Blinatumomab or Inotuzumab Which to use? Max S.Topp M.D. Universitätsklinikum Würzburg Germany

Transcription

1 Blinatumomab or Inotuzumab Which to use?. Max S.Topp M.D. Universitätsklinikum Würzburg Germany

2 Disclosures Amgen: Research support, advisory board, honoraria and travel support Pfizer: Advisory board, travel support Regeneron: Research support, advisory board and travel support.

3 < Incidence (per 100,000) ALL is a relatively rare haematological malignancy 40 SEER incidence data CLL AML CML ALL 0 Age (years) ALL is relatively rare in adults, with an estimated incidence rate of per 100,000 person years 2 AML, acute myeloid leukaemia; CLL, chronic lymphocytic leukaemia; CML, chronic myeloid leukaemia; SEER, Surveillance, Epidemiology and End Results (Accessed April 2017); Katz AJ, et al. Cancer Causes Control 2015;26:

4 High CR rates are seen after front-line therapy in adults with ALL, but relapse is common Study Year N Median age, years CR rate, % OS, % at year CALGB %, 3 years SWOG 8417/ %, 8 years NILG 08/ %, 3 years JALSG %, 6 years Sweden %, 5 years GIMEMA 02/ %, 9 years MDACC %, 5 years EORTC ALL %, 6 years LALA %, 5 years GOELAL %, 6 years PETHEMA ALL %, 5 years GMALL %, 5 years MRC-ECOG NR 90 39%, 5 years CR, complete response; NR, not reported; OS, overall survival Bassan R, et al. J Clin Oncol 2011;29:

5 Despite improvements in prognosis for young adults, survival remains poor for ALL patients 5-year relative survival (%) < year relative survival (%) 100 Survival by age 1 (diagnosed between ) 100 Survival over time (by age) Age (years) 1. Katz AJ, et al. Cancer Causes Control 2015;26: ; 2. Dinmohamed AG, et al. Leukemia 2016;30: Calendar year Age (years)

6 Overall survival (%) Salvage chemotherapy at relapse is associated with very poor outcomes Overall survival with standard of care salvage chemotherapy after relapse (Ph-negative BCP-ALL, N=1618) Time from first salvage (years) Adapted from Gökbuget N, et al. Haematologica 2016;101:

7 Despite the poor prognosis of r/r ALL, a further CR significantly improves outcomes Overall survival (%) Overall survival according to achievement of CR in patients receiving first salvage therapy* Survival months CR No CR P< *Patients achieving CR but for whom no date of CR was available were excluded from analysis; Mantel Byar analysis beginning at 36 days after start of Salvage 1 (median time to CR), and therefore patients who died, or whose data were censored, before 36 days are not included Adapted from Gökbuget N, et al. Haematologica 2016;101:

8 Targets for Immunotherapy in B precursor ALL Surface antigen Expression on precursor B-ALL cells* CD % CD % CD % CD % CD52 79% Therapies targeted to specific surface antigens are a novel alternative or complementary approach to chemotherapy *Surface antigen expression with a cut-off of >20% positive leukaemia blast cells Hoelzer D. Hematology Am Soc Hematol Educ Program 2011;2011:243 9.

9 Immunotherapy Approaches in ALL Antibodies (naked antibodies or ADCs) Bispecific T-cell engager (BiTE ) antibodies CAR-T cells Naked antibodies Inotuzumab Blinatumomab CAR T cell ozogamicin a-cd22 Alemtuzumab Rituximab Epratuzumab Calicheamicin a-cd19 a-cd3 T-cell membrane ζ-chain of CD3 Complement ADCC Linker CD3 CD3 Costimulating protein CD22 MAC ADC, antibody drug conjugate Portell CA. Leuk Lymphoma 2014;55: Apoptosis

10 Inotuzumab ozogamicin: antibody drug conjugate designed to deliver cytotoxic chemotherapy to CD22-expressing cancer cells Inotuzumab ozogamicin Calicheamicin Hydrazone linker de Vries et al. Leukemia 2012;26:255 64; Beck A, et al. Discov Med 2010;10:

