, Lee SJ and Spellman SR. Directionality of non-permissive HLA-DPB1 T-cell epitope group mismatches in 8/8 matched unrelated donor hematopoietic cell transplantation. Submitted. Leukemia.

o. IB12-03 Madbouly A, Wang T, Haagenson M, Paunic V, Vierra-Green C, Fleischhauer K, Hsu KC, Verneris MR, Majhail NS, Lee SJ, Spellman SR and Maiers Investigating the association of genetic admixture and donor/recipient genetic disparity with transplant outcomes In press. Biol Blood Marrow Transplant.

p. IB15-06a Gadalla SM, Wang T, Loftus D, Friedman L, Dagnall C, Haagenson M, Spellman SR, Buturovic L, Blauwkamp M, Shelton J, Fleischhauer K, Hsu KC, Verneris MR, Krstajic D, Hicks B, Jones K, Lee SJ, Savage SA. Donor telomere length and outcomes after allogeneic unrelated hematopoietic cell transplant in patients with acute leukemia. Submitted. J Clin Oncol.

3. Research repository update and accrual tables (S Spellman) (Attachment 2) 12:25 pm

4. Future/proposed studies and discussion 12:35 pm a. Voting guidelines (Hsu K) b. PROP1612-05 The impact of HLA-DPB1 levels of expression on clinical outcomes of HCT (M Askar/M Fernandez-Viña) – (Attachment 3) c. PROP1611-158 Identification of genomic markers of post hematopoietic cell transplantation (HCT) outcomes in patients with myelofibrosis (MF): A pilot study (W Saber/S Gadalla) – (Attachment 4) d. PROP1611-41 Donor-recipient NK cell determinants associated with survival in JMML after hematopoietic stem cell transplantation (DA Lee/HG Rangarajan) (Attachment 5)

Dropped proposals (due to overlap with current studies)

3 Not for publication or presentation

a. PROP1608-02 Outcomes following umbilical cord blood transplantation for inherited metabolic disorders: do allele-level HLA matching characteristics matter? b. PROP1611-80 Using donor characteristics and matching at low expression loci to identify optimal 7of 8 HLA matched unrelated adult donors

Dropped proposals (due to limited populations) a. PROP1611-37 Do hypo-methylating agents following HLA C1/C2 mismatched or HLA/KIR mismatched transplantation improve clinical outcome in acute myeloid leukemia?

Dropped proposals (due to feasibility/lack of samples) a. PROP1611-164 Insulin-like growth factor 1 and outcome after myeloablative allogeneic HSCT

KIR/Haploidentical studies a. Discussion about current numbers, sample availability, and planning for future studies (Attachment 6) i. PROP1610-18 Role of KIR ligand mismatches for transplant outcomes in T-cell replete, non-myeloablative haploidentical transplants, inferring KIR ligand mismatches from HLA-A, -B, and -C locus typing mismatches: a CIBMTR registry analysis ii. PROP1611-36 Impact of KIR ligand mismatches on clinical outcomes following haploidentical stem cell transplantation iii. PROP1612-04 KIR and HLA Allotypes and Donor NK Alloreactivity in HLA- Haploidentical Transplantation

5. Studies in progress (Attachment 7) 1:50 pm

NK/KIR 1:50 pm (Chair: K Hsu) R02-40/R03-63 Acquisition of natural killer cell receptors in recipients of unrelated transplant (J Miller/E Trachtenberg) Ongoing – update b. R04-74 Functional significance of killer cell immunoglobulin-like receptor genes in HLA- matched and mismatched unrelated HCT (K Hsu) Ongoing – update c. IB15-02 Natural killer cell genomics and outcomes after allogeneic transplantation for chronic lymphocytic leukemia (V Bachanova, JS Miller, D Weisdorf, S Cooley) Sample typing – no update d. IB15-03 Killer Immunoglobulin Receptor (KIR) gene content and pediatric acute leukemia transplant outcomes (MR Verneris, J Miller, S Cooley) Protocol development – no update

BREAK – 30 minutes 2:15 pm

HLA GENES – CLASSICAL MATCHING 2:45 pm (Chair: K Fleischhauer) a. IB12-02C Prospective assignment of HLA-DPB1 T cell epitope (TCE) group mismatches by functional distance scores compared to the functional TCE assignment algorithm (K Fleischhauer) Manuscript preparation – update b. IB13-01 Effect of allele-level HLA-matching after UCB HCT for non-malignant diseases

4 Not for publication or presentation

in children (P Veys/M Eapen) Analysis – no update c. IB13-08 Short and long term survival assessment of post-HSCT transplantation using predictive modeling on a Bayesian network framework (R Abdi/S Haneuse/C Lee) Manuscript preparation – no update d. IB13-09 The development of machine learning based classifiers to define the alloreactivity of HLA mismatches in unrelated donor hematopoietic stem cell transplantation (Y Louzoun) Manuscript preparation – no update e. IB14-01 Impact of human leukocyte antigen haplotypes on outcomes of allogeneic transplantation for B-cell non-Hodgkin lymphomas (B William/M de Lima/M Fernandez- Vina/B Hill) (Attachment 8) Manuscript preparation – no update f. IB14-02 Structural/Functional models of HLA for data mining of permissive mismatching in allogeneic hematopoietic stem cell transplantation (L Gragert) Manuscript preparation – no update g. IB14-08 Development and validation of a clinical unrelated donor selection score (B Shaw, SJ Lee) Analysis – update h. IB15-01 The impact of single nucleotide gene polymorphisms (SNP) in the gamma block of the major histocompatibility complex (MHC) on unrelated donor hematopoietic cell transplants (HCT) for hematological malignancies (M Askar, R Sobecks) Sample typing – no update i. IB16-02 Use of HLA structure and function parameters to understand the relationship between HLA disparity and transplant outcomes (LA Baxter-Lowe) – Protocol development – no update

CYTOKINE/CHEMOKINE 3:20 pm a. IB14-03a: The prognostic impact of somatic mutations and levels of CXC chemokine ligands on post hematopoietic cell transplantation (HCT) outcomes in patients with myelodysplastic syndromes (MDS) (Attachment 9) (W Saber/B Dhakal) Sample typing – no update

OTHER GENES 3:20 pm (Chair: K Fleischhauer) a. IB06-05 Use of high-resolution HLA data from the NMDP for the International Histocompatibility Working Group in HCT (E Petersdorf) Ongoing – update b. IB09-04 D/R gene polymorphisms of drug metabolisms and innate immune response post allele matched matched unrelated donor HCT (V Rocha) Manuscript preparation – no update c. IB09-06/RT09-04 Genetic susceptibility to transplant-related mortality after unrelated donor stem cell transplant (T Hahn) (Attachment 10) Ongoing – update d. IB13-04 Discrepancy analysis of microsatellite loci as a proxy measure for ancestral differentiation between donors and recipients: correlation between high scores and poorer overall survival in high resolution matched unrelated donor transplantation (J Harvey/C Steward/V Rocha) Sample typing – no update e. IB14-04 Assessing the similarity of the T cell receptor repertoire in allogeneic hematopoietic stem cell recipients with the same single human leukocyte mismatches (EH Meyer) Analysis – no update f. IB14-05 mtDNA haplotypes and unrelated donor transplant outcomes (M Verneris/J Ross) Sample typing – no update

5 Not for publication or presentation

g. IB15-04 Clinical outcomes among hematopoietic stem cell transplant recipients as a function of socioeconomic status and related transcriptome differences (J Knight, JD Rizzo, S Cole) Protocol development – no update h. IB15-05 Secondary findings in exome sequencing data (S Savage, S Gadalla) (Attachment 11) Analysis – no update i. IB15-06c Recipient telomere length and outcomes after allogeneic unrelated hematopoietic cell transplant in patients with acute leukemia. (S Gadalla, S Savage) – (Attachment 12) Manuscript preparation – no update j. IB15-07 Functional genetic variants of the ST2 gene in pairs of recipient and donors for risk stratification of GVHD and TRM outcomes (S Paczesny) – no update k. IB16-01 The role of HLA-E compatibility in the prognosis of acute leukemia patients undergoing 10/10 HLA matched unrelated HSCT (C Tsamadou, D Fürst, J Mytilineos) Protocol development – no update l. IB16-03 Role of recipient and donor genetic polymorphisms in interferon lambda 4 (INFL4) on outcomes after unrelated allogeneic cell transplant (S Gadalla) Protocol development – no update

SENSITIZATION/TOLERANCE 4:10 pm (Chair: K Fleischhauer) a. IB11-01 Analysis of the NIMA effect on the outcome of unrelated PBSC/BM transplantation (G Ehninger/JJ van Rood/A Schmidt) Manuscript preparation – no update b. IB14-06 Donor-specific anti-HLA antibodies, allele and antigen level HLA mismatches in the outcomes of transplantation of non-malignant diseases with unrelated donors (M Fernandez-Viña/A Woolfrey) Analysis – update

MINOR HISTOCOMPATIBILITY ANTIGENS No updates.

7. Deferred studies pending accrual/funding

a. IB13-06 Role of the complement system in graft-versus-host disease (V Afshar- Kharghan/J Belmont/C Amos) Pending funding b. IB13-07 Impact of donor signal-regulatory protein alpha (SIRPα) polymorphism on outcome of allogeneic hematopoietic stem cell transplantation (allo-HSCT) (A Gassas/J Danska/S Rajakumar) Pending funding

8. Closing remarks (K Hsu) 4:25 pm

6 Not for publication or presentation Attachment 1

MINUTES CIBMTR WORKING COMMITTEE FOR IMMUNOBIOLOGY Honolulu, Hawaii Sunday, February 21, 2016, 12:15 pm– 3:45 pm

Co-Chair: Michael Verneris, MD, University of Minnesota Medical Center - Fairview Telephone: 612-626-2961; E-mail: [email protected] Co-Chair: Katharina Fleischhauer, MD; Essen University Hospital, Telephone: +49-201-723-4582; E-mail: [email protected] Co-Chair: Katharine Hsu, MD, PhD; Memorial Sloan-Kettering Cancer Center; Telephone: 646-888-2667; E-mail: [email protected] Co-Scientific Dir: Stephanie Lee, MD, MPH, Fred Hutchinson Cancer Research Center Telephone: 206-667-5160; E-mail: [email protected] Co-Scientific Dir: Stephen Spellman, MBS, CIBMTR Immunobiology Research Telephone: 763-406-8334; E-mail: [email protected] Statistical Director: Tao Wang, PhD, CIBMTR Statistical Center Telephone: 414-955-4339; E-mail: [email protected] Statisticians: Michael Haagenson, MS, CIBMTR Statistical Center Telephone: 763-406-8609; E-mail: [email protected]

1. Introduction

Dr. Michael Verneris opened the meeting at 12:15 pm. The minutes of last year’s Immunobiology Working Committee were approved as written. He then introduced the IBWC leadership, discussed goals, expectations and limitations of new studies, and then discussed the voting process. Votes were on a scale from 1 (Highest) to 9 (Lowest). Anyone with a scanned badge and attending the meeting will be added to the committee membership. Two to three proposals would be accepted this year. Decisions on the proposals will be sent to the Principal Investigators within one month after the meeting. Rules of authorship were also discussed.

2. Published or submitted papers

There were 17 manuscripts submitted, accepted and/or published this past year.

3. Research repository update and accrual tables

Mr. Steve Spellman gave an update on the accrual tables and the research sample repository. Detailed inventory information is available on the CIBMTR Web site at https://www.cibmtr.org/Samples/Inventory/Pages/index.aspx.

The sample usage policy states: 1) investigators are required to submit the interpreted results of all testing to CIBMTR, so that data are available to HCT research community and so that they eliminate or reduce duplicative testing to preserve resources/sample inventory; and 2) Data is captured in

7 Not for publication or presentation Attachment 1

CIBMTR Data Warehouse and linked to sample inventory.

Mr. Spellman also mentioned an NHLBI R21 funding opportunity with a deadline of February 19, even though more opportunities may be available in the future. This opportunity was PAR-13-009 "Secondary Dataset Analyses in Heart, Lung, and Blood Diseases and Sleep Disorders (R21)“. This grant program is well-suited to fund analyses of CIBMTR data. CIBMTR is willing to conduct feasibility assessments and provide letters of support for potential studies.

4. Future/proposed studies and discussion

a. PROP1510-05 Impact of HLA-disparity on graft-versus-lymphoma effects in patients with and without graft-versus-host-disease after allogeneic stem cell transplantation (F Ayuketang, U Bacher)

Dr. Stephanie Lee presented this proposal on behalf of the PIs. This proposal would look at lymphoma cases and the HLA disparity impacts on Graft vs. Lymphoma for patients with or without GvHD after allogeneic HCT. It has been shown that an increase in HLA-disparity results in increased incidence of GvHD and non-relapse mortality. It is unclear whether such an increase in HLA-disparity is also accompanied by stronger graft-versus-malignancy effects. It has been shown that there is an impact of both aGVHD and cGVHD on relapse rate of various lymphoma subtypes. However, the impact of HLA-disparity on relapse has not yet been shown. The primary endpoint is relapse incidence. Secondary endpoints include relapse-free survival, non-relapse mortality and overall survival. For GvHD, the proposed definitions were Grades II-IV acute and any chronic GvHD.

Inclusion criteria for the proposal were adult patients (> 18), a variety of lymphoma groups, HLA- identical siblings or 8/8 or 7/8 unrelated donor transplants, first allogeneic transplants (previous autologous allowed), and HLA-DP considerations will be included based on data availability. There were 1065 HLA-identical siblings, 818 8/8 URD and 255 7/8 URD transplants available.

This would be a retrospective analysis comparing GvHD yes vs. GvHD no in the three donor groups. Concerns raised by the Committee: 1) most of the HLA-DP would be mismatches, and it has been shown that HLA-DP mismatching has an effect on GvHD, especially when considering the TCE permissive vs. non-permissive mismatches; 2) lymphoma histologies are complicated because relapse can be problematic since often there is no full remission typically; 3) it might be better to compare Grade 0 vs. Grades I-II vs. Grades III-IV acute GvHD. These issues would be addressed during the protocol development phase, if the proposal were to be accepted.

b. PROP1511-45 The role of HLA-E compatibility in the prognosis of acute leukemia patients undergoing 10/10 HLA matched unrelated HCT (C Tsamadou, D Fürst, J Mytilineos)

Dr. Joannis Mytilineos presented the proposal. This proposal would look at HLA-E compatibility within an acute leukemia population of 10/10 matched URD HCT. Hypotheses include: 1) HLA-E is a significant modulator of NK-cytotoxicity; 2) HLA-E expression in leukemic blasts is retained regardless of potential HLA-class Ia downregulation due to leukemogenesis; 3) HLA-E allelic variation gives rise to distinct pHLA-E complexes that may differentially tune NK “licensing”; and 4) enhanced NK-mediated alloreactivity in an unrelated HCT setting is associated with strong

8 Not for publication or presentation Attachment 1

GvL and lower GvHD effects. This might imply that HLA-E mismatching may have an effect on HCT outcome in AML patients transplanted with a 10/10 HLA MUD.

Relevance of HLA-E/NK‘s interaction in an acute leukemia HSCT context include: 1) Protein HLA-E (pHLA-E) complexes are the main ligands to NKG2A (NK inhibition) and to NKG2C ( NK activation) NK receptors; 2) pHLA-E & NKG2A/C interaction‘s dynamics is primarily influenced by the bound peptide‘s specificity  NKG2A/C receptors exhibit “TCR-like” features as to the recognition of distinct peptide repertoires; 3) “Unconventional“ pHLA-E complexes have been found to go ”unnoticed“ by NKG2A (leukemogenesis could generate such complexes as HLA-class Ia molecules, basic source of HLA-E peptide repertoire, are often downregulated); 4) These complexes, however, are recognized by the activating NKG2C receptors; and 5) The dominant receptor of newly reconstituted NKs after HCT is NKG2A.

Preliminary German study results on 552 acute leukemia patients suggest that HLA-E mismatched cases are associated with better OS, NRM and chronic GVHD than HLA-E matched cases. HLA-E mismatched cases made up about 37% of the cohort (N=203). The goal is to validate these results in a CIBMTR cohort. A preliminary inventory query identified over 3,300 cases available in the CIBMTR repository for a validation cohort. Dr. Mytilineos’ laboratory can perform the necessary typing for HLA-E and has secured local funding to support the project.

Questions and concerns raised by the Committee: 1) Are the amino acid changes in HLA-E in the antigen binding domain? Yes, the mismatch is in exon 2. 2) Is HLA-E a minor histocompatibility antigen? It is possible. 3) Would it be worth looking at CMV reactivation with the NKG2c interaction? Yes, and it hasn’t been evaluated. 4) Did the preliminary study look at mismatch vector effects? No, but it may be worth evaluating. 5) The distribution of the genotypes is about 50/50. 6) Are there expression differences? Yes; there are reports of this, but it wasn’t found in prior analysis. 7) Were there ATG effects? They did not find any. 8) What is the rationale for a mismatch effect in the NK setting? Acute leukemia can result in down-regulation of Class I; then, the NK cell effects may take over. 9) Was there any acute GvHD effect in the preliminary analysis? They did not have the data to evaluate this outcome. 10) Do the results imply a match vs. mismatch or a donor vs. recipient genotype difference? It is believed to be match vs. mismatch. 11) Will you include mismatched related donors? This would be a validation set with CIBMTR and previous analysis was URD, so no haploidentical HCTs for this cohort are desired. 12) Is there any linkage to other HLA? No strong linkage is expected.

c. PROP1511-134 Use of HLA structure and function parameters to understand the relationship between HLA disparity and transplant outcomes (LA Baxter-Lowe)

Lee Ann Baxter-Lowe presented this proposal. The hypothesis is that a scoring system for HLA disparities based upon fundamental knowledge of allorecognition is superior to conventional approaches for defining and ranking HLA mismatches. Two key elements to this scoring system are: 1) similarity of amino acids important for docking of T-cell receptors (TCR); and 2) HLA differences that affect peptide recognition. To calculate the score, the first step is to determine the amino acid differences. The second step is to determine the score with 4 steps: a) score each amino acid difference using a matrix (modified matrix from Li et al, Front. Genetics 2013); b)

9 Not for publication or presentation Attachment 1

determine the total score for docking; c) determine score for peptide binding/orientation; and d) determine structure/function score.

The study design will include a discovery data set and validation data set, matched for important donor, recipient, disease and transplant characteristic with comparable outcomes. The primary outcome will be overall survival while the secondary outcomes include aGVHD, TRM, neutrophil and platelet engraftment, cGVHD, DFS and relapse. The preliminary population identified for this study includes 846 HLA-A single mismatch cases, 393 HLA-B single mismatch cases, 1037 HLA-C single mismatch cases and 6852 8/8 HLA-matched cases.

Comments and questions from the Committee: 1) Isn’t a limitation of the approach that the docking residue differences are not relevant in the indirect allorecognition pathway? Theindirect pathway may be relevant because the matched alleles could still presentthe degraded mismatched HLA as a minor Histocompatibility antigen. True, but the indirect mHAg presentation will be a less robust effect overwhelmed that the direct allorecognition predicted by the scoring system. 2) Will engraftment be impacted by the model? Maybe, could impact it. 3) Are you planning to look at all 3 loci in the training set? Yes.

d. PROP1511-139 Investigating influence of minor histocompatibility antigens/alleles and previously uncharacterized KIR interacting factors on outcomes after HLA matched HSCT (CH Roberts, A Madbouly)

Mr. Spellman presented this proposal in the PIs’ absence. This proposal would take an existing data set of 995 cases from IB12-03 that were genotyped using the Illumina OmniExpress platform to perform some bioinformatics discovery work. The PIs intend to use the Genome Wide Association Scan for Histocompatibility antigens (GWASH) methodology developed by Dr. Roberts to search for previously uncharacterized minor histocompatibility antigens (mHAgs) that influence transplant outcome. They will also attempt to identify novel KIR ligands through statistical epistasis screening. The IB12-03 cohort would be used as a discovery dataset with the potential to use Terri Hahn’s IB09-06 data set for future validation.

Questions and concerns include: Will they consider Copy Number Variation (CNV) for KIR? It is not clear whether the methodology planned for use in the mHAG evaluation, GWASH, will be able to discern CNV.

Dropped proposals (due to limited populations) a. PROP1508-02 Mutational analysis in T-cell acute lymphoblastic leukemia (GA Challen) -

Dropped due to insufficient sample size and characterization of the samples. Dr. Challen is working with Dr. Verneris to identify a potential cohort in the University of Minnesota Research Repository for a pilot study.

b. PROP1511-52 HLA disparity and post-transplant outcomes in AML patients after T-replete HLA- haploidentical hematopoietic cell transplantation: a retrospective analysis (R Romee, A Rashidi)

Dropped due to insufficient sample size.

10 Not for publication or presentation Attachment 1

c. PROP1511-122 Impact of human leukocyte antigen haplotypes on outcomes of allogeneic transplantation for multiple myeloma (E Malek, M de Lima)

Dropped due to insufficient sample size.

5. Studies in progress (Attachment 7)

HLA GENES – CLASSICAL MATCHING (Chair: K Fleischhauer) a. IB12-02B Comparative evaluation of the immunogenicity and the diversity model of HLA- DPB1 TCE group matching in unrelated HCT (K Fleischhauer) Manuscript preparation – no update b. IB12-02C Prospective assignment of HLA-DPB1 T cell epitope (TCE) group mismatches by functional distance scores compared to the functional TCE assignment algorithm (K Fleischhauer) (Attachment 8) Manuscript preparation – update

Dr. Katharina Fleischhauer presented this study update. She showed how the Crivello assignments of the T-cell epitope (TCE) groups using functional distance scores compares to the functional TCE assignment algorithm designed using the Pidala or Zino scores. She showed that the concordance was fairly good, but many of the permissive in the Pidala/ZIno scoring came out as non-permissive in the Crivello scoring. Overall there was a reassignment of 725/3276 (22.1%) pairs. With this reassignment to Crivello scoring, the DFS comparisons were no longer significant, but the chronic GvHD and Grades II-IV acute GvHD came out as highly significant. The conclusions were that the main associations of OS and TRM remain unchanged using either the Pidala/Zino or Crivello assignments, but there were significant differences between cases that changed assignment from permissive to non-permissive for GvHD.

Questions and concerns include: Is the model closer to the expression level developed by Effie Petersdorf? To a degree. Functional distance scoring only accounts for disparity in exon 2. There were concerns about the overall message and how the different versions of the algorithm should be used for prospective donor selection. c. IB12-03 Effect of genetic ancestry matching on HSCT outcomes (A Madbouly/M Maiers/N Majhail) (Attachment 9) Manuscript preparation – no update d. IB13-01 Effect of allele-level HLA-matching after UCB HCT for non-malignant diseases in children (P Veys/M Eapen) Data file preparation – no update e. IB13-08 Short and long term survival assessment of post-HSCT transplantation using predictive modeling on a Bayesian network framework (R Abdi/G Alterovitz/G McDermott) Analysis – no update f. IB13-09 The development of machine learning based classifiers to define the alloreactivity of HLA mismatches in unrelated donor hematopoietic stem cell transplantation (Y Louzoun) Analysis – no update g. IB14-01 Impact of human leukocyte antigen haplotypes on outcomes of allogeneic transplantation for B-cell non-Hodgkin lymphomas (B William/M de Lima/M Fernandez- Vina/B Hill) Data file preparation – no update h. IB14-02 Structural/Functional models of HLA for data mining of permissive mismatching in allogeneic hematopoietic stem cell transplantation (L Gragert) Analysis – update

11 Not for publication or presentation Attachment 1

Dr. Loren Gragert presented this study update. Several retrospective HCT outcome studies have identified single amino acid (AA) mismatches in HLA proteins that confer worse than average outcome. However, these studies have been weakly powered due to few transplant pairs relative to HLA variants. Powerful machine learning approaches such as random forest have been applied to identify mismatches that do not reach significance in standard multivariate analysis. In Dr. Gragert’s approach of testing HLA variation, he wants to go beyond testing for single amino acid (AA) mismatches and examine HLA associations with groups of AA positions at varying levels of mismatch specificity (directional (GvH vs HvG), nondirectional, and nonspecific (any AAs)). His goal was to improve statistical power by having more cases per HLA mismatch category, yet retain the useful specificity of HLA categories.

Dr. Gragert applied a variety of machine learning approaches, like PLS, BayesNet and Random Forest. He built nested models on the 80% training set and left the remaining 20% as a test set. He used a “reduced” model using clinical covariates as well as an “HLA” model using HLA variants and covariates. He then classified the training and test sets using those two models, and measured the incremental predictive power of HLA. Preliminary results suggested that regardless of machine learning method, use of specific HLA variants did not significantly improve prediction of 1-year overall survival in test dataset.

Dr. Gragert had hoped to improve upon the prior findings from Dr. Susana Marino by testing HLA variation differently. However, failing to find a signal, he attempted to replicate their study using the same conditions. Some amino acid substitutions (AAS) overlapped, but there was very high variation. Problems included: 1) high correlation in AAS predictors for model training that lead to unstable importance scores, and 2) model performance did not improve outcomes in the test set. In addition, the Marino et al. did not report ROC in their analysis making it difficult to run fully compare complete results.Using machine learning techniques, the analyses are very complicated. However, if there are significant results, they should hold regardless of techniques. The AAS results in these analyses were unstable.

i. IB14-08 Development and validation of a clinical unrelated donor selection score (B Shaw, SJ Lee) – update

Dr. Bronwen Shaw presented this study update. The purpose of this study was to develop a donor selection score to assist in donor selection, based on the impact of various donor characteristics on overall survival. The study population included 5,952 8/8 HLA-matched adults with AML/ALL/CML/MDS, where 2/3 were myeloablative and the rest reduced intensity, 2/3 were PBSC and 1/3 bone marrow, and the years of transplant were from 1999 through 2011. The final statistical model was adjusted for age, race, Karnofsky score, disease and status, use of TBI and recipient CMV, and was stratified on conditioning intensity, recipient gender, graft type, and GVHD prophylaxis due to non-proportional hazards. The donor factors considered in the study were age, gender, parity, CMV, ABO, DPB1 TCE andDQB1.

In the training set, the donor variables that affected the donor score were non-permissive HLA-DPB1 matching, donor age by decade, mismatched CMV serostatus, and ABO mismatch. Some interactions and consistencies were 1) Graft type and DPB1 non-permissive mismatch (p=0.004): DPB1 effect only in BM transplants (HR 1.30); 2) ABO more impact in the earlier time period: 1999-2006 HR 1.13; 2007-2011 HR 1.03; and 3) CMV mismatch status increase risk, irrespective of Recipient status: R+ HR 1.08; R- HR

12 Not for publication or presentation Attachment 1

1.11.

However, the results did not validate in the validation set. Next steps are to find a way to validate somehow. Questions and concerns from the Committee: 1) using recipient CMV vs. CMV mismatching (although recipient CMV did not come out significant on its own); 2) grouping ABO mismatch altogether when Stanford showed it was ABO minor mismatch that had mattered at one time. 3) There was also a concern that there were too many ABO mismatches in this data set, and would ABO for the NMDP cohort be collected on the TED forms? That would allow for expansion of the cohort 4) there wasn’t enough power in the validation and there was not as strong of an effect; 5) there was more bone marrow in earlier years; and 6) there might be a change in practice causing the differences. j. IB15-01 The impact of single nucleotide gene polymorphisms (SNP) in the gamma block of the major histocompatibility complex (MHC) on unrelated donor hematopoietic cell transplants (HCT) for hematological malignancies (M Askar, R Sobecks) Sample typing – no update

CYTOKINE/CHEMOKINE No updates.

BREAK – 30 minutes

NK/KIR (Chair: K Hsu) a. R02-40/R03-63 Acquisition of natural killer cell receptors in recipients of unrelated transplant (J Miller/E Trachtenberg) Ongoing – no update b. R04-74 Functional significance of killer cell immunoglobulin-like receptor genes in HLA- matched and mismatched unrelated HCT (K Hsu) Ongoing – update

Dr. Katharine Hsu presented this study update. This study has shown that KIR3DL1 interactions with HLA-Bw4 influence outcomes. 40% of HLA-B alleles exhibit the Bw4 epitope withinIsoleucine at position 80 (I80) or threonine at position 80 (T80). KIR3DL1 is highly polymorphic and occurs in >95% of individuals. As for the activating KIR3DS1, it is a KIR3DL1 allele, 3DS1 is >95% homologous to 3DL1, 3DS1 has no demonstrable interaction with HLA-Bw4, and it occurs in 40% of individuals.

Dr. Hsu showed a table of KIR3DL1 allele frequencies in a cohort of 426 individuals, and it showed that 12 alleles make up 97.5% of all KIR3DL1 alleles. KIR3DL1 has three main functional subtypes: high affinity, low affinity and null affintiy. Prior studies have shown that patients with high affinity alleles have more relapse; low affinity has low relapse; and null affinity has the least amount of relapse.

The current study population included 1,328 AML patients where 46% were 9/10 and 54% 10/10, 95% were HLA-B allele matched and 99.5% were HLA-B KIR ligand matched, and approximately 50% received bone marrow. KIR gene presence/absence typing was obtained by Luminex-based SSO typing. KIR3DL1 allele typing was performed by “intermediate resolution” PCR-SSP to define high, low and null affinity variants. Outcomes were OS, relapse, NRM and GvHD, and analyses were adjusted for patient age, disease severity, HLA matching, type of disease and T-cell depletion. The analyses showed that KIR3DL1 and HLA-B subtype combinations impacted AML relapse and survival where strongly-interacting/high affinity pairs have worse OS and relapse rates.

