The International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) just published the final version of the harmonized drug-drug interaction (DDI) guideline (ICH M12). The guideline is pending adoption by regional regulatory health agencies. ICH released a companion Q&A document in parallel to the guidance. ICH M12 recommends in vitro and in vivo DDI studies for new medicinal products.
The guideline is generally relevant to small molecule drug development. It also gives limited consideration to therapeutic proteins and antibody-drug conjugates (ADCs; read about the FDA ADC clin pharm guidance here).
The document isn’t as comprehensive as some regional guidelines which also consider food effects, DDIs with oral contraceptives, and absorption-related DDIs including pH-mediated ones. Thus, regional guideline(s) remain valid. Consult them for the more focused aforementioned topics. ICH made minimal changes to the draft version of ICH M12 (July 2022) during the revision process for the final guideline.
Time to review and update your DDI packages!
We recommend reviewing all available in vitro (and in vivo) DDI data with a fresh perspective.
- Do your in vitro studies fulfill the ICH M12 guideline requirements?
- After implementing the ICH M12 guideline, will a negative DDI risk remain so or be considered positive or inconclusive? Will your planned in vivo DDI study no longer be required as your assumed positive in vitro signal is now considered negative?
- Are the upcoming changes already factored into your program timelines? Could health authorities put your drug development program on clinical hold next year due to a lack of relevant in vitro data?
As an example, the guideline has less strict criteria (cutoffs, R-values) i.e. for when an in vitro signal is considered to be relevant in vivo, for a time-dependent-inhibitor (TDI) and more strict criteria when evaluating enzyme induction. Look closely at the ICH M12 DDI guideline to determine whether your in vitro data or planned studies will reach the desired standard.
Changes in the ICH guideline
The DDI guideline is very detailed. Hence, we have summarized the key changes below and in the table for quick reference.
Nomenclature: While the draft guidance used the terms “victim” and “perpetrator” drugs, the final document now refers to “object” (a substrate of an enzyme or transporter) and “precipitant” (a drug that can induce or inhibit an enzyme or a transporter). ICH added a glossary to the guidance’s appendix.
Protein Binding: A new section was added describing the suitability of a given methodology which allows using unbound human plasma fraction less than 0.01 (fu <0.01) for a DDI assessment using basic methods.
Biomarkers: In contrast to the draft version, ICH now presents a biomarker approach. The guidance mentioned plasma coproporphyrin I (hepatic OATP1B1/3), plasma and urine N-methylnicotinamide and N-methyladenosine (renal OCT2, MATE1, MATE2K), plasma pyridoxic acid (renal OAT1/3), and plasma 4β-hydroxycholesterol/cholesterol ratio and urine 6β-hydroxycortisol/cortisol ratio for CYP3A. This approach is very useful, and PBPK models can leverage it.
For more details, read our blog on using Coproporphyrin-1 (CP-1) as an endogenous biomarker for predicting OATP1B DDIs.
Table: Summary of key changes in ICH M12 guideline versus previous regional guidances
| In vitro studies |
| Topic/Investigation |
Final ICH M12 (differences from the draft version are in bold font) |
| General Principles (Timing) |
- In vitro reaction phenotyping (enzymes) generally before Phase 1 in patients
- In vitro precipitant effects on CYP and transporter generally before initiating larger studies in patients.
- In vitro interactions for major/active metabolites may be later in development
|
| Substrate of Metabolizing Enzymes |
- Clarification for most common CYP enzymes: CYP3A (CYP3A4 and CYP3A5)
- List of 11 potential UGTs to screen if the drug is mainly eliminated by direct glucuronidation.
- N-acetyltransferases and glutathione S-transferases added as “other Phase 2 enzymes” list
|
| CYP Reversible Inhibition |
- In some situations, the measured drug-free fraction in plasma can be used to estimate the risk with the basic method for highly bound drugs (≥99.9%)
|
| CYP450 Time-dependent Inhibition (TDI) |
- Less stringent basic static risk assessment
- Clarification that dilution and non-dilution assays may be used.
- Turnover rate constant (Kdeg) of major CYPs now specified
|
| UGT Inhibition |
- Perform if the drug is mainly eliminated by direct glucuronidation or is to be used with another drug that is mainly glucuronidated.