11 Inotuzumab ozogamicin vs SOC chemotherapy in adults with CD22-positive R/R ALL (INO-VATE) Eligibility R/R CD22+ ALL Due for 1st or 2nd salvage therapy 5% leukaemic BM blasts Ph-negative ALL Ph-positive ALL if failing 1 2nd generation TKI 1:1 randomisation (N = 326) Stratification factors: Duration of CR1 ( 12 vs <12 months) Salvage status (2 vs 1) Age ( 55 vs <55 years) INO Starting dose 1.8 mg/m 2 /cycle 0.8 mg/m 2 on Day 1; 0.5 mg/m 2 on Days 8 and 15 of a day cycle ( 6 cycles) SC chemotherapy (investigator choice) FLAG Cytarabine plus mitoxantrone High-dose cytarabine (HiDAC) 4 cycles SOC, standard of care Kantarjian HM, et al. N Engl J Med 2016;375:

12 INO-VATE Study endpoints Primary Overall survival Rate of CR/CRi Secondary Incidence of AEs Duration of remission (DOR) Progression-free survival (PFS) Allogeneic HSCT realisation MRD negativity* *Defined as below the threshold of <10-4 bone marrow blasts, by multiparameter flow cytometry Kantarjian HM, et al. N Engl J Med 2016;375:

13 INO-VATE baseline characteristics and prior therapies Characteristic INO (N=109) SOC (N=109) Median age, y (range) 47 (18 78) 47 (18 79) CD22 expression <90%, n (%)* CD22 expression 90%, n (%)* 24 (22) 74 (68) 24 (22) 63 (58) Karyotype, n (%)* Normal t(9;22)/bcr-abl1-positive t(4;11)/mll-af4-positive Other abnormalities 27 (25) 14 (13) 3 (3) 49 (45) 23 (21) 18 (17) 6 (6) 46 (42) BM blasts <50%, n (%)* BM blasts 50%, n (%)* 30 (28) 77 (71) 29 (27) 78 (72) Salvage status, n (%)* (67) 35 (32) 69 (63) 39 (36) Duration of CR1 <12 months, n (%) Duration of CR1 12 months, n (%) 62 (57) 47 (43) 71 (65) 38 (35) Previous HSCT, n (%) 17 (16) 22 (20) Number of previous induction therapies, n (%) (69) 33 (30) 1 (1) 69 (63) 39 (36) 1 (1) *Values may not total 100% due to missing assessments Kantarjian HM, et al. N Engl J Med 2016;375:

14 INO-VATE Response to therapy Response INO (N=109) SOC (N=109) P-value CR, % [95% CI] 35.8 [ ] 17.4 [ ] CRi, % [95% CI] 45.0 [ ] 11.9 [ ] <0.001 MRD negativity among responders, % [95% CI] 78.4 [ ] 28.1 [ ] <0.001 Subsequent HSCT, % In those achieving CR/CRi, % < Kantarjian HM, et al. N Engl J Med 2016;375:

15 Probability of OS INO-VATE Overall survival Intention-to-treat population Hazard ratio*: 0.77 (97.5% CI, ) P=0.04 Patients at risk: INO SOC Months *Hazard ratio for death The co-primary objective of longer OS with INO vs SOC was not met Kantarjian HM, et al. N Engl J Med 2016;375:

16 INO-VATE Serious adverse events Safety population, regardless of causality Serious TEAEs, observed in 3 patients from either arm, occurring between first dose and 42 days after last dose 5/15 cases of VOD developed shortly after treatment 10/15 developed after transplantation (2 fatal events) Median time to development of VOD after transplant in INO patients was 16 days Kantarjian HM, et al. N Engl J Med 2016;375: INO (N=139) SOC (N=120) Any Grade Grade 3 Any Grade Grade 3 Any serious TEAE, n (%) 67 (48) 64 (46) 55 (46) 52 (43) Febrile neutropenia 16 (12) 15 (11) 22 (18) 21 (18) Veno-occlusive disease (VOD) 15 (11) 13 (9) 1 (1) 1 (1) Sepsis 3 (2) 3 (2) 6 (5) 6 (5) Pyrexia 4 (3) 2 (1) 3 (2) 1 (1) Disease progression 5 (4) 5 (4) 2 (2) 2 (2) Pneumonia 5 (4) 5 (4) 1 (1) 0 (0) Neutropenic sepsis 3 (2) 3 (2) 3 (2) 3 (2) Respiratory failure 1 (1) 1 (1) 4 (3) 4 (3) Abdominal pain 3 (2) 2 (1) 1 (1) 1 (1) Hyperbilirubinemia 0 (0) 0 (0) 3 (2) 3 (2) Hypotension 0 (0) 0 (0) 3 (2) 3 (2)

