JAK inibitori nella policitemia vera Francesco Passamonti Università dell’Insubria Varese - Italy

Prediction of prognosis in PV after diagnosis Bad factors: Hematocrit values over 45% Inadequately controlled PV

365 JAK2-pos. PV pts treated with phlebotomy  HU Which is the most safe hematocrit level in PV? the prospective CYTO-PV trial Primary composite end point: Death from CV cause or major thrombotic events (stroke, acute coronary syndrome, transient ischemic attack, PE, abdominal thrombosis, DVT, or peripheral arterial thrombosis) Secondary end point: Primary end point + superficial-vein thrombosis Treated to Hct <45% n = 182 365 JAK2-pos. PV pts treated with phlebotomy  HU R 1:1 Treated to Hct 45-50% n = 183 Post hoc analysis of two large randomized clinical trials (PVSG-01 and ECLAP) failed to show a different incidence of major thrombosis when HCT levels were kept in the range between 40% and 50%. Thus, there was a need for a randomized clinical trial to provide evidence-based data that would assess the usefulness of tight HCT control in reducing thrombosis. Studio multicentrico italiano: 365 paz. di 26 centri – arruolamento: maggio 2008-febbraio 2012 We randomly assigned 365 adults with JAK2-positive polycythemia vera who were being treated with phlebotomy, hydroxyurea, or both to receive either more intensive treatment (target hematocrit, <45%) (low-hematocrit group) or less intensive treatment (target hematocrit, 45 to 50%) (high-hematocrit group). The primary composite end point was the time until death from cardiovascular causes or major thrombotic events. The secondary end points were cardiovascular events, cardiovascular hospitalizations, incidence of cancer, progression to myelofibrosis, myelodysplasia or leukemic transformation, and hemorrhage. An intention-to-treat analysis was performed. After a median follow-up of 31 months, the primary end point was recorded in 5 of 182 patients in the low-hematocrit group (2.7%) and 18 of 183 patients in the high-hematocrit group (9.8%) (hazard ratio in the high-hematocrit group, 3.91; 95% confidence interval [CI], 1.45 to 10.53; P=0.007). The primary end point plus superficial-vein thrombosis occurred in 4.4% of patients in the low-hematocrit group, as compared with 10.9% in the high-hematocrit group (hazard ratio, 2.69; 95% CI, 1.19 to 6.12; P=0.02). Progression to myelofibrosis, myelodysplasia or leukemic transformation, and bleeding were observed in 6, 2, and 2 patients, respectively, in the low-hematocrit group, as compared with 2, 1, and 5 patients, respectively, in the high-hematocrit group. There was no significant between-group difference in the rate of adverse events. 67% HR 25% history of thrombosis 50% diagnosed within 2 y ASA in all without contraindications Median f-up 31 months (range 1.5-48) Marchioli R et al, N Engl J Med 2013;368(1):22-33

Primary endpoint achieved, n (%): Secondary endpoint achieved, n (%): To maintain the hematocrit below 45% is associated with less vascular events Treated to Hct <45% n = 182 Treated to Hct 45-50% n = 183 R 1:1 Primary endpoint achieved, n (%): 5* (2.7) 18* (9.8) Secondary endpoint achieved, n (%): 8 (4.4) 20 (10.9) After a median follow-up of 31 months, the primary end point was recorded in 5 of 182 patients in the low-hematocrit group (2.7%) and 18 of 183 patients in the high-hematocrit group (9.8%) (hazard ratio in the high-hematocrit group, 3.91; 95% confidence interval [CI], 1.45 to 10.53; P=0.007). The primary end point plus superficial-vein thrombosis occurred in 4.4% of patients in the low-hematocrit group, as compared with 10.9% in the high-hematocrit group (hazard ratio, 2.69; 95% CI, 1.19 to 6.12; P=0.02). Progression to myelofibrosis, myelodysplasia or leukemic transformation, and bleeding were observed in 6, 2, and 2 patients, respectively, in the low-hematocrit group, as compared with 2, 1, and 5 patients, respectively, in the high-hematocrit group. There was no significant between-group difference in the rate of adverse events. About 75% were maintained per protocol Hct level. Higher frequency and dose of HU in high Hct group. Significantly higher WBC in high Hct group. No difference in PLT count. No difference in total number of deaths, fibrotic/leukemic transformation, bleeding, second cancers. Significant differences might have been limited to: Females Older age group HR pts not on systemic anticoagulation Probability of remaining event-free * 0 deaths in the Hct <45% arm, 4 deaths in the Hct 45-50% arm Marchioli R et al, N Engl J Med 2013;368(1):22-33

