Inserm GENE THERAPY OF HEMOGLOBIN DISORDERS

Transcription

1 Inserm GENE THERAPY OF HEMOGLOBIN DISORDERS Institute of Emergent Diseases and Innovative Therapies (imeti) Laboratory of Gene and Cell Therapy (LTGC) CEA, Fontenay aux Roses

2 Human forms of Hemoglobin Globular protein with four polypeptide units, two alpha and two beta chains Each subunit has a nonpolypeptide component (heme) with an iron atom that binds to oxygen 2 types of chains: (embryonic) et (adult) 4 types of chains: (embryonic), (fetal) and, (adult) Several types of hemoglobin tetramers Globin switch [yolk sac][ liver ] [ bone marrow ] Embryonic Fetal Adult Gower1 ( 2 2 ) HbF ( 2 2 ) HbA ( 2 2 ) Gower2 ( 2 2 ) HbA2 ( 2 2 ) Portland ( 2 2 )

3 Hemoglobin disorder frequency 5% of the population carry a significant variant!! (including 40% S and 80% in Africa) conceptions with SS, SC, S/ conceptions with severe thalassemia (mostly 3.4% of mortality in children under 5y due to hemoglobinopathies Modell et Darlison ; Bull WHO 86: 480 (2008)

4 -thalassemia major: a lethal disease In France, ~ 400 -thal patients 1 birth / (7 / year), 70%TM Median age = 20! Modell et Darlison ; Bull WHO 86: 480 (2008) I Thuret et C. Badens ; haematologica 95: 724 (2009) ~ 4% of the 350 million Southern China population carry a E -or a 0 - allele. 0.16% of 350 millions =

5 GEOGRAPHICAL REPARTITON OF THE -HEMOGLOBINOPATHIES Hb C: 6 Glu lys Hb S: 6 Glu val Hb E: 26 Glu lys (5% of the world population ( people) carries an abnormal HBB gene. Populations from Africa, Mediterranean basin, Middle East and Southeast Asia

6 SICKLE CELL DISEASE (HbS) HBB sequence in normal adult hemoglobin (HbA) Nucleotide CTG ACT CCT GAG GAG AAG TCT Amino acid Leu Thr Pro Glu Glu Lys Ser HBB sequence in mutant adult hemoglobin (HbS) Nucleotide CTG ACT CCT GTG GAG AAG TCT Amino acid Leu Thr Pro Val Glu Lys Ser 3 6 9

7 How do cell sickle? t d = K/C 15 t d t d Adachi K., JBC, 1979

8 Hydrophobic interactions stabilize the fiber

9 Hypoxia Fever dehydratation Consequences Polymerization of deoxyhb S Decreased cell elasticity Hypoxia Ischemia Vessel occlusion Tissue damages Anemia and fatigue and +/- Pain Crises Dactylitis (swelling and inflammation of the hands and/or feet) Bacterial Infections Splenic Sequestration (sudden pooling of blood in the spleen) Lung and Heart Injury Leg Ulcers Bone Infarcts (death of portions of bone) Eye Damage Other Features

10 Beta Thalassemia Major Intermediate Minor From the Greek, (thalassa) the sea and (haima) the blood, meaning "blood from the sea", in reference to the Mediterranean Sea where the disorder was first thought to originate.

11 Pathophysiology of thalassemia No ( 0 ) or low ( + ) hemoglobin Normal hemoglobin proerythroblast Basophilic Polychromatophilic Orthochromic Reticulocyte Erythrocyte Free hemoglobin chains (unpaired) Precipitation (Heinz bodies) Oxidation and cell death Inefficient erythropoiesis and hemolysis

12 Normal erythropoiesis -thalassemic inefficient erythropoiesis Progenitors Erythroblasts Reticulocytes 120 days hemolysis RBC are daily produced

13 Pathophysiology of thalassemia (2) Inefficient erythropoiesis and hemolysis Anemia erythropoietin Inefficient erytroid cell mass Blood transfusions Intestine absorption of iron Iron overload Cardiopathy, renal dysfunction, endocrin disorders, liver failure, pulmonary hypertension, neurological complications, bone pain

14 -thalassemia major Symptoms Severe anemia Hb:3 to 7g/dL Massive hepatosplenomegaly Severe growth retardation Body deformities Supportive therapy Blood transfusion dependency Iron overload => chelation Rund D, NEJM 2005

15 Curative therapy : Bone marrow transplantation Indications : -thalassemia major: Transfusion dependency SCD: High-risk, severely-symptomatic patients Cerebral vasculopathy, multiple and severe vaso-occlusive crisis (>3/year), acute chest syndrome, multiple osteonecrosis, RBC alloimunization

16 Bone marrow transplantation in -thalassemia Favorable outcome (Lucarelli et al.) : 1. fully matched related donor 2. Class I and II + young (below 16) Under standard regimen BU + CY Disease free survival: I and II: 91 and 83% III: 58% (28% rejection and 19% mortality) Risk factors: Non compliance with iron chelation Hepatomegaly Liver fibrosis Class I: none; II: (1 or 2) III: (all three) Event free survival (%) Event: rejection, recurence of thalassemia, death Lucarelli et al. NEJM 1990; 322:417

