Early Phase Clinical Trials in Patients with Hepatic or Renal Impairment: From Design to Data Analysis
FDA [1, 2] and EMA [3, 4] guidances are intended to help companies evaluate the need for conducting PK studies in Renal Impaired and Hepatic Impaired patients and to provide guidance on how to best assess the influence of renal impairment (RI) or hepatic impairment (HI) on the pharmacokinetics of an investigational drug. They provide recommendations on when studies should be conducted, how to design such studies, and on how such studies should be carried out.
Most of drugs are likely to be administered to RI or HI patients and should be investigated in that population, however, the available guidances for industry are relatively old (EMA and FDA guidances for HI, 2005 and 2003, respectively), still draft version (FDA guidance for RI, draft since 2010) or not enough detailed (EMA guidance for RI, 2014). And although many recommendations about the design, conduct and data analysis of early phase RI or HI trials are available in the guidances, some key questions still need to be investigated case-by-case at a very early stage of the clinical development.
Key Open Questions
Full or reduced study design?
A decision tree for determining when and how RI or HI studies should be conducted is given in the FDA guidances [1, 2].
As a general approach, when renal or hepatic elimination is not the major route of elimination of parent drug and/or active metabolites, a “reduced” study can be sufficient in RI and HI, respectively. So, the sponsor has the option of conducting a “reduced” or “full” study, and the choice should be based on the pharmacokinetic (PK) properties of the drug including all data from in vitro, pre-clinical and clinical studies.
The FDA also advices that if a “reduced” PK study shows a substantial effect on PK in the RI patients, a “full” RI study should be conducted. While for HI in that case, the findings in the moderate category in “reduced” study would be applied to patients with a mild Child-Pugh category, and dosing in the severe category would generally be contraindicated.
Moreover, FDA and EMA guidances are not consistent in the definition of “reduced” design for RI studies: according to FDA End Stage Renal Disease (ESRD) subject not under dialysis should be compared to matched healthy volunteers (HVs), while according to EMA severe RI patients should be compared to matched HVs.
What about recruitment of ESRD patients not under dialysis? This type of patients are quite difficult to find, therefore, often “full” study (including all the stages of RI) is preferred by sponsors in order to meet the timelines and also to avoid possible restart of a “full” study in case of a substantial effect on PK in RI patients in the “reduced” study. Additionally, the minimum number of subjects required in “reduced” design (at least 8 per group) is higher than in “full” design (at least 6 per group).
Statistical Considerations and Matching Strategy
The control group of HV should be comparable (matched) to the RI/HI patient population with respect to age, gender, race and weight. Depending on the drug, other factors with significant potential to affect the PK should also be taken into consideration. Matching prevents confounding by increasing precision of estimates (reduction in standard errors). When sample size is small, like in RI/HI studies, without a good matching, control for confounding in the analysis will result in many strata with sparse data. There is no advice in the guidances about the matching method, but two strategies are known to be used:
When individual matching is used, controls are matched to patients one by one for each of matching criteria. The advantages of this method are the simultaneous recruitment of pairs of RI/HI patient-HV and the possibility of reuse of HV data among RI/HI groups. The most important advantage is the statistical robust results guarantee against the only disadvantage of the need of more HVs compared to other matching strategies.
The second common matching is group matching. In this approach when all RI/HI patients are completed, HV will be selected (matched) based on the defined distribution (percentiles) of each matching criteria. The advantage of this method is that you need a lower number of HVs, which will be equal to the number of patients in one group of RI/HI patients. However, this method cannot be considered as the first choice in early phase RI/HI trials because it provides less statistical robustness when comparing PK results between RI/HI groups and HVs.
Particularities of Dialysis Effect Assessment in RI Trials
Dialysis may affect the PK of a drug to an extent that dosage adjustment is needed. In general, a study of the effect of dialysis on PK may be omitted if the drug is unlikely to be administered to ESRD patients treated with dialysis, or if the dialysis procedure is unlikely to result in significant elimination of drug or active metabolites.
If dialysis is assessed, PK should be studied in such patients under both dialysis (DTP) and non-dialysis (IDTP) conditions to determine the extent to which dialysis contributes to the elimination of the drug and potentially active metabolites. Venous blood samples during IDTP and blood from both arterial and venous sides of fistula during DTP should be collected. Drug concentration in dialysate, total plasma proteins and drug free fraction in arterial and venous sides should be measured at several time points during the DTP. Some key hemodialysis parameters should also be reported, like blood flow through the dialyser (QB) and rate of ultrafiltration (QUF).