13 Not for publication or presentation Attachment 1

A concern was raised about if Bw4 was looked at in the context of HLA-A alleles. It has been but the HLA- A alleles have very low Bw4 expression resulting in low affinity interactions. c. IB11-05 KIR genotyping and Immune function in MDS patients prior to unrelated donor transplantation (E Warlick/J Miller) Analysis – update

Dr. Jeffrey Miller presented this study update. This study has analyzed pre-transplant viable samples from MDS patients evaluating methods to overcome the effects of myeloid derived suppressor cells that inhibit NK cell function. Analyses have focused on NK functional studies and resulted in a couple of basic science publications. d. IB12-04A Determining the effects of HLA-C KIR ligand expression on outcomes of unrelated hematopoietic stem cell transplantation (M Verneris/J Venstrom) Manuscript preparation – update e. IB12-04B Effect of HLA-C allele matching in the context of recipient HLA-C-encoded KIR ligand grouping (C1 or C2) on the outcome of unrelated hematopoietic stem cell transplantation (HSCT) (M Verneris/J Fischer/M Uhrberg) Manuscript preparation – update

IB12-04A and IB12-04B are being combined and written up as one manuscript by a medical student from Dr. Verneris’ laboratory because the original PIs were non-responsive and failed to make progress. CIBMTR encourages all PIs to complete assignments in a timely manner or risk the re-assignment of projects to more motivated and often junior investigators. f. IB15-02 Natural killer cell genomics and outcomes after allogeneic transplantation for chronic lymphocytic leukemia (V Bachanova, JS Miller, D Weisdorf, S Cooley) Protocol development – no update g. IB15-03 Killer Immunoglobulin Receptor (KIR) gene content and pediatric acute leukemia transplant outcomes (MR Verneris, J Miller, S Cooley) Protocol development – no update

OTHER GENES (Chair: K Hsu) a. IB06-05 Use of high-resolution HLA data from the NMDP for the International Histocompatibility Working Group in HCT (E Petersdorf) Ongoing – update

Dr. Effie Petersdorf was unable to attend due to a conflict with the Presidential Symposium. The Committee was directed to her recent publication in the New England Journal of Medicine for updates on her project. b. IB09-04 D/R gene polymorphisms of drug metabolisms and innate immune response post allele matched matched unrelated donor HCT (V Rocha) Manuscript preparation – no update c. IB09-06/RT09-04 Genetic susceptibility to transplant-related mortality after unrelated donor stem cell transplant (T Hahn) (Attachment 10) Manuscript preparation – update

Dr. Theresa Hahn presented this study update. This study was supported by an R01 and an R03 grant. The R01 grant supported analysis of approximately 3,500 recipient/donor pairs of AML/ALL/MDS patients that were matched at least 8/8 HLA-matched with their unrelated donortransplanted between

14 Not for publication or presentation Attachment 1

2000 and 2011. They analyzed recipient genotypes, donor genotypes and mismatching at individual loci between recipient and donor. Outcomes were OS, PFS, TRM, disease and death due to GvHD, Infection and/or organ failure. There was an interaction with conditioning regimen so it was analyzed as BuCy vs CyTBI and as myeloablative vs. reduced intensity. There was replication of published candidate genes for the outcomes of OS, PFS, TRM, disease and for death due to GvHD. They are also performing exome chip analysis for rare variants in recipient and donor with the same outcomes analyzed in the GWAS. They are collaborating with Dr. John Hansen and Dr. Paul Martin from FHCRC to replicate Dr. Hahn’s findings in the FHCRC’s GWAS.

The R03 grant funds a case control study of pediatric, adolescent/young adult and adult onset ALL to look for susceptibility variants with developing ALL, a case control study of AML to look for susceptibility variants with developing AML, and GEMM (Genes & Environment in Myeloid Malignancies) International Consortium to perform a meta-GWAS of de novo and therapy related AML and MDS.

There have been (or will be) several presentations and publications from this study. a) ASHG 2015 - Case control study of ALL (results by age); b) ASH 2015 – b1) GWAS OS/PFS in R and D, b2) GWAS Mismatching R-D outside the MHC – TRM/Disease, and b3) Case control study of ALL (results by sex and cytogenetics); c) Tandem 2016 – c1) GWAS Cause specific death (GVHD, Inf, Org Failure) in R and D, c2) GWAS R and D within the MHC and OS, TRM, Disease; d) BBMT 2015 Cause of death adjudication (Hahn T et al); and e) Curr Hem Reports 2015 Review of candidate gene studies and survival outcomes, justification for the GWAS (Sucheston-Campbell, et al). d. IB10-01b Telomere length telomerase polymorphisms in SAA (S Gadalla/S Savage) Manuscript preparation – no update e. IB10-01c Telomere length telomerase polymorphism in SAA-exome analysis and mosaicism (S Gadalla/S Savage) Manuscript preparation – no update f. IB10-04 A validation study of the role of base excision repair pathway as a predictor of outcomes after HCT (B Thyagrajan/M Arora) Manuscript preparation – update

Mr. Spellman mentioned that this study was an attempted validation of a single center analysis and did not validate in the CIBMTR cohort. The manuscript has been submitted as a letter to the editor to BBMT. g. IB13-04 Discrepancy analysis of microsatellite loci as a proxy measure for ancestral differentiation between donors and recipients: correlation between high scores and poorer overall survival in high resolution matched unrelated donor transplantation (J Harvey/C Steward/V Rocha) Sample typing – no update h. IB13-05 The impact of MHC Class I chain-related gene A donor-recipient mismatches and MICA-129 polymorphism on unrelated donor hematopoietic stem cell transplants for hematological malignancies (M Askar/R Sobecks) Manuscript preparation – update

Dr. Medhat Askar presented this study update. The analyses showed that: 1) at p=0.01 significance level, there was no association between any outcome and MICA mismatch or MICA-129 polymorphism; 2) there was a suggestion that MICA mismatches were associated with acute GVHD grades II-IV (HR 1.4, 95% CI 1.1-1.9, p= 0.013); and 3) there was also a suggestion of an association between donor MICA- 129 non-VV genotypes and slower platelet engraftment with HR 1.4 (95% CI 1.109-1.985, p= 0.02). It was also shown that there is a low linkage disequilibrium among HLA-B and MICA. This study is in its final stages of manuscript preparation and will be submitted by June 2016.

15 Not for publication or presentation Attachment 1 i. IB14-03 The prognostic impact of somatic mutations and levels of CXC chemokine ligands on post hematopoietic cell transplantation outcomes in patients with myelodysplastic syndromes (W Saber/RC Lindsley/BL Ebert) Analysis – update

Mr. Spellman gave a brief study update. This study is characterizing somatic mutations in approximately 1,500 MDS patients. The analysis should be completed in the next two months, and the goal is an ASH 2016 abstract submission. j. IB14-04 Assessing the similarity of the T cell receptor repertoire in allogeneic hematopoietic stem cell recipients with the same single human leukocyte mismatches (EH Meyer) Analysis – update

Dr. Everett Meyer presented this study update. The study included pre- and post-transplant samples from 98 patients who underwent single or double HLA mismatch HCTfrom an unrelated donor. They hypothesized that individuals with the same HLA mismatch would show greater than expected TCR sequence relatedness. The method of analysis is to compare clonotype distributions based on TCR V CDR3 amino acid sequences: First, between groups of subjects defined by having either the same or a different HLA mismatch; and then second, at 30- and 60-days post-transplant. Note that they will estimate the similarity of clonotype distributions using the Bhattacharyya coefficient.

They completed sequencing and evaluation of 30 HLA mismatched recipients. The distribution of similarity coefficients exhibits: a) somewhat less variability for the group with same HLA mismatch compared to the group with different HLA mismatches at 30 days post-transplant; and b) a shift towards less variability for the group with same HLA mismatch at 60 days post-transplant.

Next steps are to complete the sequencing for the full cohort in the next 3 months and re-analyze. Then, they will investigate if there is any evidence of TCR convergence by shared HLA mismatch. They will consider a relatedness by less stringent CDR3 identity (hamming distance) as well as a cluster analysis.

Questions and concerns were: 1) Are samples available from donors that donated to more than one recipient to allow for evaluation of TCR repertoire from the same donor in different recipient backgrounds? No. 2) Has the outcome data been analyzed yet? No. k. IB14-05 mtDNA haplotypes and unrelated donor transplant outcomes (M Verneris/J Ross) Sample typing – update

Dr. Verneris gave a brief study update. He has the initial wave of samples and will begin sequencing them soon. l. IB15-04 Clinical outcomes among hematopoietic stem cell transplant recipients as a function of socioeconomic status and related transcriptome differences (J Knight, JD Rizzo, S Cole) Protocol development – no update m. IB15-05 Secondary findings in exome sequencing data (S Savage, S Gadalla) Analysis – no update n. IB15-06 Donor telomere length and outcomes after hematopoietic stem cell transplantation for acute leukemia (S Gadalla, S Savage, DJ Loftus, E Hytopoulos, M Yeager, C Dagnall) Analysis – no update

16 Not for publication or presentation Attachment 1 o. IB15-07 Functional genetic variants of the ST2 gene in pairs of recipient and donors for risk stratification of GVHD and TRM outcomes (S Paczesny) – update

Dr. Sophie Paczesny presented this study update. ST2 is the IL-33R and is the only known ligand. It has been shown that ST2 has both a membrane-bound form (mST2) expressed on mast cells, Th2, Tregs, ILCs, and a soluble form (sST2) which acts as a decoy receptor for IL-33. It has also been show that sST2 levels are increased in patients with IBD, cardiovascular diseases, cardiac allo-rejection, and GVHD. In a recent GWAS study of 2,991 Framingham Cohort participants, multiple SNPs within IL1RL1 (the gene encoding ST2) demonstrated associations with plasma sST2 concentrations. For example, the SNP most significantly associated with sST2 (rs950880) has a minor allele frequency of 39% and if we assume Hardy-Weinberg equilibrium in the study population, approximately 15% of the HCT donors/patients will be homozygous for the minor allele. The hypothesis of this study is that the 16 SNPs associated with sST2 expression will correlate with risk of developing acute GVHD and TRM following allogeneic HCT.

The innovations/significance of this study are: 1) there would be a large allele frequency of the 16 ST2 SNPs in the population; 2) there could be a correlation with plasma levels of sST2; 3) these would be non-HLA SNPs; 4) this would be a well-controlled cohort of unrelated donor well-matched (high resolution HLA match 8/8 receiving standard GVHD prophylaxis (CsA + others or Tacrolimus + others); and 5) this could lead to the ability to identify patients and donors at high risk for acute GVHD and TRM before the transplant.

Patient eligibility would be first myeloablative transplants between 2011-2014 and with available whole blood for DNA extraction for the donor/recipient pair, treated for ALL, AML and MDS (low/intermediate risk) with an unrelated donor high resolution 8/8 HLA match receiving standard GVHD prophylaxis (CsA + others or Tacrolimus + others). The population will not include cord blood or haploidentical transplants, nor will it include ex-vivo nor in-vivo T-cell depleted (ATG or Campath) patients. A power calculation performed by Dr. Tao Wang estimates the need for approximately 1,000 donor/recipient pairs. A TaqMan® OpenArray® will be used to type for the 16 ST2 SNPs.

An Open Array on a pilot of 400 samples has already started and will be completed by the end of February 2016. This will verify the SNP frequencies in the population and be used to perform a power calculations.

It was suggested to work with Dr. Hahn’s group for validation of the ST2 SNP frequencies.

SENSITIZATION/TOLERANCE (Chair: K Fleischhauer) a. IB11-01 Analysis of the NIMA effect on the outcome of unrelated PBSC/BM transplantation (G Ehninger/JJ van Rood/A Schmidt) (Attachment 11) Manuscript preparation – no update b. IB14-06 Donor-specific anti-HLA antibodies, allele and antigen level HLA mismatches in the outcomes of transplantation of non-malignant diseases with unrelated donors (M Fernandez-Viña/A Woolfrey) Sample typing – no update

MINOR HISTOCOMPATIBILITY ANTIGENS No updates.

17 Not for publication or presentation Attachment 1

6. Deferred studies pending accrual/funding

a. IB13-06 Role of the complement system in graft-versus-host disease (V Afshar- Kharghan/J Belmont/C Amos) Pending funding b. IB13-07 Impact of donor signal-regulatory protein alpha (SIRPα) polymorphism on outcome of allogeneic hematopoietic stem cell transplantation (allo-HSCT) (A Gassas/J Danska/S Rajakumar) Pending funding

7. Other business

No other business was discussed.

8. Closing remarks (K Hsu)

Dr. Hsu thanked everyone for attending and for voting on the proposals. She adjourned the meeting at 3:45 PM.

18 Not for publication or presentation Attachment 1

WORKING COMMITTEE OVERVIEW PLAN for 2016-2017 STUDIES IN PROGRESS IB09-04 D/R gene polymorphisms of drug metabolisms and innate immune response post allele matched unrelated donor HCT (V Rocha) – The goal for June 2016 is to have the manuscript submitted. The manuscript will be published by June 2017.

IB10-01b Telomere length telomerase polymorphisms in SAA (S Gadalla/S Savage) – The goal for June 2016 is to have the manuscript submitted. The manuscript will be published by June 2017.

IB10-01c Telomere length telomerase polymorphism in SAA-exome analysis and mosaicism (S Gadalla/S Savage) - The goal for June 2016 is to have the manuscript submitted. The manuscript will be published by June 2017.

IB11-01 Analysis of the NIMA effect on the outcome of unrelated PBSC/BM transplantation (G Ehninger) – The manuscript will be finalized and submitted by June 2016. The manuscript will be published by June 2017.

IB12-02B Comparative evaluation of the immunogenicity and the diversity model of HLA-DPB1 TCE group matching in unrelated HCT (K Fleischhauer) – The goal for June 2016 is to have the manuscript submitted (after IB12-02C is submitted). The manuscript will be published by June 2017.

IB12-02C Prospective assignment of HLA-DPB1 T cell epitope (TCE) group mismatches by functional distance scores compared to the functional TCE assignment algorithm (K Fleischhauer) – The goal for June 2016 is to have the manuscript submitted. The manuscript will be published by June 2017.

IB12-03 Effect of genetic ancestry matching on HSCT outcomes (A Madbouly) – The manuscript will be finalized and submitted by June 2016. The manuscript will be published by June 2017.

IB12-04A Determining the Effects of HLA-C KIR Ligand Expression on Outcomes of Unrelated Hematopoietic Stem Cell Transplantation (M Verneris/G Hoff/J Venstrom) / IB12-04B Effect of HLA-C allele matching in the context of recipient HLA-C-encoded KIR ligand grouping (C1 or C2) on the outcome of unrelated hematopoietic stem cell transplantation (HCT) (J Fischer/M Uhrberg) – The combined manuscript will be finalized and submitted by June 2016. The manuscript will be published by June 2017.

IB13-01 The effect of allele-level HLA-matching on survival after umbilical cord blood transplantation for non-malignant diseases in children. (P Veys/M Eapen) –The analysis will be completed in May 2016 and the manuscript in preparation by June 2016. The manuscript will be submitted by June 2017.

IB13-04 Discrepancy analysis of microsatellite loci as a proxy measure for ancestral differentiation between donors and recipients: correlation between high scores and poorer overall survival in high resolution matched unrelated donor transplantation (J Harvey/C Steward/V Rocha) – The sample typing and datafile preparation will be complete by June 2016.

19 Not for publication or presentation Attachment 1

The analysis will be complete by December 2016 and the manuscript in preparation by June 2017.

IB13-05 The Impact Of MHC Class I Chain-Related Gene A (MICA) Donor-Recipient Mismatches and MICA-129 Polymorphism On Unrelated Donor Hematopoietic Stem Cell Transplants (HSCT) For Hematological Malignancies (M Askar/R Sobecks) – The manuscript will be finalized and submitted by June 2016. The manuscript will be published by June 2017.

IB13-08 Interaction between SNPs and clinical data using predictive modeling on a Bayesian network framework/ Short and long term survival assessment of post HSCT transplantation using predictive modeling on a Bayesian network framework. (R Abdi/G Alterovitz/D McDermott) – The manuscript will be under preparation by June 2016. The manuscript will be submitted by June 2017.

IB13-09 The development of Machine Learning based classifiers to define the alloreactivity of HLA mismatches in unrelated donor hematopoietic stem cell transplantation (Y Louzoun) - The manuscript will be under preparation by June 2016. The manuscript will be submitted By June 2017 .

IB14-01 Impact of human leukocyte antigen haplotypes on outcomes of allogeneic transplantation for B-cell non-Hodgkin lymphomas (B William/M de Lima/M Fernandez-Vina/B Hill) – The analysis will be completed by June 2016. The manuscript will be under preparation by November 2016 and submitted by June 2017.

IB14-02 Structural/Functional models of HLA for data mining of permissive mismatching in allogeneic hematopoietic stem cell transplantation (L Gragert) – The manuscript will be under preparation by June 2016. The manuscript will be submitted by June 2017.

IB14-03 The prognostic impact of somatic mutations and levels of CXC chemokine ligands on post hematopoietic cell transplantation outcomes in patients with myelodysplastic syndromes (W Saber/ RC Lindsley/ B Ebert) – The analysis will be completed by May 2016 and manuscript preparation initiated by June 2016. The manuscript will be finalized and submitted by June 2017.

IB14-04 Assessing the similarity of the T cell receptor repertoire in allogeneic hematopoietic stem cell recipients with the same single human leukocyte mismatches (E Meyer) – The goal by June 2016 is to complete the sample typing and continue the sequencing analysis. The outcome correlation analysis will be completed by December 2016 and the manuscript will be under preparation by June 2017.

IB14-05 Mitochondrial DNA (mtDNA) haplotypes and unrelated donor transplant outcomes (M Verneris/ J Ross) – The goal by June 2016 is to continue the sample typing. The datafile preparation will be completed by December 2016 and analysis initiated by June 2017.

IB14-06 Donor-specific anti-HLA antibodies, allele and antigen level HLA mismatches in the outcomes of transplantation of non-malignant diseases with unrelated donors (M Fernandez- Viña/A Woolfrey) – Sample testing and antibody identification will continue through June 2016.

20 Not for publication or presentation Attachment 1

The datafile will be finalized in November 2016, the analysis completed in April 2017 and manuscript preparation initiated by June 2017.

IB14-08 Development and validation of a clinical unrelated donor selection score (B Shaw, SJ Lee) – The goal by June is to be in manuscript preparation. The goal by November 2016 is to have the manuscript submitted and the goal by June 2017 is for the manuscript to be published.

IB15-01 The Impact of Single Nucleotide Gene Polymorphisms (SNP) in the Gamma Block of the Major Histocompatibility Complex (MHC) on Unrelated Donor Hematopoietic Cell Transplants (HCT) For Hematological Malignancies (M Askar/R Sobecks) – The SNP typing will be completed by May 2016 and analysis initiated by June 2016. The analysis will be completed by December 2016 and the manuscript submitted by June 2017.

IB15-02 Natural killer cell genomics and outcomes after allogeneic transplantation for chronic lymphocytic leukemia (V Bachanova/J Miller/D Weisdorf/S Cooley) - The protocol will be finalized by May 2016 and sample typing initiated by June 2016. The sample typing and datafile preparation will be completed by January 2017 and the analysis initiated by June 2017.

IB15-03 Killer Immunoglobulin Receptor (KIR) Gene Content and Pediatric Acute Leukemia Transplant Outcomes (M Verneris/J Miller/S Cooley) – The protocol will be finalized by May 2016 and sample typing initiated by June 2016. The sample typing and datafile preparation will be completed by January 2017 and the analysis initiated by June 2017.

IB15-04 Clinical outcomes among hematopoietic stem cell transplant recipients as a function of socioeconomic status and related transcriptome differences (J Knight/JD Rizzo/S Cole) - The protocol will be finalized by May 2016 and samples shipped for typing by June 2016. The sample typing and datafile preparation will be completed by April 2017 and the analysis initiated by June 2017.

IB15-05 Secondary Findings in Exome Sequencing Data (S Gadalla/S Savage) - The goal by June 2016 is to be in manuscript preparation. The goal by June 2017 is to have the manuscript submitted.

IB15-06 Donor Telomere Length and Outcomes after Hematopoietic Stem Cell Transplantation for Acute Leukemia (S Gadalla/S Savage/D Loftus/E Hytopoulos) - The analysis will be completed by May 2016 and manuscript preparation initiated by June 2016. The manuscript will be finalized and submitted by June 2017.

IB15-07 Functional genetic variants of the ST2 gene in pairs of recipient and donors for risk stratification of GVHD and TRM outcomes (S Paczesny) – The goal for June 2016 is to be in sample typing. The goal for June 2017 is to be in manuscript preparation.

21 Not for publication or presentation Attachment 1

IB16-01 The role of HLA-E compatibility in the prognosis of acute leukemia patients undergoing 10/10 HLA matched unrelated HSCT (C Tsamadou / D Fürst / J Mytilineos) – The protocol will be developed and finalized by December 2016. Samples will be selected and shipped for typing by February 2017. The final testing results will be completed and merged with the datafile by June 2017.

IB16-02 Use of HLA structure and function parameters to understand the relationship between HLA disparity and transplant outcomes (LA Baxter-Lowe) – The protocol will be developed and finalized by April 2017. The analysis will be initiated by June 2017.

DEFERRED STUDIES

IB13-06 Role of the complement system in graft-versus-host disease (V Afshar-Kharghan /J Belmont/C Amos) – awaiting funding

IB13-07 Impact of donor signal-regulatory protein alpha (SIRPα) polymorphism on outcome of allogeneic hematopoietic stem cell transplantation (allo-HCT) (A Gassas/J Danska/S Rajakumar) – awaiting funding

PROPOSALS NOT ACCEPTED

PROP1510-05 Impact of HLA-disparity on graft-versus-lymphoma effects in patients with and without graft-versus-host-disease after allogeneic stem cell transplantation (F Ayuketang, U Bacher) –Not accepted because the primary outcome will be difficult to define and population will consist of limited sample sizes in some HLA match groups.

PROP1508-02 Mutational analysis in T-cell acute lymphoblastic leukemia (GA Challen) - Dropped due to insufficient sample size and characterization of the samples. Dr. Challen is working with Dr. Verneris to identify a potential cohort in the University of Minnesota Research Repository for a pilot study.

PROP1511-52 HLA disparity and post-transplant outcomes in AML patients after T-replete HLA- haploidentical hematopoietic cell transplantation: a retrospective analysis (R Romee, A Rashidi) – Dropped due to insufficient sample size when characterizing HLA disparity among a smaller cohort.

PROP1511-122 Impact of human leukocyte antigen haplotypes on outcomes of allogeneic transplantation for multiple myeloma (E Malek, M de Lima) – Dropped due to insufficient number of cases for haplotypes considered as main focus.

PROP1511-139 Investigating influence of minor histocompatibility antigens/alleles and

22 Not for publication or presentation Attachment 1

previously uncharacterized KIR interacting factors on outcomes after HLA matched HSCT (CH Roberts, A Madbouly) – Not accepted due to lack of a specific defined hypothesis for evaluation in the CIBMTR cohort. CIBMTR will collaborate with the investigators and share a de-identified dataset for bioinformatics methods development.

OVERSIGHT ASSIGNMENTS FOR WORKING COMMITTEE LEADERSHIP

Michael Verneris IB11-01 Analysis of the NIMA effect on the outcome of unrelated PBSC/BM transplantation

IB14-05 mtDNA haplotypes and unrelated donor transplant outcomes

IB14-06 Donor-specific anti-HLA antibodies, allele and antigen level HLA mismatches in the outcomes of transplantation of non-malignant diseases with unrelated donors

IB15-04 Clinical outcomes among hematopoietic stem cell transplant recipients as a function of socioeconomic status and related transcriptome differences

IB15-05 Secondary findings in exome sequencing data

IB15-06 Donor telomere length and outcomes after hematopoietic stem cell transplantation for acute leukemia

IB15-07 Functional genetic variants of the ST2 gene in pairs of recipient and donors for risk stratification of GVHD and TRM outcomes

Katharina Fleischhauer IB12-02B Comparative evaluation of the immunogenicity and the diversity model of HLA-DPB1 TCE group matching in unrelated HCT

IB12-02C Prospective assignment of HLA-DPB1 T cell epitope (TCE) group mismatches by functional distance scores compared to the functional TCE assignment algorithm

IB12-03 Effect of genetic ancestry matching on HSCT outcomes

IB13-01 Effect of allele-level HLA-matching after UCB HCT for non- malignant diseases in children

IB13-08 Short and long term survival assessment of post-HSCT transplantation using predictive modeling on a Bayesian network framework

IB13-09 The development of machine learning based classifiers to define the alloreactivity of HLA mismatches in unrelated donor hematopoietic stem cell transplantation

23 Not for publication or presentation Attachment 1

IB14-01 Impact of human leukocyte antigen haplotypes on outcomes of allogeneic transplantation for B-cell non-Hodgkin lymphomas

IB14-02 Structural/Functional models of HLA for data mining of permissive mismatching in allogeneic hematopoietic stem cell transplantation

IB14-08 Development and validation of a clinical unrelated donor selection score

IB15-01 The impact of single nucleotide gene polymorphisms (SNP) in the gamma block of the major histocompatibility complex (MHC) on unrelated donor hematopoietic cell transplants (HCT) for hematological malignancies

Katharine Hsu IB12-04A Determining the effects of HLA-C KIR ligand expression on outcomes of unrelated hematopoietic stem cell transplantation

IB12-04B Effect of HLA-C allele matching in the context of recipient HLA- C-encoded KIR ligand grouping (C1 or C2) on the outcome of unrelated HCT

IB15-02 Natural killer cell genomics and outcomes after allogeneic transplantation for chronic lymphocytic leukemia

IB15-03 Killer Immunoglobulin Receptor (KIR) gene content and pediatric acute leukemia transplant outcomes

IB09-04 D/R gene polymorphisms of drug metabolisms and innate immune response post allele matched matched unrelated donor HCT

IB10-01b Telomere length telomerase polymorphisms in SAA

IB10-01c Telomere length telomerase polymorphism in SAA-exome analysis and mosaicism

IB13-04 Discrepancy analysis of microsatellite loci as a proxy measure for ancestral differentiation between donors and recipients: correlation between high scores and poorer overall survival in high resolution matched unrelated donor transplantation

IB13-05 The impact of MHC Class I chain-related gene A donor-recipient mismatches and MICA-129 polymorphism on unrelated donor hematopoietic stem cell transplants for hematological malignancies

IB14-03 The prognostic impact of somatic mutations and levels of CXC chemokine ligands on post hematopoietic cell transplantation outcomes in patients with myelodysplastic syndromes

24 Not for publication or presentation Attachment 1

IB14-04 Assessing the similarity of the T cell receptor repertoire in allogeneic hematopoietic stem cell recipients with the same single human leukocyte mismatches

25 Not for publication or presentation Attachment 2

Accrual Summary for Immunobiology Working Committee

Comprehensive Report Form (CRF) data

CIBMTR HLA- CIBMTR CIBMTR CIBMTR identical alternative unrelated unrelated sibling related (non-US) (US) Number of CRF patients 45973 8299 9161 40209 Number of centers 505 433 225 198 Recipient age at transplant 0-9 6215 (14) 2145 (26) 2096 (23) 6686 (17) 10-19 7557 (16) 1357 (16) 1515 (17) 4825 (12) 20-29 7915 (17) 1183 (14) 1387 (15) 4651 (12) 30-39 8583 (19) 1059 (13) 1528 (17) 5153 (13) 40-49 7988 (17) 1024 (12) 1331 (15) 6224 (15) 50-59 7707 (17) 1528 (18) 1303 (14) 12669 (32) Unknown 8 (N/A) 3 (N/A) 1 (N/A) 1 (N/A) Median (range) 32 (0-82) 25 (0-83) 27 (0-76) 38 (0-83) Recipient race / ethnicity Caucasian, non-Hispanic 35325 (79) 5940 (75) 6990 (78) 31250 (80) African-American, non-Hispanic 1998 (04) 618 (08) 63 (01) 2962 (08) Asian, non-Hispanic 4357 (10) 815 (10) 1386 (16) 1251 (03) Pacific Islander, non-Hispanic 54 (<1) 12 (<1) 43 (<1) 76 (<1) Native American, non-Hispanic 78 (<1) 37 (<1) 35 (<1) 141 (<1) Hispanic, Caucasian 1025 (02) 264 (03) 288 (03) 2610 (07) Hispanic, African-American 61 (<1) 15 (<1) 14 (<1) 102 (<1) Hispanic, Asian 11 (<1) 3 (<1) 3 (<1) 16 (<1) Hispanic, Pacific Islander 6 (<1) 1 (<1) 0 10 (<1) Hispanic, Native American 18 (<1) 1 (<1) 3 (<1) 32 (<1) Hispanic, race unknown 144 (<1) 28 (<1) 20 (<1) 734 (02) Other 1421 (03) 224 (03) 82 (01) 100 (<1) Unknown 1475 (N/A) 341 (N/A) 234 (N/A) 925 (N/A) Recipient sex Male 26856 (58) 5014 (60) 5429 (59) 23510 (58) Female 19117 (42) 3285 (40) 3732 (41) 16699 (42) Karnofsky score 10-80 12413 (27) 2631 (32) 2484 (27) 11619 (29) 90-100 31958 (70) 5146 (62) 6370 (70) 26591 (66) Missing 1602 (03) 522 (06) 307 (03) 1999 (05) High-resolution HLA matches available out of 8 ≤5/8 33 (04) 519 (79) 199 (14) 4701 (15) 6/8 9 (01) 37 (06) 169 (12) 3100 (10) 7/8 31 (04) 60 (09) 387 (28) 6275 (21) 8/8 662 (90) 43 (07) 648 (46) 16448 (54) Unknown 45238 (N/A) 7640 (N/A) 7758 (N/A) 9685 (N/A)

