Clinical trials with JAK inhibitors for essential thrombocythemia and polycythemia vera

Clinical trials with JAK inhibitors for essential thrombocythemia and polycythemia vera Alessandro M. Vannucchi, MD Laboratorio Congiunto MMPC Department of Experimental and Clinical Medicine University of Florence, Italy

Disclosures for Alessandro Vannucchi, MD Royalty Receipt of intellectual property/ Patent holder Consulting fee Speakers bureau Fees for non-cme services Contracted research Ownership interest (stocks, stock options) Other N/A N/A N/A N/A Novartis Novartis N/A N/A N/A = Not Applicable (no conflicts listed) Presentation includes discussion of off-label or unapproved use of a drug or medical device

Natural History of Polycythemia Vera (PV) and Essential Thrombocythemia (ET) Diagnosis 10 yr 20 yr Thrombosis Hemorrhage Diagnosis 10 yr 20 yr PPV- or PET- Myelofibrosis Leukemia

Goals of Therapy in Patients with PV and ET To avoid first occurrence and/or recurrence of thrombotic and bleeding complications To minimize the risk of acute leukemia and post-pv or post-et myelofibrosis To control systemic symptoms Barbui T, et al. JCO 2011; 29:761-70

Current ( conventional ) Therapeutic Landscape for PV and ET Treatment is based on the stratification of risk (for thrombosis) Phlebotomy Low-dose aspirin Cytotoxic drugs (hydroxyurea, busulphan) Interferon * * Not approved for these indications

Criteria for Resistance / Intolerance to Hydroxyurea RESISTANCE INTOLERANCE Hydroxyurea (HU) Need of phlebotomy to maintain Hct <45% Platelets (PLT) >400x10 9 /L and WBC >10x10 9 /L Spleen reduction by <50% or No complete relief of spleen-related symptoms ANC <10 9 /L or PLT <100x10 9 /L or Hb <100g/L Leg ulcers or Other unacceptable HU-related toxicities* After 3 months of at least 2 g/day HU At the lowest dose required to achieve complete or partial hematologic response (ELN criteria) At any dose of HU *Mucocutaneous, gastrointestinal, pneumonitis, fever Barosi G, et al. Br J Haematol. 2010;148(6):961-963.

Resistance to Hydroxyurea Adversely Affects Survival and Disease Progression in PV Overall survival Transformation to PPV-MF and AL Copyrighted graph Copyrighted graph Resistance and intolerance to hydroxyurea occurred in 11% and 13% of 261 PV patients Resistance to hydroxyurea implied a 5.6 fold increase in the risk of death and 6.8 fold increase in the risk of transformation PPV-MF, post-pv myelofibrosis; AL, acute leukemia Alvarez-Larrán A, et al. Blood. 2012;119(6):1363-1369.

Possible Different Strategies to Inhibit JAK2 Canonical Non canonical Alternative Inhibitors of the JAK2 tyrosine kinase activity (TKI) Non-TKI that inhibits JAK2 by alternative mechanisms Telomerase inhibitors

The HDAC Inhibitor Givinostat Has Direct Inhibitory Activity on JAK2 V617F Cells GVS is more cytotoxic on JAK2 V617 than JAK2 WT cell lines. Low doses of GVS inhibit proliferation and erythroid differentiation of primary MPN cells. GVS does not affect JAK2 mrna levels but impairs JAK2 half-life. Low doses of GVS impair JAK/STAT signaling by down-regulating JAK2 and p-jak2 levels A pilot study in 13 PV and 1 ET patients has provided evidence of clinical activity (Hct control, reduction of palpable spleen, symptomatic control) with mild GI toxicity. Guerini V, et al. Leukemia. 2008;22(4):740-747. Amaru Calzada A, et al. Exp Hematol. 2012;40(8):634-645. Rambaldi A, et al. Br J Haematol. 2010;150(4):446-455

Synergism of Givinostat with Hydroxyurea Copyrighted material HU synergizes with GVS in reducing proliferation of JAK2 V617F-mutated cell lines Synergism occurs through caspase 3 activation and inhibition of p21cdkn1a induction Amaru Calzada A, et al. Exp Hematol. 2013;41(3):253-260.

Synergism of Givinostat with Hydroxyurea Copyrighted material Copyrighted material HU synergizes with GVS in reducing proliferation of JAK2 V617F-mutated cell lines Synergism occurs through caspase 3 activation and inhibition of p21cdkn1a induction Finazzi G, et al. Br J Haematol. 2013;161(5):688-694. Amaru Calzada A, et al. Exp Hematol. 2013;41(3):253-260.