Special to hemodialysis PK parameters (like extraction coefficient by the dialyser (E), hemodialysis clearance (CLHD), amount of drug removed by dialysis (AR)), will allow answering to the main questions:
- Was the drug removed through the hemodialyser membrane?
- How effectively can the dialyser remove the drug from the blood?
- Is it removed through the membrane?
Common PK parameters have to be calculated during the IDTP, as well during the DTP using drug concentration data from venous side. This will allow comparing the PK behavior of investigational drug between IDTP and DTP.
Dosing Adjustment Recommendations
The principal objective of an early phase RI/HI study is to advice on dosing recommendations. When applicable, it is also important to point out in dosing recommendations that HI/RI impairment does not alter a drug’s PK. If the effect of HI/RI on the PK of the drug is obvious, dosage adjustments should be recommended in the Clinical Study Report (CSR) of the early phase study. It is also possible that the effect of HI/RI on the PK is only partial, i.e. significantly altered only in one or more groups but not in all groups compared to HVs.
To reach a conclusion about the significance of altered PK, a confidence interval approach, rather than a significance test, is preferred in HI studies according to FDA guidance . The no effect boundaries are defined prior to the conduct of the studies, based on information available for the investigational drug PK, or a standard 90% confidence interval of 80-125% for AUC and Cmax is chosen. FDA recognizes that providing evidence that a PK parameter remains within an 80-125% no effect boundary would be very difficult given the small numbers of subjects usually entered into HI studies. If a wider boundary is fixed, it should be supported clinically (no safety issue) and confirmed by formal sample size calculation based on inter-subject variability observed in HVs. However, it may be more probable to conclude that there is no need for dose adjustment with the wider boundaries; therefore, this choice should be well discussed in function of the PK properties of the drug.
For RI studies significance test is used without possible statistical flexibility, which can make it often difficult to draw conclusions about the RI effect on the drug PK with the low number of subjects. The conclusion in this case should be evidence-based, and if needed confirmed by a PK/PD study in a much larger patients population.
Extra caution is needed when developing a drug with narrow therapeutic index (NTI). Some general approaches to estimate the exact dose adjustment are available using relatively simple equations. Since some conditions should be fulfilled to apply those equations and since they are still not optimized, modeling and simulation (M&S) are highly recommended. M&S takes into account drug PK/PD properties, administration particularities and propose the best approach of dosing adjustment.
Recommendations to Correctly Interpret the Results from Early Phase RI/HI Trials
Several scientific review articles concluded that there is remarkable variation in definitions and recommendations and lack of scientifically relevant interpretation and used method’s description, making the available results from early phase RI and HI trials not really suited for clinical use. RI and HI are very complex pathologies so, the particular patho physiological changes in these patients together with the PK properties of the drug should be taken carefully into account when interpreting the observed results.
The following are some scientific advices to be considered when interpreting the PK results:
At early stage of drug development, studies in RI/HI special populations are almost always required for drugs with systemic absorption. Conclusions from early phase RI/HI trials are very difficult, variable and often not clear. The best solution to the success is the development of a robust study design by clinical and scientific experts. The study design as well the interpretation of results should be based on PK/PD properties of the drug and with understanding of the fundamental PK principles related to RI and HI pathologies.
- Guidance for Industry “Pharmacokinetics in Patients with Impaired Renal Function”, FDA, March 2010, Rev. 1 (Draft)
- Guidance for Industry “Pharmacokinetics in Patients with Impaired Hepatic Function”, FDA, May 2003
- “Guideline on the evaluation of the pharmacokinetics of medicinal products in patients with impaired hepatic function”, EMA, August 2005
- “Guideline on the evaluation of the pharmacokinetics of medicinal products in patients with decreased renal function”, EMA, August 2014
- Pharmacokinetics and dosage adjustment in patients with renal dysfunction Roger K. Verbeeck, Eur J Clin Pharmacol (2009) 65:757–773
- Pharmacokinetics and dosage adjustment in patients with hepatic dysfunction Roger K. Verbeeck, Eur J Clin Pharmacol (2008) 64:1147–1161
Narine Baririan, Pharm.D
SGS Life Science Services - Clinical Research