26 Not for publication or presentation Attachment 2

CIBMTR HLA- CIBMTR CIBMTR CIBMTR identical alternative unrelated unrelated sibling related (non-US) (US) Graft type Bone marrow 31318 (68) 5412 (65) 5219 (57) 16207 (40) Peripheral blood 14119 (31) 2737 (33) 2219 (24) 14560 (36) Umbilical cord blood 174 (<1) 41 (<1) 1688 (18) 9228 (23) BM+PBSC 228 (<1) 71 (01) 4 (<1) 8 (<1) BM+UCB 87 (<1) 11 (<1) 2 (<1) 0 PBSC+UCB 3 (<1) 6 (<1) 6 (<1) 166 (<1) Others 44 (<1) 21 (<1) 23 (<1) 40 (<1) Conditioning regimen Myeloablative 38141 (83) 6354 (77) 6967 (76) 27138 (67) RIC 2907 (06) 524 (06) 1084 (12) 6795 (17) Nonmyeloablative 1737 (04) 765 (09) 485 (05) 3851 (10) Other 3188 (07) 656 (08) 625 (07) 2425 (06) Donor age at donation To be determined / N/A 1474 (03) 326 (04) 1351 (15) 1571 (04) 0-9 5415 (12) 502 (06) 1314 (14) 8222 (20) 10-19 7440 (16) 911 (11) 123 (01) 876 (02) 20-29 8160 (18) 1614 (19) 1840 (20) 10632 (26) 30-39 8466 (18) 2006 (24) 2407 (26) 10141 (25) 40-49 7677 (17) 1536 (19) 1696 (19) 6887 (17) 50+ 7341 (16) 1404 (17) 430 (05) 1880 (05) Median (range) 31 (0-85) 34 (0-81) 32 (0-80) 30 (0-69) Disease at transplant AML 12097 (26) 2106 (25) 2261 (25) 12789 (32) ALL 7123 (15) 1434 (17) 1923 (21) 6544 (16) Other leukemia 828 (02) 111 (01) 155 (02) 1219 (03) CML 7862 (17) 1017 (12) 1776 (19) 4414 (11) MDS 3791 (08) 684 (08) 948 (10) 6075 (15) Other acute leukemia 360 (01) 95 (01) 117 (01) 405 (01) NHL 3171 (07) 569 (07) 326 (04) 2960 (07) Hodgkins Lymphoma 394 (01) 106 (01) 45 (<1) 648 (02) Plasma Cell Disorders, MM 1431 (03) 212 (03) 64 (01) 396 (01) Other malignancies 343 (01) 68 (01) 31 (<1) 93 (<1) Breast cancer 77 (<1) 26 (<1) 1 (<1) 7 (<1) SAA 4343 (09) 544 (07) 478 (05) 1272 (03) Inherited abnormalities erythrocyte diff fxn 3061 (07) 396 (05) 310 (03) 852 (02) SCIDs 644 (01) 702 (08) 352 (04) 1051 (03) Inherited abnormalities of platelets 24 (<1) 8 (<1) 13 (<1) 62 (<1) Inherited disorders of metabolism 267 (01) 163 (02) 233 (03) 906 (02) Histiocytic disorders 118 (<1) 51 (01) 117 (01) 429 (01) Autoimmune disorders 19 (<1) 4 (<1) 4 (<1) 24 (<1) Other 20 (<1) 3 (<1) 7 (<1) 62 (<1) Unknown 0 (N/A) 0 (N/A) 0 (N/A) 1 (N/A)

27 Not for publication or presentation Attachment 2

CIBMTR HLA- CIBMTR CIBMTR CIBMTR identical alternative unrelated unrelated sibling related (non-US) (US) Disease status at transplant Early 17169 (37) 2057 (25) 2904 (32) 12420 (31) Intermediate 6212 (14) 1411 (17) 1978 (22) 7876 (20) Advanced 5701 (12) 1358 (16) 1279 (14) 7359 (18) Other 16891 (37) 3473 (42) 3000 (33) 12554 (31) Donor / recipient CMV serostatus Negative / negative 10318 (22) 1805 (22) 2080 (23) 8986 (22) Negative / positive 6982 (15) 1201 (14) 1850 (20) 9572 (24) Positive / negative 4235 (09) 984 (12) 1059 (12) 3708 (09) Positive / positive 17196 (37) 2792 (34) 2109 (23) 6443 (16) Unknown 7242 (16) 1517 (18) 2063 (23) 11500 (29) GvHD Prophylaxis Ex vivo T-cell depletion 3405 (07) 1904 (23) 674 (07) 3384 (08) CD34 selection 470 (01) 347 (04) 87 (01) 787 (02) Cyclophosphamide 494 (01) 722 (09) 40 (<1) 309 (01) Tacrolimus + MMF ± others 1114 (02) 395 (05) 131 (01) 5560 (14) Tacrolimus + MTX ± others (except MMF) 4045 (09) 352 (04) 546 (06) 10516 (26) Tacrolimus + others (except MTX, MMF) 598 (01) 44 (01) 66 (01) 1641 (04) Tacrolimus alone 311 (01) 58 (01) 27 (<1) 814 (02) CSA + MMF ± others (except Tacrolimus) 1487 (03) 141 (02) 890 (10) 5218 (13) CSA + MTX ± others (except Tacrolimus, 20970 (46) 2087 (25) 4833 (53) 7691 (19) MMF) CSA + others (except Tacrolimus, MTX, 3661 (08) 285 (03) 1010 (11) 2002 (05) MMF) CSA alone 5046 (11) 452 (05) 421 (05) 387 (01) Other GVHD prophylaxis 3091 (07) 307 (04) 66 (01) 511 (01) Missing 1281 (03) 1205 (15) 370 (04) 1389 (03) Donor / recipient sex match Male / male 8011 (33) 1620 (36) 2632 (38) 12648 (38) Male / female 5327 (22) 749 (17) 1558 (22) 7877 (23) Female / male 6343 (26) 1093 (24) 1481 (21) 6952 (21) Female / female 4859 (20) 1000 (22) 1281 (18) 6124 (18) Unknown 21433 (N/A) 3837 (N/A) 2209 (N/A) 6608 (N/A) Year of transplant 1964-1985 4813 (10) 888 (11) 42 (<1) 11 (<1) 1986 1374 (03) 260 (03) 14 (<1) 18 (<1) 1987 1464 (03) 249 (03) 31 (<1) 34 (<1) 1988 1622 (04) 246 (03) 53 (01) 95 (<1) 1989 1850 (04) 258 (03) 100 (01) 185 (<1) 1990 1945 (04) 318 (04) 136 (01) 297 (01) 1991 1891 (04) 255 (03) 173 (02) 417 (01) 1992 1982 (04) 276 (03) 228 (02) 493 (01) 1993 1996 (04) 285 (03) 236 (03) 598 (01)

28 Not for publication or presentation Attachment 2

CIBMTR HLA- CIBMTR CIBMTR CIBMTR identical alternative unrelated unrelated sibling related (non-US) (US) 1994 1846 (04) 272 (03) 254 (03) 750 (02) 1995 1936 (04) 342 (04) 343 (04) 905 (02) 1996 1992 (04) 339 (04) 436 (05) 1047 (03) 1997 1684 (04) 310 (04) 413 (05) 1131 (03) 1998 1543 (03) 228 (03) 474 (05) 1165 (03) 1999 1386 (03) 216 (03) 469 (05) 1219 (03) 2000 1500 (03) 217 (03) 516 (06) 1287 (03) 2001 1484 (03) 238 (03) 518 (06) 1382 (03) 2002 1425 (03) 203 (02) 481 (05) 1569 (04) 2003 1217 (03) 175 (02) 512 (06) 1723 (04) 2004 1453 (03) 151 (02) 623 (07) 1934 (05) 2005 1497 (03) 184 (02) 598 (07) 2106 (05) 2006 1261 (03) 155 (02) 493 (05) 2424 (06) 2007 753 (02) 95 (01) 332 (04) 2673 (07) 2008 1099 (02) 241 (03) 316 (03) 2323 (06) 2009 928 (02) 162 (02) 268 (03) 2533 (06) 2010 528 (01) 61 (01) 151 (02) 1823 (05) 2011 342 (01) 67 (01) 109 (01) 1458 (04) 2012 353 (01) 84 (01) 176 (02) 1389 (03) 2013 688 (01) 344 (04) 212 (02) 2133 (05) 2014 1005 (02) 452 (05) 232 (03) 2370 (06) 2015 856 (02) 509 (06) 151 (02) 2128 (05) 2016 260 (01) 219 (03) 71 (01) 589 (01) Follow-up among survivors, months N Eval 22083 3398 4038 15077 Median (range) 93 (0-465) 56 (0-498) 62 (0-316) 68 (0-343)

29 Not for publication or presentation Attachment 2

Accrual Summary for First Allogeneic Transplants in CRF and TED with Samples Available for Recipient and/or Donor through the NMDP for Adult Unrelated Donor Transplants

Samples Available for Samples Samples Recipient and Available for Available for Donor Recipient Only Donor Only Number of patients 26393 5867 4389 CRF 18009 (68) 4007 (68) 2971 (68) TED 8384 (32) 1860 (32) 1418 (32) Number of centers 228 199 268 Recipient age at transplant 0-9 2742 (10) 607 (10) 656 (15) 10-19 2828 (11) 620 (11) 547 (12) 20-29 3118 (12) 665 (11) 572 (13) 30-39 3413 (13) 751 (13) 597 (14) 40-49 4367 (17) 985 (17) 707 (16) 50-59 9925 (38) 2239 (38) 1310 (30) Median (range) 43 (0-84) 43 (0-79) 37 (0-76) Recipient race / ethnicity Caucasian, non-Hispanic 22015 (85) 4874 (85) 3521 (84) African-American, non-Hispanic 1250 (05) 295 (05) 233 (06) Asian, non-Hispanic 559 (02) 134 (02) 126 (03) Pacific Islander, non-Hispanic 38 (<1) 6 (<1) 4 (<1) Native American, non-Hispanic 83 (<1) 20 (<1) 13 (<1) Hispanic, Caucasian 1345 (05) 275 (05) 164 (04) Hispanic, African-American 42 (<1) 14 (<1) 5 (<1) Hispanic, Asian 7 (<1) 2 (<1) 1 (<1) Hispanic, Pacific Islander 8 (<1) 0 0 Hispanic, Native American 10 (<1) 2 (<1) 1 (<1) Hispanic, race unknown 440 (02) 109 (02) 98 (02) Other 46 (<1) 21 (<1) 20 (<1) Unknown 550 (N/A) 115 (N/A) 203 (N/A) Recipient sex Male 15300 (58) 3491 (60) 2613 (60) Female 11093 (42) 2376 (40) 1776 (40) Karnofsky score 10-80 8003 (30) 1887 (32) 1196 (27) 90-100 17185 (65) 3778 (64) 2715 (62) Missing 1205 (05) 202 (03) 478 (11) High-resolution HLA matches available out of 8 ≤5/8 761 (03) 24 (01) 27 (01) 6/8 1449 (06) 56 (02) 98 (04) 7/8 5433 (22) 740 (21) 537 (22) 8/8 16778 (69) 2730 (77) 1757 (73) Unknown 1972 (N/A) 2317 (N/A) 1970 (N/A)

30 Not for publication or presentation Attachment 2

Samples Available for Samples Samples Recipient and Available for Available for Donor Recipient Only Donor Only Graft type Bone marrow 11821 (45) 2832 (48) 2352 (54) Peripheral blood 14566 (55) 3035 (52) 2034 (46) BM+PBSC 6 (<1) 0 3 (<1) Conditioning regimen Myeloablative 16053 (61) 3586 (61) 2789 (64) RIC 3227 (12) 739 (13) 508 (12) Nonmyeloablative 1181 (04) 208 (04) 179 (04) Other 5932 (22) 1334 (23) 913 (21) Donor age at donation To be determined / N/A 3717 (14) 3041 (52) 2035 (46) 10-19 475 (02) 32 (01) 26 (01) 20-29 8318 (32) 895 (15) 646 (15) 30-39 7310 (28) 1064 (18) 882 (20) 40-49 5090 (19) 660 (11) 621 (14) 50+ 1483 (06) 175 (03) 179 (04) Median (range) 33 (0-69) 35 (18-65) 36 (3-59) Disease at transplant AML 9204 (35) 2118 (36) 1435 (33) ALL 4414 (17) 918 (16) 818 (19) Other leukemia 1040 (04) 196 (03) 144 (03) CML 3033 (11) 755 (13) 542 (12) MDS 3463 (13) 761 (13) 537 (12) Other acute leukemia 272 (01) 70 (01) 38 (01) NHL 2071 (08) 427 (07) 268 (06) Hodgkins Lymphoma 175 (01) 36 (01) 35 (01) Plasma Cell Disorders, MM 201 (01) 56 (01) 30 (01) Other malignancies 40 (<1) 10 (<1) 11 (<1) Breast cancer 5 (<1) 0 0 SAA 931 (04) 207 (04) 179 (04) Inherited abnormalities erythrocyte diff fxn 505 (02) 93 (02) 79 (02) SCIDs 484 (02) 97 (02) 119 (03) Inherited abnormalities of platelets 30 (<1) 5 (<1) 6 (<1) Inherited disorders of metabolism 232 (01) 57 (01) 74 (02) Histiocytic disorders 243 (01) 48 (01) 60 (01) Autoimmune disorders 10 (<1) 4 (<1) 2 (<1) Other 40 (<1) 9 (<1) 12 (<1) Disease status at transplant Early 9500 (36) 2138 (36) 1389 (32) Intermediate 4700 (18) 1078 (18) 862 (20) Advanced 4107 (16) 928 (16) 696 (16) Other 8086 (31) 1723 (29) 1442 (33)

31 Not for publication or presentation Attachment 2

Samples Available for Samples Samples Recipient and Available for Available for Donor Recipient Only Donor Only Donor/Recipient CMV serostatus Negative / negative 6847 (26) 1310 (22) 891 (20) Negative / positive 7405 (28) 1339 (23) 984 (22) Positive / negative 2483 (09) 503 (09) 345 (08) Positive / positive 4796 (18) 871 (15) 569 (13) Unknown 4862 (18) 1844 (31) 1600 (36) GvHD prophylaxis Ex vivo T-cell depletion 2013 (08) 503 (09) 416 (09) CD34 selection 439 (02) 85 (01) 75 (02) Cyclophosphamide 307 (01) 76 (01) 58 (01) Tacrolimus + MMF ± others 2978 (11) 589 (10) 356 (08) Tacrolimus + MTX ± others (except MMF) 10105 (38) 2146 (37) 1288 (29) Tacrolimus + others (except MTX, MMF) 1213 (05) 300 (05) 138 (03) Tacrolimus alone 532 (02) 139 (02) 76 (02) CSA + MMF ± others (except Tacrolimus) 1567 (06) 304 (05) 295 (07) CSA + MTX ± others (except Tacrolimus, MMF) 5569 (21) 1362 (23) 1281 (29) CSA + others (except Tacrolimus, MTX, MMF) 558 (02) 117 (02) 154 (04) CSA alone 217 (01) 30 (01) 75 (02) Other GVHD prophylaxis 298 (01) 75 (01) 41 (01) Missing 597 (02) 141 (02) 136 (03) Donor / recipient sex match Male / male 9129 (40) 1830 (41) 1303 (40) Male / female 5630 (25) 1073 (24) 737 (22) Female / male 4138 (18) 814 (18) 660 (20) Female / female 3973 (17) 734 (16) 595 (18) Unknown 3523 (N/A) 1416 (N/A) 1094 (N/A) Year of transplant 1987 1 (<1) 0 1 (<1) 1988 54 (<1) 10 (<1) 9 (<1) 1989 147 (01) 4 (<1) 6 (<1) 1990 205 (01) 19 (<1) 30 (01) 1991 297 (01) 44 (01) 57 (01) 1992 324 (01) 40 (01) 92 (02) 1993 356 (01) 72 (01) 117 (03) 1994 444 (02) 115 (02) 112 (03) 1995 486 (02) 184 (03) 119 (03) 1996 533 (02) 231 (04) 127 (03) 1997 620 (02) 228 (04) 147 (03) 1998 578 (02) 268 (05) 132 (03) 1999 677 (03) 219 (04) 136 (03) 2000 795 (03) 155 (03) 133 (03) 2001 767 (03) 130 (02) 144 (03)

32 Not for publication or presentation Attachment 2

Samples Available for Samples Samples Recipient and Available for Available for Donor Recipient Only Donor Only 2002 653 (02) 120 (02) 367 (08) 2003 799 (03) 149 (03) 325 (07) 2004 1032 (04) 238 (04) 218 (05) 2005 1176 (04) 203 (03) 185 (04) 2006 1338 (05) 210 (04) 201 (05) 2007 1477 (06) 161 (03) 208 (05) 2008 1411 (05) 176 (03) 95 (02) 2009 1328 (05) 235 (04) 79 (02) 2010 1453 (06) 237 (04) 72 (02) 2011 1477 (06) 308 (05) 66 (02) 2012 1462 (06) 229 (04) 209 (05) 2013 1711 (06) 418 (07) 293 (07) 2014 1294 (05) 543 (09) 232 (05) 2015 2366 (09) 697 (12) 318 (07) 2016 1132 (04) 224 (04) 159 (04) Follow-up among survivors, months N Eval 10919 2422 1743 Median (range) 51 (0-319) 37 (0-319) 56 (3-339)

33 Not for publication or presentation Attachment 2

Accrual Summary for First Allogeneic Transplants in CRF and TED with Samples Available for Recipient and/or Cord Blood Unit(s) through the NMDP for Cord Blood Transplants

Samples Available for Samples Samples Recipient Available for Available for and Donor Recipient Only Donor Only Number of patients 3649 1222 529 CRF 3185 (87) 909 (74) 380 (72) TED 464 (13) 313 (26) 149 (28) Number of centers 132 119 126 Recipient age at transplant 0-9 1124 (31) 415 (34) 196 (37) 10-19 566 (16) 145 (12) 76 (14) 20-29 368 (10) 100 (08) 43 (08) 30-39 334 (09) 112 (09) 37 (07) 40-49 359 (10) 126 (10) 48 (09) 50-59 898 (25) 324 (27) 129 (24) Median (range) 24 (0-81) 24 (0-74) 19 (0-74) Recipient race / ethnicity Caucasian, non-Hispanic 2095 (60) 670 (58) 315 (63) African-American, non-Hispanic 496 (14) 166 (14) 71 (14) Asian, non-Hispanic 195 (06) 81 (07) 30 (06) Pacific Islander, non-Hispanic 12 (<1) 4 (<1) 6 (01) Native American, non-Hispanic 24 (01) 9 (01) 1 (<1) Hispanic, Caucasian 656 (19) 208 (18) 72 (14) Hispanic, African-American 13 (<1) 12 (01) 1 (<1) Hispanic, Asian 1 (<1) 3 (<1) 0 Hispanic, Pacific Islander 2 (<1) 2 (<1) 0 Hispanic, Native American 16 (<1) 2 (<1) 1 (<1) Hispanic, race unknown 1 (<1) 3 (<1) 1 (<1) Other 0 0 1 (<1) Unknown 138 (N/A) 62 (N/A) 30 (N/A) Recipient sex Male 1999 (55) 694 (57) 289 (55) Female 1650 (45) 528 (43) 240 (45) Karnofsky score 10-80 645 (18) 139 (11) 78 (15) 90-100 2244 (61) 652 (53) 270 (51) Missing 760 (21) 431 (35) 181 (34) High-resolution HLA matches available out of 8 ≤5/8 1800 (57) 532 (56) 224 (53) 6/8 719 (23) 238 (25) 102 (24) 7/8 394 (13) 127 (13) 67 (16) 8/8 230 (07) 59 (06) 33 (08) Unknown 506 (N/A) 266 (N/A) 103 (N/A)

34 Not for publication or presentation Attachment 2

Samples Available for Samples Samples Recipient Available for Available for and Donor Recipient Only Donor Only Graft type Umbilical cord blood 3621 (99) 1188 (97) 515 (97) BM+UCB 0 0 1 (<1) PBSC+UCB 28 (01) 34 (03) 13 (02) Conditioning regimen Myeloablative 2424 (66) 791 (65) 331 (63) RIC 530 (15) 121 (10) 50 (09) Nonmyeloablative 379 (10) 156 (13) 69 (13) Other 316 (09) 154 (13) 79 (15) Donor age at donation To be determined / N/A 971 (27) 1198 (98) 523 (99) 0-9 2562 (70) 0 0 10-19 105 (03) 13 (01) 4 (01) 20-29 0 1 (<1) 0 30-39 7 (<1) 4 (<1) 1 (<1) 40-49 0 2 (<1) 0 50+ 4 (<1) 4 (<1) 1 (<1) Median (range) 3 (0-68) 3 (0-68) 3 (0-52) Disease at transplant AML 1426 (39) 478 (39) 181 (34) ALL 815 (22) 278 (23) 131 (25) Other leukemia 85 (02) 24 (02) 6 (01) CML 113 (03) 24 (02) 12 (02) MDS 257 (07) 82 (07) 30 (06) Other acute leukemia 73 (02) 13 (01) 13 (02) NHL 186 (05) 80 (07) 30 (06) Hodgkins Lymphoma 15 (<1) 7 (01) 0 Plasma Cell Disorders, MM 7 (<1) 5 (<1) 2 (<1) Other malignancies 7 (<1) 0 0 SAA 62 (02) 17 (01) 9 (02) Inherited abnormalities erythrocyte diff fxn 102 (03) 50 (04) 13 (02) SCIDs 172 (05) 60 (05) 36 (07) Inherited abnormalities of platelets 12 (<1) 5 (<1) 4 (01) Inherited disorders of metabolism 215 (06) 73 (06) 42 (08) Histiocytic disorders 83 (02) 21 (02) 14 (03) Autoimmune disorders 9 (<1) 2 (<1) 3 (01) Other 10 (<1) 3 (<1) 3 (01) Disease status at transplant Early 1359 (37) 405 (33) 166 (31) Intermediate 779 (21) 272 (22) 118 (22) Advanced 396 (11) 126 (10) 47 (09) Other 1115 (31) 419 (34) 198 (37)

35 Not for publication or presentation Attachment 2

Samples Available for Samples Samples Recipient Available for Available for and Donor Recipient Only Donor Only Donor / recipient CMV serostatus Negative / negative 550 (15) 79 (06) 41 (08) Negative / positive 926 (25) 115 (09) 48 (09) Positive / negative 225 (06) 72 (06) 28 (05) Positive / positive 531 (15) 195 (16) 53 (10) Unknown 1417 (39) 761 (62) 359 (68) GvHD Prophylaxis Ex vivo T-cell depletion 4 (<1) 2 (<1) 0 CD34 selection 44 (01) 31 (03) 19 (04) Cyclophosphamide 2 (<1) 1 (<1) 2 (<1) Tacrolimus + MMF ± others 1006 (28) 356 (29) 112 (21) Tacrolimus + MTX ± others (except MMF) 143 (04) 47 (04) 21 (04) Tacrolimus + others (except MTX, MMF) 163 (04) 51 (04) 28 (05) Tacrolimus alone 69 (02) 23 (02) 11 (02) CSA + MMF ± others (except Tacrolimus) 1650 (45) 520 (43) 216 (41) CSA + MTX ± others (except Tacrolimus, MMF) 61 (02) 29 (02) 18 (03) CSA + others (except Tacrolimus, MTX, MMF) 227 (06) 100 (08) 61 (12) CSA alone 34 (01) 8 (01) 8 (02) Other GVHD prophylaxis 69 (02) 19 (02) 7 (01) Missing 177 (05) 35 (03) 26 (05) Donor / recipient sex match Male / male 937 (29) 250 (29) 89 (26) Male / female 756 (24) 186 (22) 76 (22) Female / male 820 (26) 239 (28) 95 (28) Female / female 687 (21) 185 (22) 80 (24) Unknown 449 (N/A) 362 (N/A) 189 (N/A) Year of transplant 2000 0 2 (<1) 4 (01) 2001 5 (<1) 16 (01) 4 (01) 2002 1 (<1) 11 (01) 2 (<1) 2003 10 (<1) 23 (02) 9 (02) 2004 19 (01) 24 (02) 8 (02) 2005 69 (02) 28 (02) 15 (03) 2006 130 (04) 79 (06) 16 (03) 2007 312 (09) 21 (02) 62 (12) 2008 389 (11) 19 (02) 49 (09) 2009 478 (13) 46 (04) 37 (07) 2010 529 (14) 64 (05) 39 (07) 2011 459 (13) 110 (09) 33 (06) 2012 265 (07) 164 (13) 38 (07) 2013 322 (09) 180 (15) 40 (08) 2014 199 (05) 135 (11) 47 (09)

36 Not for publication or presentation Attachment 2

Samples Available for Samples Samples Recipient Available for Available for and Donor Recipient Only Donor Only 2015 321 (09) 222 (18) 90 (17) 2016 141 (04) 78 (06) 36 (07) Follow-up among survivors, months N Eval 1760 615 258 Median (range) 44 (1-149) 25 (0-170) 36 (0-193)

37 Not for publication or presentation Attachment 2

Accrual Summary for First Allogeneic Transplants in CRF and TED with Samples Available for Recipient and/or Donor through the NMDP for Adult Related Donor Transplants

Samples Samples Samples Available for Available for Available Recipient and Recipient for Donor Donor Only Only Number of patients 4473 595 239 CRF 1577 (35) 144 (24) 74 (31) TED 2896 (65) 451 (76) 165 (69) Number of centers 76 59 45 Recipient age at transplant 0-9 402 (09) 32 (05) 31 (13) 10-19 430 (10) 33 (06) 24 (10) 20-29 349 (08) 65 (11) 22 (09) 30-39 345 (08) 63 (11) 16 (07) 40-49 656 (15) 93 (16) 38 (16) 50-59 2291 (51) 309 (52) 108 (45) Median (range) 51 (0-78) 51 (0-76) 48 (0-74) Recipient race / ethnicity Caucasian, non-Hispanic 2922 (67) 351 (62) 141 (61) African-American, non-Hispanic 515 (12) 47 (08) 31 (13) Asian, non-Hispanic 200 (05) 42 (07) 11 (05) Pacific Islander, non-Hispanic 15 (<1) 2 (<1) 1 (<1) Native American, non-Hispanic 13 (<1) 1 (<1) 0 Hispanic, Caucasian 644 (15) 120 (21) 45 (20) Hispanic, African-American 19 (<1) 0 0 Hispanic, Asian 1 (<1) 0 0 Hispanic, Pacific Islander 7 (<1) 0 0 Hispanic, Native American 5 (<1) 2 (<1) 1 (<1) Unknown 132 (N/A) 30 (N/A) 9 (N/A) Recipient sex Male 2614 (58) 349 (59) 141 (59) Female 1859 (42) 246 (41) 98 (41) Karnofsky score 10-80 1400 (31) 202 (34) 66 (28) 90-100 2909 (65) 374 (63) 163 (68) Missing 164 (04) 19 (03) 10 (04) Graft type Bone marrow 1156 (26) 117 (20) 62 (26) Peripheral blood 3266 (73) 464 (78) 162 (68) Umbilical cord blood 12 (<1) 8 (01) 10 (04) BM+PBSC 5 (<1) 2 (<1) 0 BM+UCB 15 (<1) 3 (01) 1 (<1) PBSC+UCB 4 (<1) 0 2 (01) Others 15 (<1) 1 (<1) 2 (01)

38 Not for publication or presentation Attachment 2

Samples Samples Samples Available for Available for Available Recipient and Recipient for Donor Donor Only Only Conditioning regimen Myeloablative 1579 (35) 227 (38) 84 (35) RIC 491 (11) 40 (07) 28 (12) Nonmyeloablative 197 (04) 31 (05) 14 (06) Other 2206 (49) 297 (50) 113 (47) Donor age at donation To be determined / N/A 2919 (65) 458 (77) 174 (73) 0-9 118 (03) 10 (02) 5 (02) 10-19 114 (03) 3 (01) 9 (04) 20-29 183 (04) 15 (03) 6 (03) 30-39 175 (04) 16 (03) 6 (03) 40-49 246 (05) 25 (04) 9 (04) 50+ 718 (16) 68 (11) 30 (13) Median (range) 48 (0-81) 50 (0-79) 47 (2-73) Disease at transplant AML 1452 (32) 209 (35) 68 (28) ALL 657 (15) 100 (17) 35 (15) Other leukemia 119 (03) 24 (04) 6 (03) CML 171 (04) 17 (03) 10 (04) MDS 704 (16) 84 (14) 28 (12) Other acute leukemia 57 (01) 8 (01) 3 (01) NHL 528 (12) 65 (11) 33 (14) Hodgkins Lymphoma 103 (02) 16 (03) 10 (04) Plasma Cell Disorders, MM 145 (03) 25 (04) 11 (05) Other malignancies 14 (<1) 0 0 Breast cancer 1 (<1) 0 0 SAA 182 (04) 18 (03) 13 (05) Inherited abnormalities erythrocyte diff fxn 222 (05) 24 (04) 14 (06) SCIDs 75 (02) 3 (01) 5 (02) Inherited abnormalities of platelets 6 (<1) 1 (<1) 0 Inherited disorders of metabolism 8 (<1) 1 (<1) 0 Histiocytic disorders 22 (<1) 0 1 (<1) Autoimmune disorders 1 (<1) 0 1 (<1) Other 6 (<1) 0 1 (<1) Disease status at transplant Early 1540 (34) 221 (37) 72 (30) Intermediate 455 (10) 51 (09) 21 (09) Advanced 620 (14) 88 (15) 29 (12) Other 1858 (42) 235 (39) 117 (49) Donor / recipient CMV serostatus Negative / negative 652 (15) 68 (11) 24 (10) Negative / positive 768 (17) 77 (13) 36 (15)

39 Not for publication or presentation Attachment 2

Samples Samples Samples Available for Available for Available Recipient and Recipient for Donor Donor Only Only Positive / negative 346 (08) 29 (05) 17 (07) Positive / positive 1336 (30) 162 (27) 67 (28) Unknown 1371 (31) 259 (44) 95 (40) GvHD Prophylaxis Ex vivo T-cell depletion 50 (01) 5 (01) 6 (03) CD34 selection 50 (01) 13 (02) 2 (01) Cyclophosphamide 410 (09) 49 (08) 14 (06) Tacrolimus + MMF ± others 617 (14) 44 (07) 28 (12) Tacrolimus + MTX ± others (except MMF) 1959 (44) 219 (37) 93 (39) Tacrolimus + others (except MTX, MMF) 406 (09) 143 (24) 30 (13) Tacrolimus alone 53 (01) 8 (01) 1 (<1) CSA + MMF ± others (except Tacrolimus) 139 (03) 14 (02) 9 (04) CSA + MTX ± others (except Tacrolimus, MMF) 349 (08) 31 (05) 26 (11) CSA + others (except Tacrolimus, MTX, MMF) 39 (01) 8 (01) 3 (01) CSA alone 37 (01) 4 (01) 1 (<1) Other GVHD prophylaxis 48 (01) 9 (02) 2 (01) Missing 316 (07) 48 (8) 24 (10) Donor / recipient sex match Male / male 1390 (33) 198 (36) 65 (31) Male / female 917 (22) 97 (18) 46 (22) Female / male 1088 (26) 127 (23) 55 (26) Female / female 863 (20) 131 (24) 43 (21) Unknown 215 (N/A) 42 (N/A) 30 (N/A) Year of transplant 2007 23 (01) 4 (01) 0 2008 147 (03) 10 (02) 8 (03) 2009 203 (05) 28 (05) 6 (03) 2010 267 (06) 39 (07) 9 (04) 2011 396 (09) 55 (09) 27 (11) 2012 528 (12) 74 (12) 36 (15) 2013 677 (15) 107 (18) 31 (13) 2014 870 (19) 120 (20) 51 (21) 2015 824 (18) 97 (16) 40 (17) 2016 538 (12) 61 (10) 31 (13) Follow-up among survivors, months N Eval 2753 348 138 Median (range) 24 (0-113) 24 (3-101) 24 (3-95)

40 Not for publication or presentation Attachment 3

Proposal 1612-05

Title: The impact of HLA-DPB1 level of expression on clinical outcomes of HCT

Medhat Askar: Baylor University Medical Center Marcelo Fernandez-Viña: Stanford University

Hypothesis: In this study we will tested the hypothesis that the GVHD risk correlates with the level of expression of HLA-DPB1 antigen predicted by one or more of multiple polymorphisms of the mismatched HLA-DPB1 allele(s) of the recipients and/or their unrelated donors.

