RT-100 technical summary (as of March 2015) Table of contents

Transcription

1 RT-100 technical summary (as of March 2015) Table of contents 1. Major unmet clinical need 2. Gene therapy 3. Adenylyl cyclase and heart failure 4. Clinical congestive heart failure 5. lntracoronary adenovirus vector delivery in patients 6. Selection of vector - Why adenovirus? 7. Examples of pre-clinical efficacy data associated with AC6 expression 8. Biodistribution, toxicology and cardiac expression of transgene AC6 in pigs 1 of 16

2 1. Major unmet clinical need Congestive heart failure (CHF) is a chronic disease characterized by the inability of the heart to pump sufficient blood to meet the body's demands. It is a progressive disease, and symptoms worsen over time. CHF can be caused by many conditions that damage the heart muscle, including heart attacks, infections, alcohol or drug abuse and conditions such as high blood pressure, valve disease, thyroid disease, kidney disease, diabetes, or heart defects present at birth. People with severe CHF may need a mechanical heart pump. Prevalence and severity Heart failure afflicts approximately 6 million people in the US, with 870,000 new cases every year; it affects 23 million people globally In the US, CHF is the most frequent cause of hospitalization for people ages 65 and older Even with optimal therapy, mortality in patients with moderately severe symptoms is worse than most cancers: 50% are dead five years after diagnosis In 2012, the annual cost of heart failure in the United States was $31 billion, with projections suggesting that rising prevalence will increase medical costs to $70 billion in 2030 CHF current standard of care Diuretics (water or fluid pills) ACE inhibitors Aldosterone antagonists Angiotensin receptor blockers Beta blockers LVADs (Left Ventricular Assist Devices) 2 of 16

3 2. Gene therapy The gene selected is human adenylate cyclase type 6 (AC6), the dominant AC type in heart muscle cells. AC6 is a 130 kd membrane protein that catalyzes the conversion of ATP to camp, a second messenger that is an important determinant of heart function. However, AC6 also has pronounced effects on calcium handling in the heart and consequently influences heart function independently of camp. The beneficial effects of AC6 predominantly reflect effects on calcium handling 1, the penultimate determinant of the heart's contractile force. The amount and function of AC6 is reduced in failing hearts and this is a major cause for poor heart function and admission to hospital, and contributes to increased mortality. We have modified a human adenovirus-5 (Ad5) to carry the AC6 gene into the patient's heart to increase their heart's function. This vector, which is comprised of the inactivated adenovirus and the human AC6 gene, is delivered into the arteries that feed the heart muscle (coronary arteries). This is done during an outpatient procedure by a cardiologist during cardiac catheterization, a commonly performed procedure. 2.1 Duration of expression Adenovirus provides high yield gene transfer, but expression is extrachromosomal, and therefore only one daughter cell continues to express the gene after cell division. However, our target cell is the cardiac myocyte, which does not divide so that transgene expression is not diluted in the heart over time, but is in other organs (see Figure 2). As a consequence of targeting a non-dividing cell and the absence of inflammation (see below) increased AC6 function has been shown to persist for at least 18 weeks - there was no difference in effect week 2 versus week 18 after a single administration. 2 We have documented persistent left ventricular (LV) functional improvement twelve weeks after intracoronary delivery of adenovirus vectors. 3,4 Finally, Ad5.hAC6 DNA, in doses similar to those used in the current clinical trial, persists and AC6 transgene is expressed in LV samples 12 weeks after delivery. 2.2 Promoter selection We have tested several cardiac-specific promoters, but all are substantially weaker than our preferred promoter (human cytomegalovirus promoter, CMV). The University of Florida conducted an independent biodistribution and toxicology study prior to FDA approval of the current clinical trial. This study examined AC6 expression in 11 organs 7, 28 and 70 days following intracoronary Ad5.hAC6 delivery in 36 pigs. Cardiac AC6 expression was increased and sustained at the same levels from day 7 to day 70. In contrast, over this same interval AC6 expression was reduced by 100% in lung, by 99.6% in liver, and by 89% in spleen (Figure 2). Indeed, the expression of AC6 (across all time points) was higher in heart than in any other organ. Therefore we elected to not use a cardiac-specific promoter. 2.3 Immune response (see also Section 6, Table 3). Cardiac inflammation has not been encountered following intracoronary adenovirus delivery in: a) 450 patients that received Ad5.FGF4 5,7 ; b) 62 pigs in two independent GLP toxicology studies; c) in hundreds of pigs receiving intracoronary adenovirus in studies at UC San Diego since Finally, the 3 of 16