17 Bone marrow transplantation in class III young patients (<17) Objective: to enhance immune suppression and eradicate the thalassemic clones: New regimen BU + CY + hydroxyurea + Fludarabine + Azathioprine + hypertransfusion + chelation DAYS before BMT RBC RBC RBC RBC RBC RBC RBC RBC RBC RBC RBC RBC Suppress erythropoiesis Eradicate marrow G + Epo G + Epo G + Epo G + Epo G + Epo G + Epo G + Epo G + Epo G + Epo G + Epo Stimulate cell proliferation to facilitate HU Fluda Fluda Fluda Fluda Fluda BUx4 BUX4 BUx3 BUx3 CY CY CY CY Survival: 93% Disease free survival: 85% 100mg/m 2 14mg/kg 160mg/kg Sodani et al. Blood : 1201

18 Bone marrow transplantation in class III older patients G + Epo G + Epo G + Epo G + Epo G + Epo G + Epo G + Epo G + Epo G + Epo G + Epo BUx4 BUX4 BUx3 BUx3 CY CY CY CY MTX MTX CSA CSA CSA CSA CSA CSA CSA CSA RBC RBC RBC RBC RBC RBC RBC RBC RBC RBC RBC RBC Old New Gaziev et al. Ann NY Acad Sci : 196

19 Introduction of ATG ATG: Anti-thymocyte globulin (obtained from animal serum after immunization with human thymocytes) to reduce the number of circulating T-lymphocytes and reduce the graft-versus-host disease Survival w/o rejection 82,5% +/-10 55,1% +/-14 p=0.002 Survival p=0.43 French experience (108 patients, , BU-CY +/- ATG) I. Thuret (Marseille)

20 ATG in Sickle cell disease Rejectionl probability (%) Rejections at 5 y: 25.3% (SE: 11) 2.4% (SE: 1.4) 0: no ATG 1: with ATG Years post-transplant Number at risk Group: Group: Bernaudin et al. Blood :2749

21 Alternative to allogeneic bone marrow transplantation with fully matched sibling donor Percentage of patients with such an option < 25-30% - Unrelated bone marrow donor with high resolution molecular typing for HLA class I and class II molecules (BU-CY + CSA + MTX) - Umbilical stem cell transplantation (mainly for children) from matched sibling donor Probability Survival=70% Mortality=30% Rejection=4% EFS Probability 100% survival Years post-bmt Very similar results with sibling donors Days post-bmt La Nasa BMT :971 ( -thalassemia) Locatelli Blood : 2137 ( -thalassemia and SCD)

22 Alternative treatments? Genetic modifier factors in -thalassemias (and SCD) / non Minor Major HBB locus ( deletion, -promoters, others ) Klf1 Bcl11A Myb DNA Methylation Histone deacetylation thal - 5-Azacytidine, Decitabine - Butryrate - Epo, HU

23 Proposed mechanisms of fetal hemoglobin modulation BCL11A, KLF1 (EKLF) MYB (cell cycle?) Thein S L et al. Hum. Mol. Genet. 2009;18:R216-R223

24 Hemoglobin switching model At low Klf1 conc, weakly expressed Bcl11A can not repress -glo. Klf1 activates -glo expression through LCR recruitment. At high Klf1 conc, highly expressed Bcl11A repress -glo. Klf1 activates -glo expression through LCR recruitment. Cao et al. Pediatric Report 2011; 3:e17 Regions bound by Bcl11A (ChIP-chip)

25 Bcl11A as a potential therapeutic target, proof of principle in normal and sickle cell disease transgenic mice Tg(LCR G A ) + KI(Bcl11A fl/fl ) Tg(Ery(EpoR)CRE) +/- +/- Tg(Ind(Mx1)CRE) Interference with -globin gene repression during development Reactivation of -globin in adulthood Xu et al. Science Express 2011, 13th of October

26 Tg(LCR G A S ) KO( ) + KI(Bcl11A fl/fl ) +/- Tg(Ery(EpoR)CRE) /( + S ) (%) HbA ou HbS HbF Xu et al. Science Express 2011, 13th of October

27 Important criteria in evaluation of potential targets for therapeutic induction of HbF BCL11A KLF1 Epigenetic / cell cyle regulators Effect on HbF: Large Large Variable Erythroid, nonglobin functions +/- +? Non erythroid functions «Drugability» Chalenging (to target Tr fact.) Chalenging (to target Tr fact.) + Xu et al. The Hematologist; Sept

28 Rationale for the treatment of the -hemoglobinopathies by gene therapy Allogenic bone marrow transplantation is the only hope of definitive cure for -thalassemic and sickle cell patients But most patients do not have geno-identical donnor for HSC transplantation and those who do face increased morbidity and mortality caused by graft vs. host disease (GVHD): Conventional therapies (mostly paliative) are poorly efficient so far Many -thal patients die before 30 years old despite blood transfusions The overall life expectancy of SCD patients is often shortened by several decades Gene therapy with autologous transplantation and non-immunosuppressive regimen

29 Milestones towards gene therapy with -globin vectors We had to achieve High level erythroid expression, anti-sickling effect and stem cell targeting are prerequisite for efficient gene therapy of - hemoglobin disorders Evaluation of efficacy in mouse models of -thalassemia and sickle cell disease Evaluation of safety in a pre-leukemic mouse model

