This analysis aimed to identify all strong drug-drug interactions (DDIs) associated with drugs approved by the US Food and Drug Administration (FDA) in 2021.
Presented at the ISSX/MDO Meeting, September 2022
Jingjing Yu, Yan Wang, and Isabelle Ragueneau-Majlessi
The mechanistic evaluation of enzyme- and transporter-based drug-drug interactions (DDIs) during drug development is critical to support management strategies in the clinic.
The objectives of the study were to review pharmacokinetic-based clinical DDI data available in the new drug application (NDA) reviews for drugs approved by the FDA in 2021, and to understand the main mechanisms that mediate interactions resulting in label recommendations.
Food effect (FE) and gastric pH-dependent drug-drug interactions (DDIs) are both absorption-related. Here, we evaluated if Biopharmaceutics Classification System (BCS) classes may be correlated with FE or pH-dependent DDIs. Trends in FE data were investigated for 170 drugs with clinical FE studies from the literature and new drugs approved from 2013 to 2019 by US Food and Drug Administration. A subset of 38 drugs was also evaluated to determine whether FE results can inform the need for a gastric pH-dependent DDI study. The results of FE studies were defined as no effect (AUC ratio 0.80-1.25), increased exposure (AUC ratio ≥1.25), or decreased exposure (AUC ratio ≤0.8). Drugs with significantly increased exposure FE (AUC ratio ≥2.0; N=14) were BCS Class 2 or 4, while drugs with significantly decreased exposure FE (AUC ratio ≤0.5; N=2) were BCS Class 1/3 or 3. The lack of FE was aligned with the lack of a pH-dependent DDI for all 7 BCS Class 1 or 3 drugs as expected. For the 13 BCS Class 2 or 4 weak base drugs with an increased exposure FE, 6 had a pH-dependent DDI (AUC ratio ≤0.8). Among the 16 BCS Class 2 or 4 weak base drugs with no FE, 6 had a pH-dependent DDI (AUC ratio ≤0.8). FE appears to have limited correlation with BCS classes except for BCS Class 1 drugs, confirming that multiple physiological mechanisms can impact FE. Lack of FE does not indicate absence of pH-dependent DDI for BCS Class 2 or 4 drugs.
New drug application reviews contain critical drug interaction study results with newly approved drugs tested both as victims and as perpetrators of drug-drug interactions (DDIs). Pharmacokinetic-based DDI data for drugs approved by the US Food and Drug Administration in 2013–2017 (N = 137) were analyzed using the University of Washington Drug Interaction Database. For the largest metabolism- and transporter-based drug interactions, defined as a change in exposure ≥ 5-fold in victim drugs, the mechanisms and clinical relevance were characterized. Consistent with the major role of CYP3A in drug disposition, CYP3A inhibition and induction explained a majority of the observed interactions (new drugs as victims or as perpetrators). However, transporter-mediated interactions were also prevalent, with OATP1B1/1B3 playing a significant role. As victims, 17 and 4 new molecular entities (NMEs) were identified to be sensitive substrates of enzymes and transporters, respectively. When considered as perpetrators, three drugs showed strong inhibition of CYP3A, one was a strong CYP3A inducer, and two showed strong inhibition of transporters (OATP1B1/1B3 and/or BCRP). All DDIs with AUC changes ≥ 5-fold had labeling recommendations in their respective drug labels, contraindicating or limiting the coadministration with known substrates or perpetrators of the enzyme/transporter involved. The majority of sensitive substrates or strong inhibitors were oncology and antiviral treatments, suggesting a significant risk of DDIs in these patient populations for whom therapeutic management is already complex due to poly-therapy. Pharmacogenetic studies and physiologically based pharmacokinetic models were commonly used to assess the drug interaction potential in specific populations and clinical scenarios. Finally, absorption-based DDIs were evaluated in approximately 30% of drug applications, and 14 NMEs had label recommendations based on the results.
