A Systematic Approach for Post Hoc Subgroup Analyses With Applications in Clinical Case Studies

BACKGROUND

The analysis of subgroups in clinical trials is essential to assess differences in treatment effects for distinct patient clusters, that is, to detect patients with greater treatment benefit or patients where the treatment seems to be ineffective.

METHODS

The software application subscreen (R package) has been developed to analyze the population of clinical trials in minute detail. The aim was to efficiently calculate point estimates (eg, hazard ratios) for multiple subgroups to identify groups that potentially differ from the overall trial result. The approach intentionally avoids inferential statistics such as P values or confidence intervals but intends to encourage discussions enriched with external evidence (eg, from other studies) about the exploratory results, which can be accompanied by further statistical methods in subsequent analyses. The subscreen application was applied to 2 clinical study data sets and used in a simulation study to demonstrate its usefulness.

RESULTS

The visualization of numerous combined subgroups illustrates the homogeneity or heterogeneity of potentially all subgroup estimates with the overall result. With this, the application leads to more targeted planning of future trials.

CONCLUSION

This described approach supports the current trend and requirements for the investigation of subgroup effects as discussed in the EMA draft guidance for subgroup analyses in confirmatory clinical trials (EMA 2014). The lack of a convenient tool to answer spontaneous questions from different perspectives can hinder an efficient discussion, especially in joint interdisciplinary study teams. With the new application, an easily executed but powerful tool is provided to fill this gap.

A Systematic Approach for Post Hoc Subgroup Analyses With Applications in Clinical Case Studies

BACKGROUND

The analysis of subgroups in clinical trials is essential to assess differences in treatment effects for distinct patient clusters, that is, to detect patients with greater treatment benefit or patients where the treatment seems to be ineffective.

METHODS

The software application subscreen (R package) has been developed to analyze the population of clinical trials in minute detail. The aim was to efficiently calculate point estimates (eg, hazard ratios) for multiple subgroups to identify groups that potentially differ from the overall trial result. The approach intentionally avoids inferential statistics such as P values or confidence intervals but intends to encourage discussions enriched with external evidence (eg, from other studies) about the exploratory results, which can be accompanied by further statistical methods in subsequent analyses. The subscreen application was applied to 2 clinical study data sets and used in a simulation study to demonstrate its usefulness.

RESULTS

The visualization of numerous combined subgroups illustrates the homogeneity or heterogeneity of potentially all subgroup estimates with the overall result. With this, the application leads to more targeted planning of future trials.

CONCLUSION

This described approach supports the current trend and requirements for the investigation of subgroup effects as discussed in the EMA draft guidance for subgroup analyses in confirmatory clinical trials (EMA 2014). The lack of a convenient tool to answer spontaneous questions from different perspectives can hinder an efficient discussion, especially in joint interdisciplinary study teams. With the new application, an easily executed but powerful tool is provided to fill this gap.

Investigation into the age-dependence of the pharmacokinetics of cyproterone acetate in healthy male volunteers

OBJECTIVE

To investigate a possible age-dependence of the pharmacokinetics of cyproterone acetate following single oral administration.

METHODS

Twenty eight healthy men between 22 and 74 years of age received a single oral dose of 100 mg cyproterone acetate. The pharmacokinetic parameters, area under the serum concentration-time curve, apparent volume of distribution, apparent clearance, terminal half-life and the concentration ratio of 15 beta-hydroxycyproterone acetate/cyproterone acetate were examined for possible age-dependence using regression analysis.

RESULTS

The values of area under the serum level-time curve showed high interindividual variability and were not related to age. With regard to apparent clearance and volume of distribution, decreasing and increasing values, respectively, were observed with increasing age. There was also a clear dependence of the terminal half-life on age. Elderly men had values about two times higher (95 h) than men belonging to the younger age groups (45 h). The mean concentration ratio of 15 beta-hydroxycyproterone acetate/cyproterone acetate was 0.8 (0.3) and showed no age-dependent change.