4 method is effective even in the presence of neutralizing antibodies to Ad5, and repeat injections remain therapeutic and safe. 8 The results reviewed above were sufficiently compelling that the US Federal Drug Administration (FDA) approved our clinical trial of AC6 gene therapy in human subjects with severe CHF (IND 13761; ClinicalTrials.gov NCT ). 4 of 16

5 3. Adenylyl cyclase and heart failure Molecular cloning studies indicate that there are at least ten different isoforms of AC, each with different structure, tissue, and chromosomal distribution and regulation. Cardiac myocytes appear to express AC5 and AC6 predominantly, 9,11 two AC isoforms that are related in structure and function, but which fulfill unique biological roles. AC plays a pivotal role in contractile responsiveness, is tightly linked to left ventricular (LV) function, and is functionally impaired in heart failure. Cardiac-directed expression of AC6 has favorable effects on LV contractility Among AC isoforms, AC6 has features that make it a logical candidate to increase LV function in the setting of heart failure. As a nominal adenylyl cyclase, a clinical concern would be the potential role of AC6 in increasing intracellular camp, which latter can be counterproductive in improving cardiac function in the setting of CHF. However, a variety of evidence suggests this is NOT the case for AC6: i) First, the inhibitory GTP-binding protein, Gai, inhibits camp generation via AC5 and AC6, 11 the dominant AC isoforms in cardiac myocytes. This feature may provide protection against b-adrenergic receptor (bar) overstimulation. ii) Second, submicromolar levels of intracellular calcium, which stimulate several AC isoforms, either have no effect or inhibit AC6 and AC5, 11 thereby providing an additional modulation of bar activation through AC. iii) We found that pronounced increased AC6 content does not alter basal intracellular camp production. 16 iv) Finally, recent studies have shown that increased AC6 content has favorable effects on cardiac myocytes that are independent of camp generation but which would increase contractile function. 1 For example, AC6 gene transfer increases sarco(endo)plasmic reticulum Ca 2+ -ATPase 2a (SERCA2a) calcium uptake, 17 reduces phospholamban (PLB) expression 18 and phosphorylation, and increases Akt activation through inhibition of an Akt-specific phosphatase, PH domain and leucine rich repeat protein phosphatase (PHLPP). 19 These unique features make the prospect of AC6 gene transfer fundamentally different from other agents that increase intracellular camp (dobutamine, milrinone), but which have failed to prolong life in CHF. 20 Our group and others have reported down regulation of myocardial AC5 9 and AC6 9,10 in animal models of heart failure. Might restoration of cardiac AC6 content increase contractile function in CHF? AC6 may circumvent the deleterious features of stimulation of proximal elements of this pathway by remaining under regulation by Gai and intracellular calcium, and through increased calcium handling, PLB phosphorylation, and reduced apoptosis (Akt activation), effects which are not dependent on camp. The relative advantages of AC5 vs AC6 in this regard are not precisely known. However, when these transgenes are expressed in murine cardiomyopathy, only AC6 increases heart function and prolongs life, 13,14,21 and targeted deletion of AC5 is associated with increase LV dp/dt, 22 indicating that increased cardiac AC5 content may decrease LV function. In contrast, AC6 deletion is associated with marked reduction in calcium handling and reduced LV function of 16

6 Two hallmarks of low ejection fraction heart failure of diverse etiologies are decreased ability of cardiac myocytes to generate camp and depressed myocardial contractility. Previous treatments for clinical heart failure have focused on increasing myocardial camp content using pharmacological agents that stimulate the bar (dobutamine) or decrease the breakdown of camp through phosphodiesterase inhibition (milrinone). In general these efforts have failed, perhaps because of deleterious consequences of unrelenting stimulation of the bar pathway. Indeed current approaches embrace the use of bar antagonists over agonists for management of compensated heart failure. Paradoxically, increasing AC6 in cardiac myocytes does not result in sustained camp generation. 1,2,12-17 A goal of ours has been to determine how AC6 confers protective effects to the heart while other elements in this signaling pathway do not , Our preclinical data 13,14,27 supports the idea that AC6, unlike other elements in this pathway, will be beneficial to patients with heart failure. These data and additional unpublished data provided the rationale for the current clinical trial of AC6 gene transfer in CHF. 6 of 16