Scientific impact: Disparities at HLA-DPB1 locus among hematopoietic cell transplantation (HCT) donor/recipient pairs exist more than any other HLA locus. It is estimated that only about 10 - 15 % of 8/8 HLA matched donors are matched at HLA-DPB1 due to weak linkage disequilibrium between HLA-DPB1 and the other HLA class II loci1. This makes traditional allele level at HLA-DPB1 matching in a fashion similar to other loci not feasible and calls for alternative algorithms that predict which mismatches are more permissive and which ones are not and should be avoided. Specific aims: • Investigate the association between HCT outcomes (aGVHD [any & III-IV], cGVHD [any & extensive], Relapse, TRM, neutrophil & platelet reconstitution, complete donor chimerism & OS) in a large cohort of HLA-DPB1 typed by full gene NGS. • Investigate the association of each variant published or identified by the PI in association with HLA-DPB1 level of expression with the study outcomes. • Analyze the concordance among different genetic variance predicting HLA-DPB1 level of expression • Investigate whether certain HLA-DPB1 alleles are consistently carrying sequences associated with low or high expression

Scientific justification: Numerous published studies investigating the impact of HLA-DPB1 mismatches on clinical outcomes of HCT suggested that not all such mismatches are equally deleterious. A large CIBMTR registry study has shown that T-cell-epitope matching defines permissive and non-permissive HLA-DPB1 mismatches2. According to this study, avoidance of an unrelated donor with a non-permissive T-cell-epitope mismatch at HLA-DPB1 was shown to lowering the risks of mortality after unrelated-donor HCT. A recent study has shown that the HLA-DPB1 rs9277534 expression marker influences the HLA-DPB1 mismatching associate risk of GVHD3. A Genome wide association study using the Affymetrix Genome-wide single nucleotide polymorphisms (SNPs) Array 6.0 identified three novel susceptibility loci in the HLA-DP region of recipient genomes that were associated with III–IV acute GvHD (rs9277378, rs9277542, and rs9277341). Of these three SNPs, rs9277378 and rs9277542 are located in non-coding regions of the HLA-DPB1 gene and the two are in strong linkage disequilibrium with two other published SNPs associated with acute GvHD, rs2281389 and rs9277535. In addition Fernandez-Vina et al have identified genetic variations in intron 2 including an InDel at position 9136^9137 (deletion predicts high expression) and 4 nucleotide (AAGG) tandem repeat in the region 9161…9216 (short number of repeats predicts high expression) (Unpublished data). In the proposed study we will perform whole gene sequencing of donor/recipient pairs using next generation sequencing (NGS) where all SNPs and other genetic variants (InDels, STRs)

41 Not for publication or presentation Attachment 3

previously published in association with GVHD will be validated in the context of known unambiguous allele level HLA-DPB1 typing of donors and recipients.

Study population: This study will include HCT donor recipient pairs reported to CIBMTR that are 10/10 & 9/10 HLA matches. Number of pairs to be included in the study will be estimated by power calculation conducted by CIBMTR biostat group.

Data requirements: No supplemental data is required

Sample requirements: • DNA samples of HCT donor/recipient pairs is needed to conduct whole gene HLA-DPB1 genotyping using NGS • Both Co-PIs are directors of fairly large full service transplant immunology labs that serve multiple high volume HCT and solid organ transplants with extensive experience in HLA typing by NGS. Both labs were among the 1st labs in the nation to be accredited for HLA typing by NGS for clinical purposes by the American Society for Histocompatibility & Immunogenetics (ASHI). • Biosketches of both Co-PIs is included

Study design: DNA samples of HCT donors and recipients that meet the eligibility criteria of the study and fulfill the sample size estimated by statistical power calculation will be requested from the CIBMTR biorepository. Whole gene HLA-DPB1 genotyping will be performed by NGS on these samples. Testing will be performed in the clinical transplant immunology lab at Baylor University Medical Center &/or Stanford University using internal institutional funds. Multivariable statistical models will be constructed to analyze the association between HLA-DPB1 donor recipient matching with aGVHD (any & III-IV), cGVHD (any & extensive), Relapse, TRM, neutrophil & platelet reconstitution, complete donor chimerism & OS. In addition the association of each variant published in association with HLA-DPB1 level of expression with these outcomes will be investigated too. Subset analyses that include specific combinations of each of the tested polymorphisms and different TCE permissiveness groups will also be attempted, e.g.: • TCE1+Short (high expression and immunogenic1) • TCE1+long (low expression and immunogenic1) (DPB1*17:01) • TCE2+Short (high expression and immunogenic2) • TCE2+long (low expression and immunogenic2) • TCE3+Short (high expression and non-immunogenic3) • TCE3+long (low expression and non-immunogenic3) • DP matched group

Non-CIBMTR data source: No non-CIBMTR data is needed

References: 1. Hurley CK, Baxter-Lowe LA, Begovich AB, Fernandez-Vina M, Noreen H, Schmeckpeper B, et al. The extent of HLA class II allele level disparity in unrelated bone marrow transplantation: analysis of 1259

42 Not for publication or presentation Attachment 3

National Marrow Donor Program donor-recipient pairs. Bone Marrow Transplant 2000;25(4):385-393. e- pub ahead of print 2000/03/21. 2. Fleischhauer K, Shaw BE, Gooley T, Malkki M, Bardy P, Bignon JD et al. Effect of T-cell-epitope matching at HLA-DPB1 in recipients of unrelated-donor haemopoietic-cell transplantation: a retrospective study. Lancet Oncol 2012;13(4):366-374. 3. Petersdorf EW, Malkki M, O'HUigin C, Carrington M, Gooley T, Haagenson MD et al. High HLA-DP Expression and Graft-versus-Host Disease. N Engl J Med 2015;373(7):599-609.

43 Not for publication or presentation Attachment 3

Table 1. Summary for CRF sample pairs with 9/10 or 10/10 HLA matching in hematological malignancies for adult unrelated donors available for proposal

Variable N (%) Number of patients 7444 Number of centers 190 Recipient age at transplant 0-9 423 ( 6) 10-19 615 ( 8) 20-29 910 (12) 30-39 927 (12) 40-49 1357 (18) 50-59 3212 (43) Median (range) 47 (0-78) Recipient race/ethnicity Caucasian, non-Hispanic 6410 (88) African-American, non-Hispanic 267 ( 4) Asian, non-Hispanic 124 ( 2) Pacific islander, non-Hispanic 12 (<1) Native American, non-Hispanic 19 (<1) Hispanic, Caucasian 371 ( 5) Hispanic, African-American 8 (<1) Hispanic, Asian 2 (<1) Hispanic, Native American 2 (<1) Hispanic, race unknown 57 ( 1) Other 9 (<1) Unknown 163 (N/A) Recipient sex Male 4178 (56) Female 3266 (44) Karnofsky score 10-80 2290 (31) 90-100 4649 (62) Missing 505 ( 7) High-resolution HLA matches available out of 8 7/8 1668 (22) 8/8 5776 (78) High-resolution HLA matches available out of 10 9/10 2097 (28) 10/10 5347 (72) Graft type

44 Not for publication or presentation Attachment 3

Variable N (%) Bone marrow 2478 (33) Peripheral blood 4965 (67) BM+PBSC 1 (<1) Conditioning regimen Myeloablative 5274 (71) RIC 1495 (20) Nonmyeloablative 350 ( 5) Other 325 ( 4) Donor age at donation To be determined / N/A 322 ( 4) 18-19 175 ( 2) 20-29 2679 (36) 30-39 2334 (31) 40-49 1492 (20) 50+ 442 ( 6) Median (range) 33 (18-62) Disease at transplant AML 3558 (48) ALL 1397 (19) CML 754 (10) MDS 1735 (23) Disease status at transplant Early 3359 (45) Intermediate 1637 (22) Advanced 2001 (27) Other 447 ( 6) Donor / recipient CMV serostatus Negative / negative 2149 (29) Negative / positive 2432 (33) Positive / negative 738 (10) Positive / positive 1441 (19) Unknown 684 ( 9) GvHD Prophylaxis Tacrolimus + MMF ± others 1171 (16) Tacrolimus + MTX ± others (except MMF) 3494 (47) Tacrolimus + others (except MTX, MMF) 339 ( 5) Tacrolimus alone 189 ( 3) CSA + MMF ± others (except Tacrolimus) 519 ( 7) CSA + MTX ± others (except Tacrolimus, MMF) 1354 (18)

45 Not for publication or presentation Attachment 3

Variable N (%) CSA + others (except Tacrolimus, MTX, MMF) 82 ( 1) CSA alone 70 ( 1) Other GVHD prophylaxis 65 ( 1) Missing 161 ( 2) Donor / recipient sex match Male / male 2837 (40) Male / female 1894 (27) Female / male 1154 (16) Female / female 1246 (17) Unknown 313 (N/A) Year of transplant 2000 341 ( 5) 2001 374 ( 5) 2002 349 ( 5) 2003 402 ( 5) 2004 575 ( 8) 2005 714 (10) 2006 770 (10) 2007 929 (12) 2008 674 ( 9) 2009 443 ( 6) 2010 377 ( 5) 2011 231 ( 3) 2012 214 ( 3) 2013 434 ( 6) 2014 307 ( 4) 2015 141 ( 2) 2016 169 ( 2) Follow-up among survivors, months N Eval 2806 Median (range) 73 (1-196)

46 Not for publication or presentation Attachment 4

Proposal 1611-158

Title: Identification of genomic markers of post hematopoietic cell transplantation (HCT) outcomes in patients with myelofibrosis (MF): A pilot study

Wael Saber: Medical College of Wisconsin Shahinaz M Gadalla: National Cancer Institute

Collaborators: Sharon Savage: Clinical Genetics Branch, NCI Youjin Wang: Clinical Genetics Branch, NCI Meredith Yeager: Cancer Genomics Research Laboratory, NCI Weiyin Zhou: Cancer Genomics Research Laboratory, NCI

Hypothesis: • Somatic mutation profiling of MF will identify risk stratification markers for patient outcome after HCT • Genetic variants in telomere biology genes and/or telomere length are important in MF pathogenesis and correlate with patient outcomes after HCT. • Chromosomal abnormalities in pre-HCT blood DNA of patients with MF are associated with patients outcomes after HCT

Specific aims: • Identify somatic mutations in patients who underwent HCT for MF and assess their contribution to patient outcomes • Evaluate the contribution of genetic variation in telomere biology genes in MF susceptibility, and determine their association with patient outcomes after HCT • Evaluate the association between pre-HCT telomere length in patients with MF and patient outcomes after HCT • Describe clonal chromosomal alterations in pre-HCT blood DNA of MF patients, and determine the effect of such alterations on post-HCT outcomes in those patients.

Scientific justification: Myelofibrosis (MF) belongs to the BCR-ABL1-negative myeloproliferative neoplasms (MPN).1 The disease is characterized by reactive bone marrow fibrosis, extramedullary hematopoiesis leading to marked hepatosplenomegaly, and abnormal cytokine expression leading to constitutional symtoms.2 Myelofibrosis (MF) can develop in patients with pre-existing essential thrombocythemia (ET) or polycythemia vera (PV). The criteria of diagnosing post-ET/PV-MF have been published.3 The median life expectancy of patients with MF is 4-5 years, but the prognosis varies considerably.4 Patients experience progressive marrow failure, and are at high risk of developing acute myelogenous leukemia (AML). The latter is associated with dismal prognosis with 98% mortality is expected at a median of 2.5 months after transformation.5 Several prognostic systems have been developed and validated to classify patients into low, intermediate, and high-risk group based on overall survival and leukemia-free survival.6-9 Genomic studies in MF identified somatic mutations in JAK2, CALR, and MPL genes as disease triggers; approximately 60%, 25%, and 7% of tumors are positive for those mutations, respectively.10 Other

47 Not for publication or presentation Attachment 4

described somatic mutations include LNK, CBL, TET2, ASXL1, IDH, IKZF1, EZH2, DNMT3A, TP53, SF3B1, SRSF2, or U2AF1.2,11-13 Those studies led to the development of JAK inhibitors that proved to relief patient symptoms, and may improve survival. Telomeres are long nucleotide repeat at chromosome ends maintaining their stability; their role in PMF has been suggested in few small studies. A study of 139 MPN patients (n primary MF=18) reported shorter granulocyte telomere length (TL) in patients than controls, independent of JAK2 mutations status.14 Another study of 239 patients with MPN, showed that telomere shortening in MPN is predominantly observed in patients with polycythemia vera and MF.15 In a large study of 529 MPN patients and 433 controls, TERT variant rs2736100 A>C SNP was associated with MPF, independent of disease subtype and molecular profile.16 Similar results were noted in a smaller study of 237 MPN patients, 26 of them had MF.17 Recently, imetelstat, a telomerase inhibitor showed promising therapeutic effect in a subset of MF patients.18 Chromosomal abnormalities in MPN have been described in few small studies. A study of 13 patients with MPN identified CN-LOH of 9p involving JAK2, and 9p gain as the most common alterations in those patients.19 Another study of 43 MPN patients, detected 9p CN-LOH in 26% of the patients; in MF patients, deletions of alterations in the region containing RB (13q14) or NF1 (17q11) were the most commonly observed (n=4/16; 25%).20 Chromosomal abnormalities in MPN have shown to affect patient risk to AML progression (OR=5.9, 95% CI 1.2-27.7, P = 0.006), and mortality (HR=18, 95% CI 1.9-164, P = 0.01).21 Unfavorable karyotype (defined as complex karyotype or single or two abnormalities including +8, −7/7q-, i(17q), −5/5q-, 12p-, inv(3) or 11q23 rearrangement) was associated by worse overall and leukemia-free survival in PMF patients.9 Allogeneic HCT remains the only curative therapy for primary MF, and post-ET/PV-MF. Because of the high risk associated with HCT, the optimal timing remains an open question, particularly in the presence of new targeted therapies. In a recent publication, investigators retrospectively compared survival of 188 patients with MF post-allogeneic HCT with that of 255 patients who only received conventional treatments. The analysis suggested that HCT for patients with high-risk disease (intermediate-2 and high risk DIPSS) is associated with superior survival; whereas a strategy of delaying HCT for those with low-risk disease (low risk and intermediate-1 DIPSS) can confer a survival advantage.22 A recent review paper suggested a role of genomic profiling in identifying MF patients who will benefit the most from HCT. The authors recommended a change to HCT eligibility to include a subset of patients with intermediate-1 risk. They identified this subset based on their high risk of AML transformation and death and define it as: 1) being JAK2, CALR, and MPL mutation-negative; 2) ASXL1, SRSF2, EZH2, or IDH1/2 mutation-positive; 3) having severe thrombocytopenia, or severe anemia requiring transfusions; 4) having high peripheral blood blasts percentage; or 5) high-risk cytogenetic findings. 23 However, whether allogeneic HCT can overcome the proposed adversity of having a poor risk disease defined at the molecular level is currently unknown. An analysis of a cohort of MF patients undergoing allogeneic HCT and stratified by molecular analysis will allow us to determine the prognostic impact of having these mutations at the time of HCT and might provide insight regarding HCT practices that are especially effective in overcoming the poor prognosis conferred by these molecular features. In this study, we will identify the pre-HCT genomic characteristics of a cohort of patients with primary MF and post-PV/ET MF and assess their relationship with patient survival outcomes. The proposed study will provide important preliminary data that will shape future questions about the value of HCT in genetically defined MF patients.

Study population: Patients eligible for this study are all patients who underwent HCT for chronic idiopathic myelofibrosis or myelofibrosis with myeloid metaplasia, and who have a pre-HCT blood sample available for research at

48 Not for publication or presentation Attachment 4

the NMDP/CIBMTR repository. The Immunobiology Committee most recent accrual reports indicate that there are 316 patients fulfilling the eligibility criteria.

Data requirements: No supplemental data are needed.

Study design: Retrospective cohort study.

Laboratory methods: All laboratory work will be performed at the Cancer Genomics Research Laboratory (CGR) at the National Cancer Institute. Whole exome sequencing: To identify somatic mutations in patient samples, we will use whole exome sequencing. Samples will be sequenced using the Illumina HiSeq; sequencing will be performed at an average depth of ~160x and minimum coverage of >80% at 15x. Reads will be aligned to the hg19 reference genome (Novoalign, Picard). GATK UnifiedGenotyper, GATK HaplotypeCaller, and Freebayes will be used to call variants. MuTect software will be used to identify candidate somatic point mutations. Indelocator software will be used to identify short somatic insertion and deletions. Genotyping: Genotyping will be performed on the HumanOmniExpress-12v1.1 beadchip according to the manufacturers’ protocols. This platform contains approximately 733,202 markers per sample for SNP and copy number variation (CNV) analyses. Only samples with >90% completion threshold will be included in the analysis. Log R ratio (LRR) and B allele frequency (BAF) will be used to assess copy number variation as well as to calculate the proportion of abnormal cells (allelic imbalance). LRR calculates the log base 2 of the ratio of observed total signal intensities to expected signal intensities for a SNP. LRR > 0 indicate copy gain, whereas < 0 indicate loss. The BAF is calculated as the ratio of signal intensity between two alleles at each SNP in relation to estimated genotype clusters and will be used to calculated proportion of cells with a deletion, duplication, and copy number neutral loss of heterozygosity. BAF values for heterozygous SNPs that deviate from 0.5 are indicative of copy-neutral changes associated with uniparental disomy. Telomere length measurement: We will use monoplex qPCR assay to measure leukocyte relative telomere length in extracted DNA. The qPCR assay calculates the ratio between telomeric repeat copy number (T) and that of a single reference gene (beta-globin gene; 36B4) (S). Relative T/S is calculated in relation to a reference curve and final measurements are exponentiated to assure normality. All telomeric and 36B4 reactions are measured in triplicate, and the average is used for final calculations.

Outcomes: Primary outcomes: • Overall survival • Disease-free survival Secondary outcomes: • AML incidence • Transplant related mortality

49 Not for publication or presentation Attachment 4

Variables to be analyzed: • Patient related: age, gender, Karnofsky score. • Disease related: DIPSS at diagnosis and at HCT. • Cytogenetic abnormalities: as defined in DIPSS-PLUS.9 • Transplant related: time from diagnosis to HCT, donor-recipient sex match, donor-recipient cytomegalovirus [CMV] serology, conditioning regimen intensity, stem cell source, GVHD prophylaxis regimen, pre-HCT chemotherapy or other therapies, and year of HCT.

Study design: We will use Kaplan-Meier estimator to calculate 1, 3, and 5-year survival probability; log-rank test will be used to compare survival across categories of interest. For the incidence of TRM and AML, we will use cumulative incidence estimator to account for competing risks. Cox regression hazard models will be used for multivariable analyses. Separate models will be performed for each specific aim. Identify somatic mutations in patients who underwent HCT for MF and assess their contribution to patient outcomes: Our primary analysis will focus on previously published genes (JAK2, CALR, MPL, ASXL1, SRSF2, EZH2, IDH1/2, LNK, CBL, TET2, IKZF1, DNMT3A, TP53, SF3B1, and U2AF1). In a subsequent analysis we will seek to identify new gene mutations contributing to the pathogenesis of primary MF and test their effect on HCT outcomes. We will first test the effect of each gene separately. Then, we will use information from generated models to classify patients into favorable, moderate, and high risk based on their mutation burden from all tested genes Evaluate the contribution of genetic variation in telomere biology genes in MF susceptibility, and determine their association with patient outcomes after HCT: We will focus on a published list of 743 SNPs in 43 telomere biology genes.24 To identify whether telomere-biology genes play a role in MF, we will use unconditional logistic regression to calculate odds ratios and 95% confidence intervals (CI) for each SNP comparing MF cases to controls. Controls will be cancer-free individuals with genotyping data available at CGR. We will also perform gene-level and pathway-level analyses as described in a previous study.25 For outcome analysis, we will use the same statistical analysis presented above first including all SNPs, then restricting on SNPs with a p-value of 1*10-5 in the MF case-control comparison. Evaluate the association between pre-HCT telomere length in patients with MF and patient outcomes after HCT: Telomere length will be assessed as a continuous variable and in tertile categories. We hypothesize that shorter pre-HCT telomere length in those patients will be associated with worse outcomes. Describe clonal chromosomal alterations in pre-HCT blood DNA of MF patients, and determine the effect of such alterations on post-HCT outcomes in those patients: We will describe large gains, losses, and CN-LOHs in all autosomes. Patients will be classified as having none or any alteration. First, we will test the effect of the alterations included in the DIPPS plus predictive score for PMF outcomes,9 then we will evaluate the effect of other alterations to better describe their relationship with outcomes. It is important to note that the alterations included in the DIPPS plus score were identified using clinical cytogenetics that can’t detect loss of heterozygozity. Power calculation: Several components of this study are still in the exploration phase. The calculations below present minimal detectable relative risk (RR) with the following assumptions: type I error=0.05, power=80%, and sample size=300

50 Not for publication or presentation Attachment 4

Probability in patients with longer Minimal % Exposed telomere length detectable RR 20% 30% 0.44-1.65 40% 0.53-1.50 50% 0.60-1.40 60% 0.67-1.31 30% 30% 0.52-1.55 40% 0.60-1.43 50% 0.66-1.34 60% 0.72-1.27

References: 1. Vardiman JW, Thiele J, Arber DA, et al. The 2008 revision of the World Health Organization (WHO) classification of myeloid neoplasms and acute leukemia: rationale and important changes. Blood 2009;114(5):937-951. 2. Tefferi A. Primary myelofibrosis: 2014 update on diagnosis, risk-stratification, and management. Am J Hematol 2014;89(9):915-925. 3. Barosi G, Mesa RA, Thiele J, et al. Proposed criteria for the diagnosis of post-polycythemia vera and post-essential thrombocythemia myelofibrosis: a consensus statement from the International Working Group for Myelofibrosis Research and Treatment. Leukemia. 2008;22(2):437-438. 4. Deeg HJ, Bredeson C, Farnia S, et al. Hematopoietic Cell Transplantation as Curative Therapy for Patients with Myelofibrosis: Long-Term Success in all Age Groups. Biol Blood Marrow Transplant 2015;21(11):1883-1887. 5. Mesa RA, Li CY, Ketterling RP, Schroeder GS, Knudson RA, Tefferi A. Leukemic transformation in myelofibrosis with myeloid metaplasia: a single-institution experience with 91 cases. Blood 2005;105(3):973-977. 6. Dupriez B, Morel P, Demory JL, et al. Prognostic factors in agnogenic myeloid metaplasia: a report on 195 cases with a new scoring system. Blood 1996;88(3):1013-1018. 7. Cervantes F, Dupriez B, Pereira A, et al. New prognostic scoring system for primary myelofibrosis based on a study of the International Working Group for Myelofibrosis Research and Treatment. Blood 2009;113(13):2895-2901. 8. Passamonti F, Cervantes F, Vannucchi AM, et al. A dynamic prognostic model to predict survival in primary myelofibrosis: a study by the IWG-MRT (International Working Group for Myeloproliferative Neoplasms Research and Treatment). Blood 2010;115(9):1703-1708. 9. Gangat N, Caramazza D, Vaidya R, et al. DIPSS plus: a refined Dynamic International Prognostic Scoring System for primary myelofibrosis that incorporates prognostic information from karyotype, platelet count, and transfusion status. J Clin Oncol 2011;29(4):392-397. 10. Tefferi A, Pardanani A. Myeloproliferative Neoplasms: A Contemporary Review. JAMA Oncol 2015;1(1):97-105. 11. Tefferi A. Novel mutations and their functional and clinical relevance in myeloproliferative neoplasms: JAK2, MPL, TET2, ASXL1, CBL, IDH and IKZF1. Leukemia 2010;24(6):1128-1138. 12. Tefferi A, Finke CM, Lasho TL, et al. U2AF1 mutations in primary myelofibrosis are strongly associated with anemia and thrombocytopenia despite clustering with JAK2V617F and normal karyotype. Leukemia 2014;28(2):431-433.

51 Not for publication or presentation Attachment 4

13. Tefferi A, Thiele J, Vannucchi AM, Barbui T. An overview on CALR and CSF3R mutations and a proposal for revision of WHO diagnostic criteria for myeloproliferative neoplasms. Leukemia 2014;28(7):1407-1413. 14. Elena C, Rumi E, Portolan M, Della Porta MG, Pascutto C, Passamonti F. Flow-FISH evaluation of telomere length in Philadelphia-negative myeloproliferative neoplasms. Haematologica 2011;96(8):1236-1238. 15. Ruella M, Salmoiraghi S, Risso A, et al. Telomere shortening in Ph-negative chronic myeloproliferative neoplasms: a biological marker of polycythemia vera and myelofibrosis, regardless of hydroxycarbamide therapy. Exp Hematol 2013;41(7):627-634. 16. Trifa AP, Banescu C, Tevet M, et al. TERT rs2736100 A>C SNP and JAK2 46/1 haplotype significantly contribute to the occurrence of JAK2 V617F and CALR mutated myeloproliferative neoplasms - a multicentric study on 529 patients. Br J Haematol 2016. 17. Oddsson A, Kristinsson SY, Helgason H, et al. The germline sequence variant rs2736100_C in TERT associates with myeloproliferative neoplasms. Leukemia 2014;28(6):1371-1374. 18. Tefferi A, Lasho TL, Begna KH, et al. A Pilot Study of the Telomerase Inhibitor Imetelstat for Myelofibrosis. N Engl J Med 2015;373(10):908-919. 19. Hahm C, Huh HJ, Mun YC, Seong CM, Chung WS, Huh J. Genomic aberrations of myeloproliferative and myelodysplastic/myeloproliferative neoplasms in chronic phase and during disease progression. Int J Lab Hematol 2015;37(2):181-189. 20. Kawamata N, Ogawa S, Yamamoto G, et al. Genetic profiling of myeloproliferative disorders by single-nucleotide polymorphism oligonucleotide microarray. Exp Hematol 2008;36(11):1471-1479. 21. Rumi E, Harutyunyan A, Elena C, et al. Identification of genomic aberrations associated with disease transformation by means of high-resolution SNP array analysis in patients with myeloproliferative neoplasm. Am J Hematol 2011;86(12):974-979. 22. Kroger N, Giorgino T, Scott BL, et al. Impact of allogeneic stem cell transplantation on survival of patients less than 65 years of age with primary myelofibrosis. Blood 2015;125(21):3347-3350; quiz 3364. 23. Viswabandya A, Devlin R, Gupta V. Myelofibrosis-When Do We Select Transplantation or Non- transplantation Therapeutic Options? Curr Hematol Malig Rep 2015. 24. Mirabello L, Yu K, Kraft P, et al. The association of telomere length and genetic variation in telomere biology genes. Hum Mutat 2010;31(9):1050-1058. 25. Yu K, Li Q, Bergen AW, et al. Pathway analysis by adaptive combination of P-values. Genet Epidemiol 2009;33(8):700-709.