Inadequately controlled PV Criteria for clinical trials HU-non hematological toxicity (any dose) Leg ulcers, aftosis Gastrointestinal symptoms Fever HU- hematological toxicity (lowest effective dose) Neutrophil < 1x 109/l Platelet < 100x 109/l Hemoglobin <10 g/dl HU-resistance (maximum tolerated dose) Phlebotomies requirment Platelet > 400x 109/l Uncontrolled leukocytosis HU-resistance (maximum tolerated dose) Unresponsive /Progressive splenomegaly Progressive systemic symptoms Marchioli R, et al. N Engl J Med. 2013;368:22-33. Prchal JT, et al. Polycythemia vera. Williams Hematology. 8th edition. Najean Y, et al. Blood. 1997;90(9):3370-3377. Mesa RA. Blood. 2009;113:5697-5698. Alvarez-Larrán A, et al. Blood. 2012;119(6):1363-1369. Barosi G et al, Br j Hematol. 2010;148(6)961-63.

Recorded in 137 patients (15.4%): Inadequately controlled or HU resistant & intolerant PV The size of the problem in 890 patients Recorded in 137 patients (15.4%): Need for phlebotomies (3.3%) Uncontrolled myeloproliferation (1.6%) Failure to reduce massive splenomegaly (0.8%) Cytopenia at the lowest HU-dose to achieve response (1.7%) Extra-haematological toxicity (9%) Cytopenia affected survival, progression to MF, AML Splenomegaly affected MF Alvarez-Larran et al; Br J Haematol. 2016 Mar;172(5):786-93

Implications of need of phlebotomies in PV Spanish retrospective study: 533 PV patients under HU; 16% of them required 3 or more phlebotomy per year Implications respect to lesser need: Higher doses of HU; Higher rate of thrombosis (20.5% vs 5.3%) ECLAP prospective study: 793 PV treated with HU No differences in thrombotic events per n. of phlebotomy Alvarez Larran et al; Haematologica 2016; Barbui et al; Haematologica 2017

Thrombosis/patients (%) Implication of leukocytosis permanence during PV treatment: an higher risk of thrombosis data from the prospective CytoPV study WBC (x109/L) Thrombosis/patients (%) Hazard ratio (95%CI), p-value <7.0 4/100 (4.0) 1.00 7.0-8.4 4/84 (4.8) 1.58 (0.39-6.43), 0.52 8.5-11.0 8/88 (9.1) 2.69 (0.80-9.05), 0.11 ≥11.0 12/93 (12.9) 3.90 (1.24-12.3), 0.023 Barbui et al; Blood. 2015 Jul 23;126(4):560-1.

ESMO Guidelines for PV Vannucchi et al, Ann Oncol. 2015 Sep;26 Suppl 5:v85-99

Second line treatment in PV: interferons French PVN1 Study N=37, newly diagnosed CHR in 95% (82% CR) No thrombosis in 6 yrs CMR in 8 (28%) Sustained improvements after discontinuation 89% with AE’s (grade 1/2) Discontinuation rate: 8% at 1yr 1, 24% overall Ropeginterferon N=51 Median 17 ms from dx, 33% treated with HU (not HU resistant/intolerant) 47% CR, 43% PR CMR in 21% 1 TIA, 1 DVT during study period 88% with AE’s 20% discontinued (pruritus, fatigue, arthralgia, HA, diarrhea, flu-like sx most common) MDACC N=43; ~50% treat naive; Median follow-up: 83 months ORR: 79% CHR 76% Median time to CR: 40 days Failure to achieve CMR: more likely to have/acquire non-driver mutations G3/4 AEs: 10-17% of pts/yr, over long term (fatigue, cytopenias, depression) Thrombosis and progression can occur Discontinuation rate: 22% Kiladjian et al; Blood 2008 Turlure, et al ASH 2011 Abs 280 Clinical, hematological, morphological, and molecular responses Turlure: 13/20 dc’s received no further therapy, and 10/13 still in hematological CR off therapy after 28 months—29% could stop w/o requiring further therapy Gissingler et al; Blood 2015 Quintas-Cardama et al, Blood 2013 Masarova et al Lancet Haem 2017 Front-line studies (Proud PV, ASH 2016 Abs 479) and (MPD-RC 111, ASH 2016 Abs 475) showed non-inferiority vs. HU

