George Y. Wong

Effect of granulocyte colony-stimulating factor on neutropenia and associated morbidity due to chemotherapy for transitional-cell carcinoma of the urothelium

We evaluated the ability of human recombinant granulocyte colony-stimulating factor (rhG-CSF) to prevent chemotherapy-induced neutropenia or to accelerate recovery from this complication and thus allow patients to receive full doses of antineoplastic agents on time, according to protocol design. Twenty-seven patients with transitional-cell carcinoma of the urothelium who were undergoing treatment with methotrexate, doxorubicin, vinblastine, and cisplatin were given rhG-CSF (up to 60 micrograms per kilogram of body weight per day) before their first cycle of combination chemotherapy, during the first cycle, or at both points. Treatment with rhG-CSF before chemotherapy resulted in a dose-dependent increase in the absolute neutrophil count. Treatment with rhG-CSF after chemotherapy significantly reduced the number of days (91 percent) per patient on which the absolute neutrophil count was 1000 per microliter or less (P = 0.0039), reduced the number of days (1 vs. 35) on which antibiotics were used to treat fever and neutropenia, and significantly increased the percentage (100 vs. 29 percent) of patients qualified to receive planned chemotherapy on day 14 of the treatment cycle (P = 0.0015). In addition, the incidence of mucositis was significantly decreased (11 vs. 44 percent, P = 0.041), as was its severity. These findings demonstrate that rhG-CSF is a potent stimulus of normal neutrophil proliferation and maturation. In addition, its administration can reduce both the hematopoietic and oral toxicity of chemotherapy.

The Hierarchical Logistic Regression Model for Multilevel Analysis

Abstract A hierarchical logistic regression model is proposed for studying data with group structure and a binary response variable. The group structure is defined by the presence of micro observations embedded within contexts (macro observations), and the specification is at both of these levels. At the first (micro) level, the usual logistic regression model is defined for each context. The same regressors are used in each context, but the micro regression coefficients are free to vary over contexts. At the second level, the micro coefficients are treated as functions of macro regressors. An empirical Bayes estimation procedure is proposed for estimating the micro and macro coefficients. Explicit formulas are provided that are computationally feasible for large-scale data analyses; these include an algorithm for finding the maximum likelihood estimates of the covariance components representing within— and between—macro-equation error variability. The methodology is applied to World Fertility Survey data...

Contextual analysis through the multilevel linear model.

A general linear multilevel model and its estimation are described and illustrated empirically. The specification of a multilevel linear model within the covariance component framework was rendered. This model is appropriate for a wide range of applications in contextual analysis because it allows for macro as well as micro errors. Then an estimation procedure the restricted maximum likelihood/Bayes (REML/Bayes) was proposed and described for this multilevel model. Finally an extended empirical example was presented with the primary purpose of illustrating the use of the statistical methodology in conjunction with a meaningful substantive problem not to carry out a critical test of the underlying substantive theory and not to demonstrate the general superiority of the proposed estimation procedure. The example suggests that the REML/Bayes estimation procedure may be appropriate inasmuch as the results extracted with REML/Bayes are more consistent with theoretical anticipations than those extracted with ordinary least squares or estimated generalized least squares. The methodology presented by no means exhausts the subject of multilevel estimation. There is a need for estimation procedures to handle discrete micro response variables systems of micro-structural equations and other generalizations. Work is currently proceeding on some of these extensions of the multilevel linear model as the goal of sound estimation procedures for generalized multilevel models is worthwhile and by no means esoteric. It may be hard for researchers to find contextual effects unless they use efficient and appropriate estimation techniques.

Stochastic Blockmodels for Directed Graphs

Abstract Holland and Leinhardt (1981) proposed the p 1 model for the analysis of binary directed graph data in network studies. Such a model provides information about the “attractiveness” and “expansiveness” of the individual nodes in the network, as well as the tendency of a pair of nodes to reciprocate relational ties. When the nodes are a priori partitioned into subgroups based on attributes such as race and sex, the density of ties from one subgroup to another can differ considerably from that relating another pair of subgroups, thus creating a situation called blocking in social networks. The p 1 model completely ignores this extra piece of information and is, therefore, unable to explain the block structure. Blockmodels that are simple extensions of the p 1 model are proposed specifically for such data. An iterative scaling algorithm is presented for fitting the model parameters by maximum likelihood. The methodology is illustrated in detail on two empirical examples.

Prognostic factors in patients with metastatic malignant melanoma: A multivariate analysis

Background. Current methods to predict survival in patients with advanced, metastatic melanoma are limited. To determine clinical prognostic factors that accurately predict survival in patients with metastatic melanoma, a retrospective analysis was performed.

Current Management of Male Breast Cancer

Between 1975 and 1990, 104 male patients with a total of 106 breast cancers were treated at Memorial Hospital or the Ochsner Clinic and their records reviewed. The patients were followed for a median of 67 months (range, 0.5 to 14.4 years). Analysis of the frequency distribution by stage showed that 16 (17%) patients were stage 0 and 26 (27%) patients were stage I. The median duration of symptoms before diagnosis was 18 weeks (mean, 5 weeks; range, 1 to 156 weeks). Modified radical mastectomy was undertaken in 71 (67%) patients. The actuarial 5-year relapse-free survival for the entire group was 68% and the actuarial 5-year overall survival was 85%. Relapse-free survival at 5 years for axillary node-negative patients was 87% and for node-positive patients was 30% (p

Ototoxicity of high-dose cisplatin by bolus administration in patients with advanced cancers and normal hearing.

