Development of a panel of 15 human ovarian cancer xenografts for drug screening and determination of the role of the glutathione detoxification system

Abstract

Objectives: We have established a panel of 15 human ovarian cancer xenografts grown subcutaneously in the flank of the nude mouse. Similar to the clinic, the xenografts show differences in histological subtype and volume doubling time. We determined whether the panel is useful for drug screening by testing the sensitivity to six conventional anticancer agents. In addition, we investigated whether the glutathione detoxification system affects sensitivity to cisplatin and cyclophosphamide, major drugs in the treatment of ovarian cancer.

Methods: Mice bearing well-established tumors were treated at maximum tolerated doses as defined by a reversible weight loss up to 15% of their initial weight: cisplatin 5 mg/kg iv weekly x2, cyclophosphamide 150 mg/kg ip 2-weekly x2, doxorubicin 8 mg/kg iv weekly x2, hexamethylmelamine ip 150 mg/kg every other day x4, methotrexate ip 150 mg/kg weekly x2, and 5-fluorouracil 60 mg/kg ip weekly x4. Glutathione levels and the activities of three different glutathione-dependent enzymes were measured in untreated xenograft tissues.

Results: Growth inhibition >75% was reached for cisplatin in 40%, for cyclophosphamide in 27%, and for doxorubicin in 20% of the xenografts. Methotrexate and 5-fluorouracil did not induce growth inhibition of importance. Hexamethylmelamine showed >75% growth inhibition in 53% of the xenografts, which may have been caused by the favorable metabolism of the drug in mice when compared with that in patients. Glutathione levels varied 3.6-fold in the xenografts and did not show a relation with sensitivity to cisplatin, cyclophosphamide, or doxorubicin. No relation was found between the activities of glutathione S-transferase and glutathione peroxidase and the sensitivities to the three anticancer agents. Glutathione reductase activity, however, showed a weak, inverse relation with the efficacy of cisplatin and cyclophosphamide (r values of -0.55 and -0.58, respectively).

Conclusions: The sensitivity to the six anticancer agents of our panel of 15 human ovarian cancer xenografts reflects the response rates known for similar drugs in ovarian cancer patients. In that respect, the panel may be useful for drug screening as well as studies on the relevance of drug resistance features in vivo. The various components of the glutathione detoxification system did not predict for primary drug resistance which confirms clinical data in ovarian cancer.