Introducing Genes to the Heart
Recent clinical gene therapy trials for the treatment of heart failure (HF) have failed to meet primary efficacy end points and have tampered enthusiasm for the future application of cardiac gene therapy. These results have brought to light the difficulty of efficiently introducing genes into the human heart and have focused on potential problems that need to be addressed before further clinical applications. These trials, however, have established the safety of gene delivery vectors for cardiac targeting in humans. The sinusoidal trajectory of gene therapy continues, and despite these setbacks, the future of the field is promising.
Recent Results of Cardiac Gene Therapy Clinical Trials
In the past year, the results of 3 recent phase II clinical gene therapy trials targeting HF became available.1–3 These serial publications, however, all failed to meet primary efficacy end points. In the CUPID IIb trial (Calcium Up-Regulation by Percutaneous Administration of Gene Therapy in Cardiac Disease Phase 2b), the efficacy of intracoronary-administered recombinant adeno-associated virus (AAV) carrying sarco/endoplasmic reticulum Ca2+-ATPase (SERCA2a) at a dose of 1×1013 DNase-resistant particles was examined in 250 patients.1 In contrast to a significant reduction of clinical events in the phase I/IIa trial, the number of adverse events in this trial was similar between the treated and the control groups. The STOP-HF trial (Stromal Cell-Derived Factor-1 Plasmid Treatment for Patients With Heart Failure) examined the efficacy of endocardial direct injection of plasmid stem cell–derived factor-1 with doses of 15 and 30 mg in 93 patients.2 The primary end point was a composite score of a 6-minute walk distance and a quality-of-life questionnaire. This study again reported similar outcomes in primary end points between the treated and the control groups. The most recent report was the trial of adenovirus 5–mediated adenylyl cyclase 6 gene therapy using intracoronary delivery in 56 patients.3 The primary efficacy end point was a combination of exercise time, echocardiography, and pressure-derived functional parameters before and after dobutamine challenge. Although the composite end point score was not reported, none of the parameters included in the primary efficacy end point was reported to be significant when comparing treated and control patients.3 These results certainly frustrate the field; however, there remains some hope because the latter 2 trials have reported a potential benefit of gene therapy in the subanalyses.
Why Did They Fail to Show Efficacy?
There are several possible reasons that these recent trials failed to meet the primary predefined clinical efficacy end points. These include a large placebo effect found in the control groups, diverse comorbidities in patients that offset positive effects of gene therapy, difficulty in determining the optimal end points prior to initiating the trial, and insufficient power to detect the difference between the groups. Invasive procedures to deliver the genes to the heart, as well as narrow patient inclusion criteria, limited the enrollment of a large number of patients in the trials, which was a major limitation to address some of the above problems. However, the most convincing explanation is that the therapeutic efficacy was unfortunately not as robust as we initially expected.
Gene Delivery: Critical Step
For any gene therapy to work, there are 2 principal factors that determine the success of the therapeutic approach: gene introduction to the cells and the function of the transduced genes. Without effective gene transduction, therapeutic genes have no chance to work in the target cells. Meanwhile, even if the transduction is robust, genes with minimal or deleterious effects on cells or organs will not improve the outcomes. For the CUPID IIb trial, the most likely reason for the failure is that gene delivery vectors did not transduce the human hearts as effectively as they did in preclinical animal models. The viral uptake in the heart from patients who underwent cardiac transplantation only presented 20 to 561 copies of vector per milligram of DNA.4 This titer is significantly less compared with the viral uptake observed in the animal models that have consistently demonstrated therapeutic efficacy of SERCA2a gene transfer (20 000–350 000 copies of vector per milligram of DNA). Thus, considering the amount of viral vector found in the heart, although only from a portion of patients, the neutral result in the CUPID IIb trial was probably because of the failure in transduction, and the trial was unlikely to have examined the effects of SERCA2a gene function. For examining the beneficial effects of SERCA2a overexpression observed in animal models indeed translates to humans require future trials by using a more efficient gene delivery system and a much higher dose. In the other 2 trials, it is unclear how much gene transduction was actually achieved in their treated population, partly because of the short-term expression of plasmid and adenoviral vectors. It would be informative if the investigators of the adenovirus 5–mediated adenylyl cyclase 6 gene therapy trial report the adenoviral titer retained in the heart of patients who underwent cardiac transplantation in the treatment arm.