Pharmacokinetic-based drug-drug interaction (DDI) data for drugs approved by the U.S. Food and Drug Administration in 2017 (N = 34) were analyzed using the University of Washington Drug Interaction Database. The mechaniDrug-drug interaction (DDI) data for small molecular drugs approved by the U.S. Food and Drug Administration in 2020 (N = 40) were analyzed using the University of Washington Drug Interaction Database. The mechanism(s) and clinical relevance of these interactions were characterized based on information available in the new drug application reviews. About 180 positive clinical studies, defined as mean area under the curve ratios (AUCRs) {greater than or equal to} 1.25 for inhibition DDIs or pharmacogenetic studies and {less than or equal to} 0.8 for induction DDIs, were then fully analyzed. Oncology was the most represented therapeutic area, including 30% of 2020 approvals. As victim drugs, inhibition and induction of CYP3A explained most of all observed clinical interactions. Three sensitive substrates were identified: avapritinib (CYP3A), lonafarnib (CYP3A), and relugolix (P-gp), with AUCRs of 7.00, 5.07, and 6.25 when co-administered with itraconazole, ketoconazole, and erythromycin, respectively. As precipitants, three drugs were considered strong inhibitors of enzymes (AUCR {greater than or equal to} 5): cedazuridine for cytidine deaminase, and lonafarnib and tucatinib for CYP3A. No drug showed strong inhibition of transporters. No strong inducer of enzymes or transporters was identified. As expected, all DDIs with AUCRs {greater than or equal to} 5 or {less than or equal to} 0.2 and almost all those with AUCRs of 2-5 and 0.2-0.5 triggered dosing recommendations in the drug label. Overall, all 2020 drugs found to be either sensitive substrates or strong inhibitors of enzymes or transporters were oncology treatments, underscoring the need for effective DDI management strategies in cancer patients often receiving poly-therapy. Significance Statement This minireview provides a thorough and specific overview of the most significant pharmacokinetic-based DDI data observed (or expected) with small molecular drugs approved by the U.S. Food and Drug Administration in 2020. It will help to better understand mitigation strategies to manage the DDI risks in the clinic.
To best promote drug tolerability and efficacy in the clinic, data from drug-drug interaction (DDI) evaluations and subsequent translation of the results to DDI prevention and/or management strategies must be incorporated into the US Food and Drug Administration (FDA) product labeling in a consistent manner because differences in language might result in varied interpretations. This analysis aimed to assess the consistency in DDI labeling language in New Drug Applications (NDAs).
Presented virtually at 24th North American ISSX Meeting, September 2021
Jingjing Yu,Yan Wang, Cheryl Wu, and Isabelle Ragueneau-Majlessi
Patients with infectious diseases in low-income countries (LICs) are often at risk of pharmacokinetic (PK) drug-drug interactions (DDIs). To assist in silico mechanistic modeling and simulations to predict DDI liability and guide optimal management of DDIs, a knowledgebase of anti-infective drugs, specifically treatments for malaria and tuberculosis, has been established.
Presented virtually at the 24th North American ISSX Meeting, September 2021
Jingjing Yu, Yan Wang, and Isabelle Ragueneau-Majlessi
The aim of the present work was to review pharmacokinetic drug-drug interaction (DDI) data available in New Drug Applications (NDAs) for drugs approved by the US Food and Drug Administration in 2020 and analyze the mechanisms mediating interactions in order to facilitate an optimal management of DDIs in the clinic.
Although the use of excipients is widespread, a thorough understanding of the drug interaction potential of these compounds remains a frequent topic of current research. Not only can excipients alter the disposition of coformulated drugs, but it is likely that these effects on co-administered drugs can reach to clinical significance leading to potential adverse effects or loss of efficacy. These risks can be evaluated through use of in silico methods of mechanistic modeling, including approaches, such as population pharmacokinetic (PK) and physiologically-based PK modeling, which require a comprehensive understanding of the compounds to ensure accurate predictions. We established a knowledgebase of the available compound (or substance) and interaction-specific parameters with the goal of providing a single source of physiochemical, in vitro, and clinical PK and interaction data of commonly used excipients. To illustrate the utility of this knowledgebase, a model for cremophor EL was developed and used to hypothesize the potential for CYP3A- and P-gp-based interactions as a proof of concept.
Evaluating the potential of new drugs and their metabolites to cause drug‐drug interactions (DDIs) is critical for understanding drug safety and efficacy. Although multiple analyses of proprietary metabolite testing data have been published, no systematic analyses of metabolite data collected according to current testing criteria have been conducted. To address this knowledge gap, 120 new molecular entities approved between 2013 and 2018 were reviewed. Comprehensive data on metabolite‐to‐parent area‐under‐the‐curve ratios (AUCM/AUCP), inhibitory potency of parent and metabolites, and clinical drug‐drug interactions (DDI) were collected. 64% of the metabolites quantified in vivo had AUCM/AUCP≥25% and 75% of these metabolites were tested for cytochrome P450 (CYP) inhibition in vitro, resulting in 15 metabolites with potential DDI risk identification. While 50% of the metabolites with AUCM/AUCP<25% were also tested in vitro, none of them showed meaningful CYP inhibition potential. The metabolite % plasma total radioactivity cutoff of ≥10% did not appear to add value to metabolite testing strategies. No relationship between metabolite versus parent drug polarity and inhibition potency was observed. Comparison of metabolite and parent maximum concentration (Cmax) divided by inhibition constant Ki values suggested that metabolites can contribute to in vivo DDIs and hence, quantitative prediction of clinical DDI magnitude may require both parent and metabolite data. This systematic analysis of metabolite data for newly approved drugs supports an AUCM/AUCP cutoff of ≥25% to warrant metabolite in vitro CYP screening to adequately characterize metabolite inhibitory DDI potential and support quantitative DDI predictions.