52 Not for publication or presentation Attachment 4

Table 1. Bone marrow or peripheral blood stem cell unrelated recipient samples available for all subdiseases of myelofibrosis

Samples Available for Samples Available Recipient and Donor for Recipient Only N (%) N (%) Number of patients 324 69 Number of centers 80 37 Recipient age at transplant 0-9 3 ( 1) 1 ( 1) 10-19 4 ( 1) 1 ( 1) 20-29 8 ( 2) 0 30-39 19 ( 6) 3 ( 4) 40-49 67 (21) 15 (22) 50-59 223 (69) 49 (71) Median (range) 54 (1-75) 57 (9-73) Recipient race/ethnicity Caucasian, non-Hispanic 287 (90) 60 (90) African-American, non-Hispanic 14 ( 4) 4 ( 6) Asian, non-Hispanic 6 ( 2) 0 Hispanic, Caucasian 12 ( 4) 3 ( 4) Hispanic, Asian 1 (<1) 0 Unknown 4 (N/A) 2 (N/A) Recipient sex Male 179 (55) 42 (61) Female 145 (45) 27 (39) Karnofsky score 10-80 107 (33) 18 (26) 90-100 127 (39) 22 (32) Missing 90 (28) 29 (42) High-resolution HLA matches available out of 8 ≤5/8 4 ( 1) 0 6/8 9 ( 3) 0 7/8 69 (22) 12 (22) 8/8 225 (73) 42 (78) Unknown 17 (N/A) 15 (N/A) High-res typed and audited Yes 200 (100) 2 (100) Unknown 124 (N/A) 67 (N/A) Graft type Bone marrow 75 (23) 12 (17) Peripheral blood 249 (77) 57 (83)

53 Not for publication or presentation Attachment 4

Samples Available for Samples Available Recipient and Donor for Recipient Only N (%) N (%) Conditioning regimen Myeloablative 160 (49) 31 (45) RIC 86 (27) 16 (23) Nonmyeloablative 24 ( 7) 4 ( 6) Other 54 (17) 18 (26) Donor age at donation To be determined / NA 27 ( 8) 48 (70) 10-19 8 ( 2) 0 20-29 100 (31) 4 ( 6) 30-39 101 (31) 9 (13) 40-49 61 (19) 7 (10) 50+ 27 ( 8) 1 ( 1) Median (range) 34 (19-60) 37 (23-56) Disease at transplant MDS 324 (100) 69 (100) Subdisease at transplant Polycythemia vera 36 (11) 11 (16) Essential thrombocythemia 48 (15) 5 ( 7) Myelofibrosis with myeloid metaplasia 1 (<1) 0 Primary myelofibrosis 239 (74) 53 (77) Disease status at transplant Early 162 (50) 32 (46) Advanced 29 ( 9) 2 ( 3) Other 133 (41) 35 (51) GvHD Prophylaxis Ex vivo T-cell depletion 6 ( 2) 1 ( 1) CD34 selection 8 ( 2) 1 ( 1) Cyclophosphamide 1 (<1) 0 Tacrolimus + MMF ± others 50 (15) 7 (10) Tacrolimus + MTX ± others (except MMF) 154 (48) 36 (52) Tacrolimus + others (except MTX, MMF) 13 ( 4) 7 (10) Tacrolimus alone 4 ( 1) 3 ( 4) CSA + MMF ± others (except Tacrolimus) 32 (10) 7 (10) CSA + MTX ± others (except Tacrolimus, MMF) 46 (14) 5 ( 7) CSA + others (except Tacrolimus, MTX, MMF) 1 (<1) 0 CSA alone 2 ( 1) 0 Other GVHD prophylaxis 2 ( 1) 0 Missing 5 ( 2) 2 ( 3)

54 Not for publication or presentation Attachment 4

Samples Available for Samples Available Recipient and Donor for Recipient Only N (%) N (%) Donor / recipient CMV serostatus Negative / negative 114 (35) 21 (30) Negative / positive 75 (23) 10 (14) Positive / negative 38 (12) 15 (22) Positive / positive 59 (18) 9 (13) Unknown 38 (12) 14 (20) Donor/Recipient sex match Male / male 124 (42) 13 (36) Male / female 70 (23) 5 (14) Female / Male 42 (14) 9 (25) Female / female 62 (21) 9 (25) Unknown 26 (N/A) 33 (N/A) Donor age at transplant To be determined / NA 27 ( 8) 48 (70) 10-19 8 ( 2) 0 20-29 100 (31) 4 ( 6) 30-39 101 (31) 9 (13) 40-49 61 (19) 7 (10) 50+ 27 ( 8) 1 ( 1) Median (range) 34 (19-60) 37 (23-56) Year of transplant 1990 1 (<1) 0 1994 3 ( 1) 0 1995 1 (<1) 0 1996 4 ( 1) 0 1997 5 ( 2) 2 ( 3) 1998 5 ( 2) 1 ( 1) 1999 5 ( 2) 1 ( 1) 2000 5 ( 2) 2 ( 3) 2001 9 ( 3) 2 ( 3) 2002 12 ( 4) 1 ( 1) 2003 18 ( 6) 1 ( 1) 2004 14 ( 4) 1 ( 1) 2005 26 ( 8) 5 ( 7) 2006 29 ( 9) 4 ( 6) 2007 26 ( 8) 1 ( 1) 2008 36 (11) 5 ( 7) 2009 32 (10) 8 (12)

55 Not for publication or presentation Attachment 4

Samples Available for Samples Available Recipient and Donor for Recipient Only N (%) N (%) 2010 18 ( 6) 5 ( 7) 2011 17 ( 5) 5 ( 7) 2012 24 ( 7) 7 (10) 2013 7 ( 2) 4 ( 6) 2014 0 3 ( 4) 2015 27 ( 8) 11 (16) Follow-up among survivors, months N Eval 134 42 Median (range) 62 (1-240) 37 (3-168)

56 Not for publication or presentation Attachment 5

Proposal 1611-41

Title: Donor-recipient NK cell determinants associated with survival in JMML after hematopoietic stem cell transplantation

Dean A. Lee: Nationwide Children’s Hospital Hemalatha G. Rangarajan: Nationwide Children’s Hospital

Hypothesis: We hypothesize that immunologic mechanisms are a major component of protection from relapse after HCT in patients with JMML, and specifically that determinants of greater NK cell function (e.g., KIR mismatch, KIR A/B genotype) are associated with reduced relapse. Given the modest impact of NK cells on GvHD, we hypothesize that determinants of greater NK cell function will be also be associated with increased overall, event-free, and GvHD-free/relapse-free survival (GRFS) survival.

Specific aims: Compare outcomes of JMML patients treated with HCT as stratified by NK-related factors (e.g., KIR Ligand mismatch, Donor KIR B genotype) that have previously been associated with survival of other myeloid malignancies. The specific outcomes to be ascertained include: • 2 year EFS, OS, RFS, and GRFS Secondary Objectives: • Incidence of acute and chronic graft-versus-host-disease (GvHD)

Scientific justification: The long-term relapse-free survival that is afforded by hematopoietic cell transplantation (HCT) for myeloid malignancies is associated with immunologic parameters that predict NK cell function and alloreactivity. The impact of NK cells on survival has been associated with increased numbers of NK cells in the stem cell graft, rapid recovery of NK cells in the early post-transplant period, alloreactivity of NK cells in the GvH direction predicted by KIR-ligand (HLA) mismatch, more refined measures of mismatch that include KIR genotyping, activating KIR gene content, assigned KIR genotypes, and presence or absence of specific KIR genes. Juvenile myelomonocytic leukemia (JMML) is an aggressive myeloproliferative neoplasm of young children characterized by activating mutations in the RAS/RAF/MAPK signaling pathway. HCT is the only proven curative therapy and is considered the standard of care for the disease; however, the 5-year event-free survival is only 52%. [1] Although the role of natural killer (NK) cells in the treatment of other myeloid leukemias has been studied [2], there are no reports of the impact of these NK cell factors on survival of JMML. Since no curative therapies have been described for JMML other than HCT, and dose intensity has not clearly been associated with cure for this disease, we hypothesize that the benefit of allogeneic HCT may be mediated by the graft-vs-leukemia effect. In in vitro studies, we found that the mature monocytic population of JMML cells express similar profiles of NK cell ligands as normal healthy donor monocytes and are relatively resistant to NK cell cytotoxicity. In contrast, JMML colony-forming units were significantly reduced following incubation of JMML cells with NK cells, suggesting they might express ligands for activating receptors of NK cells (Fig. A). We found that JMML stem cells, defined as Lin-CD34+CD38-, express ligands for NKG2D, NKp30, and NKp44 at levels equal or greater than that of AML stem cells (Fig. B)[1] They also express similar levels of HLA Class I and HLA-E, suggesting that KIR mismatch may be relevant for JMML in a manner similar to AML.

57 Not for publication or presentation Attachment 5

Figure: A) Relative decrease in colony-forming units as a measure of killing by NK cells. Spleen mononuclear cells from patients with JMML or cord blood mononuclear cells were assessed for colony-formation with or without co-culture with NK cells. B) Mass cytometry for NK cell ligands on CD34+ cells in samples from AML and JMML patients.

Study population: Inclusion criteria: • Diagnosis: JMML • Years: 2000-2014 Exclusion criteria: • Second or subsequent transplants

Study design: Retrospective We will collect and analyze variables as follows: • Donor source: MSD, MUD, MMUD, Haploidentical • Graft/product type: bone marrow vs peripheral blood vs cord blood • Donor/Recipient HLA type • Donor KIR genotype (activating KIR content, KIR A/B genotype, inhibitory KIR) • T cell replete transplant Y/N • Molecular markers

Data requirements: CIBMTR form - Pre and Post AlloHCT data

Variables to be analyzed: Patient Related: • Age • Sex • Race • Performance Score • Comorbidities Donor Related: • KIR genotype and B content score

58 Not for publication or presentation Attachment 5

Disease Related: • Clinical features at diagnosis: • Adenopathy, hepatomegaly, presence of Neurofibromatosis (NF), skin involvement and splenomegaly. • WBC count • Hemoglobin • Platelets • Monocyte percent • Blasts in peripheral blood • LD • Fetal hemoglobin • Hypersensitivity to GM CSF • Bone marrow blasts and monocytes in bone marrow • Cytogenetic abnormality -7, +8, t (2,8), t 99,22) other • Molecular markers: BCR/ABL, CBL, KRAS, NRAS, PTNP11 Pre HCT therapy: • Specify therapy • Splenectomy • Remission status pre-transplant: CR, PR, Marginal response, stable disease, progressive disease or relapse • Transformation to AML Y/N First HCT-related: • Time from diagnosis to first transplantation • Conditioning chemotherapy • In vivo T-Cell depletion • GvHD prophylaxis: CNI (FK506/CsA), +/- MTX, other • Viral reactivations • GvHD: acute (none vs grade 1 vs 2 vs 3-4); chronic: none vs yes • Incidence of relapse, aGvHD, cGvHD, 2 year EFS, 2-year OS, 100-day, 1 year, and 2 year TRM • Chimerism on Days +30, +60, +100, +180 and +360

Statistical consideration: The continuous variables will be summarized by mean and standard deviation; the categorical variables were summarized by percentage. 2 year EFS and OS will be calculated by using Kaplan Meier Estimates. Univariate and multivariate analysis will be performed for risk factors including donor KIR B content score with specified outcomes (2 year EFS, RFS, GRFS, aGvHD and cGvHD).

References: 1. Satwani P, Kahn J, Dvorak CC. Juvenile myelomonocytic leukemia. Pediatr Clin North Am 2015:62(1):95-106. 2. Cooley S, Weisdorf DJ, Guethlein LA, et al. Donor selection for natural killer cell receptor genes leads to superior survival after unrelated transplantation for acute myelogenous leukemia. Blood 2010; 116(14):2411-2419. 3. Lee D, Martinez S, Senyukov V, et al. High-dimension single-cell mass cytometry and colony-forming assays to determine the susceptibility of juvenile myelomonocytic leukemia to NK cell immunotherapy. Pediatric Blood & Cancer 2013; 60:S41-S41.

59 Not for publication or presentation Attachment 5

Table 1. Characteristics of JMML patients from 2000-2014 for unrelated, related and cord blood transplants using both CRF and TED retrievals

Variable N (%) Number of patients 465 Number of centers 125 Recipient age at transplant 0-9 455 (98) 10-19 9 ( 2) 50-59 1 (<1) Median (range) 2 (0-59) Recipient race / ethnicity Caucasian, non-Hispanic 271 (64) African-American, non-Hispanic 24 ( 6) Asian, non-Hispanic 45 (11) Pacific islander, non-Hispanic 6 ( 1) Native American, non-Hispanic 2 (<1) Hispanic, Caucasian 48 (11) Hispanic, African-American 1 (<1) Hispanic, race unknown 6 ( 1) Other 19 ( 5) Unknown 43 (N/A) Recipient sex Male 311 (67) Female 154 (33) Karnofsky score 10-80 71 (15) 90-100 277 (60) Missing 117 (25) KIR typing currently available No 421 (91) Yes 44 ( 9) URD HLA-A B DRB1 groups - low res A and B: high res DRB1 5/6 8 (14) 6/6 49 (86) Unknown / to be determined 145 (N/A) URD High-resolution HLA matches available out of 8 ≤5/8 2 ( 4) 6/8 4 ( 7) 7/8 13 (23) 8/8 38 (67)

60 Not for publication or presentation Attachment 5

Variable N (%) Unknown / to be determined 145 (N/A) Related HLA typing Unknown / to be determined 101 (N/A) Cord blood HLA-A B DRB1 groups - low res A and B: high res DRB1 4/6 8 (24) 5/6 23 (68) 6/6 3 ( 9) Unknown / to be determined 128 (N/A) Cord blood high-resolution HLA matches available out of 8 ≤5/8 19 (56) 6/8 11 (32) 7/8 3 ( 9) 8/8 1 ( 3) Unknown / to be determined 128 (N/A) Unrelated donor indicator Related 101 (22) Unrelated 202 (43) Umbilical cord blood 162 (35) Retrieval TED 217 (47) CRF 248 (53) Donor type and retrieval Unrelated donor and CRF 102 (22) Related donor and CRF 29 ( 6) Cord blood and CRF 117 (25) Unrelated donor and TED 100 (22) Related donor and TED 72 (15) Cord blood and TED 45 (10) Graft type Bone marrow 226 (49) Peripheral blood 76 (16) Umbilical cord blood 162 (35) Others 1 (<1) High-resolution released and audited Yes 91 (100) Unknown 374 (N/A) Conditioning regimen Myeloablative 387 (83) RIC 3 ( 1)

61 Not for publication or presentation Attachment 5

Variable N (%) Nonmyeloablative 1 (<1) Other 74 (16) Donor age by decade To be determined / N/A 1 (17) 0-9 1 (17) 20-29 2 (33) 40-49 2 (33) Unknown 459 (N/A) Disease at transplant AML 2 (<1) Other leukemia 43 ( 9) MDS 418 (90) Other acute leukemia 1 (<1) Other 1 (<1) Subdisease at transplant AML other, specify as JMML 2 (<1) Juvenile CML / JMML 398 (86) Other leukemia, specified as JMML 43 ( 9) MDS, NOS, but said JMML in other specified field 1 (<1) Other MDS, specified as JMML 15 ( 3) TED other Acute Leukemia, specified as JMML 1 (<1) Other MDS and other specified says JMML 4 ( 1) Other specified says JMML 1 (<1) Donor / recipient CMV serostatus Negative / negative 81 (17) Negative / positive 60 (13) Positive / negative 38 ( 8) Positive / positive 72 (15) Unknown 214 (46) GvHD prophylaxis No GVHD prophylaxis (forms need review) 4 ( 1) Ex vivo T-cell depletion 18 ( 4) CD34 selection 5 ( 1) Tacrolimus + MMF ± others 22 ( 5) Tacrolimus + MTX ± others (except MMF) 59 (13) Tacrolimus + others (except MTX, MMF) 10 ( 2) Tacrolimus alone 2 (<1) CSA + MMF ± others (except Tacrolimus) 37 ( 8) CSA + MTX ± others (except Tacrolimus, MMF) 171 (37)

62 Not for publication or presentation Attachment 5

Variable N (%) CSA + others (except Tacrolimus, MTX, MMF) 70 (15) CSA alone 25 ( 5) Other GVHD prophylaxis 8 ( 2) Missing 34 ( 7) Donor / recipient sex match Male / male 147 (35) Male / female 69 (16) Female / male 140 (33) Female / female 70 (16) Unknown 39 (N/A) Year of transplant 2000 29 ( 6) 2001 26 ( 6) 2002 35 ( 8) 2003 45 (10) 2004 35 ( 8) 2005 33 ( 7) 2006 35 ( 8) 2007 35 ( 8) 2008 31 ( 7) 2009 25 ( 5) 2010 36 ( 8) 2011 23 ( 5) 2012 30 ( 6) 2013 28 ( 6) 2014 19 ( 4) Follow-up among survivors, months N Eval 261 Median (range) 63 (0-188)

63 Not for publication or presentation Attachment 6

Proposal 1610-18

Title: Role of KIR ligand mismatches for transplant outcomes in T-cell replete, non-myeloablative haploidentical transplants, inferring KIR ligand mismatches from HLA-A, -B, and -C locus typing mismatches: a CIBMTR registry analysis

Muthu Kumaran Veeraputhiran: University of Arkansas for Medical Sciences Terry O Harville: University of Arkansas for Medical Sciences Marcelo Fernandez-Viña: Stanford University Medical Center

Hypothesis: In a limited cohort of T-cell replete haploidentical transplants from two transplant centers, improved survival was noted with Killer Immunoglobulin Receptor (KIR) Gene Mismatches compared with matched KIR genes.1 The improved outcome of KIR and HLA genotyping has been shown both in matched and mismatched hematopoietic transplants.2,3 We hypothesize that using surrogate HLA-A, -B, and -C locus typing results of donor and recipients from the CIBMTR database, we can infer the matching or mismatching of KIR ligands, and that appropriate mismatching will correlate with better short and long-term transplant outcomes.

Specific aims: We plan to conduct a retrospective review of CIBMTR database to analyze the effects of KIR ligand mismatching, using surrogate HLA-A, -B, and -C locus typing and transplant outcomes in T-cell replete haploidentical transplants. KIR ligand mismatching in the graft-versus-host disease (GVHD) direction is defined as absence of expression of specific class I HLA-A, -B, and/or -C inhibitory alleles and expression of specific class I HLA-A, -B, and/or -C activating alleles on patient cells capable of stimulating donor NK cells.3

The donor-recipient pairs will be divided into two groups: the first without and the second with A/B/C/KIR ligand mismatch in the GVHD direction.

The two groups will be compared for: 1. Overall survival 2. Overall disease-free survival 3. Non-relapse mortality 4. GVHD (acute: II-IV, and chronic: mild, moderate and severe) 5. Primary graft failure

Additionally, impact of donor parameters other than HLA disparity will be considered: 1. Familial relationship of donor with recipient 2. CMV status (donor and recipient) 3. Blood group differences 4. Donor-specific anti-HLA antibodies (DSA)

Scientific justification: Haploidentical transplants continue to increase in recent years due to advancement in conditioning techniques and GVHD prophylaxis, tapping into an important alternative donor source, especially for minority populations who otherwise would not have fully-matched donors.

64 Not for publication or presentation Attachment 6

HLA mismatches in haploidentical transplants have impact on the development of acute and chronic GVHD, transplant-related mortality (TRM), incidence of relapse, and overall survival.5,6 In typical T-cell replete haploidentical transplants, the alloreactive T cells are thought to more responsible for the observed level of GVHD/GVL (graft versus leukemia), than the alloreactive NK cells.7 However, the use of post-transplantation cyclophosphamide greatly reduces the alloreactive T cells, allowing the alloreactive NK cells to become more relevant. KIR has the important role in defining the alloreactivity of donor- derived NK cells in GVHD, as well as, against leukemia blasts.3 KIR binds to specific HLA-A, -B, and -C proteins expressed on patient to either activate or inhibit the donor NK cells. By using different models to determine NK alloreactivity, specific KIR mismatches have been correlated with lower relapse and higher overall survival in the non-myeloablative haploidentical transplant setting using bone marrow as the hematopoietic stem cell source. A retrospective analysis of the use of T-cell replete graft from peripheral blood showed no significant difference in outcomes in terms of acute GVHD, non-relapse mortality (NRM), and overall survival compared with a bone marrow source.8 The other factors in the donor selection process, including ABO blood type, cytomegalovirus (CMV) serologic-status of the donor and recipient, gender (and parity in females), and donor age are the same in haploidentical donor selection as with HLA-matched donor selection. There are, however, criteria that are unique to the haploidentical transplant setting, which include donor-specific anti-HLA antibodies (DSA), donor relationship and HLA mismatch (siblings, children, etc., which involve such issues as non-inherited maternal antigens [NIMA], non-inherited paternal antigens [NIPA], etc), and the NK cell alloreactivity due to the extent of KIR/HLA matching. Algorithms for haploidentical-donor selection have been suggested by Wang, et al5 and Van Rood, et al6 and and HLA-A, -B, or -DR mismatch did not appear to influence transplant outcomes in several studies. Because the conditioning regimens and GVHD prophylaxis used in these studies differ from those used in transplants with HLA haplo- identical donors and the administration of Fludarabine-Cyclophosphamide, along with post-transplant cyclophosphamide that currently is widely used, the role of HLA-mismatch and disparity needs to be reassessed. In particular, the role of KIR-ligand mismatch on the donor selection algorithm needs to be fully assessed; using surrogate HLA-A, -B, and -C locus typing as indicators of activating or inhibitory KIR ligands expressed on donor and recipient cells that determine NK KIR ligand competence and KIR ligand mismatch. The information provided by this study could have a significant impact in the development algorithms for donor selection in the Haplo-identical transplantation setting.

Study population: 1. Patients ≥ 18 year of age. 2. Non-myeloablative, related-donor, partially HLA-mismatched HCT with conditioning using Flu/Cy/TBI, and with post-transplant cyclophosphamide; with either BM or PB stem cell source. 3. Post-transplant GVHD prophylaxis with tacrolimus and mycophenolate mofetil. 4. Diagnoses include: AML, MDS, ALL, Chronic myeloproliferative disorders including CMML, polycythemia vera, TKI refractory CML, non-blast crisis CML beyond first chronic phase and other relapsed/ refractory Hodgkin’s lymphoma, non-Hodgkin’s lymphoma, chronic lymphocytic leukemia, and multiple myeloma. (Patients with different disease categories will be subsetted as possible for analysis. Patients with high-risk versus lower-risks diseases will be subsetted as possible for analysis.)

65 Not for publication or presentation Attachment 6

Variables to be analyzed: All data will be obtained from CIBMTR database (See Table-1 and 2 for preliminary custom data from NMDP). No supplemental data or biologic samples will be required. The following are the variables of interest: Patient Related: • Age • Gender • Ethnicity • Disease status • Disease type • CMV serology • HLA Loci at A, B, C, DR for patient, donor and additional family members (if available) • Graft type • CD34 cell count of the graft • Blood type • Familial relationship of patient and donor • Donor specific anti-HLA antibodies Transplant Related: • Acute GVHD by grades I-IV • Neutrophil engraftment • Non-relapse mortality rate by day 100, 1 year, 2 years, and beyond • Primary Graft failure • Chronic GVHD: Type, grade and prevalence • Overall disease free survival • Overall survival

Study design: This is a retrospective study. End points: 1. HLA mismatch in Graft versus Host Disease (GVHD) direction and Host versus Graft (HVG). GVHD would be analyzed in GVHD direction and graft rejection would be analyzed according to HVG direction mismatching. 2. Disease and survival outcomes will be analyzed according to either direction or particular direction as specified. 3. KIR ligand status will be inferred by HLA typing results and using the immune polymorphism database. 4. The KIR-ligand mismatch (≥1) would be compared to the KIR-ligand matched group for the GVHD, TRM, Relapse, and overall survival. 5. The HLA-mismatch, Donor-recipient relationship, and gender matching will be compared using univariate and multivariate analysis with outcomes of AGVHD, NRM, relapse, and overall survival. 6. Statistical methods: overall survival and event-free survival (EFS) will be calculated from day of transplant by KM method and compared with Cox-proportional hazards models. The simultaneous effect of 2 or more factors would be studied with multivariate proportional hazards models. Cumulative incidences of relapse, NRM, acute grade II-IV GVHD, and chronic GVHD, measured from the day of transplant, would be estimated using competing risk analyses and prognostic factors for these endpoints were analyzed using proportional hazards models for

66 Not for publication or presentation Attachment 6

competing risks. For competing risk analyses, death without relapse, or progression would be considered a competing risk for relapse; relapse would be a competing risk for GVHD. For evaluating differences in group characteristics, Wilcoxon test could be used for continuous variables and a Chi-squared test or Fisher’s exact test for cell counts <5 for categorical variables. P values are two sided and unadjusted for multiple comparisons.

References: 1. Fuchs, et al. Improved Survival with Inhibitory Killer Immunoglobulin Receptor (KIR) Gene Mismatches and KIR Haplotype B Donors after Non-myeloablative, HLA-Haploidentical Bone Marrow Transplantation. Biol Blood Marrow Transplant 2010;16:533-542. 2. Hsu, et al. Improved outcome in HLA-identical sibling hematopoietic stem-cell transplantation for acute myelogenous leukemia predicted by KIR and HLAgenotypes. Blood 2005;(105):4878-4884. 3. Ruggeri, et al. Effectiveness of Donor Natural Killer Cell Alloreactivity in Mismatched Hematopoietic Transplants. Science 2002;2097-2100. 4. Ruggeri L, Mancusi A, Capanni M, Urbani E, Carotti A, Aloisi T,Stern M, et al. Donor natural killer cell allo recognition of missing self in haploidentical hematopoietic transplantation for acute myeloid leukemia: challenging its predictive value. Blood 2007;110:433-440. 5. Wang, et al. Who is the best donor for a related HLA haplotype-mismatched transplant? Blood 2014;124(6);843. 6. Van Rood, et al. Effect of tolerance to non-inherited maternal antigens on the occurrence of graft- versus-host disease after bone marrow transplantation from a parent or an HLA-haploidentical sibling. Blood 2002;99(5);1572. 7. Lowe, et al. T-cell alloreactivity dominates natural killer cell alloreactivity in minimally T-cell-depleted HLA-non-identical paediatric bone marrow transplantation. British Journal of Haematology 2003;123;323–326. 8. Castagna, et al. Bone Marrow Compared with Peripheral Blood Stem Cells for Haploidentical Transplantation with a Non-myeloablative Conditioning Regimen and Post-transplantation Cyclophosphamide. Biol Blood Marrow Transplant 2014;20:724e729.

67 Not for publication or presentation Attachment 6

Proposal 1611-36

Title: Impact of KIR ligand mismatches on clinical outcomes following haploidentical stem cell transplantation.

Folashade Otegbeye: University Hospitals Cleveland Medical Center Marcos de Lima: University Hospitals Cleveland Medical Center

Hypothesis: When KIR ligands (HLA Class I epitopes) expressed by the haploidentical stem cell donor are absent in the transplant recipient, the potential for NK cell donor alloreactivity will be associated with improved clinical outcomes.

Specific aims: Primary Aim: To analyze the impact of KIR ligand mismatch in the graft-versus-host (GVH) direction on disease-free survival (DFS) of adult patients with hematologic malignancies following haploidentical stem cell transplantation. Secondary Aims: • To assess the impact of KIR ligand mismatch in the GVH direction on relapse rate, treatment related mortality, acute and chronic graft versus host disease (GVHD), and overall survival. • To assess the relationship between KIR ligand mismatch in the host-mediated graft rejection direction and engraftment failure.

Scientific justification: Over the past decade, various strategies have improved the safety of HLA-Haploidentical stem cell transplantation particularly with regards to prevention of GVHD and use with non-myeloablative conditioning in elderly patients. This has increased donor availability for patients with hematologic malignancies who do not otherwise have HLA-matched donors available. Where multiple haploidentical donors are available for a patient, criteria to prioritize donor selection will be important. One potential criterion is the prediction of clinically beneficial donor NK-cell mediated alloreactivity. Alloreactivity in the graft versus tumor direction would be mediated by donor derived NK cells whose killer cell immunoglobulin-like receptors (KIRs) are uninhibited upon tumor target recognition due to absent corresponding HLA class I ligands in the recipient. From small, single center studies, the absence of HLA C1, C2 or Bw4 epitopes in the transplant recipient when present in the haploidentical donor appears to predict donor NK alloreactivity and correlate with improved HSCT outcomes.1-5 HSCT recipients that do not express donor HLA C1 or C2 epitopes have been shown to have NK cell-mediated graft versus leukemia reactions.1 Ruggeri et al2 demonstrated the presence of alloreactive donor NK clones in 17 of 24 recipients of KIR/HLA mismatched transplants at 3 months post-transplant. This was associated with lower relapse rates, longer event-free survival and reduced risk of death in recipients from NK-alloreactive donors. KIR ligand incompatibility has also been shown to correlate with improved clinical outcomes in unrelated donor stem cell transplantation.6-10 From these studies, there is a strong hint that a haploidentical donor(s) who expresses a KIR ligand that is missing in the intended transplant patient may be the preferred donor to select when multiple haplo donors are available. We propose analyzing the large, multi-institution database of haploidentical transplants in the CIBMTR to test the association between missing KIR ligands and transplant outcomes for a variety of hematologic malignancies.

68 Not for publication or presentation Attachment 6

Study population: Inclusion Criteria: • All patients with hematologic malignancies who have received allogeneic HSCT from HLA- haploidentical donor sources will be included for evaluation. This will include acute myeloid leukemia and myelodysplastic syndrome, acute lymphoblastic leukemia, myeloproliferative neoplasms, chronic lymphocytic leukemia and lymphoproliferative neoplasms. • Transplant between year 2000 and 2015.

Variables to be analyzed: Patient Related: • Age • Gender • Race • HLA Typing and antigen assessment (Form 2005): HLA A, Bw4 and C-antigen specificity for HLA C1 and HLA C2. Disease Related: • Date of hematologic malignancy diagnosis • Disease status at last evaluation prior to start of preparative regimen for HSCT • Induction / consolidation therapy • Laboratory assessment at time of transplant: WBC, Blast % (blood and bone marrow) • Cytogenetics risk group Treatment Related: • Date of haploidentical HSCT • Donor type: parent, sibling, child • Donor HLA Typing and antigen assessment: HLA A, Bw4 and C-antigen specificity HLA C1 and HLA C2. • Preparative regimen and preparative regimen intensity • Stem cell source: peripheral blood, bone marrow • GVHD prophylaxis Post-Transplant Assessment: • Time to engraftment • Immune reconstitution 100-days post HSCT (Lymphocyte analyses) • Acute GVHD: Date, Grade, Treatment • Donor lymphocyte infusion • Disease relapse • Therapy given for disease relapse • Current disease status and Date assessed • Death: Date and Cause

Study design: Retrospectively detailed information on each patient’s clinical characteristics, presence or absence of recipient/donor KIR ligand mismatch (HLA Bw4, C1 and C2) and clinical outcome will be tabulated and compared utilizing the CIBMTR and EBMT registries.

