Analyses of Dose-Response and Mechanistic Action of Different Anti-Cancer Drugs for Neuroendocrine Tumor Cell Lines

Sammanfattning: Cancer is a disease with poor response rates on available treatments. Problems with resistance and intolerance against cancer drugs are major reasons for failure of the drugs. The need to discover new cancer drugs is important. In this thesis screening of new cancer drugs and evaluation of their mechanism of action are discussed. The aim of the thesis was to find new compounds active against neuroendocrine tumors (NETs). In paper I, we screened 1280 substances on two bronchial carcinoid cell lines and one pancreatic carcinoid cell line. Eleven of these compounds were found to have antitumor activity at low concentrations. The most active agents were brefeldin A, emetine, bortezomib and idarubicin, having IC50 values (the concentration of the drug where > 50% of the cells die) < 1μM. In addition, sanguinarine, Bay11-7085, mitoxantrone, doxorubicin, β-lapachone, NSC 95397 and CGP- 74514A were active with IC50 values < 10 μM.  In paper II, additional studies have been undertaken to investigate the combination effect of the most active drugs with conventional cytotoxic drugs used in clinical practice. If synergistic or additive effects are found, drugs with different mechanism of action and toxicity profiles may be combined, making it possible to reduce the toxic effects yet maintaining the antitumor activity. In paper III, studies were undertaken to find the mechanistic action, apoptosis or necrosis, of the drugs NSC 95397, brefeldin A, bortezomib and sanguinarine in NETs. All four drugs were shown to induce caspase-3 activity and nuclear fragmentation/condensation in the neuroendocrine tumor cell lines, indicating that their antitumor activity was induction of apoptosis. In paper IV, the mechanism of action was studied for CGP-74514A and emetine. Both drugs worked by induction of apoptosis. In addition, their cytotoxic activity was studied in a three-dimensional model, the in vitro hollow fiber model. The Hollow Fiber model permits more realistic simulation of in vivo drug effects in a controlled system providing data that more accurately reflects biological responses. Our results showed that the hollow fiber model may be suitable for studies of new drugs in the neuroendocrine tumor cell lines.