Retinoids that regulate cell growth, differentiation, and apoptosis have shown promising results in preclinical studies and in a few clinical trials of cancer chemoprevention and therapy. However, the clinical use of retinoids is limited by resistance of certain malignant cells to their antitumor effects and by side effects. To identify more potent retinoids, we examined the effects of heteroarotinoids (Hets), new synthetic retinoids with reduced toxicity, on the growth of human head and neck squamous cell carcinoma (HNSCC) lines. Six Hets with different retinoic acid receptor activation potentials were found to exhibit distinct efficacies. The most potent among the Hets examined, SHetA2, [[(4-nitrophenyl)amino][2,2,4,4-tetramethyl thiochroman-6-yl)amino] methane-1-thione], was more effective than either all-trans- or 9-cis-RA. The growth of UMSCC38, the most sensitive among the eight HNSCC cell lines examined, was suppressed by ShetA2 in a dose- and time-dependent fashion. SHetA2-induced apoptosis in UMSCC38 cells was comparable with N-(4-hydroxyphenyl)retinamide (4HPR). Reactive oxygen species (ROS) generation in the UMSCC38 cells was increased by SHetA2, and this effect was suppressed by the antioxidant butylated hydroxyanisol, which also suppressed SHetA2-induced apoptosis. SHetA2 suppressed mitochondrial permeability transition and enhanced cytochrome c release from mitochondria. Both of these effects were prevented by cyclosporin A, which also decreased SHetA2-induced apoptosis. SHetA2 increased caspase-3-like activity, and a caspase-3 inhibitor diminished SHetA2-induced apoptosis. Several retinoid receptor antagonists failed to prevent apoptosis induction by SHetA2. These results demonstrate that SHetA2 is a potent, receptor-independent, apoptosis inducer that acts on the mitochondria in HNSCC cells. Further investigation of the potential of SHetA2 in prevention and therapy of HNSCC is warranted also because of much lower toxicities compared with receptor active retinoids.