High dose bicalutamide for androgen independent prostate cancer: effect of prior hormonal therapy.

PURPOSE: A pilot study of the antiandrogen bicalutamide at 150 mg. a day for androgen independent prostate cancer was performed. This study was based on the possibility that androgen independent cases might display responses to additional hormonal agents. MATERIALS AND METHODS: The study included 31 androgen independent cases with an increasing prostate specific antigen (PSA) and progressive disease. PSA measurements were used as the primary method of assessing response. However, PSA decline was also correlated with clinical status. RESULTS: Seven patients demonstrated PSA declines of greater than 50% for 2 months or more, for an overall response rate of 22.5%. Responses were observed almost exclusively in patients treated with long-term flutamide as part of a complete androgen blockade regimen (43% response rate) in contrast to patients treated with androgen deprivation without flutamide (6% response rate). Of the 7 PSA responding patients bicalutamide resulted in a significant improvement in performance status and a decrease in analgesic requirement in 4 and 3 remained asymptomatic. Bicalutamide at 150 mg. a day was well tolerated, with the most frequent side effect being mild exacerbation of hot flashes. CONCLUSIONS: Bicalutamide at this dose is modestly effective for some patients with androgen independent prostate cancer, particularly for those previously treated with long-term flutamide. This study indicates that previous antiandrogen therapy alters the response to subsequent hormonal agents.

High-dose cyclophosphamide, carmustine, and etoposide with autologous transplantation in Hodgkin's disease: a prognostic model for treatment outcomes.

PURPOSE: To identify clinical factors predictive of treatment outcome after high-dose chemotherapy (HDC) for Hodgkin's disease and to develop a prognostic model for progression-free and overall survival. PATIENTS AND METHODS: 102 patients with relapsed or refractory Hodgkin's disease were treated with high-dose cyclophosphamide, carmustine, and etoposide and autologous marrow and/or peripheral blood progenitor cell support. Median follow-up of survivors is 4.1 years (1.8-7.5 years). Factors potentially important for treatment outcome were examined in univariate analysis, and Cox regression with forward selection was performed. A prognostic model was developed. RESULTS: Poorer progression-free and overall survival were associated with nodular sclerosis histology, abnormal performance status, progressive disease at HDC, more than one extranodal site of disease, and shorter time from initial diagnosis to HDC. These factors and the presence of B symptoms at relapse also predicted for decreased overall survival. Progressive disease immediately prior to HDC, more than one extranodal disease site, and abnormal performance status retained significance for both progression-free and overall survival in multivariate analysis. Progression-free and overall survival are 42% (95% confidence interval, CI, 34 to 53) and 65% (95% CI 54 to 73) at three years. A model based on number of risk factors present divides patients into low, intermediate, and high risk groups with three-year actuarial survival of 82%, 56%, and 19% respectively. Treatment outcome for patients treated with HDC at first chemotherapy relapse was not significantly different from that of the group overall (p > 0.3). CONCLUSIONS: Asymptomatic patients with Hodgkin's disease involving at most one extranodal site whose disease is controlled by conventional dose chemotherapy or radiation therapy at the time of HDC have good outcomes after this therapy. Presence of increasing numbers of risk factors are associated with poorer outcomes. Results of HDC compare favorably to those of standard dose salvage therapy. These data can be used to estimate likely outcomes in patients undergoing HDC for Hodgkin's disease, to identify potential candidates for innovative therapies, and to evaluate strategies for the optimal use of HDC in Hodgkin's disease.

Preapheresis peripheral blood CD34+ mononuclear cell counts as predictors of progenitor cell yield.

BACKGROUND: Peripheral blood progenitor cells, harvested by apheresis after mobilization, provide rapid hematologic recovery after high-dose chemotherapy. However, because harvesting these cells is expensive and time-consuming, there has been much interest in optimizing collection protocols. An investigation was made to determine whether, in this clinical setting, peripheral blood progenitor cell yields may be predicted from preapheresis progenitor cell counts, allowing the length of each procedure to be "fine tuned" to achieve specific target goals. STUDY DESIGN AND METHODS: Preapheresis peripheral blood CD34+ cell and total colony-forming cell counts were assessed before 78 peripheral blood progenitor cell collections from 13 consecutive patients were performed. Preapheresis counts were correlated with actual progenitor cell yields. Factors affecting this correlation were analyzed. RESULTS: With the use of linear regression analysis preapheresis progenitor cell counts were found to correlate significantly but weakly with actual yields per kg of body weight per liter of blood processed (CD34+ cells: r = 0.43; colony-forming cells: r = 0.56). Further analysis revealed two possible causes: 1) circulating progenitor cell concentrations fluctuate widely during harvest, which implies that preapheresis counts are not representative of actual concentrations during apheresis, and 2) the efficiency with which apheresis machines extract mononuclear cells varies greatly between procedures. CONCLUSION: Preapheresis CD34+ and colony-forming cell counts correlated poorly with subsequent yields in this clinical setting, which suggests that it is not practical to use such counts to predict with certainty the length of apheresis needed to achieve a target yield.