30 Milestone towards gene therapy Strategy for achieving High-level erythroid expression of -globin transgene

31 Specificity and High Expression of the Therapeutic Globin Gene HS: HS: 1 G A LCR 21 kb 5 kb 12 kb HS: DNAseI Hypersensitive Site Grosveld et al., Cell gene: erythroid specificity +++ LCR: high expression level +++ LCR: position independent expression? Stability of retroviral vectors [ -globin/lcr] retroviral vectors are very unstable with multiple rearrangements upon transmission of the proviral structure

32 Long term gene therapy: hematopoietic stem cell targeting Pluripotent stem cell Self-renewal Target Stem Cell BONE MARROW BLOOD Lymphoid Stem Cell Lymphoblast Sustained expression and correction B & T Lymphocytes BFU-E CFU -E Proerythroblast Erythroblast Reticulocyte Erythrocyte BFU-MK CFU-MK Mégakaryoblast Megakaryocyte Myeloid Stem Cell CFU-M CFU-GM Monoblast CFU-G Myeloblast CFU-B Myeloblast CFU-Eo Myeloblast Eosinophilic metamyelocyte Promonocyte Neutrophilic metamyelocyte Basophilic metamyelocyte Platelets Monocyte neutrophilic basophilic eosinophilic Granulocytes Precursors Committed Progenitors Mature Cells

33 Quels vecteurs intégratifs? La propriété majeure des vecteurs lentiviraux Vecteur murin MuLV Vecteur VIH 1 (1983) S intègrent uniquement dans les cellules en division (1996) S intègrent dans les cellules non en division Naldini et al. Science 1996.

34 DU PLASMIDE AU VECTEUR TRANSGENE + Elt CIS + LTRs LTR GAG p17 p24 P9 P6 PR RT R POL H INT VIF VPU VPR gp120 TAT REV ENV Gp4 1 NEF LTR DELETION AMPI R ORI

35 + Transgène ARN PRODUCTION Protéines virales ARN vecteur + Transgène ADN vecteur dbrin ARN vecteur ADNdbrin Transfection ARN Gag/pol, env Transgène + Gag/pol Env Cellule productrice

36 VIRUS / VECTEUR Rétrovirus (RNA + protéines) infection Rétrovecteur plasmide, DNA transfection LTR Gag-Pol Env LTR LTR Transgène LTR Cellules permissives production Rétrovirus (RNA + protéines) infection Cellules d encapsidation production Rétrovecteur (RNA + protéines) transduction LTR Gag-Pol Env LTR LTR Transgène LTR Cellules permissives Cellules cibles production Rétrovirus (RNA + protéines)

37 Local: Laboratoire confiné L3

38 5 LTR RRE RNS1: TNS9: E3 Enhancer 3 Oncoretro vs lentiviral vector Enhancer 3 E2 E1 E3 E2 E1 P LCR: 1kb LTR LCR: 3.2 kb 5 LTR RRE P HS2 HS3 HS4 3 LTR 840 bp 1308 bp 1069 bp Normalized Hu / Mu RNA (%) RNS1 TNS9 0 Week 22 Ter119+ sple May et al., Nature 2000

39 -thalassemic (intermedia) animal models Hbb th-1/th-1 : homozygout deletion of two maj genes min min maj maj Hbb th-3/+ : heterozygout deletion of one locus normal thal Hc (%) 47.4 ± ± 2.2 RBC (x10 6 /mm 3 ) 11.8 ± ± 0.4 Hb (g/dl) 14.7 ± ± 0.5 Retic (%) 3.6 ± ± 2.6 MCV (fl) 40.5 ± ± 1.7

40

41 Improvement of the -thalassemic phenotype Hbb th3+/- RBC (x10 6 /µl) Réticulocytes (%) Hbb mu Hbb hu Hematocrit (%) Hb (g/dl) Hbb +/+ -GFP Hbb th3/+ -GFP Hbb th3/+ -TNS9

42 LENTI-GLOBIN VECTOR BMT Hematopoietic cells

43 Proportion of Human -Globin Positive Red Blood Cells in Normal Transplanted Mice (7 Months Post-Transplantation) Human -globin Number RBC Mock transplanted mouse 99.6% Proportion of Human -Globin Positive Red Blood Cells (%) [ -globin/lcr] transplanted mouse N = 5 0 Human 96.8% ± 2.5 (SEM) (highest = 99.6%)

44 Improvement of the -thalassemic phenotype Hbb th1/th1 Hbb th1/th1 /Lenti-GFP Hbb th1/th1 /Lenti- A 1 0 Hbb th1/th1 /Lenti- A 2 0 C57/Bl6 Beads C57/B6 Hbb th1/th1 Hbb th1/th1 /Lenti- A RBC Density Gradients 1º Transplants 2º Transplants

45 Correction of Splenomegaly Hbb th1/th1 Lenti- -globin-lcr Normal control mouse Hbb th1/th1 Hbb th1/th1 Mock transplanted

