Abstract

Cyclooxygenase-2 (COX-2) oftentimes is highly expressed in cancer tissues, where it supports tumor development and angiogenesis. Over the past 15 years, newly developed non-steroidal anti-inflammatory drugs (NSAIDs) that are able to highly selectively inhibit this enzyme were hoped to become therapeutic tools for cancer prevention and therapy. However, while chemopreventive effects of certain selective COX-2 inhibitors indeed have been documented, their efficacy for therapy of already established cancers has been unimpressive so far. Intriguingly, the investigation of compounds such valdecoxib, rofecoxib, and in particular celecoxib, has revealed molecular targets besides COX-2, and it appears that some of these non-COX-2 targets may be critically involved in mediating the pro-apoptotic effects of these compounds without any apparent involvement of COX-2. In fact, investigations of a series of close structural analogs of celecoxib demonstrated that it is possible to separate COX-2 inhibitory function from apoptosis-stimulatory function within the molecule. For example, 2,5-dimethyl-celecoxib (DMC) has lost COX-2 inhibitory function, yet still exerts profound cytotoxic potency.

This review will summarize pertinent results from the exploratory therapeutic use of NSAIDs, in particular celecoxib, in preclinical and clinical studies of malignant glioma. Several COX-2 independent targets will be presented, and it will be discussed how DMC has helped to delineate their relevance for the surmised COX-2 independent tumoricidal effects of celecoxib.Angiogenesis, azetazolamide, carbonic anhydrase, celecoxib, diclofenac, 2,5-dimethyl-celecoxib (DMC), endoplasmic reticulum stress, etoricoxib, glucose-regulated protein 78 (GRP78), meloxicam, nimesulide, 3- phosphoinositide-dependent protein kinase-1 (PDK1), rofecoxib, temozolomide, valdecoxib.