Our institution undertook a phase I trial to define the toxicity of high‐dose (150 to 225 mg) bolus administration (every 3 to 4 weeks) of cisplatin in patients with advanced cancers. All patients reported had baseline normal hearing. Hearing levels were measured prior to each course of chemotherapy. Audiological monitoring included conventional assessment of pure tone sensitivity at 500 to 8,000 Hz and assessment of ultra high frequencies (9,000 to 20,000 Hz). After one to two doses, 100% of patients failed to respond at 9,000 Hz and above. In the 2,000 to 8,000 Hz range, repeated administration of the drug effected successively lower frequencies with progressive loss, until a maximum threshold shift or plateau was reached at each frequency between 3,000 and 8,000 Hz. The plateau for cisplatin ototoxicity appears to fall within the moderate hearing loss range (40 to 60 dB HL) in the high frequencies. All patients complained of tinnitus and difficulty understanding speech in the presence of background noise. The pattern of pure tone audiometric alteration is consistent in all patients, all dosages, and each method of administration. The ultra high frequency alteration is prompt and dramatic.

Dose-ranging study of Indole-3-Carbinol for breast cancer prevention

Sixty women at increased risk for breast cancer were enrolled in a placebo‐controlled, double‐blind dose‐ranging chemoprevention study of indole‐3‐carbinol (I3C). Fifty‐seven of these women with a mean age of 47 years (range 22–74) completed the study. Each woman took a placebo capsule or an I3C capsule daily for a total of 4 weeks; none of the women experienced any significant toxicity effects. The urinary estrogen metabolite ratio of 2‐hydroxyestrone to 16α‐hydroxyestrone, as determined by an ELISA assay, served as the surrogate endpoint biomarker (SEB). Perturbation in the levels of SEB from baseline was comparable among women in the control (C) group and the 50, 100, and 200 mg low‐dose (LD) group. Similarly, it was comparable among women in the 300 and 400 mg high‐dose (HD) group. Regression analysis showed that peak relative change of SEB for women in the HD group was significantly greater than that for women in the C and LD groups by an amount that was inversely related to baseline ratio; the difference at the median baseline ratio was 0.48 with 95 % confidence interval (0.30, 0.67). No other factors, such as age and menopausal status, were found to be significant in the regression analysis. The results in this study suggest that I3C at a minimum effective dose schedule of 300 mg per day is a promising chemopreventive agent for breast cancer prevention. A larger study to validate these results and to identify an optimal effective dose schedule of I3C for long‐term breast cancer chemoprevention will be necessary. J. Cell. Biochem. Suppls. 28/29:111–116.

Prognostic significance of HER-2/neu expression in advanced epithelial ovarian cancer: A multivariate analysis†‡

Objective: We tested the hypothesis that there is prognostic significance to the level of expression of the protooncogene HER-2/ neu in advanced ovarian cancer, as prior studies have suggested. Study Design: We determined expression of HER-2/ neu by immunohistochemistry, with monoclonal antibody 9G6 and the indirect immunoperoxidase technique, on frozen tumor specimens from 105 patients with stage III or IV epithelial ovarian cancer. All patients were treated at Memorial Sloan-Kettering Cancer Center, and no patient was lost to follow-up. Median follow-up among surviving patients is 34 months. HER-2/ neu expression was scored as negative, weak, 1+, 2+, or 3+. The staining pattern of normal ovarian epithelium was scored negative to 1+. Multivariate analysis was performed to evaluate the prognostic significance of HER-2/ neu expression. Results: Twenty-five of the 105 patients (24%) showed strong membrane staining (3+); the other tumor specimens showed weaker membrane staining or no immunoreactivity. There was no correlation of HER-2/ neu expression with any of a variety of clinical factors, including stage, grade, cell type, and residual tumor. No significant survival difference was found between patients with levels of staining intensity similar to those of normal ovarian epithelium and those with increased expression (3+). Median survival times were 36 and 27 months, respectively, for the two groups (95% confidence intervals 29 to 45 and 18 to 39 months). Multivariate analysis of possible prognostic factors showed that HER-2/ neu overexpression conferred a marginal worsening of survival ( p = 0.09) for the subgroup of patients in whom a negative surgical reassessment was not achieved after chemotherapy. Conclusion: Her-2/ neu expression does not appear to be an important prognostic factor in patients with advanced epithelial ovarian cancer.

Treatment of Superior Sulcus Tumor (Pancoast Tumor)

It appears that combined preoperative radiation and surgery continue to offer the best survival results in patients with superior sulcus tumors. Patients with involvement of the brachial plexus, Horners syndrome, rib invasion, and ipsilateral neck node metastases are still candidates for combined modality therapy, with expectations of survival of about 30 to 40 per cent. However, those presenting with invasion of vertebrae, involvement of subclavian vessels, and mediastinal lymph node metastases do poorly. In this latter group, treatment by high-dose external radiation alone may prove to be as effective as combined modality treatment.