46 Anti-Sickling Property of β 87 Gln (Q) residue Anti-sickling effect of -globin Axial Contact Val 6 Val 6 Val 6 Lateral Contact Phe 85 Leu 88 Phe 85 Leu 88 Phe 85 Leu 88 Hereditary persistence of -globin Hydroxyurea globin HLDDLKGTFA Q LSELHCDKLHVDPENF : -globin HLDNLKGTFA T LSELHCDKLHVDPENF : -globin HLDNLKGTFS Q LSELHCDKLHVDPENF V F L T V F Q L polymerization destabilization

47 Polymerisation delay times for deoxygenated mixtures of human Hb 0,5 Absorbance (700 nm) 100 % HbS 25 % HbA ; 75 % HBS 25 % HbF ; 75 % HBS Time (min) 90 Hb concentration = 60 mg/dl McCune PNAS, 1994

48 Construction of a Variant Human A -Globin Gene with Potent Anti-Sickling Activity and Normal Oxygen Binding Affinity A-T87Q promoter Exon 1 Exon 2 Exon 3 87Thr 87Gln p50 (Torr) HbS (100%) C SAT = 17.0 g/dl HbS:Hb A-T87Q (50%: 50%) HbS:HbA (50%: 50%) C SAT = 29.5 g/dl C SAT = 23.5 g/dl HbS:HbF (50%: 50%) C SAT = 30.5 g/dl Hb Concentration (g/dl) p50 of Hb in whole blood (mm Hg) A-T87Q A

49 LentiGlobin vector AT87Q cppt/flap human -globin gene III II I 2.7 Kb -LCR 266bp644bp 845bp 1153bp p HS2 HS3 HS4 ppt HIV LTR RRE 3 enhancer Thr(T) HIV LTR Gln (Q)

50 Testing the A-T87Q /LCR Lentivirus in Two Transgenic Mouse Models of SCD SAD Model "Berkley" Model Trudel et al., EMBO (1991). Paszty et al., Science (1997). Mice express mouse - and -globins, human -globin and human "supersickling" SAD -globin: S ( A-E6V ), Antilles ( A-V23I ), D Punjab ( A-E121Q ). Mice express only human -globin and S -globin. Hb SAD accounts for 19% of total hemoglobin. The sickle cell disease is mild: red blood cell (RBC) sickling and dehydration, irreversibly sickled cells (ISCs), reduced life expectancy. The sickle cell disease is severe: RBC sickling, anemia and reticulocytosis, ISCs, multi-organ pathology, reduced life expectancy.

51 Prevention of RBC Sickling and HbS Polymerization 3 Months Post-Transplantation BERK MOCK BERK A-T87Q po 2 = 5% * p=0.01 p=0.03 Percentage of sickled red blood cells n= BERK MOCK 5% p0 2 13% p0 2 * BERK 87Q SAD MOCK SAD 87Q Log 1/dt Log [c] mg/ml 0.45 SS patient SAD MOCK BERK MOCK AS patient SAD A-T87Q BERK A-T87Q

52 Hematological Parameters in Transplanted BERK Mice Reticulocytes (%) RBC (10 6 per mm 3 ) BERK BERK/GFP BERK/ A-T87Q C57BL/6 * * Hematocrit (%) Hemoglobin (g/dl) BERK BERK /GFP BERK / A-T87Q C57BL/6 * *

53 Elimination of Irreversibly Sickled Cells (ISCs) and Normalization of Urine Concentration 4000 ISCs (No./1000 Red Blood Cells) Urine Concentration (mosm) BERK MOCK BERK A-T87Q 0 Normal (n=22) BERK MOCK (n=4) BERK A-T87Q (n=2)

54 CD34+ cells of a SS patient 62 % transduced progenitors (PCR) Detection of HbA T87Q in 87 % of the transduced BFU-e (HPLC) SCD control SCD transduced SCD transduced 27% 71% Hb 87Q = 14 % adult Hb

55 CD34+ cells of a Thal M patient Complete correction of cellular -Thal M phenotype (in vitro) 4d 13d 17d (Puthenveetil et al., Blood 2004)

56 Gene Transfer and Expression in Primary cord blood HSC (Imren et al., J Clin Invest 2004) ~ 50% transduction of SCID/NOD- CB-derived normal human HSCs 75 > 30% of total human -globin in normal erythroid cells (up to 59%)

57 Conclusions Long term expression (> 10 months) of human A-T87Q -globin in transplanted mice with an average of: 96% RBCs expressing human A-T87Q -globin by FACS (normal mice) 71% expression of human A-T87Q -globin mrna as compared to mouse single mrna (normal mice) High A-T87Q -globin chain expression in human RBCs from transduced CD34+ cord blood cells (~ 50%) Inhibition of RBC sickling and dehydration with correction of hematological abnormalities, prevention of urine concentrating defect and organ pathology (SAD/BERK mice).