7 4. Clinical congestive heart failure Even with optimal management, heart failure is an inexorable disease and is associated with excessive morbidity and substantial short-term mortality. In addition to high mortality, dilated systolic heart failure is associated with reduced LV contractile function, LV dilation, reduced ejection fraction, and elevations in pulmonary artery wedge pressure. Hallmarks of CHF include exercise intolerance, breathlessness, leg swelling (edema) and pulmonary congestion (fluid in the lungs). The current clinical trial focuses on exercise tolerance, hemodynamic measurements and LV contractile function to assess benefit of Ad5.hAC6 therapy. We have included a control group to estimate variability. This will assist in our interpretation of clinical indicators regarding safety and potential efficacy. The protocol is designed to determine whether Ad5.hAC6 can attenuate progression and reverse some of the manifestations of disease in patients with stable but severe CHF. In the trial, patients have an implantable cardiac defibrillator (ICD) in place. Heart transplantation has an 80% 10-year survival rate, an improvement over medical therapy, but very few patients are eligible for transplantation, and donor hearts are not readily available. Only 2500 procedures are performed in the US each year, which represents <0.1% (1 per 2000) of 6 million patients with severe CHF in US. Patients on bar antagonists who enrolled in the trial continued such therapy. We anticipate that AC6 will provide additive beneficial effects in these patients due to increased contractility via campindependent mechanisms, 1 including reduced expression and function of phospholamban, improved calcium handling, 1,17,19 and Akt activation of 16

8 5. lntracoronary adenovirus vector delivery in patients 5.1 Rationale The adenovirus vector will be delivered into the coronary arteries to expose the heart to the highest possible concentration of vector. 2,27 We believe that the portal of entry into the cardiac interstitium is through the coronary capillary endothelial cells via vesicular transport (transcytosis). By volume, the majority cell type in the heart is the cardiac myocyte. Gene transfer is necessary for the proposed studies because the rationale is to increase cardiac content of a specific AC that will then increase contractile function. 1,2,12,17-19,27,28 This cannot be achieved by infusion of the protein itself, which is too large to enter the cardiac myocyte. 5.2 Safety There is no previous experience with intracoronary delivery of Ad5.hAC6 in patients. There have been clinical trials in which an E1-deleted adenovirus encoding human fibroblast growth factor-4 (Ad5.hFGF4), an angiogenic gene, was delivered into the coronary arteries of 450 patients with angina (aggregate number from four randomized, double-blind clinical trials), 5,7 in doses ranging from 3.3x10 8 to 3.3x10 10 vp. AGENT-3 & AGENT-4 (combined) enrolled 177 PBStreated, and 355 Ad5.hFGF4-treated patients (10 9 vp, n=166; vp, n=165) in a randomized, double-blind clinical trial. 7 The occurrence of adverse events from initial administration to a minimum of 23 months of follow-up was no different between PBS-treated and Ad5.hFGF4-treated patients and serial measurements of cardiac troponin - the best serum indicator of myocarditis - were normal in the 450 subjects randomized, indicating that clinically a significant myocardial injury did not occur. There are three differences between the Phase 2b/3 clinical trial of Ad5.hFGF4 (AGENT-3) and the current trial (Table 1): 1) the current trial uses a higher maximal dose of adenovirus vector; 2) the current trial includes use of intracoronary nitroprusside, which increases the extent of cardiac gene transfer; 30 and 3) we will be treating a different disease with a different transgene. These differences, combined with our preclinical data, provide a rationale for the proposed clinical trial. AGENT Current Highest Nitroprusside? CV Yes 8 of 16

