RHAPSODY-2: ZZ-3K3A-301: A Multicenter, Randomized, Placebo-Controlled, Double-Blinded, Phase 3 Study To Evaluate The Efficacy And Safety Of 3K3A-APC, A Recombinant Variant Of Human Activated Protein C, In Combination With Tissue Plasminogen Activator, Me

NIH StrokeNet Concept Synopsis Date: 14 November 2019 Title: ZZ-3K3A-301: A multicenter, randomized, placebo-controlled, double-blinded, Phase 3 study to evaluate the efficacy and safety of 3K3A-APC, a recombinant variant of human activated protein C, in combination with tissue plasminogen activator, mechanical thrombectomy, or both in subjects with moderate to severe acute ischemic stroke Principal Investigator: Institution: Patrick D. Lyden, MD, FAAN, FAHA Department of Neurology Cedars-Sinai Medical Center 8700 Beverly Blvd, AHSP 8318 Los Angeles, CA 90048 Ph: + 1-310-248-6652 Email: [email protected] Project Description: Aspect of cerebrovascular disease targeted: (Check all that apply) Primary or secondary prevention Emergent management or acute treatment Recovery and rehabilitation Biomarker-validation study Ancillary study to ongoing NIH StrokeNet trial The primary goal of the NIH StrokeNet network is to maximize efficiencies to develop, promote and conduct a balanced portfolio of high-quality, multi-site exploratory Phase 1, 2, and confirmatory Phase 3 clinical trials in stroke prevention, treatment, and recovery. Such trials will be focused on key interventions, as well as on biomarker-validation studies that are immediately preparatory to trials and ancillary studies to existing NIH StrokeNet trials. In one paragraph, please state the question that you wish to explore in this study: Study ZZ-3K3A-301 (RHAPSODY-2) is a multicenter, randomized, placebo-controlled, double-blind, Phase 3 study to evaluate efficacy and safety of 3K3A-APC (a recombinant variant of human activated protein C [APC] in which 3 lysine residues [191-193] are replaced by 3 alanine residues) following administration of tissue plasminogen activator (tPA), mechanical thrombectomy, or both in subjects with moderate to severe acute ischemic stroke. A maximum of 400 subjects will be randomized during a lead-in phase to 3K3A-APC or placebo. During this lead-in phase, a Bayesian adaptive approach will evaluate 10, 15, or 30 mg or placebo. The study will transition to a selected 3K3A-APC dose when a single dose proves superior to all other doses. The study will stop if no dose proves superior to placebo. If all doses prove superior to placebo, but none are superior to each other by the end of the lead-in phase, the lowest dose will be evaluated in the definitive efficacy phase of the study. Intervention (drug/biologic/device/behavioral): Subjects will be randomized to 3K3A-APC (10, 15, or 30 mg during lead-in dose-finding phase to 3K3A-APC (10, 15, or 30 mg) or placebo. No later than 60 minutes following completion of tPA infusion or initiation of mechanical thrombectomy (arterial puncture), whichever is sooner, adult subjects will receive study treatment given as a 15-minute infusion. Subjects will receive another 15-minute infusion of 3K3A-APC every 12 hours (± 1 hour) for up to 5 total doses. StrokeNet Concept Synopsis Version 4.0 – 4/1/2019 Page 1 of 22 Primary Aim: To evaluate the effect of 3K3A-APC on 3-month disability (utility-weighted modified Rankin score [mRS] analysis) and on bleed-free survival at Day 30. Primary Outcomes: Utility-weighted analysis of Day 90 mRS and proportion of subjects alive and without intracerebral hemorrhage at 30 days after ischemic stroke. Secondary Aims/Outcomes: ? To evaluate the effect of 3K3A-APC on bleeding (yes/no) at Day 30 in those subjects with magnetic resonance imaging (MRI) scans. Outcome: Day 30 bleeding (yes/no) or death. ? To evaluate the effect of 3K3A-APC on death (yes/no) at the time of the Day 30 visit in all subjects. Outcome: Day 30 death (yes/no). ? To evaluate the effect of 3K3A-APC on the Day 30 volume of bleeding in the brain as determined by MRI. Outcome: Day 30 mean volumes of bleeding. ? To