58 Safety issues: oncogenesis, insertional mutagenesis

59 Clonal T cell proliferation in SCID patients transplanted with cells modified by c expressing retroviral vectors. Oncogenic safety, insertional mutagenesis LMO2 gene locus (acute lymphoblastic leukemia) -3kb 0kb 5kb 10kb 20kb 30kb Science 2003 P5 C C P normal Patient 4 normal Patient 5 lymphoblasts (P4) P4 P5 LMO2 Actin

60 Insertional mutagenesis /LTR driven oncogene activation MFG c 1 MFG c 2 MFG c P5 (34) P4 (30) P10 (33) LMO2, chr11 BMI pb 1 4 MFG c pb P10 SPAG6, chr10p12.31 MFG c P7 (68) pb CCND2 Intergenic, chr12p13.32

61 Gene therapy and oncogenic risk Insertional mutagenesis : Transcriptional deregulation of a gene, following integration of DNA Target gene -LMO2 Hacein-Bey et al., Science Evi 1 Métais et al., Mol. Ther Transgene - c? - Wood et al., Nature HOXB4 Zhang et al., JCI 2008 Cancer Vector -SIN? -Transcriptional regulatory elements - Origine (onco, lenti) Dose effect -number of grafted cells -MOI -transgene expression level Target cells - Intrinsic factors (proliferation state) -ex vivo manipulation Others Recipient predisposition -immunodeficiency - Shou et al., PNAS 2006

62 Factors favoring insertional oncogenesis Oncongenic potential of IL2R c (Woods et al., Nature 2006) In X-SCID mice ( c -/- ) Non-transduced LV-GFP LV-IL2R c LV-LMO2 6 mois

63 Factors favoring insertional oncogenesis Oncongenic potential of IL2R c Immunodeficiency (Shou Y, PNAS 2006) 1 2 mice Modified cells mouse gene Arf-/- c-/- Arf-/- c-/- GFP c C -/- C Arf-/- c+/+ c 1 2 Bone marrow cells of mice genetically prone to development of lymphoma by deletion of the Arf tumor suppressor gene. C +/+ C C -/- GFP

64 Gene therapy of hemoglobin disorders and evaluation of oncogenic safety - No immunodeficiency - -globin gene, non oncogenic - Regulated, erythroid expression of the transgene - Limited proliferation of modified cells / low or no selection

65 Gene therapy and oncogenic risk Insertional mutagenesis : Transcriptional deregulation of a gene, following integration of DNA Target gene -LMO2 Hacein-Bey et al., Science Evi 1 Métais et al., Mol. Ther Transgene - c? - Wood et al., Nature HOXB4 Zhang et al., JCI 2008 Cancer Vector -SIN? -Transcriptional regulatory elements - Origine (onco, lenti) Dose effect -number of grafted cells -MOI -transgene expression level Target cells - Intrinsic factors (proliferation state) -ex vivo manipulation Others Recipient predisposition -immunodeficiency - Shou et al., PNAS 2006

66 LentiGlobin vector [ A-T87Q -globin/lcr] with anti-sickling properties, selfinactivation & chromatin insulators Increased safety and potential decrease in clonal HSC selection chs4 insulator HIV LTR human -globin gene cppt/flap 266bp 644bp 845bp 1153bp III II I p HS2 HS3 HS4 3 enhancer RRE 2.7 Kb -LCR ppt SIN deletions Chromatin Insulator chs4 Insulator Reverse transcription & chromosomal integration chs4 Insulator cppt/flap human -globin gene III II 266bp 644bp 845bp 1153bp I p HS2 HS3 HS4 ppt SIN + Insulator RRE 3 enhancer SIN + Insulator

67 LentiGlobin vector. Oncogenic risk in -thalassemic mice BMT after lethal irradiation BMT after BFX conditioning 105,0 110,0 100,0 100,0 Survival (%) 95,0 90,0 85,0 85 % Survival(%) 90,0 80,0 70,0 80,0 75, Jours 78 % 60,0 50,0 0,0 100,0 200,0 300,0 Jours Untransduced (n=29) transduced (n=58) Untransduced (n=15) Transduced (n=10)

68 LentiGlobin vector. Oncogenic risk in -thalassemic mice Eight months after bone marrow transplantation of genetically modified cells - No tissue damage (heart, lung, kidney, spleen, liver, thymus) - No cancer cells - No liquid or solid tumor But, only eight months, how to improve sensitivity? Three years in patients ( c), six years in non human primates (hoxb4)

69 How to detect oncogenic events in mice? 1.Use a pre-leukemic mouse model with mutations leading to easily detectable (measurable) liquid tumors 2. Secondary transplant to activate proliferation and reveal cells with oncogenic propensity The lentiglobin vector can activate genes in erythroid cells The animal model must be predisposed to erythroleukemia development

70 Mol Cell Biol Overexpression of the Spi1 oncogene -Around 4 months : clonal erythroleukemia in 50 % of animals -1 st step (Spi1) : differentiation blockage of erythroid precursors -2 nd step (secondary oncogenic events) : uncontrolled proliferation of non- Epo dependent erythroid precursors. Rapid death

71 Experimental protocol Lin - cells BMT I BMT II Eight months 53 souris tg Spi-1 Ly5.2 - untransduced : 11 mice Ex vivo Transduction - vector w/o INS : 12 mice (MOI 10) - vector with INS : 10 mice (MOI 10) 33 syngenic mice Ly5.2 x Ly5.1 1 million transduced Lin - cells 33 syngenic mice Ly5.2 x Ly5.1 cellules femurs + tibias Erythroleukemia frequency and delay LTR Globine pg LCR LTR -INS Globine pg LCR + INS