17 The world of bispecific antibody constructs Fynomers Zybodies Kontermann, mabs (2012)

18 Bispecific antibody constructs in clinical development for B-cell malignancies Structural comparison BiTE 1,2 DART 1,2 TandAb 3,4 Native format bispecific antibody 5,6 N N C N N N N Common light chain Constant region Specificity 1 Specificity 2 Linker C N Monomer Bivalent C C N Heterodimer Bivalent 1. Moore PA, et al Blood 2011;117: ; 2. Rader C. Blood 2011;117:4403 4; 3. Reusch U, et al. MAbs 2015;7: ; 4. Rothe A, et al. Blood 2015;125: ; 5. Smith EJ, et al. Sci Rep 2015;5:17943; 6. Sun LL, et al. Sci Transl Med 2015;287:287ra70. N C Homodimer Tetravalent C C C C Heterotetramer Bivalent Disulphide bridge Different on off rates to respective targets for both leukaemia antigens and T cells

19 Additional T-cell engaging modalities for patients with ALL Modality and specificity Dual affinity re-targeting (DART ) anti- CD19/CD3 1 Bispecific anti-cd19/cd3 tandem diabody (TandAb ) 2,3 Name Duvortuxizu mab Bispecific native format anti-cd20/cd3 4,5 REGN1979 Disease setting ALL, NHL, CLL Developm ent Stage Phase 1 AFM11 ALL, NHL Phase 1 ALL, NHL, CLL Phase 1 Bispecific anti-cd19/cd47 6 NI-1701 ALL, DBCL Pre-clinical Azymetric TM bispecific anti-cd19/cd3 7 ZW38 ALL, NHL Pre-clinical Sleeping beauty-transduced anti-cd19 CAR-T cells 8 CD19RCD2 8 ALL, NHL Phase Ferlin WG, et al. ASH 2016; Abstract 44 and oral presentation; 7. Ng G, et al. ASH 2016; Abstract 1841 and poster presentation; 8. Kebriaei P, et al. J Clin Invest 2016;126:

20 Mode of action of Blinatumomab Bassan, Blood: 120; 2012

21 Administration of blinatumomab Continuous IV (civ) administration via pump Inpatient & ambulatory; 24/7 for 4 weeks, 2 weeks off Administration in inpatient and ambulatory home settings Inpatient: COE, academic centre, community Outpatient: hosp. OP, clinic, home Hospital Minimum of: Days 1 9, cycle 1* Days 1 2, cycle 2 COE, centre of excellence; OP, outpatient Home civ infusion bag changes every 4 days by home healthcare service or in outpatient setting *14 days in patients with a history or presence of clinically relevant CNS pathology BLINCYTO Summary of Product Characteristics Amgen BV, 2015

22 Summary of phase 2 studies of Blinatumomab in r/r B precursor ALL Adults with R/R ALL (n=36) 1 Blinatumomab 5 15 μg/m 2 /d civ, 4 weeks on, 2 weeks off identified in dose-finding phase, used in extension phase Key efficacy results 69% CR/CRh* rate after 2 cycles (1 endpoint) 88% molecular CR rate 52% allosct realisation rate Adults with R/R ALL (n=189) 2 Blinatumomab 9 28 mg/d civ, 4 weeks on, 2 weeks off Key efficacy results 43% CR/CRh* rate after 2 cycles (1 endpoint) 82% molecular CR rate 42% allosct realisation rate in CR/CRh patients Key safety results Leukopenia most common grade 3 AE (14% pts) Neurological events requiring treatment interruption/discontinuation in 6 pts (17%) Disease burden associated with risk of grade 3 CRS (n=2); no grade 3 CRS with stepped dosing + prephase dexamethasone and/or cyclophosphamide for pts with BMB 50% Key safety results Febrile Neutropenia most common grade 3 AE (14% pts) Neurological events requiring treatment interruption/discontinuation in 13% pts 3 CRS events in 3 out of 189 pts (1,6%) * CRh, CR with partial haematological recovery (platelets >50,000/μL, haemoglobin >7 g/dl, ANC >500/μL) Dexamethasone 10 mg/m 2 (up to 5 days) and/or cyclophosphamide 200 mg/m 2 (up to 3 days); BMB, bone marrow blast; civ, continuous IV infusion 1. Topp MS et al. J Clin Oncol 2014;32: Topp MS, et al. Lancet Oncol 2015;16:57 66

