Tamoxifen sensitivity-testing of glioblastomas: Comparison of in vitro andin vivo results

Background. Only less than half of the patients with malignant gliomas resp ond to a continuous high dose Tamoxifen (TAM) and/or Carboplatin (CP)-treat ment. Therefore, a method for predicting the efficacy of TAM-treatment woul d be desirable. Methods. Paralleling a clinical study, the predictive value of in vitro-sen sitivity testing of TAM and TAM's metabolite 4-OH-TAM in primary cultures o f tumour explants from 15 of a total of 50 patients was examined. Additiona lly, the influence of TAMI 4-OH-TAM, and CP on the proliferation of establi shed glioblastoma cell lines and of those explanted from athymic nude mice and reestablished in cell culture was investigated. Human glioblastomas xen otransplanted subcutaneously into athymic nude mice and subsequently treate d with TAM and/or CP were examined in a parallel in vivo-study. Findings. TAM-chemosensitivity-testing of glioblastomas failed to predict t he clinical response to TAM-treatment in our patients and did not correlate with the in vivo-TAM-response of tumours xenotransplanted into nude mice. TAM's and 4-OH-TAM's ability to inhibit growth of various glioblastoma cell lines in vitro in very similar concentrations was shown to be a consistent phenomenon which seems to be independent of the in vivo response in either patients or mice as previous hosts. However, CP's antiproliferative effect on glioblastomas in vivo was paralleled by respective in vitro results. Wh ereas TAM showed to mediate its in vitro antiproliferative effect by induci ng apoptosis in most cell lines examined, CP-treatment lead to necrosis of cells. Interpretation. Combining the results obtained from our human and mouse stu dies. it has to be postulated that host factors other than the sensitivity to TAM of the individual cell, determine the efficacy of TAM-treatment in v ivo.