Detection of circulating tumor cells in men with localized prostate cancer.

PURPOSE: Using prostate-specific antigen (PSA) mRNA as a marker for prostatic epithelial cells, we have developed a sensitive technique that involves reverse transcription and polymerase chain reaction (RT-PCR) to detect circulating tumor cells in the peripheral blood of men with prostatic carcinoma (CaP). PATIENTS AND METHODS: A sensitive RT-PCR assay was used to evaluate the peripheral blood of 135 men with a history of CaP. Fourteen men with benign prostate disease, many of whom had elevated serum PSA levels, were used as a control group. RESULTS: All patients with benign prostate disease had a negative result in the RT-PCR assay. Of particular interest was a subgroup of 65 patients with clinically localized CaP evaluated before definitive local therapy. Five of these patients had detectable PSA mRNA by RT-PCR, suggesting circulating tumor cells. Within this group, systemic disease was detected by RT-PCR in some men with PSA levels less than 10 ng/mL and clinical stage B disease. Blood from men with hormone-refractory and progressive CaP demonstrated a higher frequency of PSA mRNA detectable by RT-PCR (10 of 20 patients). In contrast, none of seven patients with newly diagnosed metastatic prostate cancer and only one of seven patients with metastatic, hormone-responsive disease had blood that was positive for PSA mRNA by RT-PCR. CONCLUSION: Circulating tumor cells can be detected in the blood of a subset of patients with clinically localized CaP and a larger subset of patients with progressive metastatic disease.

Heterogeneity of binding subunits of the human 150K insulin-like growth factor binding protein.

Models for the structure of the GH-dependent 150K insulin-like growth factor-binding protein (IGF-BP) complex include 1) a binding subunit of 40-60K mol wt associated with a larger nonbinding component, and 2) an oligomeric structure simply made up of six 25-28K monomeric IGF-BP complexes. To evaluate these alternative models we examined the IGF-binding characteristics and behavior on an SP-Sephadex ion exchange column of BP species identified by chemically cross-linking [125I]IGF-I and [125I]IGF-II. In addition, human serum was gel filtered on Sephadex G-200 in 0.05 M NH4HCO3, pH 8.0, and the 150K BP identified by binding of [125I]IGF-II to column fractions. When [125I]IGF-I or [125I]IGF-II was cross-linked to the 150K BP with disuccinimidyl suberate and analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (10-15%) and autoradiography, four specifically labeled complexes of 20K, 24K, 33K, and 47K mol wt were identified. We examined the IGF-binding characteristics of these species by cross-linking [125I]IGF-I and [125I]IGF-II after incubation in the presence of increasing concentrations of unlabeled IGF-I or IGF-II. Formation of the 24K complex was inhibited more potently by IGF-II than IGF-I, whereas the relative potency of IGF-I vs. IGF-II for inhibition of the formation of the other complexes depended upon whether [125I]IGF-II or [125I]IGF-I was used. When the 150K BP complex generated from gel filtration on Sephadex G-200 was acid stripped, the only species seen with chemical cross-linking of either [125I]IGF-I or [125I]IGF-II was the 47K complex. By both conventional competitive binding studies and cross-linking [125I]IGF-I and [125I]IGF-II after incubation with increasing concentrations of unlabeled IGF-I or IGF-II, the formation of the 47K complex was usually more potently inhibited by IGF-I than IGF-II. When Cohn fraction IV extract was chromatographed on a SP-Sephadex column (pH 3) and cross-linking performed on the flow-through, the 47K species was intensely labeled, and the 20K, 33K, and 24K complexes were weakly labeled or not seen at all. The 20K, 33K, and 24K complexes could be identified after cross-linking [125I]IGF-II to the pH 7 eluate from the ion exchange column, but the 47K complex was not seen. These data provide additional physical evidence for separate binding sites with different relative affinities for IGF-I and IGF-II within the 150K BP complex.