Design and analysis of clinical trials in the presence of delayed treatment effect.

In clinical trials with survival endpoint, it is common to observe an overlap between two Kaplan-Meier curves of treatment and control groups during the early stage of the trials, indicating a potential delayed treatment effect. Formulas have been derived for the asymptotic power of the log-rank test in the presence of delayed treatment effect and its accompanying sample size calculation. In this paper, we first reformulate the alternative hypothesis with the delayed treatment effect in a rescaled time domain, which can yield a simplified sample size formula for the log-rank test in this context. We further propose an intersection-union test to examine the efficacy of treatment with delayed effect and show it to be more powerful than the log-rank test. Simulation studies are conducted to demonstrate the proposed methods.

A phase I study of PF-04929113 (SNX-5422), an orally bioavailable heat shock protein 90 inhibitor, in patients with refractory solid tumor malignancies and lymphomas.

PURPOSE: To determine the maximum tolerated dose (MTD), toxicities, and pharmacokinetic/pharmacodynamic profile of the Hsp90 inhibitor PF-04929113 (SNX-5422) in patients with advanced solid tumors and lymphomas. METHODS: This was a single-institution, phase I, dose-escalation study of PF-04929113 administered twice weekly. Endpoints included determination of dose-limiting toxicities (DLT), MTD, the safety profile of PF-04929113, pharmacodynamic assessment of PF-04929113 on Hsp70 induction, pharmacokinetic analysis of PF-04928473 (SNX-2112) and its prodrug PF-04929113, and assessment of response. RESULTS: Thirty-three patients with advanced malignancies were treated. Dose escalation was continued up to 177 mg/m(2) administered orally twice a week. One DLT (nonseptic arthritis) was noted. No grade 4 drug-related adverse events were seen; grade 3 adverse events included diarrhea (9%), nonseptic arthritis (3%), aspartate aminotransferase elevation (3%), and thrombocytopenia (3%). No objective responses were seen in 32 evaluable patients. Fifteen patients (47%) had stable disease; 17 patients (53%) had progressive disease. Pharmacokinetic data revealed rapid absorption, hepatic, and extrahepatic clearance, extensive tissue binding, and almost linear pharmacokinetics of the active drug PF-04928473. Pharmacodynamic studies confirmed inhibition of Hsp90 and a linear correlation between pharmacokinetic parameters and Hsp70 induction. CONCLUSIONS: PF-04929113 administered orally twice a week is well tolerated and inhibits its intended target Hsp90. No objective responses were seen, but long-lasting stabilizations were obtained. Although no clinically significant drug-related ocular toxicity was seen in this study, the development of PF-04929113 has been discontinued because of ocular toxicity seen in animal models and in a separate phase I study.

p53 autoantibodies predict subsequent development of cancer.

Because TP53 mutations can induce an immune response and can occur early in the carcinogenic process for some tumors, p53 autoantibodies may be useful biomarkers for risk of development of cancer. Using banked serum samples from an asbestosis cohort at high risk for cancer, we demonstrate for the first time a statistically significant relationship between p53 autoantibodies and the subsequent development of malignancy (hazard ratio [HR] = 5.5, 95% confidence interval [CI] = 2.8-10.9) with a positive predictive value of 0.76 and an average lead time to diagnosis of 3.5 years. p53 autoantibodies were also significantly associated with p53 alterations in the resultant tumors (kappa = 0.78, p = 0.01).

Bayesian analysis of serial dilution assays.

In a serial dilution assay, the concentration of a compound is estimated by combining measurements of several different dilutions of an unknown sample. The relation between concentration and measurement is nonlinear and heteroscedastic, and so it is not appropriate to weight these measurements equally. In the standard existing approach for analysis of these data, a large proportion of the measurements are discarded as being above or below detection limits. We present a Bayesian method for jointly estimating the calibration curve and the unknown concentrations using all the data. Compared to the existing method, our estimates have much lower standard errors and give estimates even when all the measurements are outside the "detection limits." We evaluate our method empirically using laboratory data on cockroach allergens measured in house dust samples. Our estimates are much more accurate than those obtained using the usual approach. In addition, we develop a method for determining the "effective weight" attached to each measurement, based on a local linearization of the estimated model. The effective weight can give insight into the information conveyed by each data point and suggests potential improvements in design of serial dilution experiments.