C. Logothetis

To evaluate the relative efficacy of cisplatin, cyclophosphamide, and Adriamycin (doxorubicin; Adria Laboratories, Columbus, OH) (CISCA) versus methotrexate, vinblastine, Adriamycin, and cisplatin (MVAC), a prospective randomized trial was performed in patients with advanced metastatic urothelial tumors. Patients were stratified by histologic disease type and degree of tumor dissemination. Equal distribution of the clinical characteristics was achieved. One hundred ten patients with metastatic disease of the urinary tract (86 bladder, 16 renal pelvis, seven ureter, one prostatic urethra) met eligibility criteria and were enrolled on study. These represented 82% of the total patients seen during the study period in the Section of Genitourinary Oncology who met the eligibility criteria. The combined complete and partial response rate was significantly higher for patients treated with MVAC than for those treated with CISCA (65% v 46%; P less than .05). The survival duration of MVAC-treated patients was significantly longer than that of CISCA-treated patients (mean, 62.6 weeks; median, 48.3; range, 5.0+ to 162.3+ v mean, 40.4 weeks; median, 36.1; range, 7+ to 147.1+). We conclude that MVAC chemotherapy is superior to CISCA chemotherapy, achieving a higher response rate and a longer survival for equivalent patients with metastatic urothelial tumors.

BACKGROUND: Expression of platelet-derived growth factor (PDGF) and activation (by autophosphorylation) of its receptor (PDGF-R), a tyrosine kinase, are associated with the growth of metastatic prostate tumor cells in the bone parenchyma. The tyrosine kinase inhibitor STI571 blocks the PDGF signaling pathway by inhibiting PDGF-R autophosphorylation. We examined the effects of STI571, given alone or with paclitaxel (Taxol), on tumor growth in a mouse model of prostate cancer metastasis. METHODS: Human prostate cancer PC-3MM2 cells were injected into the tibias of male nude mice. Three days later the mice (20 per group) were randomly assigned to 5 weeks of treatment with oral and injected water (control), daily oral STI571, weekly injected paclitaxel, or STI571 plus paclitaxel. Lesions in bone and the surrounding muscles were then harvested and analyzed by histology, western blotting (for PDGF-R phosphorylation), immunohistochemistry (for expression of proangiogenic molecules), and double immunofluorescence (to identify endothelial cells and apoptotic tumor cells). Growth of bone lesions was monitored by digital radiography. Bone lesions from control mice were used to establish short-term cell cultures for analysis of PDGF-R phosphorylation. All statistical tests were two-sided. RESULTS: PC-3MM2 cells cultured from bone lesions and treated in vitro with STI571 had less phosphorylated PDGF-R than untreated cells. In control mice, bone lesions expressed high levels of PDGF and activated (i.e., phosphorylated) PDGF-R, whereas lesions in the adjacent musculature did not. Activated PDGF-R was present on the surface of endothelial cells within the bone lesions but not in endothelial cells of uninjected bone. Mice treated with STI571 or STI571 plus paclitaxel had a lower tumor incidence, smaller tumors, and less bone lysis and lymph node metastasis than mice treated with water or paclitaxel alone (P<.001 for all). Mice treated with STI571 or STI571 plus paclitaxel had less phosphorylated PDGF-R on tumor cells and tumor-associated endothelial cells, less tumor cell proliferation, statistically significantly more apoptotic tumor cells (all P<.001), and fewer tumor-associated endothelial cells (P<.001) than control mice. CONCLUSIONS: Endothelial cells appear to express phosphorylated PDGF-R when they are exposed to tumor cells that express PDGF. Using STI571 to inhibit PDGF-R phosphorylation may, especially in combination with paclitaxel, produce substantial therapeutic effects against prostate cancer bone metastasis.

BACKGROUND: The clinical behavior of the tumor in patients with locally advanced bladder carcinoma is unpredictable. Current predictors of clinical behavior include depth of muscle invasion, presence of vascular invasion, proliferation rate, and loss of blood group antigens. Treatment selection would be facilitated by the development of a reliable marker of tumor progression. Functional retinoblastoma (RB) gene loss has been reported to occur in bladder carcinoma, but the significance of this loss is unknown. PURPOSE: We have evaluated the frequency of functional loss of the RB gene in locally advanced bladder carcinoma and have compared the results to known prognostic factors in the same cohort. METHODS: Forty-three study patients with pathologically well-characterized, locally advanced bladder carcinoma, who were placed in a protocol incorporating surgery and chemotherapy, were studied for known clinical and pathological prognostic indicators as well as for their Rb status. Formalin-fixed and paraffin-embedded archival primary tumor tissues were used for histological and immunohistochemical analyses. RESULTS: Altered Rb protein expression was documented in 37% of the tumor specimens. The high rate of altered Rb expression found in this cohort with advanced urothelial tumors strongly suggests that RB functional loss may be associated with tumor progression in this malignancy. Altered Rb protein expression was found to be independent of other known prognostic variables. A significantly poorer tumor-free survival rate also was noted for those patients who had a tumor with an altered Rb protein with or without vascular invasion. CONCLUSION: The high frequency of Rb alteration in locally advanced bladder carcinomas, plus the fact that a significant correlation could not be found between the Rb status and other known prognostic markers in this preliminary study, suggests that altered RB expression may be an independent prognostic marker of tumor progression in bladder cancer.

Seventy-one patients received adjuvant Cytoxan (cyclophosphamide; Bristol-Myers Co, Evansville, IN), Adriamycin (doxorubicin; Adria Laboratories, Columbus, OH), and cisplatin (CISCA) chemotherapy between March 1981 and March 1986. Patients received adjuvant CISCA chemotherapy if they had pathological findings that were thought to predict for high likelihood of relapse. These included the presence of resected nodal metastases, extravesicular involvement of tumor, lymphatic/vascular permeation of the primary tumor, or pelvic visceral invasion. Sixty-two patients at a similar high risk for recurrence did not receive adjuvant CISCA chemotherapy because they refused, had medical contraindications to therapy, or were not referred for chemotherapy. Two-hundred six patients had a cystectomy performed during the same study period but had none of the poor prognostic features suggesting a high risk for relapse. Sixty-two percent of the patients receiving adjuvant chemotherapy are alive and disease-free for a mean follow-up of 118 weeks (range, 28 to 310 weeks). A survival advantage exists for the adjuvant-treated patients when compared with those with unfavorable pathological findings who did not receive adjuvant chemotherapy (70% v 37%) (P = .00012): no difference exists in long-term disease-free survival for those with favorable pathological findings (long-term disease-free survival 76%) v those who received adjuvant chemotherapy (70%) (P = .33). Adjuvant CISCA chemotherapy prolongs the disease-free survival of some patients following a cystectomy. Patients who benefitted from adjuvant CISCA chemotherapy included those with resected nodal metastases, extra-vesicular involvement of tumor, and direct invasion of the pelvic viscera. Patients not benefitting from adjuvant CISCA chemotherapy in this analysis included those with lymphatic/vascular invasion in their primary tumor as the sole manifestation of high risk for relapse.