23 Probability of RFS Survival probability Long-term follow-up of exploratory phase 2 study of adults with relapsed/refractory ALL RFS OS 1.0 Censored 1.0 Censored Median RFS: 8.8 months (at 28.9 months median follow-up) 0.6 Median OS: 13 months (at 32.6 months median follow-up) At risk Time (months) 0.0 At risk N=10 long-term survivors at the time of data cut-off for these ad hoc analyses Zugmaier G et al. Poster presentation at ASH, San Francisco, CA; December , Abstract 2287

24 Long-term follow-up of exploratory phase 2 study of adults with relapsed/refractory ALL Overall survival by MRD response Survival probability MRD response* assessed at Day Overall survival, MRD responders Overall survival, MRD non-responders 67% risk reduction associated with achieving an MRD response (Mantel-Byar odds ratio 0.33; p = 0.009) Patients at risk (n) *MRD <10-4 by PCR testing Zugmaier G, et al. Blood 2015;126: Time (months)

25 Phase 3, randomised, open-label study of blinatumomab vs standard of care (SOC) chemotherapy in adults with r/r B-ALL (TOWER) 2:1 randomisation (blinatumomab:soc) Stratified by age (<35 or 35 years), prior salvage, and prior allogeneic HSCT Blinatumomab civ* SOC chemotherapy Induction (up to 2 cycles) Consolidation (up to 3 cycles) 4 weeks on, 2 weeks off (9 µg/day in induction Cycle 1 Week 1 and 28 µg/day thereafter) If 5% blasts Induction (up to 2 cycles) Consolidation (up to 3 cycles) Investigator choice of: FLAG ± anthracycline HiDAC-based High-dose MTX-based Clofarabine-based If 5% blasts Maintenance (up to 12 months) 4 weeks on, 8 weeks off Maintenance (up to 12 months) Continuation of SOC chemotherapy *Patients with high tumour load were to receive pre-phase dexamethasone to prevent cytokine release syndrome. Patients were to receive pre-dose dexamethasone to prevent infusion reactions; FLAG, fludarabine, high-dose cytarabine and granulocyte colony-stimulating factor; HiDAC, high-dose cytarabine; MTX, methotrexate Kantarjian H,; Topp MS. N Engl J Med 2017;376:

26 TOWER Study endpoints Primary Overall survival Key secondary CR within 12 weeks CR/CRh/CRi within 12 weeks Event-free survival Other secondary Remission duration MRD-negativity (<10-4 ) Allogeneic HSCT rate Adverse event rates Time to deterioration in Quality of Life (QoL) CRi, CR with incomplete haematological recovery: 5% BM blasts, no evidence of circulating blasts or extra-medullary disease, platelets <100,000/μL or ANC <1,000/μL but not achieving CRh Kantarjian H, et al. N Engl J Med 2017;376:836 47;.

27 TOWER Baseline patient characteristics Characteristic Blinatumomab (n=271) SOC chemotherapy (n=134) Mean age, years (range) 40.8 (18 80) 41.1 (18 78) Male, n (%) 162 (59.8) 77 (57.5) Key study inclusion criterion, n (%) Refractory to primary or salvage therapy In first relapse with CR1 <12 months In untreated second or greater relapse* Relapsed after allogeneic HSCT* Salvage status, n (%) First Second Third Fourth Fifth or later Prior allogeneic HSCT, n (%) Yes No Unknown 115 (42.4) 76 (28.0) 32 (11.8) 46 (17.0) 114 (42.1) 91 (33.6) 45 (16.6) 14 (5.2) 7 (2.6) 94 (34.7) 176 (64.9) 1 (0.4) 54 (40.3) 37 (27.6) 16 (11.9) 27 (20.1) 65 (48.5) 43 (32.1) 16 (11.9) 5 (3.7) 5 (3.7) 46 (34.3) 87 (64.9) 1 (0.7) *Patients who met none of the above inclusion criteria. Kantarjian H, et al. N Engl J Med 2017;376:

28 Proportion of patients (with upper 95% CI) TOWER Remission rates within 12 weeks 60% Blinatumomab (n=271) SOC (n=134) 50% (P<0.001) (P=0.007) 40% 44% (P<0.001) 45% 30% 34% 30% 20% 25% 10% 0% CR/CRh/CRi (Intent-to-treat) 16% 9% 2% 5% 5% (n=267) (n=109) CR CRh CRi CR/CRh/CRi (Treated patients) MRD Response in CR patients: 77% for Blinatumomab vs 52% for SOC Kantarjian H,; Topp MS. N Engl J Med 2017;376:

29 Overall survival (%) TOWER Overall survival Intent-to-treat population Blinatumomab SOC Median OS, months (95% CI) 7.7 ( ) 4.0 ( ) Median follow-up, months HR for OS (95% CI) 0.71 ( ); P=0.012* Blinatumomab SOC 0 Number of patients at risk: Years *Stratified log-rank P-value; HR, hazard ratio Kantarjian H,; Topp MS. N Engl J Med 2017;376:

30 Overall survival (%) TOWER Overall survival Censoring at the time of allogeneic HSCT HR for OS (95% CI) Blinatumomab SOC Median OS, months (95% CI) 6.9 ( ) 3.9 ( ) 0.66 ( ); P=0.004* Blinatumomab SOC 20 0 Number of patients at risk: Years *Stratified log-rank P-value Kantarjian H,; Topp MS. N Engl J Med 2017;376:836 47

31 TOWER Adverse events Event, n (%) Blinatumomab (n=267) SOC Chemotherapy (n=109) n % n % Any Leading to discontinuation of study treatment Serious AE Fatal AE Any Grade 3 AE Grade 3 AEs of interest categories* Neutropenia Infection Elevated liver enzymes Neurological events Cytokine release syndrome Tumor lysis syndrome Lymphopenia Any decrease in platelets Any decrease in WBC count *Reported for 3% of patients in either group; WBC, white blood cell Kantarjian H,; Topp MS. N Engl J Med 2017;376:

32 Key patient subgroups in INO-VATE and TOWER studies Inclusion criteria Patients due for 1st salvage, % With early relapse* TOWER 1 (N=405) INO-VATE 2 (N=218) Patients due for 2nd salvage, % Patients due for 3rd salvage, % 23 0 Refractory patients, % Patients who received prior HSCT, % Ph-positive patients, % 0 15 In both studies, baseline characteristics were balanced between treatment arms Patients in TOWER had a poorer prognosis and fewer treatment options than those in INO-VATE. Therefore, the outcomes of these trials cannot be directly compared *Duration of first CR <12 months Defined as disease refractory to primary therapy or salvage therapy (TOWER) and treatment-resistant disease (INO-VATE) 1. Kantarjian H, et al. N Engl J Med 2017;376:836 47; 2. Kantarjian H,; Topp MS. N Engl J Med 2017;376:836 47

33 SOC therapy in INO-VATE and Tower studies SOC therapy options INO-VATE 1 Available SOC regimens, % of cycles FLAG 70% HiDAC 11% AraC + mitoxantrone 19% Median treatment cycles received (range) 1 (1 4) SOC therapy options TOWER 2 Available SOC regimens, % of patients FLAG ± anthracycline 45% HiDAC 17% High-dose methotrexate 20% Clofarabine 17% Median treatment cycles received (range) 1 (1 4) 1. Kantarjian H, et al. N Engl J Med 2016;375:740 53; 2. Kantarjian H,; Topp MS. N Engl J Med 2017;376:

34 Key response and efficacy data in TOWER and INO-VATE studies Parameter CR/CRi Overall 1st salvage INO 81% 88% INO-VATE 2,3 SOC 29% 29% CR 36% 17% MRD-negativity* 78% 28% HSCT realisation 41% 11% Median DOR 4.6 months 3.1 months Median OS Overall 1st salvage 7.7 months 8.6 months 6.7 months 7.1 months Parameter CR/CRh/CRi Overall 1st salvage Blinatumomab 44% 53% TOWER 1 SOC 25% 35% CR 34% 16% MRD-negativity* 76% 48% HSCT realisation 24% 24% Median DOR 7.3 months 4.6 months Median OS Overall 1st salvage 7.7 months 11.1 months 4.0 months 5.3 months The INO-VATE and Tower studies had differing patient populations and SOC therapy. Therefore, the outcomes of these trials cannot be directly compared *In responding patients; DOR, duration of remission 1. Kantarjian H, et al. N Engl J Med 2017;376:836 47; 2. Kantarjian H,; Topp MS. N Engl J Med 2017;376: Kantarjian H, et al. EHA 2016; Abstract LB2233 and oral presentation.

