The demonstration that angiogenic growth factors can stimulate new blood vessel growth and restore perfusion in animal models of myocardial ischemia has led to the development of strategies designed for the local production of angiogenic growth factors in patients who are not candidates for conventional revascularization. The results of recent clinical trials of proangiogenesis gene therapy have been disappointing; however, significant limitations in experimental design, in particular in gene transfer strategies, preclude drawing definitive conclusions. In the REVASC study cardiac gene transfer was optimized by direct intramyocardial delivery of a replication-deficient adenovirus-containing vascular endothelial growth factor (AdVEGF121, 4 x 10(10) particle units (p.u.)). Sixty-seven patients with severe angina due to coronary artery disease and no conventional options for revascularization were randomized to AdVEGF121 gene transfer via mini-thoracotomy or continuation of maximal medical treatment. Exercise time to 1 mm ST-segment depression, the predefined primary end-point analysis, was significantly increased in the AdVEGF121 group compared to control at 26 weeks (P=0.026), but not at 12 weeks. As well, total exercise duration and time to moderate angina at weeks 12 and 26, and in angina symptoms as measured by the Canadian Cardiovascular Society Angina Class and Seattle Angina Questionnaire were all improved by VEGF gene transfer (all P-values at 12 and 26 weeks < or =0.001). However, if anything the results of nuclear perfusion imaging favored the control group, although the AdVEGF121 group achieved higher workloads. Overall there was no significant difference in adverse events between the two groups, despite the fact that procedure-related events were seen only in the thoracotomy group. Therefore, administration of AdVEGF121 by direct intramyocardial injections resulted in objective improvement in exercise-induced ischemia in patients with refractory ischemic heart disease.