72 Primary transplant Eight months after BMT mouse weight (g) Spleen weight (mg) L L L E 0 NT - INS + INS 0 NT - INS + INS NT -INS + INS L : Leukemic cells E : splenic, non leukemic erythroblastosis -3/11 blast cells (cytologie) : Control group (NT) -0/22 blast cells : LentiGlobin groups

73 Secondary transplants Higher time to follow transduced cells Favors emergence of potential malignant cells (Baum et al., 2006) Groupes Survival (two years) Detection of leukemic cells Untransduced (n=11) 36,4 % 27,3 % LentiGlobin w/o ins (n=12) LentiGlobin SIN + insulateurs (n=12) 45,5 % 25 % 45,5 % 25 %

74 Leukocytes (10 3 /µl) 20 Primary Secondary weeks post-bmt Erythrocytes (10 6 /ml) Primary Secondary weeks post-bmt 1500 Primary Secondary NT - INS + INS Platelets (10 3 /µl) weeks post-bmt

75 Survival 110,0 Secondary transplant Survival (%) 100,0 90,0 80,0 70,0 60,0 50,0 40,0 30, Days Untransduced (n=12) Modified (n=24)

76 Phase I / II Clinical Trial (ICH standards) 10 patients (5-35 yo) - 5 thalassemia major - 5 sickle cell anemia

77 Primary objective Short-term and 2-year safety and tolerability Secondary objectives To quantify gene transfer efficiency and expression To quantify the degree of hematopoietic chimerism with transduced HSCs To assess correction level of the disease phenotype (e.g., transfusion requirements)

78 Main inclusion criteria Transfusion dependency > 100 ml/kg/year Candidate for allogeneic BMT without suitable, willing, HLA-identical sibling donor High Lansky/Karnofsky performance status > 2 year follow-up history at specialized center

79 Main exclusion criteria Past viral infections (e.g., HIV, HBV, HCV) Active bacterial, fungal, viral or parasitic infections Contraindication to general anesthesia Previous malignancy or familial cancers Cytopenia or previous BMT Psychiatric disorder Pregnancy or lactation Major organ damage (e.g., Lucarelli Class III)

80 Overview of the Clinical Protocol Mobilized HSCs CD34+ cells Vector + Cytokines Maximize % Transduced HSCs or Testing and Release While Frozen Bone Marrow Harvest Busulfex Bone Marrow Conditioning Maximize Myeloablation 2x10 8 unsorted BM cells/kg kept for rescue) IV Infusion Transduced Cells (>2x10 6 CD34 + /Kg) (Spontaneous Homing)

81 The first gene therapy patient with severe E / 0 -thalassemia In Thailand, ~ 3,000 new cases born each year. ~ 4% of the 350 million Southern China population carry a E -or a 0 - allele. Severe in 50% cases (transfusion dependency, iron chelation, splenectomy, candidate for allogenic transplant when sibling donor available) SD E (codon 26 AAG>GAG) SD E SD E : Alternative splice donor with translation frameshift SD: Normal RNA splicing yielding +/- stable E (Glu26Lys) Stop 0

82 The first gene therapy patient with severe E / 0 -thalassemia 18 year old male with E/ 0-thalassemia and no HPFH mutation Transfusion dependent since age 3 (2-3 RBC packs every month, > 225 ml RBCs /kg/year for Hb > 10 g/dl) Spontaneous Hb levels as low as 4.5 g/dl Major hepato-splenomegaly (splenectomy at age 6) and growth retardation Failure of Hydroxyurea therapy (between ages 5 and 17) Desferoxamine (5 days/week) since age 8, and oral Exjade since age 18. No liver fibrosis. Moderate iron overload by liver MRI (561 mol/g). No related, genoidentical HLA-matched donor. Transplantation at age 18 on June 7, 2007 Uneventfull - Duration aplasia = 35 days

83 Lentiviral vector and transduction - transplantation protocol chs4 Insulator chs4 Insulator cppt/flap human -globin gene III II 266bp 644bp 845bp 1153bp I p HS2 HS3 HS4 ppt SIN + Insulator RRE 3 enhancer A-T87Q SIN + Insulator Large-scale clinical grade (cgmp) vector production, concentration and purification Marrow harvest and concerns about hematopoietic stem cell (HSC) content of CD34+ cell population due to hyper-erythroid hematopoiesis in thalassemia Viral titer = 1.1 x 10 8 TU / ml during CD34 + cell transduction Cryopreservation of transduced CD34 + cells with extensive QC testing Pre-transplant transduction efficiency = 0.3 vector copy / CD34 + cell Pre-translant conditioning: BUSULFEX alone (at myeloablative dose) Dose of CD34+ cells injected = 3.9 x 10 6 CD34 + cells / kg

84 Percentage of modified cells in peripheral blood leukocytes (and erythroblastes) Phase I Phase II Whole blood cells (35 months) 9.0 % Modified cells (%) Reconstitution Change in Myelo / Erythro ratio Months post-transplantation Granulo-Mono (CD15+) 17.9 % Erythroblastes (CD45- CD71+) 1.7 % B Lymphocytes (CD19+) 9.2 % T Lymphocytes (CD3+) 2.3 % Bone marrow (36 months) whole bone marow 21.0 % CD % Erythroblastes (glya+ CD71+) 34.6 % Platelets / µl 750,000 Reconstitution 25,000 EB 600,000 20, ,000 15,000 PLT 300,000 10, ,000 5,000 Neutro GM Months post-transplantation Neutro, GM, EB / µl