35 EORTC Quality of Life questionnaire QLQ-C30, a patient reported outcome instrument Core 30 questions of the EORTC QLQ questionnaire for the evaluation of patients participating in clinical cancer studies Global health status/qol Functional scales Symptom scales Better health-related QoL: Higher global health Status/QoL Higher functions scores Lower symptoms scores Physical Role Emotional Cognitive Social Fatigue Pain Nausea & vomiting Appetite loss Constipation Diarrhoea Dyspnoea Financial difficulties Insomnia Aaronson NK, et al. J Natl Cancer Inst 1993;85:365 76; Topp MS, et al. ASH 2016; Abstract 222 and oral presentation; Kantarjian HM, et al. N Engl J Med 2016;375:

36 Symptom scales Symptom scales Functional scales Functional scales Summary of Health-related QoL data in TOWER and INO-VATE studies Favoured treatment arms in TOWER and INO-VATE QLQ-C30 analysis TOWER 1 INO-VATE 2 QLQ-C30 measure Favoured arm QLQ-C30 measure Global health status/qol Blinatumomab Global health status/qol Physical Blinatumomab Physical Role Blinatumomab Role Emotional Blinatumomab Emotional Cognitive Blinatumomab Cognitive Social Blinatumomab Social Nausea/vomiting Blinatumomab Nausea/vomiting Constipation Blinatumomab Constipation Dyspnoea Blinatumomab Dyspnoea Insomnia Blinatumomab Insomnia Pain Blinatumomab Pain Appetite loss Blinatumomab Appetite loss Diarrhoea Blinatumomab Diarrhoea Financial difficulties Blinatumomab Financial difficulties Favoured arm INO INO INO SOC INO INO INO SOC INO INO INO INO INO INO The TOWER and INO-VATE studies had differing patient populations and SOC therapy. Therefore, the outcomes of these trials cannot be directly compared 1. Topp MS, et al. ASH 2016; Abstract 222 and oral presentation; 2. Kantarjian HM, et al. ASH 2016; Abstract 1599 and poster presentation. Significant (P<0.05) Not significant (P 0.05)

37 Continuous CR (%) Overall survival (%) MRD-negativity is a marker of improved outcome following induction therapy in adult ALL Impact of MRD status at Week 16 on response in Ph-negative adult ALL patients following induction/consolidation therapy Continuous CR* Overall survival* MRD-negative CCR at 5 years: 70% (n=333) MRD-negative OS at 5 years: 81% (n=333) MRD-positive CCR at 5 years: 12% (n=63) MRD-positive OS at 5 years: 33% (n=63) Time (years) Time (years) Patients MRD-negative at Week 16 had significantly higher probability of CCR and OS at 5 years (P< for both endpoints) than those who remained MRD-positive. Achievement of MRD negativity is therefore a potentially useful clinical endpoint * Patients with HSCT in CR1 excluded; MRD-negativity with an assay sensitivity of 10-4 ; Persistent quantifiable MRD positivity within the quantitative range ( 10-4 ); CCR, continuous CR; MRD, minimal residual disease; Ph, Philadelphia chromosome; HSCT, Haematopoietic stem cell transplant Adapted from Gökbuget N, et al. Blood 2012;120:

38 Probability of RFS Benefit of allohsct predominantly in patients who are MRD-positive after first induction Effect of MRD status on RFS with/without allohsct (GRAALL 2003/2005) MRD-, no HSCT MRD+, no HSCT MRD-, with HSCT MRD+, with HSCT Year HSCT largely overcomes the poor prognostic impact of MRD HSCT does not further improve outlook for MRD-negative patients Post-induction MRD status specifies which patients could benefit from HSCT MRD-, MRD1 <10-3 ; MRD+, MRD Dhédin N, et al. Blood 2015;125:

39 Initial exploratory Phase 2 studies identified efficacious dosing strategies while managing risk of dose-limiting AEs Adults with MRD+ ALL (n=21) 1 Blinatumomab 15 μg/m 2 /d civ, 4 weeks on, 2 weeks off Key efficacy results 80% molecular CR rate after 2 cycles (1 endpoint) Key safety results Lymphopenia most common grade 3 AE (33% pts) Grade 3/4 neurological events in 4 pts (19%), including seizure requiring treatment discontinuation in 1 pt (fully reversible) No cases of severe cytokine release syndrome (CRS) * CRh, CR with partial haematological recovery (platelets >50,000/μL, haemoglobin >7 g/dl, ANC >500/μL) Dexamethasone 10 mg/m 2 (up to 5 days) and/or cyclophosphamide 200 mg/m 2 (up to 3 days); Blinatumomab is not licensed for use in MRD+ ALL BMB, bone marrow blast; civ, continuous IV infusion 1. Topp MS, et al. J Clin Oncol 2011;29:2493 8;