evaluate the effect of 3K3A-APC on the Day 7/discharge National Institutes of Health Stroke Scale (NIHSS) score. Outcome: Proportion of subjects achieving a 7-point or better improvement (or a score of zero if baseline was < 7) in Day 7/discharge NIHSS score. ? To evaluate the safety of 3K3A-APC. Outcome: Adverse events (AEs). Exploratory Aims/Outcomes: ? To further evaluate the effect of 3K3A-APC on the Day 7/discharge NIHSS score. Outcome: Day 7/discharge NIHSS score. ? To evaluate the effect of 3K3A-APC on the Day 90 dichotomized mRS. Outcome: Day 90 dichotomized mRS. ? To evaluate the effect of 3K3A-APC on the Day 90 EQ-5D quality of life scale. Outcome: Day 90 EuroQoL 5 Dimensions (EQ-5D) scores. ? To evaluate the effect of 3K3A-APC on the Day 90 Barthel index (BI) score: Outcome: Day 90 BI scores. ? To evaluate the effect of 3K3A-APC on the Day 90 infarct volume as determined by MRI (or computed tomography [CT] if unable to obtain MRI). Outcome: Day 90 infarct volumes. Briefly describe the scientific rationale/premise for the study: APC is an endogenous serine protease with systemic anticoagulant activity, as well as cell-signaling actions that convey endothelial stabilizing, anti-inflammatory, and anti-apoptotic activities and that promote neurogenesis (Mosnier et al 2007a; Griffin et al 2015, Griffin et al 2016a; Griffin et al 2016b; Griffin et al 2018). 3K3A-APC is a mutant variant of recombinant wild-type APC engineered to have greatly reduced anticoagulant activity compared with wild-type APC, but fully preserved anti-apoptotic, cytoprotective and anti-inflammatory activities (Mosnier et al 2007; Guo et al 2009a and 2009b; Griffin et al 2015; Griffin et al 2018). These cell-signaling effects result in vasculoprotection and neuroprotection after ischemic stroke (Wang et al 2012; Wang et al 2013), and a decrease in intracranial hemorrhage in stroke patients receiving tPA, thrombectomy, or both (Lyden et al 2019). Wild-type APC is a crucial anti-coagulant endogenous protease that promotes anti-coagulation by inactivating coagulation factors Va and VIIIa, limiting thrombin formation (Griffin et al 2015; Griffin et al 2018). In contrast to wild-type APC, the mutations in 3K3A-APC alter the factor Va binding exosite, which decreases 3K3A-APC anticoagulant activity to < 10% of wild-type APC. However, the mutations leave the N-terminal Gla domain of 3K3A-APC intact and do not affect exosites that recognize APC cell receptors such as endothelial protein C receptor (EPCR) and protease activated receptor-1 (PAR1), which are critical for 3K3A-APC cell-signaling and cytoprotective activities (Griffin et al 2015; Griffin et al 2018). APC is normally generated in vivo from the zymogen protein C through activation by thrombin on the surface of endothelial cells. The activation requires 2 membrane receptors, thrombomodulin and endothelial protein C receptor. The anticoagulant activity of APC is independent of its direct cellular effects. Its anticoagulant activity is mediated by irreversible proteolytic degradation of factors Va and VIIIa with contributions by various cofactors, whereas its cytoprotective, cell-signaling activities StrokeNet Concept Synopsis Version 4.0 – 4/1/2019 Page 2 of 22 normally require proteolytic activation of PAR-1 (Mosnier et al 2007b; Mosnier et al 2012; Griffin et al 2015; Griffin et al 2016a; Griffin et al 2016b; Griffin et al 2018). The multiple properties (anticoagulant and cytoprotective) of APC may be useful in reversing the effects of an ischemic stroke and in protecting ischemic brain tissue from further damage. The cellular signaling by APC results in cytoprotective alterations in gene expression profiles resulting in multiple cytoprotective actions due to anti-inflammatory activity and anti-apoptotic activity, as well as reducing endothelial barrier disruption (Cheng et al 2003; Dömötör et al 2003; Joyce et al 2001; Mosnier and Griffin 2003; Riewald et al 2002). The extensive preclinical studies of the vasculoprotective and neuroprotective actions