Statistical analysis: Two-way ANOVA will be used to compare the median DFS and OS, incidence of GVHD and engraftment failure and time to engraftment between patient groups classified by type of KIR ligand mismatch (HLA

69 Not for publication or presentation Attachment 6

Bw4, C1 or C2) either in the graft versus tumor or graft rejection direction. The various disease categories will be used to further stratify DFS and OS. Event free survival will be measured from the date of transplantation to the date of death or relapse and censored at the date of last follow-up for survivors without relapse. Survival distribution will be estimated using Kaplan-Meier methods and difference between/among groups will be tested using log rank. The effect of KIR-ligand mismatch on time-to-event outcomes stratified by disease type will be analyzed by Cox model with time-dependent covariates after adjusting for the effects of other confounders.

References: 1. Ruggeri L, Capanni M, Casucci M, et al. Role of natural killer cell alloreactivity in HLA-mismatched hematopoietic stem cell transplantation. Blood 1999;94(1):333-339. 2. Ruggeri L, Mancusi A, Capanni M, et al. Donor natural killer cell allorecognition of missing self in haploidentical hematopoietic transplantation for acute myeloid leukemia: challenging its predictive value. Blood 2007;110(1):433-440. 3. Mancusi A, Ruggeri L, Urbani E, et al. Haploidentical hematopoietic transplantation from KIR ligand- mismatched donors with activating KIRs reduces nonrelapse mortality. Blood 2015;125(20):3173-3182. 4. Pende D, Marcenaro S, Falco M, et al. Anti-leukemia activity of alloreactive NK cells in KIR ligand- mismatched haploidentical HSCT for pediatric patients: evaluation of the functional role of activating KIR and redefinition of inhibitory KIR specificity. Blood 2009 Mar 26;113(13):3119-29. 5. Khanuntong S, Kuptawintu P, Upaisilpsathaporn K, et al. The effect of missing KIR ligands, activating KIR genotype and haplotype on the outcome of T-cell-replete hematopoietic stem cell transplantation from HLA-identical siblings in Thai patients. HLA 2016;87:422–431. 6. Willemze R, Rodrigues CA, Labopin M, Sanz G, Michel G, Socie G, et al. KIR-ligand incompatibility in the graft-versus-host direction improves outcomes after umbilical cord blood transplantation for acute leukemia. Leukemia 2009;23(3):492-500. 7. Ruggeri L, Capanni M, Urbani E, et al. Effectiveness of donor natural killer cell alloreactivity in mismatched hematopoietic transplants. Science 2002;295(5562):2097-2100. 8. Cooley S, Weisdorf DJ, Guethlein LA, et al. Donor killer cell Ig-like receptor B haplotypes, recipient HLA-C1, and HLA-C mismatch enhance the clinical benefit of unrelated transplantation for acute myelogenous leukemia. J Immunol 2014 May 15;192(10):4592-600. 9. Giebel S, Locatelli F, Lamparelli T, et al. Survival advantage with KIR ligand incompatibility in hematopoietic stem cell transplantation from unrelated donors. Blood 2003;102(3):814-819. 10. Park M, Seo JJ. Role of HLA in Hematopoietic Stem Cell Transplantation. Bone Marrow Research 2012;2012:680841.

70 Not for publication or presentation Attachment 6

Proposal 1612-04

Title: KIR and HLA Allotypes and Donor NK Alloreactivity in HLA-Haploidentical Transplantation

Brian Shaffer: Memorial Sloan Kettering Cancer Center Katharine C. Hsu: Memorial Sloan Kettering Cancer Center

Hypothesis: Natural killer (NK) cells are potent mediators of the graft versus leukemia phenomenon (GVT) after allogeneic hematopoietic cell transplantation (HCT). The killer Ig-like receptors are a family of NK cell surface proteins that include both activating and inhibitory isoforms that collectively calibrate NK effector function. Inhibitory KIR on donor NK cells interact with HLA class I molecules on target cells to dampen NK cytotoxicity, thereby diminishing GVT. Paradoxically, the same interactions between inhibitory KIR and class I combine to confer functional competence. The latter is particularly relevant in HLA-mismatched HCT, where donor KIR ligands absent in the recipient provoke a potent “missing self” GVT effect. Considerable polymorphism exists within the inhibitory KIR genes, generating KIR alleles with varied binding strength to their respective class I HLA ligands. Because these interactions are stereotypical and predictable, one can estimate the degree of NK inhibition by considering KIR allele typing among prospective donors with the HLA class I typing of the donor and recipient. Robust data now exists supporting the role of KIR gene and allele typing in HLA-matched unrelated donor selection to maximize donor NK reactivity and minimize NK inhibition. The global increase in haploidentical HCT provides a clinical framework in which to study the impact of KIR and HLA allele interactions in an HLA-mismatched environment, in which the donor and the recipient may be either KIR ligand-matched or KIR ligand-mismatched, and at times, mixed for both. In KIR ligand-mismatched HCT fulfilling missing self, donors with the strongest allelic KIR-HLA interactions would have the strongest alloreactive NK response to tumor in comparison to donors with the weakest interactions. In contrast, in KIR ligand-matched HCT, donors with the weakest KIR-HLA interactions would be least inhibited by ligand in the recipient in comparison to donors with the strongest interactions. We therefore hypothesize that donor KIR gene and allele typing, when taken into consideration with donor and recipient HLA, can predict the degree of NK mediated GVT, and thus the risk of relapsed malignancy in the recipient after HLA-haploidentical transplantation. The current proposal will go beyond inquiries that focus on donor-recipient KIR ligand match/mismatch and/or donor KIR genotyping on HCT outcome; instead, we aim to address the role of KIR-HLA allotype interaction in defining the strength of KIR-mediated NK cell activity and inhibition.

Scientific impact: Relapse is the most common cause of death after HLA haploidentical allogeneic hematopoietic cell transplantation. One potentially effective way to reduce relapse is to improve our understanding of factors that guide selection of stem cell donors, leading to a greater degree of graft versus tumor effects without increasing graft versus host disease. NK effector function may be enhanced in this HLA disparate environment and furthermore is not linked to the development of graft versus host disease. Optimizing donor NK function after transplantation is an efficient method to reduce relapse after transplantation. KIR genotyping is widely available and KIR allotyping can be readily scaled to immediately increase access to this technology should the results of this study demonstrate an association with post- transplant outcomes. If positive, this study would be of high impact to the field of HLA haploidentical transplantation and could be immediately applied to the donor selection process.

71 Not for publication or presentation Attachment 6

Specific aims: We will perform KIR gene and allele typing on HLA haploidentical donors. Using known recipient and donor high resolution HLA we can address the following objectives: • To examine the incidence of relapse in transplant recipients with donors that express KIR ligands absent in the recipient (missing self) versus those without missing self i. Distinction between HLA-C and HLA-B KIR ligands may reveal which KIR ligand absence induces the strongest alloreactive effect. ii. Strong KIR-HLA allotype interactions would be predicted to have the highest missing self effect. iii. The greatest alloreactive NK impact is likely to be seen in AML patients. • To determine the incidence of relapse in HLA haploidentical transplant recipients missing one or more inhibitory KIR ligands (missing ligand effect) versus those recipients expressing all KIR ligands • To determine the incidence of relapse in HLA haploidentical transplant recipients whose KIR ligands are matched with the donor, yielding high versus low inhibitory signaling via KIR3DL1, KIR2DL1, and KIR2DL2/3 allotypes. • To explore the role of donor activating KIR and HLA in determining the incidence of relapse after HLA haploidentical transplantation • To explore the role of inhibitory KIR interaction strength in determining the incidence of acute and chronic GVHD after HLA haploidentical transplantation. • To explore the role of inhibitory KIR interaction strength in determining the incidence and severity of CMV infections after HLA haploidentical transplantation.

Scientific justification: Allogeneic hematopoietic cell transplantation (alloHCT) is an effective therapy for high-risk blood and lymph node malignancies. This is due to a recognized GVT effect mediated in part by alloreactive donor NK cells.1, 2 NK effector function is controlled by an array of cell surface receptors including activating and inhibitory killer immunoglobulin-like receptors (KIR).3 Inhibitory KIR recognize cognate epitopes in the human leukocyte antigen (HLA) complex including the HLA-C1 ligand (characterized by Asn77 and Lys80), HLA-C2 ligand (characterized by Ser77 and Lys80) in HLA-C, and the HLA-Bw4 epitope in HLA-B. Under normal conditions, interaction between inhibitory KIR and HLA prevents NK mediated autoimmune reactions. In the context of alloHCT, however, inhibitory KIR signaling dampens NK effector function against malignant cells, facilitating relapse. The importance of donor NK alloreactivity was first described in HLA-mismatched (specifically, KIR-ligand mismatched) alloHCT, where licensed NK reactivity due to “missing self” could be tested. AML patients lacking KIR ligands (HLA-C1, -C2, or -Bw4) present in their haploidentical donors had significantly lower risk for relapse and higher survival.4-7 Limiting the application of this rubric are the approximately 50% of transplant recipients that either express all KIR ligands or who lack a donor with missing self. Therefore, if inhibitory KIR interactions cannot be avoided the next logical step is to use variegations in NK inhibitory strength to select donors with KIR geno- or allotypes that minimize NK inhibition upon contact with recipient malignant cells. Recently, we have investigated the impact of KIR/HLA allotype interaction in HCT. KIR3DL1 and HLA-Bw4 are the most polymorphic of the KIR/KIR ligand pairs, and different pairings of each have been described to result in substantial differences in strength of inhibition.8-11 Highly expressed KIR3DL1 alleles (KIR3DL1hi) interact strongly with HLA-Bw4 alleles containing an isoleucine at position 80 (Bw4-I80) to produce a strong licensing/inhibition signal, while the poorly expressed KIR3DL1 alleles (KIR3DL1lo) have similar interactions with HLA-Bw4 containing threonine at position 80 (Bw4-T80).8-12 We have

72 Not for publication or presentation Attachment 6

demonstrated in a cohort of 1,328 individuals undergoing 9/10 or 10/10 HLA matched unrelated donor allogeneic transplantation for AML that use of donors with high inhibitory interaction between KIR3DL1 and HLA-Bw4 results in increased relapse and poorer overall survival (manuscript in revision). Similarly, Leung and colleagues have shown variegations in inhibitory signaling between KIR2DL1 alleles with arginine versus cysteine at position 245 (245R v. 245C).13 Finally, polymorphism at position 35 of KIR2DL2/3 may alter the inhibitory strength of interactions with HLA-C1.14 In the context of this study we propose to perform inhibitory KIR gene and allele typing on HLA- haploidentical donors for patients undergoing transplant for malignant conditions. We will then test the hypothesis that favoring donors with minimal inhibitory KIR signaling by the patients respective HLA will result in decreased relapse of malignancy after HLA haploidentical transplantation. Secondary endpoints will address whether these interactions alter the likelihood of developing graft versus host disease or control of viral infections after transplantation.

Study population: The study will include adult patients who underwent HLA-haploidentical transplantation for malignancy and who have complete donor and recipient sample and data archived as described below.

Data requirements: This study does not require collection of data outside of the existing CIBMTR forms. The proposed study will use data collected from CIBMTR forms: 2000, 2004, 2005, 2006, and form 2450.

Sample requirements: This study will require a single aliquot of 500 ng or greater of donor DNA for KIR gene and allele typing. No recipient samples are required.

Variables to be analyzed: Patient Related: • Patient age • Patient gender • Patient KPS • Disease histology, molecular markers • Disease risk, disease status at HCT • CMV serostatus • Year of transplant Donor Related: • High resolution HLA typing donor • High resolution HLA typing recipient • Graft source (BM v blood) • Conditioning intensity (MA v RIC v NMA) • Conditioning type i. TBI containing ii. Chemotherapy only • Donor age and gender • CMV serostatus • GVHD prophylaxis regimen: PT-Cy or other

KIR gene and allele typing will be performed at MSKCC as previously described.14, 15

73 Not for publication or presentation Attachment 6

For the primary endpoints the incidence of relapse and death will be determined using Kaplan & Meier estimates and compared across groups. The analysis will occur separately on all disease types, all myeloid malignancies including myelodsyplastic syndrome, and all lymphoid malignancies. For analysis of multiple KIR interactions we propose to perform Cox multivariate analysis to determine independent associations with outcome. We will use descriptive statistics to describe demographic data.

References: 1. Olson JA, Leveson-Gower DB, Gill S, Baker J, Beilhack A, Negrin RS. NK cells mediate reduction of GVHD by inhibiting activated, alloreactive T cells while retaining GVT effects. Blood 2010 May 27;115(21):4293-301. PubMed PMID: 20233969. Pubmed Central PMCID: Pmc2879101. 2. Venstrom JM, Gooley TA, Spellman S, Pring J, Malkki M, Dupont B, et al. Donor activating KIR3DS1 is associated with decreased acute GVHD in unrelated allogeneic hematopoietic stem cell transplantation. Blood 2010 Apr 15;115(15):3162-5. 3. Hsu KC, Chida S, Geraghty DE, Dupont B. The killer cell immunoglobulin-like receptor (KIR) genomic region: gene-order, haplotypes and allelic polymorphism. Immunological reviews 2002 Dec;190:40-52. 4. Giebel S, Locatelli F, Lamparelli T, Velardi A, Davies S, Frumento G, et al. Survival advantage with KIR ligand incompatibility in hematopoietic stem cell transplantation from unrelated donors. Blood 2003 Aug 1;102(3):814-9. 5. Ruggeri L, Capanni M, Casucci M, Volpi I, Tosti A, Perruccio K, et al. Role of natural killer cell alloreactivity in HLA-mismatched hematopoietic stem cell transplantation. Blood 1999 Jul 1;94(1):333-9. 6. Ruggeri L, Capanni M, Urbani E, Perruccio K, Shlomchik WD, Tosti A, et al. Effectiveness of donor natural killer cell alloreactivity in mismatched hematopoietic transplants. Science 2002 Mar 15;295(5562):2097-100. 7. Ruggeri L, Mancusi A, Capanni M, Urbani E, Carotti A, Aloisi T, et al. Donor natural killer cell allorecognition of missing self in haploidentical hematopoietic transplantation for acute myeloid leukemia: challenging its predictive value. Blood 2007;110:433-40. 8. Yawata M, Yawata N, Draghi M, Little A, Partheniou F, Parham P. Roles for HLA and KIR polymorphisms in natural killer cell repertoire selection and modulation of effector function. J Exp Med 2006;203:633-45. 9. O'Connor G, Guinan K, Cunningham R, Middleton D, Parham P, Gardiner C. Functional polymorphism of the KIR3DL1/S1 receptor on human NK cells. J Immunol 2007;178:235-41. 10. Carr W, Pando M, Parham P. KIR3DL1 polymorphisms that affect NK cell inhibition by HLA-Bw4 ligand. J Immunol 2005;175:5222-9. 11. Yawata M, Yawata N, Draghi M, Partheniou F, Little A, Parham P. MHC class I-specific inhibitory receptors and their ligands structure diverse human NK cell repertoires towards a balance of missing- self response. Blood 2008;112:2369-80. 12. Gardiner C, Guethlein L, Shilling H, Pando M, Carr W, Rajalingam R, et al. Different NK cell surface phenotypes defined by the DX9 antibody are due to KIR3DL1 gene polymorphism. J Immunol 2001;166:2992-3001. 13. Bari R, Rujkijyanont P, Sullivan E, Kang G, Turner V, Gan K, et al. Effect of donor KIR2DL1 allelic polymorphism on the outcome of pediatric allogeneic hematopoietic stem-cell transplantation. J Clin Oncol 2013;31:3782-90. 14. Bari R, Thapa R, Bao J, Li Y, Zheng J, Leung W. KIR2DL2/2DL3-E35 alleles are functionally stronger than -Q35 alleles. Scientific Reports 2016 03/31/online;6:23689. 15. Boudreau JE, Le Luduec J-B, Hsu KC. Development of a Novel Multiplex PCR Assay to Detect Functional Subtypes of KIR3DL1 Alleles. PLOS ONE 2014;9(6):e99543.

74 Not for publication or presentation Attachment 6

Table 1. Characteristics of patients undergoing first allogeneic HCT from a haploidentical donor receiving post-transplant Cyclophosphamide between 2007-2015, as reported to the CIBMTR.

MA Not MA Variable N (%) N (%) Number of patients 133 320 Number of centers 27 51 Conditioning regimen intensity Myeloablative 133 (100) 0 RIC 0 33 (10) Nonmyeloablative 0 266 (83) Other 0 21 ( 7) Low resolution HLA matching for HLA-A, -B and –DRB1 3/6 62 (65) 147 (68) 4/6 32 (33) 59 (27) 5/6 1 ( 1) 8 ( 4) 6/6 1 ( 1) 1 (<1) To be determined 37 105 Low resolution HLA matching for HLA-A, -B, -C and –DRB1 4/8 55 (59) 125 (63) 5/8 31 (33) 58 (29) 6/8 6 ( 6) 10 ( 5) 7/8 0 5 ( 3) 8/8 1 ( 1) 0 To be determined 40 122 Recipient age at transplant 18-29 32 (24) 30 ( 9) 30-39 17 (13) 23 ( 7) 40-49 20 (15) 44 (14) 50-59 34 (26) 69 (22) 60-69 28 (21) 122 (38) 70+ 2 ( 2) 32 (10) Median (range) 47 (18-71) 59 (19-78) Recipient sex Male 72 (54) 199 (62) Female 61 (46) 121 (38) Recipient race Caucasian 93 (72) 223 (71) African-American 31 (24) 74 (23) Asian 4 ( 3) 17 ( 5) Pacific Islander 2 ( 2) 0 Native American 0 1 (<1)

75 Not for publication or presentation Attachment 6

MA Not MA Variable N (%) N (%) Unknown 3 (N/A) 5 (N/A) Karnofsky score <90 70 (54) 110 (35) 90- 100 59 (46) 204 (65) Unknown 4 (N/A) 6 (N/A) Sorror Comorbidity score 0 29 (22) 71 (22) 1 32 (24) 51 (16) 2 14 (11) 42 (13) 3 18 (14) 52 (16) 4 16 (12) 32 (10) 5 11 ( 8) 22 ( 7) 6 4 ( 3) 11 ( 3) 7 1 ( 1) 9 ( 3) 8 0 6 ( 2) 10 0 1 (<1) To be determined (review needed) 8 ( 6) 23 ( 7) Disease at transplant AML 67 (50) 125 (39) ALL 32 (24) 37 (12) CML 4 ( 3) 12 ( 4) MDS 15 (11) 51 (16) MPS 4 ( 3) 19 ( 6) Other acute leukemia 1 ( 1) 1 (<1) Other leukemia 1 ( 1) 11 ( 3) NHL 8 ( 6) 51 (16) Hodgkins Lymphoma 1 ( 1) 13 ( 4) Donor / recipient sex match Male / male 49 (37) 136 (43) Male / female 38 (29) 74 (23) Female / male 23 (17) 63 (20) Female / female 23 (17) 47 (15) Donor / recipient CMV serostatus Negative / negative 26 (20) 72 (23) Negative / positive 37 (29) 88 (28) Positive / negative 11 ( 9) 41 (13) Positive / positive 53 (42) 113 (36) Unknown 6 (N/A) 6 (N/A) Graft type Bone marrow 43 (32) 207 (65) Peripheral blood 90 (68) 113 (35)

76 Not for publication or presentation Attachment 6

MA Not MA Variable N (%) N (%) Donor age by decade To be determined / N/A 1 ( 1) 0 0-9 1 ( 1) 1 (<1) 10-19 9 ( 7) 18 ( 6) 20-29 34 (26) 59 (18) 30-39 29 (22) 83 (26) 40-49 21 (16) 82 (26) 50+ 38 (29) 77 (24) Year of transplant 2007 0 2 ( 1) 2008 1 ( 1) 27 ( 8) 2009 4 ( 3) 14 ( 4) 2010 0 12 ( 4) 2011 2 ( 2) 0 2012 5 ( 4) 9 ( 3) 2013 21 (16) 82 (26) 2014 42 (32) 86 (27) 2015 58 (44) 88 (28) GvHD Prophylaxis Cy + Tac + MMF ± others 120 (90) 302 (94) Cy + Tac + MTX ± others (not MMF) 0 5 ( 2) Cy + Tac ± others (not MMF, MTX) 1 ( 1) 6 ( 2) Cy + CsA + MMF ± others (not Tac) 12 ( 9) 2 ( 1) Cy + MMF ± others (not Tac, CsA) 0 5 ( 2) Follow-up among survivors, months N Eval 82 178 Median (range) 12 (3-51) 13 (2-97)

77 Not for publication or presentation Attachment 6

Table 2. Characteristics of Patients for KIR-Haploidentical Cases with Related Donor Samples Available in the CIBMTR Biorepository

Variable N (%) Number of patients 127 Number of centers 22 Recipient age at transplant 10-19 2 ( 2) 20-29 22 (17) 30-39 4 ( 3) 40-49 16 (13) 50-59 83 (65) Median (range) 55 (20-78) Number of sample types Samples available for recipient and donor 123 (97) Samples available for donor only 4 ( 3) Recipient race / ethnicity Caucasian, non-Hispanic 77 (62) African-American, non-Hispanic 33 (26) Asian, non-Hispanic 4 ( 3) Hispanic, Caucasian 11 ( 9) Unknown 2 (N/A) Recipient sex Male 71 (56) Female 56 (44) Karnofsky score 10-80 51 (40) 90-100 75 (59) Missing 1 ( 1) Graft type Bone marrow 77 (61) Peripheral blood 49 (39) Others 1 ( 1) Conditioning regimen Myeloablative 46 (36) RIC 19 (15) Nonmyeloablative 50 (39) Other 12 ( 9) Donor age at donation 0-9 1 ( 1) 10-19 8 ( 6) 20-29 37 (29) 30-39 33 (26)

78 Not for publication or presentation Attachment 6

Variable N (%) 40-49 23 (18) 50+ 25 (20) Median (range) 35 (9-69) Disease at transplant AML 58 (46) ALL 23 (18) CML 6 (5) MDS 28 (22) NHL 8 ( 6) Hodgkins Lymphoma 4 ( 3) Disease status at transplant Early 49 (39) Intermediate 18 (14) Advanced 37 (29) Other 23 (18) Donor / recipient CMV serostatus Negative / negative 16 (13) Negative / positive 40 (31) Positive / negative 15 (12) Positive / positive 53 (42) Unknown 3 ( 2) GvHD Prophylaxis Cyclophosphamide 127 (100) Donor / recipient sex match Male / Male 53 (42) Male / Female 39 (31) Female / Male 18 (14) Female / Female 17 (13) Year of transplant 2011 1 ( 1) 2012 3 ( 2) 2013 23 (18) 2014 54 (43) 2015 46 (36) Follow-up among survivors, months N Eval 83 Median (range) 12 (3-39)

79 Not for publication or presentation Attachment 7

TO: Immunobiology Working Committee Members

FROM: Stephanie Lee, MD, MPH; Co-Scientific Director for the Immunobiology WC Stephen Spellman, MBS; Co-Scientific Director for the Immunobiology WC

RE: Studies in Progress and Publication Summary

Studies in Progress Summary

HLA GENES

IB12-02c: Prospective assignment of HLA-DPB1 T cell epitope (TCE) group mismatches by functional distance scores compared to the functional TCE assignment algorithm (K Fleischhauer) The primary objective of the study is to test whether the functional distance (dFD) scores adds to or replaces the TCE group assignment score for permissive and non-permissive HLA-DPB1 mismatches. Abstract submitted to 2016 EBMT. Additional analyses were conducted to better understand the relationship between TCE permissiveness assignments and dFD scores. Plan to be in manuscript preparation by June 2017.

IB13-01: The effect of allele-level HLA-matching on survival after umbilical cord blood transplantation for non-malignant diseases in children (P Veys/M Eapen) The primary objective of the study is to evaluate the impact of allele-level HLA matching at HLA-A, B, C and DRB1 on umbilical cord blood transplantation for non-malignant disease in children. This is a collaborative study between the CIBMTR, UK and Eurocord. The datafile was finalized in December 2016 and analysis is in progress. Plan to be in manuscript preparation by June 2017.

IB13-08: Prediction of acute graft versus host disease post HSCT transplantation using predictive modeling on an ensemble learning framework (R Abdi/S Haneuse/C Lee) The primary objective of the study is to evaluate predictors of acute GVHD following related or unrelated donor HCT for hematological malignancies using an ensemble learning framework refered to as the ”super learner”. The manuscript is in progress. Plan to submit to Blood by June 2017.

IB13-09: The development of machine learning based classifiers to define the alloreactivity of HLA mismatches in unrelated donor hematopoietic stem cell transplantation (Y Louzoun) The main goal of this study is to classify alloreactivity of HLA Class I mismatches based on peptide repetoire differences and apply Machine Learning based classifiers to evaluate the impact on HCT outcome. The manuscript is in progress. Plan to submit to BBMT by June 2017.

IB14-01: Impact of human leukocyte antigen haplotypes on outcomes of allogeneic transplantation for B-cell non-Hodgkin lymphomas (B William/M de Lima/M Fernandez-Vina/B Hill) This study will test the hypothesis that recipient HLA type will predict outcomes after 8/8 matched URD transplantation. The

80 Not for publication or presentation Attachment 7

results were presented as a poster abstract at the 2016 ASH meeting. Manuscript preparation is in progress. Plan to submit to BBMT by June 2017.

IB14-02: Structural/Functional models of HLA for data mining of permissive mismatching in allogeneic stem cell transplantation (L Gragert) The two main objectives of the study are to: 1) determine sets of HLA alleles that when mismatched lead to better overall survival and lower incidence of acute GVHD than other mismatches in allogeneic transplantation, and 2) determine specific structural/functional properties of HLA which lead to better overall survival and lower incidence of acute GVHD when mismatched in allogeneic transplantation. Structural properties include single amino acid polymorphisms (SAP) and sequence feature variant type (SFVT). SFVT categories are often annotated with functional features of these amino acid motifs on the HLA protein. The manuscript is in progress. Plan to submit to BBMT by June 2017.

IB14-08: Development and validation of a clinical unrelated donor selection score (B Shaw/SJ Lee) The aim of the study is to develop and validate a donor risk score for matched unrelated donors that helps to identify the optimal donor among those available. The donor risk score model developed from the discovery cohort failed in the initial validation cohort. A secondary validation cohort was identified and the analysis is in progress. Plan to submit to Blood by June 2017.

IB15-01: The impact of single nucleotide gene polymorphisms (SNP) in the gamma block of the major histocompatibility complex (MHC) on unrelated donor hematopoietic cell transplants (HCT) for hematological malignancies (M Askar/R Sobecks) The objectives of this study are to investigate the incidence of gamma block SNPs in matched and mismatched URD transplantation, and to correlate these SNPs with clinical outcome. The gamma block contains inflammatory and regulatory genes and mismatching is an indicator of haplotype recombination. In datafile preparation. Plan to be in manuscript preparation by June 2017.

IB16-01: The role of HLA-E compatibility in the prognosis of acute leukemia patients undergoing 10/10 HLA matched unrelated HSCT (C Tsamadou/ D Furst/ J Mytilineos) The primary objective of the study is to investigate whether HLA-E donor-recipient incompatibility is associated with better HCT outcomes in a cohort of acute leukemia patients transplanted with HLA-A, B, C, DRB1 and DQB1 matched unrelated donors. The PI secured funding from a German foundation to cover the costs of the HLA-E testing in December 2016. The protocol is under development. Plan to be in sample testing and datafile preparation by June 2017.

IB16-02: Use of HLA structure and function parameters to understand the relationship between HLA disparity and transplant outcomes (LA Baxter-Lowe) The main objective of the study is to determine the relationship between HLA disparities ranked by their impact on T cell receptor docking, peptide binding and the combination of docking and binding. The protocol is under development. Plan to be in datafile preparation by June 2017.

CYTOKINE/CHEMOKINE IB14-03a: The prognostic impact of somatic mutations and levels of CXC chemokine ligands on post hematopoietic cell transplantation (HCT) outcomes in patients with myelodysplastic syndromes (MDS) (W Saber/B Dhakal) The primary objective of the study is to evaluate the plasma CXCL4 and CXCL7 chemokine levels in patients undergoing transplantation for MDS in comparison to those with other conditions. The preliminary results of the analysis were presented as a poster at the 2016 ASH annual

81 Not for publication or presentation Attachment 7

meeting. Planning for testing of additional disease cohorts is underway with the intent to be in sample typing by June 2017.

NK/KIR

R02-40/R03-63: Choosing donors with favorable KIR B genotypes for unrelated hematopoietic cell transplantation (HCT) results in superior relapse protection and better relapse-free survival for patients with acute myeloid leukemia (AML) (J Miller) This is an ongoing study in support of Dr. Miller’s NK Biology program project grant. Ongoing.

IB15-02: Natural killer cell genomics and outcomes after HCT for chronic lymphocytic leukemia (V Bachanova/J Miller/D Weisdorf/S Cooley) The study is based on recent advances linking NK cell genetics with outcomes after allogeneic donor transplantation. The primary hypothesis is that the gene content regulating donor-derived NK cell function influences the graft-versus-leukemia reaction following allogeneic donor hematopoietic cell transplantation (HCT) for chronic lymphocytic leukemia (CLL). Sample typing in process. Plan to be in analysis by June 2017.

IB15-03: Effect of Killer immunoglobulin like receptors on allogeneic HCT for pediatric acute leukemia (M Verneris/J Miller/S Cooley) The primary aim of the study is to examine the association of donor KIR genotype (A/A vs B/x) on relapse and disease free survival (DFS) in children undergoing allogeneic transplantation (allo-HCT) with acute lymphoblastic leukemia (ALL) and acute myeloid leukemia (AML). A draft protocol is in process. Plan to be in sample typing by June 2017.

OTHER GENES IB06-05: Use of high-resolution HLA data from the NMDP for the International Histocompatibility Working Group in HCT (E Petersdorf) This study proposes to identify novel major histocompatibility complex resident SNPs of clinical importance. HLA matched pairs were genotyped for 1120 MHC SNPs and correlated with outcomes. This is a collaborative study with the IHWG. Ongoing.