- Potential UGTs to study: UGT1A1, UGT1A4, UGT1A9, UGT2B7, and UGT2B15
|
| CYP Induction |
- The concentration to use in the basic “mRNA Fold-change” method is 50x Cmaxu
- Evaluate CYP2C8, CYP2C9, and CYP2C19 when a clinical study shows that your drug induces CYP3A4. A negative clinical study can’t be extrapolated to other enzymes if there is also relevant 3A4 inhibition in vitro.
- CYP2C19 via enzyme activity (not mRNA)
- Actual unbound drug concentration in the assay should be used in general (EC50,u)
- Positive controls should have at least ≥6-fold mRNA increase for CYP1A2, 2B6, and 3A (not mandatory for CYP2Cs)
- Incubations shorter than 48 hours could be used if sufficient positive control response is maintained but should be justified.
- The RIS (relative induction scores) correlation method can be done with only 1 qualified batch of hepatocytes.
|
| Mechanistic static model |
- This model can be used to extrapolate victim DDI results with CYP or transporter inhibitors to “less potent” inhibitors. Here, the guideline allows for a semi-quantitative approach based on justifications and sensitivity analyses. There is an opening for a fully quantitative use of MSM when estimating perpetrator effects. Sufficient justification of relevant drug concentrations in the gut and liver is required.
|
| Substrate of Transporters |
- Additional transporters (MRP2, OATP2B1, OCT1) on a case-by-case basis
- Evaluate if a drug is also a substrate of OATP1B1 and OATP1B3 if the pharmacological target is in the liver.
|
| Transporter Inhibition |
- Additional transporters (BSEP, MRP2, OATP2B1, OCT1) on a case-by-case basis
- Can take into consideration if commonly co-administered drugs are substrates for these transporters.
- Harmonization of cut-off values between FDA and EMA
- Describe the use of IC50,u (vs Ki) for transporters
- Equation for a drug administered parenterally that inhibits P-gp or BCRP
- New cutoffs for positive signals based on basic static assessments, as compared to FDA (for MATEs) and EMA guidelines
|
| DDI of Metabolites |
- Metabolite as a substrate if contributes to in vivo target effect to a similar or greater extent than the parent drug.
- Metabolite as an inhibitor (CYP and transporters) if ≥ 25% of AUCparent and ≥10% of drug-related material in circulation
- Induction study of metabolite if 1) drug is a pro-drug or 2) metabolite mainly formed extra-hepatically
|
| Assay Conditions |
- Detailed guidance to ensure drug solubility, lack of cytotoxicity, and adequate recoveries/stability in assays
|
| In vivo studies |
| General Principles (Timing) |
- The human absorption, metabolism, and excretion study (hAME) results should generally be available before Phase 3 (read more about hAME studies in this blog).
- Based on the results, in vivo victim DDI studies are planned and if required, further in vitro metabolism and transport studies identified.
|
| In vivo studies |
- If the DDI effect changes over time (I.e. combined inhibitor and inducer) – evaluate the DDI early and late during co-treatment
- Added clarity for dose: “When there are safety concerns, lower doses of the substrate drug, including doses lower than the therapeutic doses, can be used.”
- A parallel, two-arm study can be appropriate when a major active metabolite has a long half-life.
- Including measurement of metabolite(s), which may help when a clinical DDI study evaluates drugs that may interact via multiple pathways.
- In general, ICH requires more extensive information on study design, considerations in choosing index drugs for DDI studies, etc.
|
| PBPK and population PK approaches |
- Many sections of the DDI assessment recommend using PBPK (dynamic models) modeling, including informing the DDI strategy and study design, translating pharmacogenomic effects to object DDIs, evaluating complex DDIs, DDIs in specific populations, replacing studies of staggered dosing between object and precipitant, supporting and/or evaluating object drugs with long half-lives, etc.
|
If you need help navigating the new requirements, our experts can help. Read our white paper to learn more about modeling and simulation approaches to predict DDIs.
The blog was first published on March 31, 2023. Eva Gil Berglund and Nathalie Rioux updated this blog in June 2024.