Ruxolitinib in PV: Phase 3 trials RESPONSE and RESPONSE 2 Ruxolitinib, 10 mg bid HU resistance or intolerance (ELN criteria) q3mo phlebotomy requirement Palpable spleen with MRI-confirmed vol. of ≥ 450 cm3 Platelet > 100K N = 110 1o Endpoint Failure Disease Progression Crossover HCT 40–45% inclusive Randomised Best Available Therapy N = 112 Week 32 Week 80 Week 28 in Response-2 NO Splenomegaly in Response-2 Primary composite endpoint: haematocrit control (phlebotomy independence from week 8 to 32, with ≤ 1 phlebotomy post randomization) in the absence of phlebotomy and 35% reduction in spleen volume at week 32 (this latter absent in Response 2) Secondary endpoints: complete haematological remission at week 32 (absence of phlebotomy requirement, PLT count ≤ 400 x 109/L, and WBC count ≤ 10 × 109/L); % of patients who maintain primary endpoint response for ≥ 48 weeks; Symptom improvement (MPN-SAF diary) and quality of life (EORTC QLQ-C30; PGIC). Vannucchi et al, N Engl J Med. 2015 Jan 29;372(5):426-35; Passamonti et al, Lancet Oncol. 2016 Dec 1. pii: S1470-2045(16)30558-7.

Individual Components of Primary Endpoint RESPONSE study: haematocrit control and 35% reduction in spleen volume at Week 32 Primary Endpoint Individual Components of Primary Endpoint P < .0001 OR, 28.64 (95% CI, 4.50-1206) 60 38 21 20 1 1 Vannucchi et al, N Engl J Med. 2015 Jan 29;372(5):426-35

RESPONSE study: durability of primary endpoint Graph is adopted from RUX15143.3011 Graphics/New Figures 2a 20/25 (80%) ruxolitinib-treated patients had a durable primary response defined as maintenance for 48 weeks after initial response 3 of the 5 without durable response were classified as nonresponders because of missing assessments The probability of maintaining the primary response in the ruxolitinib arm for at least 80 weeks from time of response was 92% Verstovsek et al. Haematologica 2016

RESPONSE-2 study: haematocrit control at w 28 OR, 7.28 (95% CI, 3.43-15.45) In patients who received interferon at any time in the BAT arm (n = 13; at randomization or after a regimen change), 4 (30.8%) achieved Hct control at week 28 Significantly more patients randomized to ruxolitinib achieved Hct control without phlebotomy (primary endpoint) compared with those randomized to BAT Passamonti et al, Lancet Oncol. 2016 Dec 1. pii: S1470-2045(16)30558-7. OR, odds ratio.

Response-2: proportion of patients receiving phlebotomies up to week 28 ≤ 2 > 4 No. of Phlebotomies More than 80% of patients in the ruxolitinib arm did not have a phlebotomy, compared with 40% in the BAT arm The total number of phlebotomies was much higher in the BAT arm than in the ruxolitinib arm (98 vs 19) Passamonti et al, Lancet Oncol. 2016 Dec 1. pii: S1470-2045(16)30558-7. 9

Time to Subsequent Phlebotomy Eligibility BAT vs Ruxolitinib Nonresponders Verstovsek S, et al. ASH 2014. Abstract 3201.

Normalization of iron status/MCV over time with ruxolitinib Verstovsek S, et al. Leuk Res. 2017;56:52–59.

