During the past few years major conceptual and technical advances have been made towards the therapeutic modulation of cardiac gene expression for the treatment of cardiac diseases. Among these are 1) the identification of new molecular therapy targets in cardiac disorders, often derived from genetic animal models. 2) A better understanding of the molecular and cellular determinants of cardiac gene transfer in vivo, in animal models and in first clinical trials. 3) The development of novel regulatable and long-term stable vector systems. This review is focused on nucleic acid-based modulation of cardiac calcium homeostasis as a paradigm for the new gene therapeutic approaches, since recent landmark papers have suggested this to be a molecular target of key importance in heart failure. In particular, the development of severe heart failure in the genetic MLP(-/-) animal model could be completely abolished by the targeted ablation of phospholamban (PL), a key regulator of cardiac calcium homeostasis. This impressive effect of permanent germline PL ablation provides-in conjunction with former important work on disturbed calcium handling in the failing human heart-a rationale for the gene therapeutic approach of ad hoc suppression of PL by antisense strategies (antisense RNAs, ribozymes, RNA interference) or PL variants. Based on the broad spectrum of methods employed to characterize this general strategy, PL-targeted approaches may be considered as a paradigm of future genetic treatments of cardiac disorders, although the differences between animal models and humans must be kept in mind. High safety of any such therapy will be a prerequisite for any possible clinical application and therefore novel methods to improve control are being devised: 1) The regulation of gene therapy vectors by biochemical abnormalities associated with the target disease itself (" Induction-by-Disease" gene therapy). 2) External control of vector activity by the employment of drug-sensitive promotors. In addition, the important goal of cardiac long-term stability of the therapeutic vectors has recently been achieved in animal models using vectors derived from adeno-associated viruses (AAVs).