of APC and 3K3A-APC have been summarized in several reviews (Amar et al 2018; Amar et al 2015; Griffin et al 2015; Griffin et al 2016a; Griffin et al 2016b; Griffin et al 2018; Zlokovic and Griffin 2011). Studies in rodent models of stroke have shown that 3K3A-APC extends the therapeutic window of tPA and reduces post-ischemic bleeding risk and intracerebral hemorrhage associated with tPA (Wang et al 2009; Wang et al 2012; Wang et al 2013). For 3K3A-APC direct vasculoprotective effects resulting in increased cerebrovascular integrity and neuroprotection in rodent preclinical ischemic stroke models, not only is PAR-1 required, but also cleavage at Arg46 in PAR-1 is specifically required, strongly supporting the concept that APC vasculoprotection and neuroprotection requires “biased” signaling initiated by the G-protein coupled receptor PAR-1 (Mosnier et al 2012; Sinha et al 2018). Finally, it is noteworthy that studies using human fetal neural stem and progenitor cells show that 3K3A-APC promotes neurogenesis in vitro as well as in vivo in the murine middle cerebral artery occlusion (MCAO) stroke model (Guo et al 2013; Wang et al 2016). In summary, an array of data from in vitro and in vivo studies provides a strong rationale for clinical trials of 3K3A-APC for ischemic stroke. The most frequently used Food and Drug Administration-approved treatment for stroke is thrombolysis with recombinant human tPA (Group NINDS 1995). Certain mechanical devices have received Food and Drug Administration approval for clot retrieval, and mechanical thrombectomy has been shown to be effective in certain patients as a stroke therapy. All patients who meet tPA or mechanical thrombectomy treatment criteria and who arrive to the hospital acutely will receive tPA (or mechanical thrombectomy or both (per local suitability criteria), as standard of care. Neuroprotection appears to work best if delivered within 6 hours and in the setting of recanalization/reperfusion (Sutherland et al 2012). Thus, the best way to study putative neuroprotectants is to combine them with recanalization treatments, which, if successful, will result in reperfusion of ischemic tissue (Amar et al 2015; Amar et al 2018). Thus, ischemic stroke patients who are eligible to receive tPA or mechanical thrombectomy or both are the most likely patient population to benefit from a neuroprotectant like 3K3A-APC, due to the need for reperfusion. In vitro studies show that there is minimal effect on 3K3A-APC coagulation and no significant effect on tPA clot lysis when both drugs are present at clinically relevant doses. Studies of wild-type APC (which has greater anti-coagulant potential than 3K3A-APC) in combination with tPA provide safety reassurance. Despite having anticoagulant properties, wild-type APC has been shown to reduce tPA-induced bleeding and neurotoxicity in preclinical models, probably due to the anti-inflammatory and anti-apoptotic activities and the ability to stabilize endothelial cell barriers previously discussed (Cheng et al 2003; Liu et al 2004; Cheng et al 2006; Shibata et al 2001; Zlokovic and Griffin 2011; Griffin et al 2015). 3K3A-APC has also shown to reduce hemorrhage related to tPA treatment in animal stroke models (Wang et al 2013). 3K3A-APC has 10% of anticoagulant activity of wild-type APC and provides greater vasculoprotection and neuroprotection than recombinant wild-type APC in mouse models of stroke; when 3K3A-APC is combined with tPA, tPA-associated intracerebral microhemorrhages are almost eliminated and infarct volumes significantly reduced (Guo et al 2009a and 2009b; Wang et al 2009; Wang et al 2012; Wang et al 2013). At the same time, behavioral outcomes in both mouse and rat models of stroke are improved. 