CONCLUSIONS

Apparent clearance and apparent volume of distribution of cyproterone acetate showed age-dependent changes. Combined, the two effects cause a clear age-dependence of the terminal half-life of cyproterone acetate. An age-related reduction in liver volume is thought to be mainly responsible for the decrease in hepatic clearance with age. Chronic daily administration of the drug to elderly men may therefore lead to somewhat higher steady-state concentrations in the serum than in young men receiving the same dose.

Radioimmunological analysis of cyproterone acetate in human serum. Comparison with a gas chromatographic/mass spectrometric method and influence of each method on the outcome of a bioequivalence trial

Cyproterone acetate (CAS 427-51-0, CPA) is a steroid hormone with antiandrogenic and progestogenic properties, which has been used in the therapy of prostate carcinoma in men, and for the treatment of severe acne and hirsutism in women. The aim of the present study was to compare two equally sensitive analytical methods, a radioimmunoassay (RIA) and a gas chromatographic/mass spectrometric method (GC/MS), for the quantitative determination of CPA in human serum samples and to assess their suitability for bioequivalence trials. To this end, serum samples which had been collected during a study on the bioequivalence oaf two CPA-containing formulations (Androcur 100 and Androcur 50) were analysed using both methods. Basic pharmacokinetic parameters, like Cmax, tmax, t1/2 and AUC were calculated from each data set and corresponding parameters were compared by statistical methods. A comparison of the drug concentration-time curves obtained with both analytical methods revealed that in particular at later sampling times (48 to 120 h post dose) concentration values generated by RIA were slightly higher by about 20-40% than those measured with GC/MS. This indicates the presence of cross-reacting metabolite(s), most likely the 15 beta-hydroxy-cyproterone acetate. Accordingly, the values of Cmax, AUC(0-120 h) and AUC were overestimated by the RIA method by about 10-20%, 5% and 10%, respectively. However, there was a high correlation between corresponding parameters derived from RIA and GC/MS analysis. Although AUC values were slightly overestimated with the RIA method was used, this had no influence on the decision about bioequivalence because the mean ratio of the target variables Cmax and AUC was not affected. The variance of the pharmacokinetic parameters which is relevant for the bioequivalence test were even lower for RIA based values than those calculated after GC/MS analysis. In conclusion, it was found that although the GC/MS method is superior to RIA in terms of specificity, both methods were equally suited to demonstrate the bioequivalence of the two CPA-containing formulations. Thus, in future studies of this kind, the more simple RIA may be used instead of GC/MS.

Elimination of the diatrizoate-induced effects on the microcirculation by the prostacyclin derivative, iloprost

RATIONALE AND OBJECTIVES

Contrast media, especially high-osmolar ionic compounds elicit drastic effects on the microcirculation. In some areas of the microcirculatory network, blood flow decreases up to total stasis; in other areas, blood velocity increases. The authors studied the effect of the stable prostacyclin derivative, iloprost, added to the diatrizoate formulation on the number of perfused microvessels in the rat intestine.

METHODS

Diatrizoate (600 mg iodine/kg) and increasing concentrations of iloprost, ranging from a total dose of 0 to 560 ng/kg, were injected intravenously into groups of four anesthetized male rats. The number of unperfused microvessels in the intestine were determined by in vivo microscopy. Each animal received two treatments at an interval of 35 minutes. One treatment consisted of diatrizoate alone, the other treatment consisted of diatrizoate plus iloprost. Two animals of each group received first diatrizoate alone and then diatrizoate plus iloprost. The other two animals first received diatrizoate plus iloprost and then diatrizoate alone.

RESULTS

Diatrizoate alone reduced the number of perfused microvessels by 45%. The addition of iloprost resulted in a dose-dependent improvement of microcirculation. At a dose of 280 ng/kg, the effects of diatrizoate were completely abolished. Increasing the dose further diminished this effect.

CONCLUSIONS

The addition of iloprost to diatrizoate eliminates the deleterious effects of diatrizoate on microcirculation.