Safety of Vectors
One encouraging point is that none of the trials reported vector-related safety issues, specifically in terms of immune responses. In the past, immune responses to adenoviral vectors have resulted in morbidities and mortalities.5 In these trials, the vector doses were kept low, and it is an important factor for the safety signal in these trials. As we envision higher doses to enhance transduction in human hearts, immune responses will need to be monitored closely because all vectors could potentially elicit a T cell response in the heart.
How Can We Improve Cardiac Gene Therapy?
To improve gene delivery, there are 3 potential strategies: (1) increasing the dose of the vectors, (2) using more efficient gene delivery methods, and (3) developing vectors with higher cardiac tropism and transduction efficacy in the myocardium.
The first strategy, increasing the dose, is the simplest approach with currently available tools. In preclinical studies in large animals, there is a clear dose–response relationship with the amount of administered AAV and the viral genomes in the heart. Thus, it is likely that a higher dose of vectors can lead to increased transduction efficacy in patients as well. AAV vector doses in large animals, with similar heart weights as humans, which showed contractile improvements, did not show clinical efficacy in patients. One explanation is that patients entering the trials had history of multiple procedures and interventions and, in some cases, extensive scarring, rendering the entry of the AAV vectors more difficult. Although higher doses are considered to achieve higher efficacy, an excessive amount of viral vectors can induce a cellular immune response against the vectors. In liver gene therapy studies using AAV, the magnitude of T cell response seemed to be associated with the amount of vector administered,6 while none of the CUPID patients (where the AAV doses were much lower) had a T cell response. Although this immune reaction could be controlled by high-dose steroids, cardiac patients need to be more closely monitored for this reaction because a T cell immune reaction in the heart might lead to fatal arrhythmias even if the reaction is mild. Thus, because clinical trials with AAVs are being planned with higher doses, patients should desirably have an implantable cardioverter-defibrillator. Similarly, a high dose of other nonviral gene delivery vectors can induce a vector-related immune response; thus, careful patient monitoring is necessary after administration.
The second strategy, using a more efficient gene delivery method, has been examined in several preclinical large animal studies. Increased efficacy is generally accompanied by a higher invasiveness of the delivery method.7 Retrograde coronary sinus delivery of vectors during anterior coronary artery flow blockade has been shown to achieve efficient gene transduction in pigs and is being considered a delivery method for future clinical trials.8 Extracorporeal recirculating devices and closed-circuit retrograde infusion during bypass result in much larger viral genomes per DNA and higher transduction efficacy overall.9 Because majority of patients who are candidates for cardiac gene therapy have impaired cardiac function, using invasive procedures need to be cautiously considered because it directly links to safety. Meanwhile, when patients need surgical procedures for coronary bypass or valve replacement surgeries, vector delivery using cardiopulmonary support device may be an effective approach.9
The third strategy, using vectors with higher transduction efficacy, is being actively explored. AAV serotype 9 emerged as a vector with high cardiac tropism and has become one of the most powerful tools to target hearts in cardiac gene therapy research, especially in rodent models. This vector, thus, deserves testing in clinical trials. In addition to AAV9, vector modification of AAVs using DNA shuffling and directed evolution have also generated efficient and more cardiotropic AAVs. A re-engineered AAV vector, AAV2i8 is a good example that transduces cardiomyocytes with high efficiency with different antigenic profiles.10 Using this vector, we recently reported that gene transfer of constitutively active form of inhibitor 1 in a pig model of heart failure results in improved cardiac function while detargeting the liver.11 The altered antigen profile seems to circumvent the humoral immune reaction in humans, and a broader population of patients can be treated using this vector compared with AAV1, a vector of which more than half of the candidate patients were inapplicable for the study enrollment in the CUPID trial because of preexisting neutralizing antibodies against it.
