The nonhalogenated double ester of prednisolone, prednicarbate (PC), is the first topical glucocorticoid with an improved benefit/risk ratio verified clinically and in vitro. To evaluate if this is due to unique characteristics of this steroid, a new compound created according to an identical concept, prednisolone 17-ethylcarbonate, 21-phenylacetate (PEP), and the new halogenated monoester desoximetasone 21-cinnamate (DCE) were tested and compared to PC, desoximetasone (DM) and betamethasone 17-valerate (BMV). Isolated foreskin keratinocytes served for in vitro investigations of anti-inflammatory processes in the epidermis, fibroblasts of the same origin were used to investigate the atrophogenic potential. Inflammation was induced by TNFα, resulting in an increased interleukin 1α (Il-1α) synthesis. As quantified by ELISA, all drugs significantly reduced Il-1α production. But PC and BMV appeared particularly potent, followed by DM and the two new congeners, which revealed minor anti-inflammatory activity. Glucocorticoid esters including PEP are rapidly degraded in keratinocytes (85% within 12 h). Hence, a ribonuclease protection assay of Il-1α mRNA was performed allowing short incubation times and thus minimizing biodegradation. This assay confirmed the anti-inflammatory potency of native PC and BMV. In contrary DCE and PEP did not reduce Il-1α mRNA to a significant extent. Therefore PEP acts as a prodrug only. In fibroblasts, Il-1α and Il-6 syntheses indicate proliferation and inflammation, respectively. Whereas PC and PEP inhibited Il-1α and Il-6 production in fibroblasts only to a minor extent, cytokine synthesis was strongly affected by the conventional glucocorticoids BMV and DM, but also by DCE. The minor unwanted effect of PC and PEP on fibroblasts was also reflected by their low influence on cell proliferation as derived from 3H-thymidine incorporation. Again, more pronounced antiproliferative features were seen with the halogenated glucocorticoids. In the following, the correlation between antiphlogistic effects in keratinocytes (suppression of Il-1α) and antiproliferative effects in fibroblasts (suppression of Il-1α and Il-6; 3H-thymidine incorporation) was analyzed. Here, PC is revealed as the only glucocorticoid with an improved benefit/risk ratio. Native PEP is shown to be almost ineffective and DCE presents exactly the opposite features of PC. It is tempting to speculate if this is due to different glucocorticoid receptor subtypes or different signaling pathways in keratinocytes and fibroblasts.