A Phase II Study of Givinostat in Combination With HU: Response and Safety 50 mg/day 100 mg/day Overall response 11 PR = 50% 1 CR + 9 PR = 45% Hct normalization 28% 42% WBC normalization 25% 36% Platelets normalization 36% 20% Spleen normalization 0% 7% Pruritus > grade 2 normalization 64% 80% AE grade 2 32% 36% Drop out 9% 14% PR, partial response; CR, complete response Finazzi G, et al. Br J Haeatol. 2m013;161(5):688-694.

A Phase II Study of Vorinostat (MK-0683) in Patients With PV and ET Week 24 Week 36 vorinostat 400 mg/day Primary objective: Clinicohematological response per the ELN response criteria at the end of the intervention and observation period Second objective: To investigate whether vorinostat influenced the JAK2 mutant allele burden Andersen CL, et al. Br J Haematol. 2013;162(4):498-508.

A Phase II Study of Vorinostat (MK-0683) in PV and ET: Clinical Responses and Safety 63 patients (PV= 44, ET= 19) were enrolled. 30 In the intervention phase, 33 (52%) discontinued mainly due to AE and SAE. At the end of study, the discontinuation rate was 60%. Main toxicities were mainly non-haematological (fatigue, renal impairment, 25 diarrhoea, hair loss, weight loss, nausea, unspecified pain, headache, and leg ulcers) Efficacy (ITT analysis): End of intervention Responses occurred in 35% of the patients (3 CR, 19 PR) The median decrease of JAK2 V617F allele burden from baseline level was 5.6% End of observation ITT analysis: 9.5% responses (2 in CR, 4 in PR) 33 38 Andersen CL, et al. Br J Haematol. 2013;162(4):498-508.

Dysregulated Telomerase Activity in MPN Upregulated telomerase activity may be involved in proliferation and replication immortality of neoplastic progenitor cells, and has been shown to occur also in MPN cells IMETELSTAT is the first telomerase inhibitor in clinical development Competitively binds to RNA template of telomerase and inhibits is activity IMETELSTAT inhibited growth of spontaneous CFU-MK from ET pts Did not inhibit cytokine-induced CFU-MK growth from healthy controls Ruella M et al, Exp Hematol. 2013 Jul;41(7):627-34; Baerlocher GM, et al. Blood Nov 2012; 120: abstr 179

Hematologic and Molecular Response in ET with Imetelstat JAK2V617F Allele Burden (%) Baerlocher GM, et al. Blood Nov 2012; 120: abstr 179

Studies with JAK2 Inhibitors in PV and ET Drug LY2784544 momelotinib (CYT387) SAR302503/TG101348 INCB018424/Ruxolitinib Diseases and studies PV-ET: phase II completed PV-ET: phase II ongoing PV-ET: phase II STOPPED PV-ET: phase II completed; PV: phase III completed

Phase II Study of Ruxolitinib in Patients With Advanced PV Eligibility criteria: Refractory or intolerant to HU or HU contraindicated Hct >45% or phlebotomy 2 times in last 6 months, with at least one phlebotomy in last 3 months 10 mg BID (n = 7) 25 mg BID (n = 8) 50 mg QD (n = 7) 10 mg BID (n = 12) Part 1 Part 2 Patients received ruxolitinib for a median of 152 weeks (range, 31-177 weeks) Verstovsek S, et al. Cancer. 2014;120(4):513-520.

Phase II Study of Ruxolitinib in Patients With Advanced PV: Main Results Hematocrit <45% without phlebotomy was achieved in 97% of patients by week 24 Among patients with palpable splenomegaly at baseline, 44% and 63%, respectively, achieved nonpalpable spleen at weeks 24 and 144 Verstovsek S, et al. Cancer. 2014;120:513-520.

Pharmacodynamic and Molecular Studies in Ruxolitinib-Treated PV Patients: LAP HV_Max HV_Min LAP= Leukocyte Alkaline Phosphatase Verstovsek S, et al. Cancer. 2014;120:513-520. Passamonti F et al, Blood 2006; 107:3676-82. Falanga A et al, Blood 2000; 96:4

Pharmacodynamic and Molecular Studies in Ruxolitinib-Treated PV Patients: LAP Elevated levels of LAP are correlated with JAK2V617F allele burden. Copyrighted material HV_Max HV_Min LAP= Leukocyte Alkaline Phosphatase In vivo leukocyte activation was associated with laboratory signs of endothelium and coagulation system Verstovsek S, et al. Cancer. 2014;120:513-520. Passamonti F et al, Blood 2006; 107:3676-82. Falanga A et al, Blood 2000; 96:4261-6.