40 Probability of no haematological relapse/death Long-term follow-up of exploratory Phase 2 study of adults with MRD+ ALL Probability of no MRD relapse Haematological RFS MRD response duration in Ph- patients who received only blinatumomab Patients with SCT Patients without SCT Time (months) Time (months) With SCT (n=9): median follow-up = 32.9 months Without SCT (n=11): median follow-up = 30.8 months Median follow-up (n=6) = 29.2 months Topp MS, et al. Blood 2012;120:5185 7

41 Follow-up Open-label, single-arm, multicentre, Phase 2 study of blinatumomab in adults with MRD-positive* B-precursor ALL (BLAST) 1,2 MRD assessment by PCR at central laboratory: timing of bone marrow biopsy Screening Day 15 (recommended) End of Cycle 1 End of Cycle 2 End of Cycle n Efficacy follow-up Screening and enrolment Adults ( 18 years) MRD+ B-ALL in haematological CR No prior HSCT Blinatumomab 15 µg/m 2 /day civ infusion 4 weeks on, 2 weeks off 1 cycle Primary endpoint assessment Blinatumomab 15 µg/m 2 /day civ infusion 4 weeks on, 2 weeks off Up to 3 cycles in responding patients Consolidation Study endpoints Primary: MRD-negative rate after 1 Cycle Secondary: Haematological RFS at 18months OS Time to haematological relapse Duration of MRD-negativity Incidence and severity of AEs Exploratory: MRD response after 1 cycle AlloHSCT offered to eligible patients when donor available 15 µg/m 2 /day dose used in this study corresponds to 28 µg/day fixed-dosing strategy 3 *Blinatumomab is not licensed for use in MRD-positive ALL; <5% blasts after three chemotherapy blocks, ANC 1,000/μL, platelets 50,000/μL, Hb 9 g/dl, with MRD 10-3 ; MRD-negative, no amplification in PCR (assay sensitivity 10-4 ); Reduction of MRD to <10-4 (assay sensitivity 10-4 ) 1. Gökbuget N, et al. ASH 2014: Abstract 379 and oral presentation; 2. Gökbuget N, et al. ASH 2015: Abstract 680 and oral presentation; 3. Wu B, et al. ASCO 2013: Abstract and poster presentation.

42 BLAST MRD-negativity within 1 cycle Patients with evaluable MRD Primary endpoint full analysis set* (n=113) n % 95% CI Primary endpoint MRD-negativity after Cycle Exploratory endpoint MRD response (MRD not detectable or only present at <10-4 ) after Cycle 1 2 patients achieving reduction of MRD to below the quantifiable limit during Cycle 1 achieved a MRD-negativity after continued treatment in Cycle 2 The lower CI bound exceeded 44% (the null hypothesis response rate), confirming that the study met its primary objective. *Patients who had an MRD analysis available with an assay sensitivity of 10-4 at the central lab Gökbuget N, et al. ASH 2014: Abstract 379 and oral presentation.

43 Survival probability BLAST RFS according to MRD-negativity Ph-negative patients in haematological CR MRD-negative at Cycle 1 (N=85); median 95% CI 23.6 (17.4 NR ) MRD-positive at Cycle 1 (N=15); median 95% CI 5.7 ( ) P=0.003* Subjects at risk, n: Time (months, landmark analysis beginning at study Day 45) 100 patients with data on MRD response according to the primary endpoint efficacy set *Log rank P-value for association between RFS and MRD response; causality not implied; underlying baseline characteristics may also influence both outcomes; The landmark analysis by MRD response included patients with overall survival of 45 days Gökbuget N, et al. ASH 2015: Abstract 680 and oral presentation.

44 Survival probability BLAST Relapse-free survival Ph-negative patients in haematological CR The key secondary endpoint was met: 18-month KM estimate of RFS = 54% (95% CI: 33% 70%) exceeding the pre-specified lower boundary of 28% Median (95% CI) follow-up: 29.9 ( ) months Subjects at risk, n: Time (months) Censoring* Not censoring HSCT in continuous CR: 67% Survival *RFS censoring at allohsct and post-blinatumomab chemotherapy (N=110); median 95% CI NR (6.3 NR); RFS not censoring at allohsct and post-blinatumomab chemotherapy (N=110); median 95% CI 18.9 ( ) Gökbuget N, et al. ASH 2015: Abstract 680 and oral presentation.