9 6. Selection of vector - Why adenovirus? In selecting from among adenovirus, AAV or lentivirus, the delivery method dictates the selection of vector, which in turn is critical in gene therapy to achieve effective gene transfer to the target cell population. Lentivirus vectors are unsuitable for intracoronary delivery because they do not readily cross endothelial cells and therefore do not gain access to the cardiac interstitium. 31 Using indirect intracoronary delivery, we obtain up to 50% transgene expression in murine heart with adenovirus vectors, 32 but we are unable to detect any gene expression with very high dose HIV1- based lentivirus vectors using the same methods (unpublished data). The direct intracoronary delivery method provides fold higher gene transfer efficiency with adenovirus than with an AAV2 vector (Table 2). Newer AAV vectors, especially AAV9, may provide increased cardiac gene transfer efficiency. 36 However, insert size is limited in AAV vectors - only 4.7 kb - enough for the AC6 cdna (3.4 kb), but insufficient to include tet-regulated expression elements (an additional 2.0 kb) to enable termination of transgene expression, which we feel is important when using vectors that provide perpetual transgene expression. Efficiency Citation ( ) 19 39% 767, 20 AAV2 J 20 Of paramount importance in considering vectors is the degree to which the vector will incite an inflammatory response in the tissue and host into which it is delivered. Most virus vectors have been modified so that they are unlikely to cause a clinical infection. However, proteins encoded by adenovirus vectors can result in cell-mediated and humoral immune responses. In some applications this has been an insurmountable problem. Recent modifications, including deletions of virus DNA so that virus protein expression is reduced, have reduced immune response. Route of delivery, dose, and targeted tissue also are important determinants of an inflammatory response. For example, with intracoronary delivery of first and second-generation adenovirus vectors (Table 3), myocardial inflammation is not reported except when very high amounts of adenovirus (7.5x10 12 vp per gram of LV) are used. 2-8, 22, 27, 33, 34, The highest dose in the current clinical trial is vp, which represents 1.3x10 11 vp per gram of LV perfused fold lower than this inflammatory dose. In contrast, when adenovirus vectors are directly injected in the LV wall, a dose-dependent inflammatory response is seen. 46 This may reflect high local concentrations of adenovirus and a direct cytopathic effect with subsequent inflammation. The absence of myocarditis after intracoronary delivery of adenovirus (Table 3) differs from studies that have shown that adenovirus vectors induce cellular immune responses in other target organs, which may account, at least in part, for observed transient gene expression The cellular immune response generally is directed toward capsid proteins and involves antigen presentation by major histocompatibility complex (MHC) class I molecules, which play a fundamental role in foreign antigen presentation to the cellular immune system. Several 9 of 16

10 elements may contribute to the apparent "immune privilege" of heart relative to other organs: 1) Antigen presenting cells are not abundant in heart; 51 2) MHC class I molecules are expressed only at low levels in myocardium; 52 3) Cardiac myocytes per se do not appear to express MHC class I or class II molecules. 52 Table 3. Direct intracoronary delivery of adenovirus ( ) 2.2x10 6 ( ) 7.1 x Yes ( ) - 1.4x Yes Ad.FGF Yes 27 Ad.FGF2 ( ) Yes P ( ) - Yes 8.3x Yes ( ) Yes Rabbit ( ) Yes ( ) ( ) ( ) ( ) Yes Ad. ARKct ( ) ; Yes Ad.V2 ( ) Yes vs ( ) Yes Yes (mild) other studies direct intracoronary delivery of - thoracotomy clamping indirect delivery) Studies order of o vessel(s) specifically in the paper, certain assumptions ventricular (L weight: 0 ventricular (L grams; is body weight per weight grams; perfuses gram myocardium; adenovirus; receptor; day; vector; relevant. 10 of 16

11 7. Examples of pre-clinical efficacy data associated with AC6 expression 7.1 Effect of long-term increased cardiac-directed AC6 expression in mice with Gq-associated cardiomyopathy (Reference 14) To investigate whether increased AC6 expression is associated with long-term improvement in heart function that will provide a survival benefit in cardiomyopathy, transgenic mice with cardiac-directed AC6 expression (a 20- fold increase) were cross-bred with transgenic mice with cardiac-directed expression of Gq (4-5-fold increase). The latter results in LV dilation and dysfunction, decreased cardiac responsiveness to catecholamines and impaired b-adrenoceptor and AC-dependent camp production. Three lines from the cross were studied in Gq/AC6 (double positive), Gq alone and control (double negative). Three groups of animals (Gq, n=24; Gq/AC6, n=12; control n=25) were housed until death or killed at 24 months. Where death occurred less than 12 hours previously, morphometric measurements were made. Ventricular myocyte size was also measure in mice at 16 months. Echocardiography was performed on anaesthetized 14 month-old mice. Gq mice exhibited the expected significant reduced LV function with reduced fractional shortening and increased end-diastolic dimension. Both parameters were substantially improved in Gq/AC6 mice towards those values of control mice. Survival was increased by cardiac-directed expression of AC6 (Figure 1E) to such an extent that Gq/AC6 mice had survival rates indistinguishable from control mice. Myocardial hypertrophy developed in older Gq mice but was abrogated by simultaneous cardiac expression of AC6 (see Figure 1A-D). Thus, whole-heart weight and wet and dry ventricular weights of hearts from Gq/AC6 mice were similar to hearts from control animals. LV cardiac myocytes from Gq animals were substantially increased in size compared with controls. Gq/AC6 derived LV cardiac myocytes were indistinguishable from myocytes from control mice. 11 of 16

