Among NSAIDs Celecoxib is one of the most efficient in triggering in vitro cancer cell death, and from this perspective has been subject of numerous studies. However, it is still controversial whether this in vitro-observed effect can also occur in vivo and contribute to the antitumor action of the drug. Moreover, besides common agreement on the involvement of COX-independent pathways, the mechanisms underlying Celecoxib toxicity are still unclear. In an attempt to shed light on these mechanisms, I found that cell death only occurs at insoluble concentrations of the drug, and follows irreversible binding and damage of the plasmamembrane by precipitates. This evidence strongly suggests that Celecoxib is devoid of true molecular toxicity. Moreover, since plasma levels reached during therapy are far below the threshold of toxic precipitation, direct cytotoxicity by Celecoxib is unlikely to occur on tumor cells in vivo. Thus the antitumor effect might be only due to COX inhibition, which requires significantly lower levels of the drug. Nonetheless, direct cytotoxicity might not be confined to an in vitro artifact, but contribute to the upper gastrointestinal side effects of Celecoxib. Overall, these findings represent an important basis for further studies on Celecoxib, where true molecular actions of the drug should be discriminated from the precipitate-dependent ones, and the relationship between in vitro and in vivo effects considered at the light of the precipitate-dependent model. Moreover, remarkably, this article indicates a model of critical analysis that can be extended to other poorly soluble drugs. J. Cell. Biochem. 114: 1434–1444, 2013. © 2013 Wiley Periodicals, Inc.