45 Probability of RFS Probability of RFS Confirmatory Phase 2 study of blinatumomab in adults with MRD-positive B-precursor ALL Role of SCT in patients with complete MRD response Transplant realisation rate: 72% Cox model analyses of SCT as time-dependent covariate for RFS First remission Second or later remission Hazard ratio: 2.26 (95% CI: 0.73, 6.97) 0.33 (95% CI: 0.11, 0.98) p Simon Makuch plot of RFS (Landmark 45 days) First remission (N=60) Second or later remission (N=25) 1.0 AlloSCT No AlloSCT 1.0 AlloSCT No AlloSCT AlloSCT prior to relapse or censoring: Yes: 15 (60%) No: 10 (20%) AlloSCT prior to relapse or censoring: Yes: 46 (77%) No: 14 (23%) Time (months, landmark analysis beginning at study Day 45) 0.0 Gökbuget N, et al. ASH 2015: Abstract 680 and oral presentation

46 Confirmatory Phase 2 study of blinatumomab in adults with MRD-positive B-precursor ALL Overview of AEs AE (n=116) Treatmentrelated AE (n=116) Any AE,* n (%) 116 (100) 112 (97) Serious AE 73 (63) 60 (52) Grade 3 AE 38 (33) 34 (29) Grade 4 AE 31 (27) 25 (22) Fatal AE 2 (2) 1 (<1) AE leading to interruption of treatment 36 (31) 33 (28) AE leading to permanent discontinuation of treatment 20 (17) 14 (12) Treatment interruption due to AE (>2% of patients): pyrexia (8%); chills, alanine aminotransferase or aspartate aminotransferase increases, overdose, tremor, aphasia, encephalopathy (3% for each) Permanent discontinuation due to AE (>2% of patients): tremor (4%); aphasia, encephalopathy and seizure (3% for each) *Including treatment and 30 days after last infusion; Investigator judged that it was reasonably possible that the event was related to treatment with investigational product Gökbuget N, et al. ASH 2015: Abstract 680 and oral presentation

47 Confirmatory Phase 2 study of blinatumomab in adults with MRD-positive B-precursor ALL Neurological AEs Any neurological event, n (%)* 61 (53) 12 (10) 3 (3) Neurological events in >2% of patients, n (%)* Tremor 35 (30) 6 (5) 0 (0) Aphasia 15 (13) 1 (1) 0 (0) Dizziness 9 (8) 1 (1) 0 (0) Ataxia 7 (6) 0 (0) 0 (0) Paraesthesia 7 (6) 0 (0) 0 (0) Encephalopathy 6 (5) 3 (3) 2 (2) Dysgraphia 5 (4) 0 (0) 0 (0) Intention tremor 4 (3) 0 (0) 0 (0) Convulsion 3 (3) 1 (1) 1 (1) Dysarthria 3 (3) 0 (0) 0 (0) Memory impairment 3 (3) 0 (0) 0 (0) Rates decreased over time (Cycles 1, 2, 3 and 4) for any neurologic event (47%, 24%, 15% and 15%) and any Grade 3 neurologic event (10%, 4%, 0% and 0%) 12 (10%) of patients had treatment interruptions or discontinuations due to Grade 3 4 neurological AEs 5 resumed treatment without another interruption 2 resumed then discontinued treatment for another neurological event Median time to resolution of any neurological events was 4 days (Q1, Q3: 2, 8 days) *Including treatment and 30 days after infusion Gökbuget N, et al. ASH 2015: Abstract 680 and oral presentation All (N=116) Grade 3 (N=116) Grade 4 (N=116)

48 Conclusions Adult ALL is an aggressive malignancy with high unmet clinical need 1 Conventional therapies produce high CR rates but relapse is frequent and, in most cases, fatal 2 Both Inotuzumab and Blinatumomab, are a viable treatment option for patients with r/r Ph-negative B- precursor ALL3 3,4 Blinatumomab is highly active in patients with MRD positive ALL The future position of inotuzumab and/or blinatumomab in the ALL treatment paradigm is currently unknown 1 1. Benjamin JE, Stein AS. Adv Ther Hematol 2016;7:142 56;; 2. Gökbuget N, et al. Haematologica 2016;101: ; 3. Kantarjian HM, et al. N Engl J Med 2016;375: Kantarjian H,; Topp MS. N Engl J Med 2017;376:836 47;