12 E Figure 1. Effect of long-term increased cardiac-directed AC6 expression in mice with Gqassociated cardiomyopathy. Assessment of hypertrophy (Panels A-D). (A) Representative hearts from Gq/AC6 and Gq mice at 11 months of age. (B) Representative cardiac myocytes from Gq/AC6 and Gq mice obtained at 15 months (Mag x160). (C) Wet ventricular-tibial length (Wet Vent/TL) ratios. (D) Left ventricular cardiac myocyte size. (E) Kaplan-Meier curve showing mortality in Gq (n=24), Gq/AC6 (n=12) and control mice (n=25). 7.2 Activation of cardiac AC6 expression in mice with failing ischemic hearts (Reference 28) Mice with cardiac-directed and regulated expression of AC6 underwent coronary artery ligation to induce severe CHF five weeks later. AC6 expression was activated in one group (AC-on, Figure 2) but not the other (AC-off). Multiple measures of LV systolic and diastolic function were obtained following a further five week period. Five weeks following activation of the transgene a 16-fold increase in LV AC6 protein was measured by Western blot (Figure 2B). Mice in the AC-off group showed continuing decline in LV ejection fraction (LVEF) between weeks five and ten. In contrast, LVEF increased over the same period in those mice with AC6-on. The increase in slope of the ESPVR (end systolic pressurevolume relationship Figure 2C), perhaps the best measure of overall contractile function of the intact heart, was four-fold. Overall, measures of systolic function 12 of 16

13 (LVEF, +dp/dt) were significantly increased with corresponding improvements in diastolic function (decreases in EDP, -dp/dt, tau) in the mice with activated AC6 expression (Figure 2D). These improvements in function were associated with normalization of troponin I phosphorylation and reduced apoptosis. About 70% of patients with severe HF have previous myocardial infarction as the cause. In our hands, the coronary artery ligation in this mouse model produces infarction of 49±3 % of the LV and septum, so this model represents a large MI with the transmural scar comprising the majority of the free wall. Despite the striking improvement in function in this study, no changes in LV dimensions were recorded after 5 weeks of treatment. (A) (B) (C) (D) Figure 2. Effect of Activation of Cardiac AC6 Expression in Mice with Failing Ischemic Hearts. (A) Protocol Schematic. (B) Western Blot showing a marked increase in AC6 protein in LV samples from mice five weeks after activation of AC6 transgene expression. (C) Basal contractile function measured in intact mice five weeks after AC6 expression. Examples of LV pressure-volume loops are shown together with end-systolic pressure volume relationship (ESPVR) mean values from all mice. (D) Changes in LV function expressed as the % change between parameters measured in the two groups of mice at the end of the study. 7.3 Pacing-induced CHF in pigs and AC6 gene transfer (Reference 27) Using the pacing model of CHF 10 we tested whether intracoronary Ad5.AC6 could increase LV function. Contractile function (LV +dp/dt) was measured in conscious pigs before and after 21 days of continuous LV pacing - used to induce severe dilated CHF. On day 7, when substantial CHF was already present, pigs received intracoronary Ad5.AC6 (1.4x10 12 vp + nitroprusside) or intracoronary 13 of 16