85 Expression of β A-T87Q - globin in whole blood (HPLC analysis) Absrobance 220 nm Absrobance 220 nm β A β E β A-T87Q β E β A 3 months α γ G γ A 12 months α γ G γ A Absrobance 220 nm βa-t87q 2.5 β E β A-T87Q 33 months α γ G γ A x 100 = % Σ (all β + γ) -5.0

86 Expression of β A-T87Q - globin in whole blood (HPLC analysis) 40 4,0 3,5 A T87Q -globin (%) p < r 2 = ,0 2,5 2,0 1,5 1,0 0,5 Hb A T87Q -globin (g/dl) 0 0, Months post-transplant

87 Conversion to transfusion independence (I) Last RBC transfusion 28 months ago 12 Bleeding (200 ml) Hemoglobin (g/dl) Months post-transplantation 9 g / dl RBC transfusions

88 Conversion to transfusion independence (II) Last RBC transfusion 28 months ago 12 Hemoglobin (g/dl) Total Hb HbA T87Q HbF HbE HbA Months post-transplantation

89 Decreased dyserythropoiesis and increased RBC lifespan Decreased circulating erythroblasts Increased RBC lifespan Erythroblasts (x10 3 /µl) Months post-transplantation RBC (x10 6 /µl) 4,0 3,5 3,0 2,5 2,0 1,5 1, Months post-transplantation 4,0 3,5 3,0 2,5 2,0 1,5 1,0 Retic (x10 5 /µl) Near normal level of A-T87Q -globin expression on a per gene basis MCH correction: Patient's average MCH months post-transplant = 28.4 pg (within normal range pg) AT87Q -globin accounts for 36.2% = 10.3 pg per gene (20.6 pg per two genes) If 100% of RBC contained the therapeutic globin chain, Gene expression output on a per gene basis and compared to normal would be: 20.6 / 30 x 100 = 69 % As it is unlikely that all the RBCs derive from modified progenitors, A-T87Q -globin gene expression output on a per gene basis = %

90 Integration site (IS) analysis by DNA pyrosequencing (Whole nucleated blood cells) 100 % Low total number of different IS (< 300) 24 IS both myeloid and lymphoid Months post-transplantation HMGA2 Relative dominance of IS at the HMGA2 locus (dominance relative to other IS, but > 90 % cells remain untransduced) E1 E2 E3 E4 E5

91 Integration site (IS) analysis by DNA pyrosequencing (Blood cell subsets) T lymphocytes (CD3+) B lymphocytes (CD19+) Granulocytes Monocytes (CD15+) Erythroblasts (CD45-/CD71+) Dominance relative of the HMGA2 IS in both granulocytes-monocytes and erythroblasts, but not lymphocytes

92 Quantification by qpcr of the number of copies of vector provirus in circulating blood cells (overall and HMGA2 IS) (qpcr for whole nucleated blood cells and purified sub-populations) Whole nucleated blood cells Modified blood cells (%) % 2-4% Months post-transplantation Total vector copy number (black) and copy number at specific HMGA2 site (blue)

93 Physiological regulation of HMGA2 expression ATG TAG Let-7 mirnas E1 E2 E3 E4 E5 Intron 3 (~ 113 kb) Expression largely restricted to embryonic tissues and adult stem cells Normal degradation of RNA by Let-7 mirnas (multiple targets in E5) Decreased expression in HSCs correlated with aging

94 Main mechanism of activation of HMGA2 correlated with benignity e.g., lipomas and Paroxysmal Nocturnal Hemoglobinuria (PNH) Expression of truncated HMGA2 mrna (E1 E3) upon rearrangement within the long Intron 3 (with or without translation into fusion protein) Amplification of the expressed truncated mrna by loss of target sites for Let-7 mirna (deleted E5) Let-7 mirnas ATG Intron 3 (~ 113 kb) E1 E2 E3 E4 E5 Main mechanism of activation of HMGA2 correlated with malignancy Expression of full-length by loss of expression of Let-7 mirnas Possible epiphenomenon since Let-7 mirnas also control the degradation of multiple oncogenic mrnas (e.g., Myc and Ras) HMGA2 was NOT identified as an oncogene in the genome-wide N. Copeland mouse oncogene screen

95 Is partial clonal dominance linked to HMGA2 activation? Relative expression Months post BMT hes HeLa E 1-2 E 3-4 E 4-5 HeLa E 1-2 E 3-4 E 4-5 hes E 1-2 E 3-4 E M E 1-2 E 3-4 E M High expression level: x10,000 (whole blood cells) by RT-qPCR Evidence of truncation of the main HMGA2 transcript (E1 E3) by staggered RT-qPCR thal patient Tr FL E1 E2 E3 LG E4 E5 Detection of the protein in erythroid cells derived from BFUes in vitro