IB09-04: Donor/recipient gene polymorphisms of drug metabolism and in innate immune response post allele-matched unrelated donor hematopoietic stem cell transplantation (HCT) (V Rocha) This study is designed to validate associations between polymorphisms in drug metabolism and innate immune response genes and outcomes previously identified in matched sibling donor HCT in the HLA- matched unrelated donor HCT setting. The manuscript is in progress. Plan to submit to BBMT by June 2017.

IB09-05: Identification of functional single nucleotide polymorphisms (SNPs) in umbilical cord blood transplant (E Petersdorf) The primary hypothesis of the study is that umbilical cord blood units and recipients differ for genome-wide single nucleotide polymorphism and gene copy number variation and that these differences may define putative transplant outcome determinants. This is a collaborative study with IHWG. Ongoing.

IB09-06/RT09-04: Genetic susceptibility to transplant-related mortality after matched unrelated stem cell transplant (T Hahn) This is a joint study with the Regimen Related Toxicity working committee and is supported by an R01 grant to Drs. Hahn and Sucheston-Campbell. This study will test for a genetic association with transplant-related and overall mortality in recipients of myeloablative and reduced

82 Not for publication or presentation Attachment 7

intensity conditioning matched unrelated donor HCT. Multiple oral abstracts were presented at the 2015, 2016 ASH and 2016, 2017 BMT Tandem meetings. Multiple manuscripts are in process. Ongoing.

IB09-07: Clinical significance of genome-wide variation in unrelated donor hematopoietic stem cell transplantation (HCT) (E Petersdorf) This study is designed to assess the impact of genome-wide variation between donors and recepients in HLA matched unrelated donor HCT. This is a collaborative study with the IHWG. Ongoing.

IB13-04: Discrepancy analysis of microsatellite loci as a proxy measure for ancestral differentiation between donors and recipients: correlation between high scores and poorer overall survival in high resolution matched unrelated donor transplantation (J Harvey/C Steward/V Rocha) The primary objective of the study is to validate a single center study noting a correlation between microsatellite variation in donors and recipients and HCT outcome. Sample testing is in progress. Plan is to still be in sample typing by June 2017.

IB14-04: Assessing the similarity of the T cell receptor repertoire in allogeneic hematopoietic stem cell recipients with the same single human leukocyte mismatches (EH Meyer) The goal of this study is to measure T cell alloreactivity following hematopoietic stem cell transplantation by examining cases where transplant recipients have the same HLA single, double or multiple HLA mismatch, to see if they develop similar alloreactivity. Control transplant recipients with the same HLA type who did not receive an HLA-mismatched transplant will be also analyzed. Sample typing in progress. Plan to be in manuscript preparation by June 2017.

IB14-05: mtDNA haplotypes and unrelated donor transplant outcomes (M Verneris/L Spector) The goal of this study is to test whether patient or donor mitochondrial haplotypes predict outcomes of unrelated donor transplantation, particularly GVHD, relapse and survival. The study is supported by an R01 grant to Drs. Verneris and Ross. Sample typing and datafile preparation is in progress. Plan to be in analysis by June 2017.

IB15-04: Clinical outcomes among hematopoietic stem cell transplant recipients as a function of socioeconomic status and related transcriptome differences (J Knight/JD Rizzo/S Cole) The primary hypothesis of the study is that increased expression of the conserved transcriptional response to adversity (CTRA) gene profile will be associated with lower socioeconomic status (SES) and worse clinical outcomes among a group of unrelated donor (URD) myeloablative (MA) acute myelogenous leukemia (AML) recipients in CR1. Sample typing in process. Plan to be in manuscript preparation by June 2017.

IB15-05: Secondary findings in exome sequencing data (S Savage/S Gadalla) The goal of the study is to incorporate exome sequencing results from the IB10-01 cohort into a larger cohort of exome sequenced cases and controls from the NCI Division of Cancer Epidemiology and Genetics to evaluate the frequency of incidental findings. Analysis in process. Plan to submit by June 2017.

IB15-06b: Validation of Telomatch for donor telomere length and outcomes after hematopoietic stem cell transplantation for acute leukemia (S Gadalla/S Savage/DJ Loftus/E Hytopoulos/M Yeager/C Dagnall) The main objective of the study was to validate the utility of the TeloMatch prognostic scoring system, preliminary results presented as an oral abstract at the 2015 ASH annual meeting, on an independent cohort. Manuscript preparation in process. Plan to submit to Blood following acceptance of the primary IB15-06a manuscript that was submitted to JCO in January 2017.

83 Not for publication or presentation Attachment 7

IB15-06c: Recipient telomere length and outcomes after hematopoietic stem cell transplantation for acute leukemia (S Gadalla/S Savage/DJ Loftus/E Hytopoulos/M Yeager/C Dagnall) The primary objective of the study was to evaluate the association between recipient telomere length and outcomes after unrelated donor HCT for acute leukemia. The results will be presented as a poster abstract and the 2017 BMT Tandem meetings and manuscript preparation is in process. Plan to submit to BBMT by June 2017.

IB15-07: Functional genetic variants of the ST2 gene in pairs of recipient and donors for risk stratification of GVHD and TRM outcomes (S Paczesny) The serum biomarker sST2 is associated with an increased risk for therapy-resistant GVHD and death. The primary hypothesis is that similar to the heritability of sST2 in the Framingham Offspring Cohort explaining sST2 as a predictor of cardiovascular risk, the 16 SNPs most associated with sST2 will determine which donor /recipient pair is at risk of developing acute graft-versus-host disease (aGVHD) and transplant-related mortality (TRM) following allogeneic hematopoietic cell transplantation in a well-controlled cohort. The study will use the GWAS typing results and dataset from the DISCOVERY-BMT (IB09-06/RT09-04) cohort to evaluate the association of 16 sST2 SNPs. Datafile preparation is underway. Plan to be in analysis by June 2017.

IB16-03: Role of recipient and donor genetic polymorphisms in interferon lambda 4 (INFL4) on outcomes after unrelated allogeneic cell transplant (S Gadalla/L Prokunina-Olsson) The primary goal of the study is to evaluate the effect of recipient and donor genetic polymorphisms in the type-III interferon, interferon lambda 4 (INFL4) on outcomes following unrelated donor HCT for SAA and acute leukemia. Sample typing is underway. Plan to be in analysis by June 2017.

SENSITIZATION and TOLERANCE IB11-01: Analysis of the NIMA effect on the outcome of unrelated PBSC/BM transplantation (G Ehninger/J van Rood/A Schmidt) The goal of this study is to determine whether matching for NIMA in the selection of unrelated donors can lead to better outcomes This is a collaborative study with the EBMT Immunobiology Working Party. Poster presented at 2015 ASH. Manuscript in progress. Plan to submit to BBMT by June 2017.

IB14-06: Donor-Specific anti HLA antibodies, allele and antigen level HLA mismatches in the outcomes of transplantation of non-malignant diseases with unrelated donors (M Fernandez-Vina/A Woolfrey) The goal of this study is to test whether pre-existing donor-specific anti-HLA antibodies are associated with higher rates of graft failure and worse outcomes in HLA-mismatched unrelated donor transplants for non-malignant diseases. Analysis in process. Plan to be in manuscript preparation by June 2017.

MINOR HISTOCOMPATIBILITY ANTIGENS IB11-08: Synergism between minor and major histocompatibility antigens (E Spierings) The goal of the study is to evaluate the interaction between HLA-DPB1 mismatching and female into male (HY mismatched) transplantation to determine whether there is a synergistic effect on transplant outcome. Multivariate analysis revealed a synergistic interaction between HLA-DP mismatching and H-Y mismatching on acute GVHD grades III-IV. This is a collaborative study with IHWG. Ongoing.

84 Not for publication or presentation Attachment 7

Publication Summary

IIB08-06 Rocha V, Ruggeri A, Spellman S, Wang T, Sobecks R, Locatelli F, Askar M, Michel G, Arcese W, Iori AP, Purtill D, Danby R, Sanz GF, Gluckman E, Eapen M. Killer cell immunoglobulin-like receptor ligand matching and outcomes after unrelated cord blood transplantation acute myeloid leukemia. Biol Blood Marrow Transplant. 2016 Jul 1; 22(7):1284-1289. doi:10.1016/j.bbmt.2016.04.007. Epub 2016 Apr 16. PMC4915071. A collaborative retrospective study between the CIBMTR, Eurocord and EBMT evaluating the role of Killer Immunoglobulin-Like Receptor (KIR) ligand matching in unrelated cord blood transplantation. There were no significant differences in non-relapse mortality (NRM), relapse, and overall mortality by KIR-ligand match status. However, among recipients of 3-5/8 HLA- matched transplants, NRM and overall mortality were higher with KIR-ligand mismatched compared with KIR-ligand matched transplants. These data do not support selecting cord blood units based on KIR-ligand match status for transplants mismatched at 1 or 2 HLA loci. Although transplants mismatched at 3 or more HLA loci are not recommended, avoiding KIR-ligand mismatching in this setting lowers mortality risks.

IB08-08 Goyal RK, Lee SJ, Wang T, Trucco M, Haagenson M, Spellman SR, Veneris M, Ferrell RE. Novel HLA-DP region susceptibility loci associated with severe acute GvHD. Bone Marrow Transplant. doi:10.1038/bmt.2016.210. Epub 2016 Sep 5. A retrospective genome-wide association study (GWAS) using samples from the CIBMTR Repository to identify recipient and donor genes associated with the risk of acute GvHD. Three novel susceptibility loci in the HLA-DP region of recipient genomes that were associated with III- IV acute GvHD. This report contributes to emerging data showing clinical significance of the HLA-DP region genetic markers beyond structural matching of DPB1 alleles.

IB10-01b Gadalla SM, Khincha PP, Katki HA, Giri N, Wong JYY, Spellman S, Yanovski JA, Han JC, De Vivo, Alter BP, Savage SA. The limitations of qPCR telomere length measurement in diagnosing dyskeratosis congenita. Molecular Genetics & Genomic Medicine. 2016 Jul 1; 4(4):475-479. doi:10.1002/mgg3.220. Epub 2016 Mar 20. PMC4947866. A retrospective study using biospecimens from the CIBMTR Repository to compare qualitative real-time polymerase chain reaction (qPCR) assessment of telomere length versus the current gold standard test using flow cytometry florescence in-situ hybridization (Flow FISH) for the diagnosis of dyskeratosis congenital (DC), a telomere biology disorder characterized by very short telomeres. The qPCR assay failed to identify 60% of patients with known DC. Flow FISH remains the gold standard for DC diagnosis.

IB10-01c Gadalla SM, Wang T, Dagnall C, Haagenson M, Spellman SR, Hicks B, Jones K, Katki HA, Lee SJ, Savage SA. Effect of recipient age and stem cell source on the association between donor telomere length and survival after allogeneic unrelated hematopoietic cell transplantation for severe aplastic anemia. Biol Blood Marrow Transplant. 2016 Dec 1; 22(12):2276-2282. doi:10.1016/j.bbmt.2016.09.012. Epub 2016 Sep 15. PMC5116252. A retrospective validation analysis using samples from the CIBMTR Repository to evaluate the association between donor leukocyte relative telomere length (RTL) and post- HCT survival in patients with severe aplastic anemia. Data from the validation cohort found no association between donor RTL and patient survival, but further analysis identified differences by recipient age and stem cell source. Analyses using data from the discovery and validation cohorts showed

85 Not for publication or presentation Attachment 7

a statistically significant survival benefit only in <40-year-old patients receiving bone marrow grafts. The study suggested that the association between donor RTL and post-HCT outcomes may vary by recipient age and stem cell source. A larger study is needed to account for multiple comparisons and to further test the generalizability of the findings.

IB10-04 Arora M, Lee SJ, Spellman SR, Weisdorf DJ, Guan W, Haagenson M, Wang T, Horowitz MH, Verneris MR, Fleischhauer K, Hsu K, Thyagarajan B. Validation study failed to confirm an association between genetic variants in the base excision repair pathway and transplant-related mortality and relapse after hematopoietic cell transplantation. Biol Blood Marrow Transplant. 2016 Aug 1; 22(8):1531-1532. doi:10.1016/j.bbmt.2016.04.020. Epub 2016 May 4. A retrospective study using biospecimens from the CIBMTR Repository to validate findings from a single institution study that identified an association between genetic polymorphism in DNA repair genes and transplant outcome. The results of the study were negative with no associations found in the larger CIBMTR cohort. The manuscript stressed the importance of validating associations between genetic variants and transplant outcomes in independent datasets prior to implementing in clinical practice.

IB12-04 Hoff GA, Fischer JC, Hsu K, Cooley S, Miller JS, Wang T, Haagenson M, Spellman S, Lee SJ, Uhrberg M, Venstrom JM, Verneris MR. Recipient HLA-C haplotypes and miRNA 148a/b binding sites have no impact on allogeneic hematopoietic cell transplantation outcomes. Biol Blood Marrow Transplant. doi:10.1016/ j.bbmt.2016.09.028. Epub 2016 Oct 13. NK cells are important in graft-versus-leukemia responses after HCT. A variety of surface receptors dictate NK cell function, including KIR recognition of HLA-C. Previous single-center studies found that HLA-C epitopes and surface expression can influence HCT outcomes. In this retrospective analysis, neither HLA-C surface expression, as determined by miR-148a/b, nor recipient HLA-C epitopes (C1, C2) were associated with any outcomes.

IB12-05/RT10-01 Kornblit B, Wang T, Lee SJ, Spellman SR, Zhu X, Fleischhauer K, Müller C, Johansen JS, Vindelov L, Garred P. YKL-40 in allogeneic hematopoietic cell transplantation after AML and myelodysplastic syndrome. Bone Marrow Transplant. doi:10.1038/bmt.2016.192. Epub 2016 Jul 18. A retrospective study using plasma samples from the CIBMTR Repository to evaluate the prognostic value of recipient and donor pre-transplant levels of YKL-40, an inflammatory biomarker, on transplant outcomes. Patient YKL-40 levels did not aid in the risk stratification, but elevated donor levels were associated with a trend towards higher rates of grades II-IV acute graft versus host disease (aGvHD). This suggests that YKL-40 may aid donor selection when multiple, otherwise equal, donors are available.

IB12-06 Bachanova V, Weisdorf DJ, Wang T, Marsh SGE, Trachtenberg E, Haagenson MD, Spellman SR, Ladner M, Guethlein LA, Parham P, Miller JS, Cooley SA. Donor KIR B genotype improves progression- free survival of non-Hodgkin lymphoma patients receiving unrelated donor transplantation. Biol Blood Marrow Transplant. 2016 Sep 1; 22(9):1602-1607. doi:10.1016/j.bbmt.2016.05.016. Epub 2016 May 21. PMC4981536. A retrospective analysis using biospecimens from the CIBMTR Repository. Donor killer immunoglobulin-like receptor (KIR) genotypes are associated with relapse protection and survival after HCT for acute myelogenous leukemia, but had not been evaluated in non-Hodgkin

86 Not for publication or presentation Attachment 7

lymphoma (NHL). In 10/10 HLA matched HCT for NHL, use of KIR B/x haplotype donors was associated with significantly reduced relapse risk and improved PFS. The relapse protection was not observed in HLA-mismatched HCT.

IB13-02 Marino SR, Lee SM, Binkowski TA, Wang T, Haagenson M, Wang H-L, Maiers M, Spellman S, van Besien K, Lee SJ, Karrison T, Artz A. Identification of high-risk amino-acid substitutions in hematopoietic cell transplantation: a challenging task. Bone Marrow Transplant. 2016 Oct 1; 51(10):1342-1349. doi:10.1038/bmt.2016.142. Epub 2016 May 23. PMC5052106. A retrospective validation analysis of 19 previously identified amino-acid substitution and position types (AASPT) conferring higher risks of transplant related mortality and GvHD following HLA mismatched unrelated donor HCT. When tested in an independent validation cohort of 3530 patients, none of the AASPT were validated as high risk, however. The analysis does not support avoidance of any specific class I AASPT for unrelated donor HCT.

IB13-03 Lazaryan A, Wang T, Spellman SR, Wang H-L, Pidala J, Nishihori T, Askar M, Olsson R, Oudshoorn M, Abdel-Azim H, Yong A, Gandhi M, Dandoy C, Savani B, Hale G, Page K, Bitan M, Reshef R, Drobyski W, Marsh SGE, Schultz K, Müller CR, Fernandez-Viña M, Verneris MR, Horowitz MM, Arora M, Weisdorf DJ, Lee SJ. Human leukocyte antigen supertype matching after myeloablative hematopoietic cell transplantation with 7/8 matched unrelated donor allografts: A report from the Center for International Blood and Marrow Transplant Research. Haematologica. 2016 Oct 1; 101(10):1267-1274. doi:10.3324/haematol.2016.143271. Epub 2016 May 31. PMC5046657. A retrospective analysis on the impact of matched and mismatched HLA-A -B, -C and -DRB1 supertypes on clinical outcomes following unrelated donor HCT for acute leukemias or myelodysplasia/myeloproliferative disorders. The analysis found no associations between supertype mismatching and survival. However, HLA-B supertype mismatches were associated with increased risk of grade II-IV acute GvHD.

IB13-05 Askar M, Sobecks R, Wang T, Haagenson M, Majhail N, Madbouly A, Thomas D, Zhang A, Fleischhauer K, Hsu K, Verneris M, Lee SJ, Spellman SR and Fernández-Viña M. MHC Class I Chain-Related Gene A (MICA) Donor-Recipient Mismatches and MICA-129 Polymorphism in Unrelated Donor Hematopoietic Cell Transplants (HCT) For Hematological Malignancies: A CIBMTR Study. Biol Blood Marrow Transplant. 2016 Dec 14. pii: S1083-8791(16)30521-3. doi: 10.1016/j.bbmt.2016.11.021. [Epub ahead of print] PubMed PMID: 27987385. Single-center studies had previously reported associations of MHC Class I Chain-Related Gene A (MICA) polymorphisms and donor-recipient MICA mismatching with graft-versus-host disease (GVHD) after unrelated donor hematopoietic cell transplantation (HCT). In this study, we investigated the association of MICA polymorphism (MICA-129, MM versus MV versus VV) and MICA mismatches after HCT with 10/10 HLA-matched (n = 552) or 9/10 (n = 161) unrelated donors from the CIBMTR Repository. No significant associations between MICA mismatching or MICA polymorphisms were observed.

IB14-03b Lindsley RC, Saber W, Mar B, Medd, Wang T, Haagenson M, Grauman P, Zhu Z, Spellman S, Lee SJ, Verneris M, Hsu K, Fleischhauer K, Cutler C, Antin JH, Neuberg D, Ebert BL. Prognostic Mutations in Myelodysplastic Syndrome After Stem Cell Transplantation. In press. New Engl J Med. Specific mutations have been associated with poor prognosis in AML and MDS. Targeted mutational analysis was performed on samples from 1514 patients with MDS who underwent

87 Not for publication or presentation Attachment 7

unrelated donor transplantation. TP53 mutations were seen in 19% of patient and associated with worse survival (p<0.001) and shorter remission (p<0.001). RAS mutations in patients 40 years and older were associated with worse survival (p=0.001) and higher relapse. JAK2 mutations in patients 40 years and older were associated with worse survival (p=0.001) and higher treatment-related mortality. In young patients, 4% had compound heterozygous mutations in the SBDS gene and accompanying TP53 mutations; this subset had a poor prognosis. This paper shows the clinical significance of mutations in patients undergoing unrelated donor transplantation for MDS, and may provide important prognostic information to help make decisions about transplantation.

IB10-01c Gadalla SM, Ballew BJ, Haagenson M, Spellman S, Hicks B, Alter BP, Zhu B, Zhou W, Yeager M, Wang T, Fleischhauer K, Hsu K, Verneris M, Freedman N, Lee SJ, Savage SA. Germline Mutations in Marrow Failure Predisposition Genes in Patients Receiving Unrelated Donor Hematopoietic Cell Transplant for Severe Aplastic Anemia. Submitted. J Clin Invest. Patients with severe aplastic anemia and idiopathic bone marrow failure syndromes may harbor undetected pathogenic mutations. Whole exome sequencing and genome-wide SNP analyses were performed on 608 patients (516 reported to CIBMTR as acquired aplastic anemia and 92 with inherited bone marrow failure syndromes [IBMFS]) who had unrelated donor transplants for marrow failure syndromes. Deleterious variants were detected in IBMFS genes in 59.8% of the IBMFS population and 10.5% of the aplastic anemia population (5% were in dyskeratosis congenita genes). Aplastic anemia patients who were 40 years old or less and had dyskeratosis congenita mutations or 2 or more pathologic variants in any tested IBMFS gene had worse survival. This study suggests that patients with IBMFS or aplastic anemia should be tested for pathogenic variants before transplantation.

IB12-02B Fleischhauer K, Ahn KW, Wang H-L, Zito L, Crivello P, Müller C, Verneris M, Shaw BE, Pidala J, Oudshorn M, Lee SJ and Spellman SR. Directionality of non-permissive HLA-DPB1 T-cell epitope group mismatches in 8/8 matched unrelated donor hematopoietic cell transplantation. Submitted. Leukemia. Non-permissive HLA-DPB1 mismatches are associated with worse outcomes in unrelated donor transplantation. This study tested whether the directionality (graft-vs.-host, host-vs.- graft, or bidirectional) of the HLA-DPB1 mismatch is associated with outcome in 3281 8/8 HLA- matched unrelated donor transplants. No differences in transplant-related mortality or overall survival were observed. This study suggests that directionality does not matter, and all non- permissive HLA-DPB1 mismatches are associated with worse outcomes.

IB12-03 Madbouly A, Wang T, Haagenson M, Paunic V, Vierra-Green C, Fleischhauer K, Hsu KC, Verneris MR, Majhail NS, Lee SJ, Spellman SR and Maiers M. Investigating the association of genetic admixture and donor/recipient genetic disparity with transplant outcomes Submitted. Biol Blood Marrow Transplant. Racial and ethnic minority status has been associated with worse outcomes after unrelated donor transplantation in some studies. Racial and ethnic identity is usually self-reported. This study tested whether assignment of racial and ethnic status by genetic testing in 1378 10/10 HLA-matched pairs improved categorization and strengthened the association with outcomes. Results suggested that increased African genetic ancestry in either patients or donor was associated with worse outcomes but that self-identification was highly correlated with genetic assignments. There was no significant association between donor-recipient genetic distance and outcomes. This study suggests that self-identification is a reasonable surrogate for genetic

88 Not for publication or presentation Attachment 7

ancestry, and confirms that African Americans experience worse outcomes with unrelated donor transplantation.

IB15-06a Gadalla SM, Wang T, Loftus D, Friedman L, Dagnall C, Haagenson M, Spellman SR, Buturovic L, Blauwkamp M, Shelton J, Fleischhauer K, Hsu KC, Verneris MR, Krstajic D, Hicks B, Jones K, Lee SJ, Savage SA. Donor telomere length and outcomes after allogeneic unrelated hematopoietic cell transplant in patients with acute leukemia. Submitted. J Clin Oncol. Longer telomere length in unrelated donors is associated with better survival in patients transplanted for severe aplastic anemia. This study tested whether a similar observation exists for patients transplanted for acute leukemia. A total of 1084 donor samples from 8/8 HLA- matched pairs were tested for relative telomere length by PCR. Patients were divided into a discovery cohort and two validation cohorts. No associations were observed between donor relative telomere length and any outcomes.

IB15-06b Buturovic L, Shelton J, Spellman SR, Wang T, Friedman L, Loftus D, Hesterberg L, Woodring T, Fleischhauer K, Hsu K, Verneris M, Haagenson M, Lee SJ. Evaluation of a Machine Learning-Based Prognostic Model for Unrelated Hematopoietic Cell Transplantation Donor Selection. To be submitted. An algorithm to select the best unrelated donor to maximize patient survival has proven elusive. This study used machine learning techniques to try to create a multivariate classifier for unrelated donor selection that maximized 5 year overall survival of patients with acute leukemia. Although the algorithm appeared promising in the discovery cohort, it failed validation in an independent cohort of 522 patients. This suggests that pre-transplant information currently available is not able to predict 5-year survival of acute leukemia patients with sufficient accuracy to be clinically useful.

89 Not for publication or presentation Attachment 8

2016 AMERICAN SOCIETY OF HEMATOLOGY (ASH) MEETINGS ABSTRACT

Impact of Human Leukocyte Antigen (HLA) Alleles on Outcomes of Allogeneic Transplantation (Allo- HCT) for B-Cell Non-Hodgkin Lymphomas (B-NHL); a CIBMTR Analysis

Basem M. William, Tao Wang, Mike Haagenson, Marcos de Lima, Marcelo Fernandez-Vina, Stephen Spellman, Stephanie J. Lee, and Brian T. Hill

Even in the modern era of targeted therapies, allo-HCT remains a strategy that offers a chance of extended survival in B-NHL who relapsed after, or are deemed ineligible for, autologous transplantation (auto-HCT). Moreover, it was shown that efficacy of auto-HCT in extending survival in patients with diffuse large B-cell lymphoma is limited in the post-rituximab era. The use of allo-HCT for salvage of relapsed/refractory B-NHL is likely to increase with the wide adoption of reduced intensity conditioning (RIC) since transplant-related mortality (TRM) is lower while preserving the benefit of graft vs lymphoma (GVL) effect. Further understanding of the factors likely to predict a more robust GVL response, and potentially better clinical outcomes, would be very useful in selecting patients with B-NHL who are likely to benefit from allo-HCT. Multiple prior studies have shown that the presence of certain human leukocyte antigen (HLA) alleles may be instrumental in modulating the outcomes of B-NHL after immunochemotherapy and/or allo-HCT. We hypothesized that certain HLA alleles may have superior ability to present antigens to cytotoxic T-cells or possibly modulate natural killer (NK) cell activity.

We undertook a retrospective analysis of CIBMTR data to evaluate whether the presence of HLA-A2, HLA-C1C1, HLA-DRB1*01:01 or HLA-DRB1*13 alleles and HLA-A1+, HLA-A2- and HLA-B44- haplotypes were associated with outcomes of patients with relapsed/refractory B-NHL after transplantation from HLA-8/8 matched sibling (MRD) or unrelated (MUD) donors. The primary outcome was time to progression. Secondary outcomes included progression-free (PFS) and overall survival (OS), TRM, and incidence and severity of acute and chronic graft versus host disease (GVHD). Multivariate analyses were performed using Cox proportional hazards models adjusting for multiple disease and transplant related characteristics. HLA-A*02 was tested individually; HLA-DRB1*0101 and HLA-DRB1*13 were tested in 4 combinations; HLA-A1+/A2-, B44- vs. A2+ vs. neither. Because of multiple comparisons, p<0.01 is considered significant.

The final analysis population was restricted to 1314 patients who received the first allo-HCT during the years of 1995-2012 for indolent (N=506), aggressive (N=514), and mantle (N=294) B-NHLs from MRD (N=708) or MUD (N=606) utilizing myeloablative; MA (N=636), RIC (N=434), non-myeloablative; NMA (N=196) or other (n=48) conditioning. The different HLA allele analysis groups were well-balanced for all disease, patient, transplant and donor characteristics. In multi-variate analysis, we have observed no significant association between any of the HLA alleles or combinations we tested with any of the primary or the secondary end-points. A power calculation suggested very large numbers would be needed to reach statistical significant based on the data analyzed so far. The marked heterogeneity of different

90 Not for publication or presentation Attachment 8

lymphoma subtypes, conditioning regimens, donor choices, GVHD prophylaxis, supportive regimens or other center-related characteristics likely limited the ability to detect effects of differences in individual recipient HLA alleles. In conclusion, this study represents the largest reported series of allo-HCT outcomes of B-cell NHL patients based on HLA type. Further study of other variables will be required to delineate the immunologic impact of donor-host interactions on outcomes of allo-HCT for B-cell NHL.

91 Not for publication or presentation Attachment 9

2016 AMERICAN SOCIETY OF HEMATOLOGY (ASH) MEETINGS ABSTRACT

Pilot Study of Prognostic Impact of Pre-Allogeneic Hematopoietic Cell Transplantation (HCT) Plasma Levels of CXC-Chemokines (CXCL-4 and CXCL-7) in Patients with Myelodysplastic Syndromes (MDS)

Binod Dhakal, Tao Wang, Michael Haagenson, Fenlu Zhu, Carolyn A. Keever-Taylor, Stephen R. Spellman, Michael R. Verneris, Katharine Hsu, Katharina Fleischhauer, Stephanie J. Lee, and Wael Saber

Background: There are a limited number of studies that have evaluated the proteome of MDS patients (pts). Two chemokines (CXCL-4 and CXCL-7) were found to be differentially depressed in patients with MDS compared to patients with other hematologic disorders (Aviado et al, Proc Nat Acad Sci 2007). One study found that normal levels of these proteins are essential for stem cell quiescence, and with their depletion stem cell exhaustion ensues (Bruns I et al, Nat Med). In the current study, we sought to confirm the MDS-specific reduction of these chemokines, and in an exploratory fashion examine the association of CXCL4/CXCL7 pre-HCT plasma levels with MDS disease status pre-HCT and subsequent post HCT outcomes.

Methods: We analyzed three cohorts of pts >18 years of age reported to CIBMTR from 2008-2011: MDS cohort (N= 80); Non-MDS cohort undergoing allogeneic HCT for either aplastic anemia or non-Hodgkin lymphoma (N= 70); and Normal donors (ND) (N= 79). The primary objectives were to compare the pre-HCT plasma levels of the two chemokines (CXCL-4 and CXCL-7) among the three cohorts, and to explore whether these chemokines correlate with post-HCT disease free survival (DFS), overall survival (OS), relapse and non-relapse mortality (NRM) in the MDS cohort. Abcam’s Human ELISA kits were used to measure pre-HCT serum CXCL4 and CXCL7 concentrations. Optimal cut-off points to classify patients as “high” vs. “low” were identified using maximum likelihood method. We also examined whether the use of hypomethylating agents (HMA) pre-HCT correlated with chemokine levels pre-HCT in the MDS cohort.