14 saline (PBS). Saline-treated animals showed progressively worsening CHF associated with marked reduction in LV contractility. The fall in LV +dp/dt was less in pigs that had received AC6 gene transfer (Figure 1). Serial echocardiography showed that Ad5.AC6 treatment was associated with increased LV function and reduced LV dilation (Figure 3, Table). LV gene transfer was confirmed by PCR. These data indicate that AC6 gene transfer increases LV function and attenuates deleterious LV remodeling in pacing-induced CHF. LV dysfunction was present before gene transfer, underscoring the clinical relevance of this study. 27 Fall in LV dp/dt - mmhg/s/ Ad.AC6 (n=7) PBS (n=7) p = Isoproterenol (µg/kg) Pre vs 21d Post PBS (9) AC6 (6) Change EDD Increase (mm) 18±2 13±2 28% 0.04 FS Decrease (% unit) 29±2 23±2 21% 0.03 Vcf Decrease (circ/s) 1.1± ±0.1 36% p Figure 3. Pacing-induced heart failure in pigs, with subsequent intracoronary Ad5.AC6 delivery (1.4x10 12 vp, with nitroprusside) vs PBS (control). Data were obtained before and 21 days after continuous LV pacing ( bpm). Gene transfer performed on Day 7, when severe CHF was already present. Data acquisition and analysis blinded to group identity of 16

15 8. Biodistribution, Toxicology and Cardiac Expression of Transgene AC6 in pigs Figure 4 shows results of the biodistribution & toxicology study of intracoronary Ad5.hAC5 with nitroprusside in pigs. This was a GLP study performed at University of Florida, independently. It showed no toxicity or inflammation in any organ even at doses of 3.2x10 12 vp examined 7, 28 and 70 days after delivery. ACS expression (mrna) was highest in LV at 28 days, and importantly, expression was persistent in LV at 70 days, but fell in liver, lung and spleen (Figure 4, Left Panel). Previously, we documented reporter gene expression in LV after intracoronary Ad5 gene transfer in pigs 4 (Figure 4, Middle Panel) and a striking increase in AC6 protein in LV 14 days after intracoronary Ad5.AC6 in pigs 2 (Figure 4, Right Panel). RNA (copies/µg) Days 3.2x10 12 vp Ad5.hAC LV Liver Lung Spleen 30X 200X Ad5.lacZ-nuc (10 12 vp): 5d after ic delivery (pig), substantial gene transfer in LV, without inflammation. - Nature Medicine 5:534-9, 1996 (AC6(((((((((lacZ ' AC6 ' GSTAC6 ' Figure 4. Left: Biodistribution/toxicology study in pigs (independent GLP laboratory) found LV AC6 expression was high and persisted to the final endpoint at 70 days, with no inflammation or toxicology in any organ. Middle: lntracoronary Ad5.lacZ resulted in LV expression (blue color) with no evidence of inflammation 4. Right: lntracoronary Ad5.AC6 delivery (1.4x10 12 vp) resulted in striking increase in AC6 protein in samples of LV (AC6 group, n=4; LacZ group, n=5; p<0.001; 2 ) 15 of 16

16 References (Hammond Lab publications: RED) 1. JMCC 50:751-8, Circ 102: , Hum Gene Ther 15:574-87, Nat Med 2:534-9, Circ 105:1291-7, JACC 42: , JACC 50: , Hum Gene Ther 17:230-8, JCI 93:2224-9, AJP 273:H707-17, Ann Rev Pharm/Tox 41:145-74, Circ 99: , Circ 99: , Circ 105: , Cardiov Res 56: , PNAS 95: , AJP 287:H , JBC 279: , BBRC 384:193-8, NEJM 320:677-83, FEBS Lett 458:236-40, BAS Res Card 101:117-26, Circulation 117:61-9, Circ Res 84:34-42, PNAS, 96: , Circulation 101: , Circulation 110:330-6, JACC 51:1490-7, Circulation 114:388-96, Hum Gene Ther 15:989-94, Circ 107: , Am J Physiol 287:H172-7, Gene Therapy 1:51-8, Circ Res 96:767-75, J Gene Med 7:316-24, Mol Ther 14:45-53, Hum Gene Ther 16: , Card Vasc Regen 1:11-21, J Thor Car Sur 120:720-28, Gene Ther 3:145-53, Circulation 101:408-14, Circulation 103:131-6, Circulation 104: , Circulation 101: , Nature Med 6:1395-8, Circulation 90: , Transplantation41:776-80, PNAS 92: , PNAS 93: , PNAS 95: , APMIS 106: , Nat Med 5: , of 16