96 Aberrant splicing of the third intron and PA within the Globin LV ATG TAG LG hes = vector Patient HMGA2 exons Alternative exon LentiGlobin vector exon Poly A β-globin Let-7 mirna target sites I II III LCR IV V ΔU3-cI-R-U5 ΔU3-cI-R-U5 I II III AAAAAAAA Sequencing of the main HMGA2 transcript Aberrant splicing within the vector

97 Erythroid-specific transcription of truncated HMGA2 RNA with added post-transcriptional mrna stabilization RT-qPCR "run-on" Only expressed in erythroid cells (not granulo-mono) in spite of similar percentage vector at HMGA2 IS (qpcr HMGA2/LV junction) Increased transcription = 371 fold (likely LCR-driven with insulator failure) Increased RNA stability = 36 fold

98 Proposed target of Globin LV-HMGA2 IS clone formation HMGA2 IS = 0% LT-HSC LTC-IC HMGA2 IS = 8 % Lymphoid-restricted LT-HSC (?) Myeloid-restricted LT-HSC (?) "α cells" "Stemness" acquisition Common Myeloid Progenitor BFU-E HMGA2 IS = 17% CFU-GM HMGA2 IS 7 % BM Erythroblast Erythroblast Granulo-Mono

99 Is hematopoietic homeostasis maintained? Normal blood and bone marrow cytology and cytoflurometry examination Normal karyotype and CGH-array chromosomal analysis Lack of cytokine-independence in vitro in CFC assays Normal LTC-IC counts Asymptotic stabilization of the clone relative dominance qpcr for HMGA2/LV DNA junction RT-qPCR for HMGA2 RNA Vector bearing whole blood cells (%) Months post-transplantation Relative expression Months post-transplantation

100 Absence of HMGA2-driven amplification of PLB s myeloid common progenitors / myeloid stem cells in transplanted NSG mice Injection of four NSG mice with a total of bone marrow CD34+ cells (three years after transplant). Week 8 : pool of bone marrow cells, depletion of mouse cells and detection of human erythroid, myeloid and lymphoid cells. Glycophorin A Erythroid Myeloid B-Lymphoid % 0.001% % CD33/15/66b CD33-APC CD19/20-PE CD19/ hu-cd hu-cd hu-cd45

101 Recent evidence of HMGA2 IS in other gene therapy trials ( hotspot or evidence of homeostatic in vivo advantage?) HMGA2 in X-SCID trial (γ-rv vector) > 15 cluster IS in HMGA2 (aggregates of patients data): - 12 in HMGA2 Intron 3-11 in same orientation - Increase abundance with time and then stabilize - 2 (at least) with truncated RNA by aberrant splicing Intron 3 into vector HMGA2 in ALD trial (LV vector) 1 IS in HMGA2 Intron 3 in patient P1: - only in B lymphocytes and 1 time-point (9 months) Interpretation of apparent polyclonal patterns of IS (ALD, SCID) HMGA2 hotspot and true lack of genetic dysregulation resulting in growth advantage? OR "Peaceful co-existence" of cells with minor genetic dysregulation, including HMGA2 and others?

102 CONCLUSIONS (I) High expression of therapeutic globin gene % normal endogenous β A -globin on a per gene basis No evidence of PEV Therapeutic gene contribute > 3.5 g / dl Hb in blood Conversion to transfusion independence for the last 16 months Major life span increase of corrected RBCs (life span > 20) Hb > 9.0 g/dl (in the absence of any transfused RBCs) Persistence reticulocytosis and erythroblastosis Much improved quality of life Excellent tolerance: constrution worker (electrician) No longer transfused Oral Exjade / Deferiprone

103 CONCLUSIONS (II) Transcriptional activation of HMGA2 gene by LCR + truncation HMGA2 mrna by aberrant splicing and polyadenylation into Globin LV Failure of single core chs4 insulator Further increase HMGA2 RNA stability by loss of Let-7 mirna control Presumptive homeostasis (benign HMGA2 RNA truncation, clonal stabilization, normal caryo-cytology) to be monitored Other cases of IS in HMGA2: SCID-X1 (RV), ALD (LV)

104 Harvard Medical School, Brigham and Women s Hospital, Boston, MA Philippe Leboulch Karen Westerman Resy cavalleso Shamil Sunyaev Robert Pawliuk Institute of Emerging Diseases and Innovative Therapies INSERM U. 962 University Paris 11 Philippe Leboulch Emmanuel Payen Olivier Negre Floriane Fusil Béatrix Gillet-Legrand Yves Beuzard Hopital Necker (AP-HP), Paris Marina Cavazzana-Calvo Salima Hacein-Bey Abina Laure Cacavelli Other University Hospitals of Paris (AP-HP): Saint-Louis, Mondor, CHIC, Tenon, Saint Vincent de Paul Eliane Gluckman Françoise Bernaudin Gérard Socié Robert Girot Jean Soulier Nathalie Cartier Patrick Aubourg Genetix Pharmaceuticals, Inc, Cambridge, MA Kathy Hehir Maria Denaro Julian Down University of Pennsylvania School of Medicine, Philadelphia, PA Frederick Bushman Gary Wang Indiana Vector Production Facility, Indianapolis, IN Ken Cornetta Scott Cross Chris Ballas

105 Science 21/12/2007