3K3A-APC also extends the therapeutic window of tPA in ischemic stroke models in rodents, supporting further development of tPA and 3K3A-APC combination therapy (Amar et al 2015; Amar et al 2018). Of note, the transient suture model of stroke that has been used in several of these StrokeNet Concept Synopsis Version 4.0 – 4/1/2019 Page 3 of 22 studies mimics well the clinical procedure of thrombectomy as recently reviewed (Amar et al 2015; Amar et al 2018). This implies the combination of thrombectomy and 3K3A-APC for focal ischemic stroke in humans may be efficacious. Describe the potential clinical, scientific and public health impact of this study: A stroke occurs when the blood supply to part of the brain is interrupted (ischemic stroke) or when a blood vessel in the brain bursts (hemorrhagic stroke), allowing blood into the spaces surrounding brain cells. Each year about 795,000 people in the United States (US) experience a new or recurrent stroke (Lloyd-Jones et al 2010). Of all strokes, 87% are ischemic and 10% are intracerebral hemorrhage strokes, whereas 3% are subarachnoid hemorrhage strokes (Benjamin et al 2017). Ischemic stroke is a leading cause of death and the most common cause of disability in industrialized nations (Thom et al 2006). Currently, the most commonly used approved treatment for acute stroke is thrombolytic therapy with tPA. tPA is approved for intravenous (IV) administration within 3 hours of onset of acute ischemic stroke in the US and for up to 4.5 hours following the stroke in Europe (Hacke et al 2008; Group NINDS 1995). Thrombolytic therapy with IV tPA up to 4.5 hours following stroke is recommended by the American Heart Association/American Stroke Association (Powers et al 2018). The primary adverse effect of tPA in clinical use is symptomatic intracranial hemorrhage (~6%); other risks include systemic bleeding, myocardial rupture (when used to treat acute myocardial infarction), and, in rare cases, anaphylaxis or angioedema (Group NINDS 1995). Although tPA is widely available in the US, it is estimated that only 10% to 20% of stroke patients receive treatment with tPA (Stemer and Lyden 2010; Grotta et al 2001), principally because they present for care > 4.5 hours after the onset of symptoms or are at increased risk for bleeding from concomitant medication use, or for other reasons. Another effective treatment for stroke, albeit less frequently used, is mechanical thrombectomy in patients with documented large vessel occlusion (Goyal et al 2015; Campbell et al 2015; Berkhemer et al 2015; Saver et al 2015). Large vessel occlusion is defined as thromboembolic blockage of the distal internal carotid artery, the M1 or proximal M2 portions of the middle cerebral artery, or the proximal anterior cerebral artery. Several well-controlled randomized clinical trials showed benefit of thrombectomy when added to IV tPA treatment of large vessel occlusion. In some of these trials, however, patients benefited who were ineligible for IV tPA and were treated with thrombectomy alone. Although thrombectomy performed late (> 6 hours after stroke onset) did not appear to be successful in some trials (Kidwell et al 2014; Ciccone et al 2013; Broderick et al 2013), more recently it appears that with careful imaging selection thrombolysis or thrombectomy may be successful as late as 16 or 24 hours after symptom onset (Albers et al 2018; Dankbaar et al 2018; Nogueira et al 2018; Thomalla et al 2018). Use of mechanical thrombectomy, with or without IV tPA treatment, is considered standard of care in patients with documented large vessel occlusion. Although recent studies have indicated patients may respond to thrombolysis or thrombectomy very late (up to 24 hours) there remains a very large unmet need in stroke treatment. Furthermore, not all patients treated with thrombolysis or thrombectomy recover full function. Thus, adjuvant neuroprotectants are needed to complement recanalization therapies in patients so-treated, or to provide a modicum of protection in patients unable to receive thrombolytic or thrombectomy therapy. StrokeNet Concept Synopsis Version 4.0 – 4/1/2019 Page 4 of 22 Briefly describe relevant evidence (pre-clinical and/or clinical) used to support the proposed study-addressing the questions below (http://grants.nih.gov/grants/guide/notice-files/NOT-NS- 11-023.html ): PRECLINICAL: Efficacy in Animal Models of Stroke 3K3A-APC has satisfied 10 out of 10 of the Stroke Treatment Academic Industry Roundtable (STAIR) suggestions for pre-clinical drug safety and efficacy assessment (Zlokovic and Griffin 2011). 