RESPONSE-2 study: WBC count over time Rux Median WBC Count, × 109/L Week Ruxolitinib, n = 74 68 65 66 69 69 67 68 BAT, n = 75 69 71 70 69 69 61 40 WBC counts in the ruxolitinib arm were ≤ 10 × 109/L from week 8 onward, whereas they remained > 10 × 109/L in the BAT arm Passamonti et al, Lancet Oncol. 2016 Dec 1. pii: S1470-2045(16)30558-7.

Thromboembolic complications with ruxolitinib in the Response studies Response: at the Week 80 analysis, the rates of thromboembolic events per 100 patient- years of exposure were 1.8 in the ruxolitinib arm vs. 8.2 in the BAT arm Response-2: there was 1 thromboembolic event in the ruxolitinib arm and 3 in the BAT arm “Thrombosis/embolism (vascular access-related)” is defined as (in CTCAE version 3.0): Grade 1: NA Grade 2: Deep vein thrombosis or cardiac thrombosis; intervention (e.g., anticoagulation, lysis, filter, invasive procedure) not indicated Grade 3: Deep vein thrombosis or cardiac thrombosis; intervention (e.g., anticoagulation, lysis, filter, invasive procedure) indicated Grade 4: Embolic event including pulmonary embolism or life-threatening thrombus Grade 5: Death When we search the online version 3.0, there are no hits for “thrombophlebitis”. However, version 4.03 (http://evs.nci.nih.gov/ftp1/CTCAE/CTCAE_4.03_2010-06-14_QuickReference_5x7.pdf) lists “superficial thrombophlebitis” as “A disorder characterized by a blood clot and inflammation involving a superficial vein of the extremities” (there is no mention of non-superficial thrombophlebitis) Vannucchi et al, N Engl J Med. 2015 Jan 29;372(5):426-35; Passamonti et al, Lancet Oncol. 2016 Dec 1. pii: S1470-2045(16)30558-7.

RESPONSE and RESPONSE -2 studies: improvement of symptomatology

Response-2: Safety Hematologic Patients, % Ruxolitinib (n = 74) BAT (n = 75) Grade 1/2 Grade 3 Grade 4 Anemia 55.4 42.7 Thrombocytopenia 8.1 22.7 1.3 Leukopenia 16.0 Other Adverse Events of Interest Up to Week 28 Patients, n (%) Ruxolitinib (n = 74) BAT (n = 75) Infections All infections 23 (31.1) 18 (24.0) Grade 3 or 4 3 (4.1)a 1 (1.3)b Herpes zoster 1 (1.4) Nonmelanoma skin cancers All NMSCs 1 (1.4)c 1 (1.3)d There was 1 AE after treatment discontinuation of ruxolitinib, grade 2 middle ear infection Passamonti et al, Lancet Oncol. 2016 Dec 1. pii: S1470-2045(16)30558-7.

Active trials/Novel agents in PV Study Study Population Preliminary Findings Rationale MPD-RC 111: Peg IFN Salvage High risk w/ HU intolerance/resistance (or prior AVT) N/A To investigate peg-IFN as salvage therapy MPD-RC 112: Peg-IFN vs HU Front-line, high risk population Non-inferiority in response rates, interim analysis To investigate peg-IFN as 1st line therapy PROUD-PV Ropeginterferon vs HU 257 Rx-naïve or HU-treated PV w/o intolerance or CHR Non-inferiority in CHR between 2 groups at 12 m To test efficacy of a novel agent, dosed bi-weekly COMBI-trial Peg-IFN with Ruxolitinib (Denmark) 30 pts with PV (N=20), post-PVMF, or primary MF   -15/20 PV pts had CHR   -Statistically significant ↓ in JAK2% Anemia (n=15), neutropenia (n=13), and TCP (n=6) Provide control of inflammatory symptoms with JAK inhibition along with anti-clonal activity with IFN RG7388 (MDM2 inhibitor) in Pts w/ PV and ET High risk, or low risk w/ sx Prior resistance/intolerance Includes pilot combo with peg-IFN if non-response to single agent RG7388 Aim is to activate the p53 pathway and induce apoptosis. Givinostat Phase 1B/2 Dose escalation Prelim efficacy PV in need of treatment To test concept of HDAC inhibition in PV