Modified mRNA and exosomes are more biological vectors that have recently emerged. By chemically modifying the nucleotide bases, modified RNA has a more stable structure and low immunogenicity in vivo.12 It also has unique rapid and short expression kinetics suited to overexpress some of the growth factors that may cause tumorgenesis when expressed long term. Exosomes are small cell–derived vesicles that contain proteins, RNAs, and lipids for transporting these materials to other cells.13 Because some exosomes seem to target specific cell types, these small vesicles can be used as a gene delivery vector to target the heart. These vectors essentially deliver mRNA or microRNAs directly to the cells and skip the transcriptional step, thus, has the potential to improve gene transfer efficiency dramatically.
In addition to these 3 approaches, endogenous expression levels of target genes may be taken into consideration. That is, the same efficiency of gene transduction may result in different levels of overexpression depending on the background expression. For example, an increase in 10 copies of mRNA by gene transfer in cells with the background of 1 copy or 100 copies of basal mRNA likely leads to different effects. It is of note that in deep sequencing data of mRNA in the heart, SERCA2a (ATP2A2) mRNA numbers are relatively high even in patients with heart failure who have significantly reduced SERCA2a mRNA levels compared with normal patients.14 Thus, SERCA2a may be a gene with a high hurdle, and targeting genes with less mRNA presence could be an alternative approach for success with current inefficient gene delivery tools. Alternatively, development of novel promoters with very high transcriptional efficiency may overcome this issue. It is also of paramount importance that new tools and methods be tested in animal models that closely reflect human disease phenotype because comorbidities, age, and immune response likely play important roles in gene transduction.
By analyzing the hearts/tissues of patients who received gene therapy AAV vectors, we now realize that introducing genes into the human heart is a formidable task. There are several approaches to improve cardiac gene therapy, and some of these approaches are expected to be incorporated in the upcoming clinical trials. Importantly, it is worth emphasizing that although improving transduction efficacy must be pursued, efficacy and safety need to be always balanced, and safety should never be compromised in these clinical trials. Developing tools/methods for improved transduction efficacy is as important in determining the effects of therapeutic genes. More resources should be focused on improving gene delivery methods, which are critical for efficiently introducing genes to the heart.15 The many failures experienced in clinical gene therapy trials in many monogenic diseases have been now reversed with more focused delivery and appropriate vectors. In fact, gene therapy has undergone an amazing rebirth in the treatment of monogenic diseases. The ups and downs we are experiencing in gene therapy for heart failure are teaching us valuable lessons and will eventually lead us to effective treatments for patients with heart failure.
This work is supported by National Institutes of Health (NIH) R01 HL117505, HL119046, HL129814, 128072, a P50 HL112324, and a Transatlantic Fondation Leducq grant. We acknowledge the Gene Therapy Resource Program (GTRP) of the National Heart, Lung, and Blood Institute, National Institutes of Health.
R.J. Hajjar is a scientific founder of Nanocor Corp. The other author reports no conflicts.