Pharmacodynamic and Molecular Studies in Ruxolitinib-Treated PV Patients: CRP CRP, C-reactive protein Verstovsek S, et al. Cancer. 2014;120(4):513-520. Barbui T et al, Haematologica

Pharmacodynamic and Molecular Studies in Ruxolitinib-Treated PV Patients: CRP Unadjusted and sequentially multivariable adjusted risk of thrombosis associated to different hs-crp CRP, C-reactive protein Verstovsek S, et al. Cancer. 2014;120(4):513-520. Barbui T et al, Haematologica 2011; 96:315-8. Obtained form http://www.haematologica.org

Percent Change From Baseline in JAK2 V617F Allele Burden (mean ± SEM) Pharmacodynamic and Molecular Studies in Ruxolitinib-Treated PV Patients: JAK2V617F 25 0 n=30 n=28 n=23 25 8 14 22 50 48 96 Weeks 144 With longer follow-up (5 yr) some patients may become JAK2 V617F negative However, TET2 mutant ancestral clone remain untouched (Pieri L et al, ASH 2014, Poster # 3185) Verstovsek S, et al. Cancer. 2014;120(4):513-520.

Randomized (1:1) RESPONSE Study Design Resistance to or intolerance of hydroxyurea (modified ELN criteria) Phlebotomy requirement Splenomegaly Pre-randomization (day -28 to day -1) Hct 40%-45% Ruxolitinib 10 mg BID n = 110 BAT n = 112 Week 32 (primary endpoint) Crossover to ruxolitinib Week 48 a Extended Treatment phase Week 80 Week 208 Week 208 Ruxolitinib-randomized patients were individually titrated for efficacy and safety (to a maximum of 25 mg BID) Investigator-selected best available therapy (BAT) as monotherapy (hydroxyurea, IFN/peg-IFN, anagrelide, pipobroman, IMIDs, or observation); BAT could be changed in case of lack of response or BATrelated toxicity requiring drug discontinuation a The primary analysis occurred after all patients completed week 48. Vannucchi A, et al. Haematologica. 2014;99(Suppl): Abstract LB2436.

Patients, % Primary Response Primary endpoint (composite): Percentage of patients who achieved both Hct control (Hct <45% and no phlebotomy) and spleen response (reduction of SV to <35% from baseline assessed by MRI) at week 32. Primary endpoint Individual components of primary endpoint 70 60 50 40 30 20 10 0 P<.0001 OR, 28.64 (95% CI, 4.50-1206) 21% 38% 1% 1% 1% 60% 20% Primary Composite Endpoint >35% Reduction in in SV SV Hct Control Ruxolitinib BAT 77% of patients randomized to ruxolitinib met at least 1 component of the primary endpoint SV, spleen volume Vannucchi A, et al. Haematologica. 2014;99(Suppl): Abstract LB2436.

Probability Duration of Primary Response Only 1 patient lost primary response 37.1 weeks after start of that response 100 80 60 Ruxolitinib Censoring times 40 20 0 0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102 108 114 120 Time From Initial Response, weeks At risk 23 23 23 22 22 21 18 15 14 14 14 10 10 10 6 6 4 1 1 0 0 Events 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 The probability of maintaining primary response for 1 year was 94% Vannucchi A, et al. Haematologica. 2014;99(Suppl): Abstract LB2436.

Patients, % Complete Hematologic Remission CHR is defined as Hct control, PLT count 400 10 9 /L, and WBC count 10 10 9 /L 50 40 30 20 10 23.6% P =.0028 a OR, 3.35 (95% CI, 1.43-8.35) 8.9% Ruxolitinib (n = 110) BAT (n = 112) 0 Complete Hematologic Remission at Week 32 88.5% of patients who achieved CHR had a durable response at week 48 a P value, odds ratio and 95% CI were calculated using stratified exact Cochran-Mantel-Haenszel test by adjusting for the WBC/PLT status (abnormal vs normal) at baseline. WBC/PLT status was defined as abnormal if WBC count was >15 10 9 /L, and/or PLT count >600 10 9 /L. Vannucchi A, et al. Haematologica. 2014;99(Suppl): Abstract LB2436.

