Urinary bladder cancer (transitional cell carcinoma, TCC) affects more than 2 million people worldwide. Even with aggressive treatment, invasive TCC (InvTCC) is lethal in up to 50% of cases. A promising new target for InvTCC treatment is cyclooxygenase-2 (COX-2). COX-2 is up-regulated in InvTCC. COX inhibitors have antitumor activity against naturally occurring invasive TCC in pet dogs and in experimentally induced bladder tumors in rodents. Mechanisms of the antitumor activity of COX inhibitors, however, remain to be fully defined, and this prevents optimal use of these drugs. One reason why mechanisms have not been defined is that COX inhibiting drugs do not have antiproliferative effects in standard in vitro assays. The purpose of this work was to define different in vitro systems for studies of COX-2 inhibitor antitumor effects, and to study the mechanism of action of a COX-2 inhibitor, celecoxib. Human InvTCC cell lines used included: HT1376 (high COX-2 expression), TCC-SUP (modest COX-2 expression), and UMUC-3 (no detectable COX-2 expression). Assay variables tested included different microenvironment (monolayer, soft agar, collagen, and poly HEMA coated culture dishes); length of treatment (72-120 hrs); presence or absence of serum and fatty acids; and concentration of the COX-2 inhibitor, celecoxib. In addition, cell proliferation was “rescued” from COX-2 inhibitor effects by the addition of prostaglandin E2 (PGE2). Soft agar assays were found to be useful for the study of COX-2 inhibitors (38% reduction in colony number) with relevant celecoxib concentrations (i.e. those safely achieved in vivo in humans; < 5 uM). UMUC3 and TCCSUP colony formation was not inhibited by celecoxib in soft agar. Collagen assays were not found to be useful due to inability to recover and count cells. Poly Hema coated plate assays were not useful due to poor cell growth. Monolayer assays were found to be useful for detecting celecoxib antiproliferative effects under specific conditions (changing media and replenishing drug every 48 hours; continuing assay for 120 hours). Under these monolayer conditions, celecoxib (5 uM) inhibited proliferation by 30%, and this inhibition was completely reversed by the addition of prostaglandin E2 to the media, implying a COX-dependent effect. The celecoxib antiproliferative effects were accompanied by reduction in pAkt expression. At high celecoxib concentrations (25 uM; much higher than would be reached safely in vivo), substantial growth inhibition was observed in all 3 cell lines, and inhibition was not reversed by the addition of celecoxib (implying COX-independent effect). In conclusion, these studies defined two types of in vitro systems (specific conditions for monolayer assay, soft agar assay) which can be used to study COX inhibitor effects (at relevant drug concentrations) in vitro. Furthermore, in these systems, the COX-2 inhibitor, celecoxib, inhibited proliferation of HT1376 InvTCC cells in a COX-dependent fashion.

[Proc Amer Assoc Cancer Res, Volume 47, 2006]