3K3A-APC has been tested for vasculoprotective and neuroprotective activity in several rodent models of stroke including both mice and rats, old and young animals, and animals with stroke risk factors. In a murine model of embolic stroke, an intact clot was placed at the origin of the middle cerebral artery, and functional neurological testing and histopathological analyses were conducted up to 7 days following the stroke (Guo et al 2009a and 2009b). In this study, a single IV dose of 0.2 mg/kg of murine 3K3A-APC, administered 4 hours after stroke as a single dose, or 12 hours after stroke followed by IV administration 1, 3, 5, and 7 days (0.2 mg/kg) within 3 and 7 days after stroke (Guo et al 2009b), significantly improved functional recovery and neurological scores, and reduced infarct and edema volume in the brain. Interestingly 3K3A-APC, similarly to APC (Zlokovic et al 2005), exhibited strong species differences with mouse recombinant 3K3A-APC being about 10 times more potent in mouse models of stroke than human 3K3A-APC (Guo et al 2009a), and human 3K3A-APC was 5 times more protective than the mouse 3K3A-APC in a human model of ischemia (i.e., oxygen- glucose deprivation of human brain endothelial cells) (Guo et al 2009a), suggesting clear species differences, which should be taken into account when designing human studies for ischemic stroke with APC mutants. In murine models of MCAO, the artery was occluded for 1 hour followed by 24 hours of reperfusion (transient MCAO model) or was permanently ligated (permanent MCAO model). Animals were assessed for bleeding risk (hemoglobin content in the ischemic brain areas, or red blood cells), neurological function, and neuropathology, for up to 7 days after stroke. In both MCAO models, wild- type APC treated animals showed increased hemoglobin content in the brain, indicative of bleeding, compared with vehicle controls, whereas 3K3A-APC-treated animals showed substantially reduced risk of bleeding as shown by normal hemoglobin content in brain and no evidence of post-ischemic red blood cells extravasation. Human recombinant 3K3A-APC significantly improved neurological function and histopathology scores relative to vehicle-control animals at single doses of 0.4 mg/kg or 2.0 mg/kg, administered 5 minutes prior to transient MCAO, and at multiple doses of 1.0 mg/kg administered 12 hours and 1, 3, 5, and 7 days following permanent MCAO (Wang et al 2009). The improvement in neurological function and histopathology scores following 3K3A-APC monotherapy was superior to that observed with comparable doses of wild-type APC. Of note, mouse recombinant 3K3A-APC has similar protective effects at a range 10 times lower than human 3K3A-APC (Guo et al 2009a). 3K3A-APC acts synergistically with tPA in both mouse and rat stroke models (Wang et al 2012). Human recombinant tPA (10 mg/kg), alone or in combination with human recombinant 3K3A-APC (2 mg/kg), was administered IV 4 hours after proximal or distal transient MCAO in mice or embolic stroke in rats, followed by 3K3A-APC for 3 to 4 consecutive days after stroke. In this delayed treatment paradigm, tPA alone led to increased post-ischemic bleeding and had no beneficial effects on infarct volume or behavior (neurological score, foot-fault, forelimb asymmetry, adhesive removal) compared with vehicle. In contrast to either therapy alone, the tPA plus 3K3A-APC combination substantially reduced tPA-associated bleeding risk and significantly reduced infarct volume at 24 hours and at 7 days following proximal or distal transient MCAO in mice and at 7 days after embolic stroke in rats, by 65%, 63%, and 52%, respectively, (p < 0.05; determined by 1-way analysis of variance followed by StrokeNet Concept Synopsis Version 4.0 – 4/1/2019 Page 5 of 22