To Estimate or to Forecast? Lessons From a Comparative Analysis of Four Bayesian Fitting Methods Based on Nonparametric Models.

ABSTRACT: Using pharmacokinetic (PK) models and Bayesian methods in dosing software facilitates the analysis of individual PK data and precision dosing. Several Bayesian methods are available for computing Bayesian posterior distributions using nonparametric population models. The objective of this study was to compare the performance of the maximum a posteriori (MAP) model, multiple model (MM), interacting MM (IMM), and novel hybrid MM(HMM) in estimating past concentrations and predicting future concentrations during therapy. Amikacin and vancomycin PK data were analyzed in older hospitalized patients using 2 strategies. First, the entire data set of each patient was fitted using each of the 4 methods implemented in BestDose software. Then, the 4 methods were used in each therapeutic drug monitoring occasion to estimate the past concentrations available at this time and to predict the subsequent concentrations to be observed on the next occasion. The bias and precision of the model predictions were compared among the methods. A total of 406 amikacin concentrations from 96 patients and 718 vancomycin concentrations from 133 patients were available for analysis. Overall, significant differences were observed in the predictive performance of the 4 Bayesian methods. The IMM method showed the best fit to past concentration data of amikacin and vancomycin, whereas the MM method was the least precise. However, MM best predicted the future concentrations of amikacin. The MAP and HMM methods showed a similar predictive performance and seemed to be more appropriate for the prediction of future vancomycin concentrations than the other models were. The richness of the prior distribution may explain the discrepancies between the results of the 2 drugs. Although further research with other drugs and models is necessary to confirm our findings, these results challenge the widely accepted assumption in PK modeling that a better data fit indicates better forecasting of future observations.

SIMulation of Medication Error induced by Clinical Trial drug labeling: the SIMME-CT study.

OBJECTIVE: To assess the impact of investigational drug labels on the risk of medication error in drug dispensing. DESIGN: A simulation-based learning program focusing on investigational drug dispensing was conducted. SETTING: The study was undertaken in an Investigational Drugs Dispensing Unit of a University Hospital of Lyon, France. PARTICIPANTS: Sixty-three pharmacy workers (pharmacists, residents, technicians or students) were enrolled. INTERVENTION: Ten risk factors were selected concerning label information or the risk of confusion with another clinical trial. Each risk factor was scored independently out of 5: the higher the score, the greater the risk of error. From 400 labels analyzed, two groups were selected for the dispensing simulation: 27 labels with high risk (score ≥3) and 27 with low risk (score ≤2). Each question in the learning program was displayed as a simulated clinical trial prescription. MAIN OUTCOME MEASURE: Medication error was defined as at least one erroneous answer (i.e. error in drug dispensing). For each question, response times were collected. RESULTS: High-risk investigational drug labels correlated with medication error and slower response time. Error rates were significantly 5.5-fold higher for high-risk series. Error frequency was not significantly affected by occupational category or experience in clinical trials. CONCLUSIONS: SIMME-CT is the first simulation-based learning tool to focus on investigational drug labels as a risk factor for medication error. SIMME-CT was also used as a training tool for staff involved in clinical research, to develop medication error risk awareness and to validate competence in continuing medical education.

Phase II Study of Pembrolizumab As First-Line, Single-Drug Therapy for Patients With Unresectable Cutaneous Squamous Cell Carcinomas.

PURPOSE: To evaluate first-line pembrolizumab monotherapy efficacy and safety in patients with unresectable cutaneous squamous cell carcinomas (CSCCs). PATIENTS AND METHODS: Patients, predominantly men, with their CSSCs' immunohistochemically determined programmed cell death-ligand 1 (PD-L1) status determined (tumor proportion score threshold, 1%), received pembrolizumab (200 mg every 3 weeks). The primary endpoint was the 39-patient primary cohort's objective response rate at week 15 (ORRW15). Secondary objectives were best ORR, overall survival (OS), progression-free survival (PFS), duration of response (DOR), safety, ORR according to PD-L1 status and health-related quality of life using Functional Assessment of Cancer Therapy-General (FACT-G) score. An 18-patient expansion cohort, recruited to power the study to evaluate the ORRW15 difference between PD-L1+ and PD-L1- patients, was assessed for ORR, disease control rate, and safety, but not survival. RESULTS: Median age of all patients was 79 years. The primary cohort's ORRW15 was 41% (95% CI, 26% to 58%), including 13 partial and 3 complete responses. Best responses were 8 partial and 8 complete responses. At a median follow-up of 22.4 months, respective median PFS, DOR, and OS were 6.7 months, not reached, and 25.3 months, respectively. Pembrolizumab-related adverse events affected 71% of the patients, and 4 (7%) were grade ≥ 3. One death was related to rapid CSCC progression; another resulted from a fatal second aggressive head and neck squamous cell carcinoma diagnosed 15 weeks postinclusion. ORRW15 for the entire population was 42%; it was significantly higher for PD-L1+ patients (55%) versus PD-L1- patients (17%; P = .02). Responders' W15 total FACT-G score had improved (P = .025) compared with nonresponders. CONCLUSION: First-line pembrolizumab monotherapy exhibited promising anti-CSCC activity, with durable responses and manageable safety. PD-L1 positivity appears to be predictive of pembrolizumab efficacy.

Hypersensitivity to oxaliplatin: clinical features and risk factors.

BACKGROUND: Oxaliplatin-based regimens induce a potential risk of hypersensitivity reaction (HSR), with incidence varying from 10% to 25% and lack of clearly identified risk factors. The present study aimed to assess incidence and risk factors in HSR. METHODS: All patients treated with oxaliplatin in the Medical Oncology Department of the Lyon Sud University Hospital (Hospices Civils de Lyon, France) from October 2004 to January 2011 were enrolled. Incidence and severity of HSR were analyzed retrospectively and the potential clinicopathological covariates were tested on univariate and multivariate analysis. RESULTS: A total of 1,221 doses of oxaliplatin were administered for 191 patients, 8.9% of whom experienced an HSR. Seventeen HSRs were observed, with 1.6% grade 3 and no grade 4 events. The first reaction appeared after a median of 3 oxaliplatin infusions. Using univariate analysis, HSR was associated with younger age (mean age, 56.2 years; p = 0.04), female gender (p = 0.01) and prior exposure to platinum salts (p = 0.02). No increased risk was associated with mean dose or with presence of atopic background. Multivariate analysis confirmed that women were at higher risk of oxaliplatin HSR than men (p < 0.05). Reintroduction of oxaliplatin was effective in 64.7% of hypersensitive patients using an appropriate premedication strategy. Patients who experienced a grade 3 HSR were not rechallenged. CONCLUSION: The risk of developing oxaliplatin HSR should not be underestimated (8.9% of patients). The medical team's vigilance should be increased with women, younger patients and patients with prior exposure to platinum salts.