Patients, % Improvement in Symptoms Percentage of patients with a 50% improvement in MPN-SAF symptom score at week 32 a 80 60 40 49 64 37 62 Ruxolitinib BAT 20 0 a In patients with scores at both baseline and week 32 MPN-SAF, Myeloproliferative Neoplasm Symptom Assessment Form 5 11 13 n = 74 81 74 80 71 80 63 71 MPN-SAF Total symptom score Cytokine symptom cluster Tiredness Itching Muscle ache Night sweats Sweating while awake Hyperviscosity symptom cluster Headache Dizziness Skin redness Vision problems Ringing in ears Concentration problems Numbness/tingling in hands/feet Vannucchi A, et al. Haematologica. 2014;99(Suppl): Abstract LB2436. 17 Splenomegaly symptom cluster Fullness/early satiety Abdominal discomfort

Patients, n (%) Other Adverse Events Ruxolitinib (n = 110) BAT (n = 111) Infections All infections 46 (41.8) 41 (36.9) Grade 3 or 4 4 (3.6) 3 (2.7) Herpes zoster Grade 3 or 4 7 (6.4) 0 0 0 Non-melanoma skin cancers a All NMSC 4 (3.6) 2 (1.8) Grade 3 or 4 3 (1.8) 1 (0.9) Progression to MF and AML b MF 2 (1.8) 1 (0.9) AML 1 (0.9) 0 a A higher rate of NMSC was reported in the ruxolitinib arm (4.7 vs 2.7 pts / 100-PY); No patients discontinued treatment for NMSC. A higher proportion of patients in the ruxolitinib arm had a prior history of NMSC or pre-cancerous skin condition compared to the BAT arm (10.9% vs 6.3%) All 4 ruxolitinib patients and 1 of the 2 BAT patients with NMSC had a history of skin cancer or precancer Patients randomized to ruxolitinib had a longer prior exposure to HU (162.9 vs 145.6 weeks). HU exposure was higher in patients who developed secondary malignancies, including NMSCs (461.3 vs 206.9 weeks) b After week 32, there was 1 transformation to MF in the ruxolitinib arm and 2 in the BAT arm (after crossover); there was 1 transformation to AML in the BAT arm (after crossover).

RESPONSE: Additional Findings The rate of thromboembolic events was lower in the ruxolitinib group (1 thrombosis compared with 6 thromboses in the BAT arm) Tolerability: 85% of patients in the ruxolitinib arm were still on treatment at a median follow-up of 81 weeks Most adverse events were grade 1/2, and few patients developed grade 3/4 cytopenias The safety profile of ruxolitinib in this study is generally consistent with that observed in the phase III COMFORT studies 1,2 of ruxolitinib for the treatment of MF Lower rates of cytopenias were observed in this PV population 1. Verstovsek S, et al. N Engl J Med. 2012;366(9):799-807. 2. Harrison CH, et al. N Engl J Med. 2012;366(9):787-798. 1. Verstovsek S, et al. N Engl J Med. 2012;366(9):799-807. 2. 2. Harrison CH, et al. N Engl J Med. 2012;366(9):787-798.

Phase II Study of Ruxolitinib in Patients with Advanced ET Eligibility criteria: ET refractory or intolerant to hydroxyurea or for whom treatment with hydroxyurea was contraindicated, as assessed by the investigator Platelets > 650 x 10 9 /L unless on therapy 10 mg BID (n = 7) 25 mg BID (n = 8) 50 mg QD (n = 7) 25 mg BID (n = 15) Part 1 Part 2 At 192 weeks of follow-up, 57% of patients with platelet count >600x10 9 /L had reduction to <600x10 9 /L Most patients obtained control of leukocytosis, resolution of splenomegaly, and improvement of symptoms Verstovsek S et al, Long-Term Results From a Phase I Open-Label Study of Ruxolitinib in Patients With Essential Thrombocythemia Refractory to or Intolerant of Hydroxyurea. ASH 2014, abstr 1847.

Conclusions Inhibition of the JAK/STAT pathway in patients with PV and ET may help control the disease manifestations and, possibly, reduce complications

Conclusions Inhibition of the JAK/STAT pathway in patients with PV and ET may help control the disease manifestations and, possibly, reduce complications But will not cure the patient

Conclusions Inhibition of the JAK/STAT pathway in patients with PV and ET may help control the disease manifestations and, possibly, reduce complications But will not cure the patient The unwanted inhibition of wild-type JAK2 causing anemia and thrombocytopenia in MF is of minor concern in PV and ET But other toxicities require careful monitoring There are additional non-canonical and alternative strategies for inhibiting JAK2 mutated cells

Final take-home message JAK2 inhibitors may represent a new therapeutic option for selected categories of patients with PV who are not satisfactorily managed with conventional therapies, while there is still too little experience in ET.