Results: Median ages were 58 yrs, 47 yrs, and 30 yrs for MDS, non-MDS, and ND, respectively. Forty-five pts (56%) in the MDS cohort had advanced stage disease at transplant and 46% had poor or very poor risk cytogenetics based on the International Prognostic

92 Not for publication or presentation Attachment 9

Scoring System-Revised (IPSS-R) criteria. Fifty pts (63%) were male and 65% had Karnofsky performance score of 90-100 in MDS cohort. Myeloablative regimen was used in 55% of MDS pts compared to 23% in non-MDS cases. Low CXCL4 levels were seen in 57 (71%) MDS vs. 38 (48%) ND vs. 69 (99%) non-MDS pts (P<0.001) (Fig A). Low CXCL7 levels were seen in 63 (79%) MDS, 47 (59%) ND and 65 (93%) non-MDS pts (p<0.001) (Fig B). In MDS pts with low CXCL4 levels, 39% had high or very high IPSS-R score compared to only 11% among those with high CXCL4 levels (p=0.07) (Table 2). Similarly, among MDS pts with low CXCL7, 36% had high or very high IPSS-R score compared to only 7% among those with high CXCL7 levels (p=0.18) (Table 2). There was no correlation between the use of HMA pre-HCT and the pre-HCT chemokine levels in MDS pts. For MDS pts >50 yrs, there was no correlation between age and CXCL4 and CXCL7 levels. In univariate analysis, neither chemokine was associated with post-HCT OS, DFS, NRM or relapse in MDS pts.

Conclusions: Our study confirms that levels of chemokines CXCL-4 and CXCL-7 in MDS pts were significantly lower compared to ND. However, in contrast to previous findings by Aivado et al.2007, we did not observe MDS specific reduction as these chemokines were significantly lower among those with NHL and AA compared to MDS patients at HCT. Consistent with previous findings by Aivado et al. we observed a trend towards further reduction of these chemokines among MDS patients with more advanced disease (defined by higher IPSS-R scores) compared to MDS pts with less advanced disease. In our exploratory analysis, we did not observe an association between pre-HCT chemokines levels and post-HCT outcomes among MDS patients. Similarly, the use of HMA among MDS pts was not associated with pre-HCT chemokines levels. An important caveat to this conclusion is that we did not have data on the duration of HMA use. In MDS pts who were >50 yrs, there was no correlation between age and chemokine levels. Large-scale prospective studies are needed to understand more fully the prognostic role of the MDS proteome as well as the complex interplay between proteomic perturbations and acquired somatic mutations these patients often display in determining outcomes.

93 Not for publication or presentation Attachment 9

94 Not for publication or presentation Attachment 9

95 Not for publication or presentation Attachment 10

2016 AMERICAN SOCIETY OF HEMATOLOGY (ASH) MEETINGS ABSTRACT

Exome Array Analyses Identify Low-Frequency Germline Variants Associated with Increased Risk of AML in a HLA-Matched Unrelated Donor Blood and Marrow Transplant Population

Alyssa I. Clay, Theresa Hahn, Qianqian Zhu, Li Yan, Leah Preus, Daniel O. Stram, Christopher A. Haiman, Loreall C. Pooler, Xin Sheng, Kenan Onel, MD, Qian Liu, Qiang Hu, Song Liu, Xiaochun Zhu, Stephen R. Spellman, Marcelo C. Pasquini, Philip L. McCarthy, and Lara E. Sucheston- Campbell

Both genome wide association studies (GWAS) of common variation and exome wide association studies (EXWAS) of rare variation have successfully identified disease susceptibility variants for a variety of diseases. One GWAS of inherited susceptibility to Acute Myeloid Leukemia (AML) has been conducted, but no EXWAS have been performed to measure risk of AML attributable to low-frequency constitutional genetic variation. We performed the first EXWAS of risk of AML as a nested case-control study in the DISCOVeRY-BMT (Determining the Influence of Susceptibility Conveying Variants Related to one-Year mortality after BMT) cohorts. The DISCOVeRY-BMT parent study examined transplant-related mortality in leukemia patients undergoing unrelated donor allogeneic BMT.

To identify low frequency variants and genes contributing to increased susceptibility to AML we used genotype data from the Illumina HumanExome BeadChip typed in the DISCOVeRY-BMT cohorts; the HumanExome BeadChip contains 242,901 variants, which are mainly protein-coding variants. The optimal sequence kernel association test (SKAT-O) was used to analyze gene-level associations with risk of AML. These gene-based tests evaluate the cumulative effects of multiple single gene variants on risk of AML. Analyses were performed in all European American AML cases and two subtypes: 1) de novo AML, 2) de novo AML with normal cytogenetics. Models were adjusted for age at transplant and principal components to control for population stratification. For gene-based tests at least 2 variants with minor allele frequency (MAF) ≤ 5%, were required to be present in the gene. This yielded a total of 13,687 genes tested, and a Bonferroni corrected significance level of P<3.65 x 10-6. Association tests were performed in 1,189 AML cases reported to CIBMTR 2000-08 (Cohort 1) and 327 AML cases reported to CIBMTR from 2009-11 (Cohort 2). Controls in Cohorts 1 (n=1,986) and 2 (n= 515) were 10/10 HLA-matched unrelated donors who passed a comprehensive medical exam and deemed healthy. We used metaSKAT to combine Cohorts 1 and 2 and obtain p-values of association with AML. We present the results of gene-level tests significant in both cohorts. The likely pathogenicity of these variants was determined in silico using SIFT, PolyPhen and MutationTaster.

Patient characteristics are in Table 1. DNMT3A, on chromosome 2, was associated in the gene- based test with risk of AML (Pmeta=1.70x10-9, Table 2). Three missense variants at MAF <1% comprise both overall AML and de novo AML gene-based association: exm177559 (Asn->Ser), exm177507 (Arg->His), and exm177543 (Arg->Trp). Normal cytogenetics de novo AML gene-

96 Not for publication or presentation Attachment 10

based assocations consisted of only 2 of these variants: exm177559 and exm177507 (Table 2). While prevalence of exm177507 is <1% for all AML cases, in de novo AML with normal cytogenetics the MAF was higher at 3%. The other 2 variants had a MAF<1% irrespective of subtype. Somatically, DNMT3A is most frequently mutated in hematologic malignancies, with >30% of de novo AML cases with a normal karyotype and >10% of MDS patients having DNMT3A mutations. Although these are germline gene associations all three of the variants found have been reported somatically in hematologic malignancies. In 200 AML cases from The Cancer Genome Atlas (TCGA) p.R882H (represented as exm177507 on the exome chip) was a frequent somatic mutation (25%). Exm177543 (p.R635W) and exm177559 (p.N501S) are reported in the Catalogue of Somatic Mutations in Cancer (COSMIC) as somatic mutations involved in hematopoietic and lymphoid tissue in both cell lines and humans. Exm177507 and exm177543 show evidence of pathogenicity in all three in silico tools, while exm177559 was reported as deleterious and disease causing by Sift and MutationTaster, respectively.

Our results show that multiple potentially pathogenic missense germline variants in DNMT3A comprise the gene-based association with AML, specifically de novo AML with normal cytogenetics. Given the functional nature of these variants it is possible germline risk stratification could be informative in determining AML risk, and subsequently development of AML harboring DNMT3A mutations. Confirmation of these findings in additional cohorts could have implications for individualized risk screening, prediction and prognosis. Additional cytogenetic subgroup analyses, including treatment-related AML, are underway.

97 Not for publication or presentation Attachment 10

2016 AMERICAN SOCIETY OF HEMATOLOGY (ASH) MEETINGS ABSTRACT

Exome Array Analyses Identify New Genes Influencing Survival Outcomes after HLA-Matched Unrelated Donor Blood and Marrow Transplantation

Qianqian Zhu, Li Yan, Qian Liu, Qiang Hu, Leah Preus, Alyssa I. Clay, Kenan Onel, Daniel O. Stram, Loreall C. Pooler, Xin Sheng, Christopher A. Haiman, Xiaochun Zhu, Stephen R. Spellman, Marcelo C. Pasquini, Philip L. McCarthy, Song Liu, Theresa Hahn, and Lara E. Sucheston-Campbell

While survival outcomes after HLA-matched unrelated donor (URD) blood and marrow transplant (BMT) have significantly improved over the last two decades, about 40% of patients die before 1-year post URD BMT. Previously we performed a genome-wide association study (GWAS) named DISCOVeRY-BMT (Determining the Influence of Susceptibility COnveying Variants Related to one-Year mortality after BMT) to investigate the contributions of common genetic variants on survival outcomes. To address the specific contributions of low frequency (≤ 1%) exonic variants on survival outcomes, we used the DISCOVeRY-BMT cohorts typed with the Illumina HumanExome BeadChip containing 242,901 variants, which are mainly protein-coding variants. We used a gene-based test (the optimal sequence kernel association test (SKAT-O)) to evaluate the cumulative effects of multiple genetic variants within a gene on overall survival (OS), transplant-related mortality (TRM), and disease-related mortality (DRM) 1 year after URD BMT. SKAT-O measures the overall significance of the gene and allows the variants to have different directions and magnitude of effects. We used a 2-stage study design: SKAT-O analysis was carried out in Cohort 1 consisting of 1,972 AML, ALL or MDS patients reported to the Center for International Blood and Marrow Transplant Research from 2000 to 2008 and their 2,006 10/10 HLA-matched URD. Genes with a suggestive association with each outcome (P < 1×10-3) entered the final meta-analysis including Cohort 1 and an independent cohort (Cohort 2), consisting of 503 recipients and 520 10/10 HLA- matched URD from 2009 to 2011. The analysis included recipients and URD of European descent due to the low frequency of other populations. All association models included age at BMT, diagnosis (AML, ALL, MDS), disease status at BMT, cell source (peripheral blood, marrow), year of BMT, and principal components from EIGENSTRAT to control for population stratification. Only coding variants (missense and nonsense) with minor allele frequency ≤ 1% were included in our analysis, which resulted in > 12,000 genes tested in our gene-based analysis in recipients -6 and donors. Exome-wide significance was set at P meta< 4.10×10 after Bonferroni correction for the total number of genes tested. The likely pathogenicity of these variants was determined in silico using SIFT and PolyPhen. In recipients, we identified OR51D1, a member of the olfactory receptor gene family, as significantly associated with both OS and TRM. The OS association with ORD51D1is driven by TRM as the same six missense variants contribute to both TRM and OS, but not to DRM. In donors, four (ALPP, EMID1, SLC44A5, LRP1), one (HHAT), and two (LYZL4, NT5E) genes were significantly associated with OS, TRM, and DRM, respectively

98 Not for publication or presentation Attachment 10

(Table). ALPP encodes an alkaline phosphatase. SLC44A5 is a member of the solute carrier gene family. LRP1 (low density lipoprotein receptor-related protein 1) encodes an endocytic receptor involved in several cellular processes, including intracellular signaling, lipid homeostasis, and clearance of apoptotic cells. HHAT encodes an enzyme that acts within the hedgehog secretory pathway, which is involved in hematologic malignancy. LYZL4 is a member of a family of lysozyme-like genes, which have a protective role in host defense. The encoded protein of NT5E, CD270, is used as a determinant of lymphocyte differentiation. The likely pathogenicity of several of the functional variants in all significant genes identified was predicted to be both (possibly or probably) damaging and deleterious (Table), thus indicating that the amino acid substitution of the variant may not be well tolerated. One damaging variant in LRP1 has been identified as a somatic mutation in hematopoietic and lymphoid tissue. Our study is the first analysis of the low-frequency coding variant contribution to URD BMT survival outcomes. We found that the missense variants contributing to many of these significant gene associations shows evidence of pathogenicity and thus it is biologically plausible these variants are contributing to survival outcomes. Further confirmation of these findings, and studies of the functional consequences of protein-coding changes in these genes, may provide more individualized risk prediction and prognosis as well as alternative donor selection strategies.

99 Not for publication or presentation Attachment 10

2016 AMERICAN SOCIETY OF HEMATOLOGY (ASH) MEETINGS ABSTRACT

Replication of Candidate SNP Survival Analyses and Gene-Based Tests of Association with Survival Outcomes after an Unrelated Donor Blood or Marrow Transplant: Results from the Discovery-BMT Study

Ezgi Karaesmen, Abbas Rizvi, Qianqian Zhu, Li Yan, Qian Liu, Qiang Hu, Leah Preus, Alyssa I. Clay, Song Liu, Daniel O. Stram, Loreall C. Pooler, Xin Sheng, Christopher A. Haiman, Xiaochun Zhu, Stephen R. Spellman, Marcelo C. Pasquini, Philip L. McCarthy, Theresa Hahn, and Lara E. Sucheston-Campbell

Over the past decade, there have been numerous published candidate gene association studies of non-HLA variants, genes and/or pathways in relation to BMT survival outcomes. Most of these studies investigated a small number of candidate single nucleotide polymorphisms (SNPs), chosen based on known gene function, in a limited number of cases without replication. Here we test the association of previously reported SNPs with survival outcomes (overall survival (OS), progression-free survival (PFS), death due to disease (DD) or transplant-related mortality (TRM)) after URD BMT using data from the DISCOVeRY-BMT (Determining the Influence of Susceptibility Conveying Variants Related to one-Year mortality after BMT) study which included 2,052 European descent recipient-unrelated donor pairs (Cohort 1) and 763 recipient-unrelated donor pairs (Cohort 1) reported to the Center for International Blood and Marrow Transplant Research (CIBMTR) between 2000-2009 and 2009-2011, respectively. DISCOVeRY-BMT is a Genome Wide Association Study (GWAS) that includes over 9 million typed or imputed SNPs. This represents the largest study to date and is well powered to test both SNP and gene-based association hypotheses.

We conducted an extensive literature review of PubMed using MESH terms related to transplant and hematologic disease which yielded 76 candidate gene association studies that investigated a total of 46 genes with any survival outcome. Of the 46 genes, 11 genes (CTLA4, CYP2B6, CYP2C19, GSTM1, GSTT1, IL23R, IL6, MTHFR, NOD2/CARD15, TNFSRF1B, and VDR) were studied in at least 2 publications. A total of 16 SNPs in 8 of these genes reported a significant (P< 0.05) association with at least one survival outcome (OS, PFS, DD or TRM). We tested these 16 SNPs for an association with the survival outcome originally reported in each article (Table 1). All association models included age at BMT, diagnosis (AML, ALL, MDS), disease status at BMT, cell source (blood, bone marrow) and year of BMT. P-values from Cohorts 1 and 2 were combined using a fixed-effects model in METAL software. Since the candidate gene studies have a gene-based hypothesis, we also performed gene-level testing using the VErsatile Gene-Based Association Study (VEGAS2) software to measure the aggregate effect of all available typed and imputed SNPs from our GWAS data in these 8 genes.

100 Not for publication or presentation Attachment 10

The candidate gene association studies included heterogeneous patient populations of autologous and/or allogeneic (related or unrelated donors) BMT for a variety of hematologic diseases. Most candidate gene studies performed genotyping on both donors and recipients, and SNPs were tested either independently for donor or recipient genomes, or considering the total number of variant alleles the recipient-donor pair had at a given locus (0-4) (Table 1).

The 16 SNPs from the previous candidate gene association studies were not significantly associated at a nominal P<0.05 to either the survival outcome reported in the original published paper, or to any survival outcome (Table 1). Likewise, the VEGAS2 gene-based analysis did not reveal any statistically significant associations for any of the 8 genes at P<0.05. Our study also corroborated prior reports of SNPs and genes that were not associated with survival outcomes.

Candidate gene association studies use a priori biological knowledge to select genes that are likely associated with outcome and subsequently SNPs within the gene thought likely to affect function based on knowledge at the time of the investigation. Our findings suggest that neither the genetic variation tested in these candidate genes, nor other individual SNPs in the gene independently or collectively are not associated with survival outcomes in recipients of URD BMT. Our failure to replicate the prior candidate gene studies may be due to these SNPs being dependent on the underlying disease (ie. AML, ALL, etc), the donor relation (related, unrelated, autologous), or that there is no true association of these SNPs with the outcomes studied.

101 Not for publication or presentation Attachment 10

102 Not for publication or presentation Attachment 10

2017 TANDEM MEETINGS ABSTRACT

Functional Genetic Variants on 14q32 Associate with Death Due to Acute Myeloid Leukemia (AML) or Myelodysplastic Syndrome (MDS) Within One Year After HLA-Matched Unrelated Donor Blood and Marrow Transplantation (DISCOVeRY-BMT Study)

Lara Sucheston-Campbell, Leah Preus, Philip L. McCarthy, Marcelo Pasquini, Kenan Onel, Xiaochun Zhu, Stephen Spellman, Christopher Haiman, Daniel Stram, Loreall Pooler, Xie Sheng, Qianqian Zhu, Li Yan, Qian Liu, Qiang Hu, Alyssa Clay, Sebastiano Battaglia, David Tritchler, Song Liu, and Theresa Hahn

The most frequent cause of death after BMT is disease. To assess the contribution of genetics to survival outcomes we measured the association of variants typed on the Illumina HumanOmniExpress-24 Chip and 1-year survival in 3,532 AML, ALL or MDS patients reported to CIBMTR from 2000-2011 and their 8/8 HLA-matched URDs (Determining the Influence of Susceptibility COnveying Variants Related to one-Year mortality after BMT (DISCOVeRY-BMT) cohorts). After quality control and imputation, almost 9 million typed and imputed SNPs were available on donor-recipient pairs in cohorts 1 (N=2,111) and 2 (N=779). Cox proportional hazard models adjusted for age, disease status, and graft source were used to estimate variant effect on survival. Meta P-values and hazard ratios were estimated using METAL software (Pmeta) and fixed effects models in R (HRmeta), respectively. We report genome-wide significant AML and MDS SNP associations with disease death (DD) and progression free survival (PFS). The donor genome has three significant DD associated regions, 5p15, 9p21 and 14q32 (Fig. 1, inner circle & Table); no significant hits are seen in recipients (Fig. 1, middle circle) or when comparing allele matching at each SNP in the recipient and donor (Fig. 1, outer circle). The most significant SNP on chromosome 5, rs32250, is in linkage disequilibrium (LD, r2>.8) with rs2215201, which likely affects binding of several nuclear receptor superfamily transcription factors. On chromosome 9 the donor contribution to risk of DD is genome-wide significant in both AML and MDS and AML only; variants near this region reside in a nuclear receptor superfamily motif (NR3C1) in multiple cell lines. The chromosome 14 region (60262026- -8 60597197 bp) contained several donor variants at Pmeta<5 x 10 that increased risk of DD in -7 patients with AML and MDS and near genome-wide significance (Pmeta<5 x 10 ) in patients with AML only. Many of these significant linked SNPs have biochemical function: a typed missense variant in PCNX4, rs150687 (60581935 bp), associated with DD at near genome-wide significance, is considered deleterious and possibly damaging by software used to assess clinical variant significance and has been identified in a cancer tissue from a patient with aerodigestive cancer; in monocytes, rs8011499, rs160235, rs160236 and rs160240 (r2> .8 with rs150687) are cis expression quantitative trait loci of DHRS7, a steroid and retinoid metabolizer gene, and are predicted likely to affect transcription factor (TF) binding of many important TF in leukemia cell

103 Not for publication or presentation Attachment 10

lines. The donor genome has one region associated with PFS (Table) in AML patients, 13q12 (23047591-23070662 bp) which has some linked variants thought likely to affect CTCF transcription factor binding. We identified several regions in the donor genome that contributes to risk of DD and reduced PFS in patients with myeloid malignancies that merit further investigation in additional cohorts.

104 Not for publication or presentation Attachment 10

Table. Most significant donor genome associations with disease death and progression free survival in AML and MDS patients by region Disease death

SNP Chromsome: Effect Allele HR (95% CI) , P-value HRmeta (95% CI) Gene Base pair / Effect Pmeta position Allele Frequency rs32250 5:12465740 T:45% 1.46 [1.26, 1.70] , 7.7 x 10-07 1.43 [1.26, 1.62] RP11-308B16.2 1.36 [1.07, 1.72] , .013 3.8 x 10-08 rs72720973 9:33048584 T:12% 1.51 [1.24, 1.84] , 3.4 x 10-05 1.61 [1.36, 1.90] SMU1 1.90 [1.39, 2.61] , 6.4 x 10-05 1.8 x 10-08 rs117104143* 14:6041296 C:7% 1.57 [1.24, 1.83] , 4.9 x 10-05 1.61 [1.36, 1.90] LRRC9 1.89 [1.36, 2.62] , 1.5 x 10-04 4.6 x 10-08 Progression free survival rs17076928¥ 13:23059577 T:1% 2.11 [1.39, 3.20], 4.4 x 10-04 2.94 [2.10, 4.11] BC035104 5.42 [3.08, 9.54], 4.9 x 10-09 1.7 x 10-09 (208,918 bp upstream from variant) *rs79076914 (60559239 bp) is also associated at genome-wide significance level and is in LD (r2=1) with this variant ¥Typed variant in LD (r2=1) with other genome-wide significant variants: rs577669014, rs17076913, rs115421327, rs73435960 and rs17076979

105 Not for publication or presentation Attachment 10

2017 TANDEM MEETINGS ABSTRACT

Novel Genetic Variants Associated with Death Due to Acute Lymphoblastic Leukemia Within One Year after HLA-Matched Unrelated Donor Blood and Marrow Transplantation (DISCOVeRY-BMT Study)

Theresa Hahn, Leah Preus, Philip L. McCarthy, Marcelo C Pasquini, Kenan Onel, Xiaochun Zhu, Stephen Spellman, Christopher A Haiman, Daniel O Stram, Loreall Pooler, Xin Sheng, Qianqian Zhu, Li Yan, Qian Liu, Qiang Hu, Alyssa Clay, Sebastiano Battaglia, David Tritchler, Song Liu, and Lara Sucheston-Campbell

Disease relapse is the most common cause of early mortality after HLA-matched unrelated donor (URD) BMT. We performed a genome-wide association study (GWAS) named DISCOVeRY-BMT of survival outcomes in 3,532 patients treated for AML, ALL or MDS, reported to CIBMTR from 2000-2011 and their 8/8 HLA-matched URDs (details previously published via ASH 2015, Tandem BMT meeting 2016). We now report outcomes for 577 ALL patients: Cohort 1 included 483 ALL recipients and their 466 HLA matched URDs, Cohort 2 included 94 recipients and 92 URDs of European-American (EA) ancestry typed at 637,655 markers and ~9 million imputed SNPs. Each SNP was measured for association with death due to disease (DD) and PFS using Cox models. PFS was defined as time to disease progression or death due to any cause with patients censored alive without ALL progression at a maximum of 1 year post- BMT. All models included age, disease status (early, intermediate, advanced), cell source (blood, marrow) and year of BMT. P -values were combined using METAL software and weighted by sample size. -8 We report results for a combined P-value (Pmeta) <5x10 , an info score >0.7 and minor allele frequency >0.01. Two regions in URD genomes containing multiple variants are associated with death due to disease (DD) [Table, Figure]. On chromosome 2, an imputed variant (rs79503405) in LRP2 significantly increased risk of DD. This variant is in complete linkage disequilibrium (LD, r2=1.0) with a genotyped missense variant -7 (rs17848149, Pmeta=4.65 x 10 ) and a synonymous coding variant (rs35114151) in LRP2. It is also in modest LD (r2=0.66) with a splice variant (rs34564141) considered deleterious, damaging and disease causing by software used to assess the clinical significance of SNPs. On chromosome 17, an imputed intergenic variant rs77618918 significantly increased DD, however it is not in LD with other SNPs of functional importance. There were no genome-wide significant SNPs in URDs associated with PFS. In ALL recipients, 3 regions were associated with DD and PFS [Table]. Three linked variants on chromosome 8 (rs79853417, rs6990973, rs145488394) and a SNP (rs60640657) on chromosome 2 were significantly associated with DD. A region on chromosome 3 containing MLH1 was significantly associated with PFS. Given this gene’s important role in mismatch repair, variants have been previously tested for association with BMT outcomes and a MLH1 variant was recently correlated with PFS in lymphoma and myeloma patients. This study has discovered several novel variants in recipients and their URDs associated with DD and PFS. The DD and PFS associations are distinct probably due to different definitions, as PFS contains death due to non-relapse mortality as well as non-fatal disease progression. Further confirmation of these

106 Not for publication or presentation Attachment 10

findings in other cohorts may yield new insight into the mechanisms of disease relapse after URD-BMT or potentially impact donor selection.

107 Not for publication or presentation Attachment 10

108 Not for publication or presentation Attachment 11

2016 AMERICAN SOCIETY OF HEMATOLOGY (ASH) MEETINGS ABSTRACT

Germline Mutations in Patients Receiving Unrelated Donor Hematopoietic Cell Transplant for Severe Aplastic Anemia

Shahinaz M Gadalla, Bari J Ballew, Michael Haagenson, Stephen Spellman, Belynda Hicks, Blanche P Alter, Bin Zhu, Weiyin Zhou, Meredith Yeager, Tao Wang, Katharina Fleischhauer, Katharine Hsu, Michael R. Verneris, Neal Freedman, Stephanie J. Lee, and Sharon A Savage

Background The correct classification of patients presenting with aplastic anemia is essential for proper clinical management. This study aimed to identify patients with unrecognized pathogenic variants associated with inherited bone marrow failure syndromes (IBMFS) in patients receiving hematopoietic cell transplant (HCT) for acquired aplastic anemia (AAA), and to evaluate their impact on patient survival.

Methods We performed whole exome sequencing (WES) and genome-wide SNP array genotyping on pre-HCT blood samples from 608 patients who received unrelated donor HCT for severe aplastic anemia in 175 centers reporting to the Center for International Blood and Marrow Transplant Research (CIBMTR) and research samples available in the CIBMTR Repository. An IBMFS was the reported diagnosis in 92/608 patients. From generated genomic data and mode of IBMFS inheritance, we developed a bioinformatic algorithm to identify deleterious mutations in 52 known IBMFS genes. Cox proportional hazard models were used for survival analyses. Variables included in the model were selected by a stepwise backward procedure with a p-threshold of 0.05 for entry and 0.1 for retention.

Results We validated our algorithm in 16 patients with genetically confirmed dyskeratosis congenita (DC, n=12) and Fanconi anemia (FA, n=4) from the NCI IBMFS cohort. Our algorithm was 88% sensitive and 100% specific in identifying the exact mutation in those patients. In HCT recipients, analysis of 52 IBMFS genes identified 130 pathogenic variants in the 92 patients with reported IBMFS and 277 in the 516 patients with reported AAA. Our algorithm combining WES and SNP array data identified genetic evidence of an IBMFS in 59.8% (55/92) of patients reported to have an IBMFS. In patients originally classified as AAA, 10.5% (54/516) had genetic evidence of an IBMFS; 88.7% of them received HCT at ≤ 40 years old, and DC was the most commonly unrecognized disorder (n=26, 5%). Other identified IBMFS by our algorithm included: Diamond-Blackfan anemia (n=8, 1.5%), and FA, myelodysplastic syndrome, or severe congenital neutropenia with 6 patients each (1.2% each), and 1 patient with amegakaryocytic thrombocytopenia (0.2%). Among patients with reported AAA who were ≤ 40 years old, the 3 years survival probability for patients with DC-associated genetic variants (n=26) was 53% (95% CI=35-73%), for patients with genetic variants in other IBMFS (n=20, 4 had FA) was 75% (95% CI=56-94%), and for those with no genetic evidence of IBMFS was 69% (95% CI=64- 74%), p log-rank= 0.02. Compared to patients with no genetic evidence of IBMFS, worse post-HCT

109 Not for publication or presentation Attachment 11

survival was noted with DC-associated genetic variants (HR=1.75, 95% CI=1.03-2.97, p=0.04), but no survival difference was noted with other IBMFS variants (HR=0.42, 95% CI=0.16-1.15, p=0.09)

Conclusions We identified a high incidence of unidentified IBMFS in patients undergoing HCT for AAA, which was associated with worse post-HCT survival in those with unrecognized DC. Our study supports the importance of comprehensive genetic testing for patients presenting with AA, given the high incidence of underlying genetic abnormalities that have implications for donor selection, family counseling and selection of the preparative regimen

110 Not for publication or presentation Attachment 12

2016 TANDEM BMT MEETINGS (ASBMT/CIBMTR) ABSTRACT

Recipient Pre-HCT Telomere Length and Outcomes after Unrelated Donor HCT for Acute Leukemia

Youjin Wang, Tao Wang, Casey Dagnall, Michael Haagenson, Stephen R. Spellman, Belynda Hicks, Kristine Jones, Stephanie J. Lee, Sharon A. Savage, Shahinaz M. Gadalla

Background Telomeres are tandem nucleotide repeats and a protein complex located at the end of the chromosomes. They are necessary for maintaining chromosomal stability. There are limited data on the role of recipient telomere length in HCT outcomes. In this study, we evaluated the association between recipient pre-HCT telomere length and outcomes after unrelated donor HCT for patients with acute leukemia.

Method We used qPCR to measure pre-transplant leukocyte relative telomere length (RTL) in 536 acute leukemia patients who underwent 8/8 HLA-matched unrelated donor HCT between 2005 and 2012 after myeloablative conditioning regimens. Patient pre-transplant blood samples and outcome data were provided by the Center for International Blood and Marrow Transplant (CIBMTR) Research Repository. We used Kaplan-Meier and competing risk estimators to calculate survival probability and cumulative incidence, respectively. Cox proportional hazard regression was used for adjusted analyses.

Results The study included 396 AML and 140 ALL patients; median age at HCT =41 years (range= 0.5-66 years), and median follow-up for survivors= 5.8 years (range= 0.5-10.3). In univariate analysis, recipient RTL was not significantly associated with disease-free survival (p=0.18), overall survival (p=0.16), relapse (P=0.58) or transplant-related mortality (p=0.16). Similarly, there were no significant associations between recipient RTL and acute GvHD and chronic GvHD. In multivariable analyses, recipient RTL was not associated with any of these HCT outcomes.

Conclusion Pre-HCT telomere length is not significantly associated with HCT outcomes in patients with acute leukemia.

111