Greenberg B, Butler J, Felker GM, Ponikowski P, Voors AA, Desai AS, Barnard D, Bouchard A, Jaski B, Lyon AR, Pogoda JM, Rudy JJ, Zsebo KM. Calcium upregulation by percutaneous administration of gene therapy in patients with cardiac disease (CUPID 2): a randomised, multinational, double-blind, placebo-controlled, phase 2b trial.Lancet. 2016; 387:1178–1186. doi: 10.1016/S0140-6736(16)00082-9.CrossrefMedlineGoogle Scholar
Chung ES, Miller L, Patel AN,. Changes in ventricular remodelling and clinical status during the year following a single administration of stromal cell-derived factor-1 non-viral gene therapy in chronic ischaemic heart failure patients: the STOP-HF randomized Phase II trial.Eur Heart J. 2015; 36:2228–2238. doi: 10.1093/eurheartj/ehv254.CrossrefMedlineGoogle Scholar
Hammond HK, Penny WF, Traverse JH,. Intracoronary gene transfer of adenylyl cyclase 6 in patients with heart failure: a randomized clinical trial.JAMA Cardiol. 2016; 1:163–171. doi: 10.1001/jamacardio.2016.0008.CrossrefMedlineGoogle Scholar
Hulot JS, Ishikawa K, Hajjar RJ. Gene therapy for the treatment of heart failure: promise postponed.Eur Heart J. 2016; 37:1651–1658. doi: 10.1093/eurheartj/ehw019.CrossrefMedlineGoogle Scholar
Raper SE, Chirmule N, Lee FS, Wivel NA, Bagg A, Gao GP, Wilson JM, Batshaw ML. Fatal systemic inflammatory response syndrome in a ornithine transcarbamylase deficient patient following adenoviral gene transfer.Mol Genet Metab. 2003; 80:148–158.CrossrefMedlineGoogle Scholar
Mingozzi F, High KA. Immune responses to AAV vectors: overcoming barriers to successful gene therapy.Blood. 2013; 122:23–36. doi: 10.1182/blood-2013-01-306647.CrossrefMedlineGoogle Scholar
Ishikawa K, Aguero J, Naim C, Fish K, Hajjar RJ. Percutaneous approaches for efficient cardiac gene delivery.J Cardiovasc Transl Res. 2013; 6:649–659. doi: 10.1007/s12265-013-9479-7.CrossrefMedlineGoogle Scholar
Pleger ST, Shan C, Ksienzyk J,. Cardiac AAV9-S100A1 gene therapy rescues post-ischemic heart failure in a preclinical large animal model.Sci Transl Med. 2011; 3:92ra64. doi: 10.1126/scitranslmed.3002097.CrossrefMedlineGoogle Scholar
Katz MG, Brandon-Warner E, Fargnoli AS, Williams RD, Kendle AP, Hajjar RJ, Schrum LW, Bridges CR. Mitigation of myocardial fibrosis by molecular cardiac surgery-mediated gene overexpression.J Thorac Cardiovasc Surg. 2016; 151:1191–200.e3. doi: 10.1016/j.jtcvs.2015.11.031.CrossrefMedlineGoogle Scholar
Asokan A, Conway JC, Phillips JL, Li C, Hegge J, Sinnott R, Yadav S, DiPrimio N, Nam HJ, Agbandje-McKenna M, McPhee S, Wolff J, Samulski RJ. Reengineering a receptor footprint of adeno-associated virus enables selective and systemic gene transfer to muscle.Nat Biotechnol. 2010; 28:79–82. doi: 10.1038/nbt.1599.CrossrefMedlineGoogle Scholar
Ishikawa K, Fish KM, Tilemann L,. Cardiac I-1c overexpression with reengineered AAV improves cardiac function in swine ischemic heart failure.Mol Ther. 2014; 22:2038–2045. doi: 10.1038/mt.2014.127.CrossrefMedlineGoogle Scholar
Chien KR, Zangi L, Lui KO. Synthetic chemically modified mRNA (modRNA): toward a new technology platform for cardiovascular biology and medicine.Cold Spring Harb Perspect Med. 2014; 5:a014035. doi: 10.1101/cshperspect.a014035.MedlineGoogle Scholar
Zhang J, Li S, Li L, Li M, Guo C, Yao J, Mi S. Exosome and exosomal microRNA: trafficking, sorting, and function.Genomics Proteomics Bioinformatics. 2015; 13:17–24. doi: 10.1016/j.gpb.2015.02.001.CrossrefMedlineGoogle Scholar
Yang KC, Yamada KA, Patel AY, Topkara VK, George I, Cheema FH, Ewald GA, Mann DL, Nerbonne JM. Deep RNA sequencing reveals dynamic regulation of myocardial noncoding RNAs in failing human heart and remodeling with mechanical circulatory support.Circulation. 2014; 129:1009–1021. doi: 10.1161/CIRCULATIONAHA.113.003863.LinkGoogle Scholar
Donahue JK. Cardiac gene therapy: a call for basic methods development.Lancet. 2016; 387:1137–1139. doi: 10.1016/S0140-6736(16)00149-5.CrossrefMedlineGoogle Scholar