Introduction of ICH M12 – A Harmonised Approach for DDI Studies

It can be challenging to navigate the requirements for drug interaction studies when each regulatory authority has independent guidelines with distinct differences in protocols and interpretation of the data. Furthermore, complying with all the guidelines can lead to inefficiency due to additional cost and extended development times.

In May 2022, the industry welcomed a harmonised draft guideline for drug interaction studies endorsed by members of the ICH Assembly (ICH M12). The guideline covers recommendations for designing, conducting and interpreting enzyme- or transporter-mediated in vitro and clinical DDI studies during the development of a therapeutic product. The draft recommendations are currently under public consultation via the regulatory authorities of the ICH regions with the deadlines extending until the end of November 2022. By aligning the requirements, it is expected that the individual guidelines from the FDA, EMA and Japanese PMDA and others will hold less weight and may become obsolete once the final M12 guidelines are released – anticipated to be April 2024.

So how does the draft M12 guidance compare to the existing FDA, EMA and Japanese PMDA guidance for in vitro DDI studies? Well, the first fact to point out is that there are no real surprises compared to existing guidance, the same key CYP isoforms and transporters are recommended.  A quick comparison is summarised below:

  • Reaction phenotyping: Data interpretation is similar to all existing guidance. Timing of in vitro studies is still prior to phase 1 (first in human) but with a strong emphasis on confirming important routes following clinical mass balance study (typically pre phase 3).
  • Reversible CYP inhibition: Data interpretation is similar to all existing guidance. Timing of in vitro studies recommended as early as possible and before trials in patients (typically phase 2).  This is a slight shift in emphasis vs FDA guidelines from 2020, which recommended studies pre phase 1 (first in human).  As previously stated in EMA guidance there is a requirement to investigate UGT1A1/UGT2B7 inhibition if glucuronidation is major route of metabolism and now also includes other relevant isoforms if co-medication is metabolised by UGTs.  In addition, there are some changes to the list of example inhibitors for CYP in vitro studies compared to FDA accepted inhibitor list.
  • Time dependent inhibition (TDI): Although the cut-off value is identical, the equation for the kobs in the M12 guidance differs to the current FDA, EMA and PMDA guidelines – includes a safety factor of 5 rather than 50 (FDA and PMDA) or no margin for EMA. There is also no separate recommendation for TDI in the GI tract as previously referred to in the 2018 Japanese PMDA guidance.
  • CYP induction: Interpretation is similar to all the previous guidance. M12 highlights the poor induction response with CYP2C19 mRNA and the suitability of using catalytic activity in this case.  In addition, it sets an expectation of ≥6 fold induction for CYP1A2, CYP2B6 and CYP3A4 for positive controls  (measured as mRNA change) across donors and CITCO is no longer included as a positive control (vs EMA guidance).  Immortalised cell lines are considered supportive data only, contrasting with the 2020 FDA guidance.  For the fold change approach M12 suggests a different safety factor 15x Cmax,u (compared to 30x for FDA and 50x for EMA previously).  Correlation methods such as RIS can be performed in just one well-performing qualified batch of hepatocytes.
  • Mechanistic static models: The models are similar with the exception that the M12 guidance suggests [I]h should be calculated from the maximal inhibitor concentration in the plasma (similar to the FDA guidance) rather than the blood as currently stated in the EMA and Japanese PMDA.
  • Transporter substrate: Data interpretation is similar to all existing guidance.
  • Transporter inhibition: For P-gp/BCRP inhibition, the recommended cut-off for orally administered drugs in the M12 guidance is the same as the current regulatory guidance documents. However, if the drug is administered parenterally or if it is a metabolite formed post-absorption that inhibits P-gp or BCRP then the cut-off is identical to the EMA but is different to other existing regulatory guidance. For OATP1B1, OATP1B3, OAT1, OAT3 and OCT2 inhibition, the cut-off in M12 reflects the existing FDA and Japanese PMDA guidance. For MATE inhibition, the cut-off in the M12 guidance reflects the Japanese PMDA and EMA guidelines. Unlike existing regulatory guidance, the M12 guidance supports the use of mechanistic static models for evaluating the potential for transporter-mediated DDIs, akin to those previously described for enzymes (net effect model). One such transporter model cited in the guidance is that of Elsby et al. (2012), which is routinely used at Cyprotex to contextualise DDI risk for clinicians; as demonstrated in our publications of Elsby et al. (2017) (metformin MATE DDIs) and Elsby et al. (2019) (transporter and enzyme DDIs with simvastatin acid).
  • Metabolites: DDI studies for metabolites still need to be considered but timing can be later in development when more information is available on exposure and activity.  Understanding of metabolic fate is required if metabolite contributes similar efficacy to parent, previously this was ≥50%.  DDI studies required if AUCmetabolite>25% AUCparent and the metabolites account for ≥10% drug related material.  Induction potential of metabolites may need to be tested if they are formed extra hepatically such as prodrugs and unlikely to be formed in induction experiment.
  • Fup<0.01: M12 guidance is open to using measured fup values <0.01 for DDI risk assessment but determination by multiple approaches with supporting method validation is required.
  • Analytical methods: For DDI studies GLP not required but a full description of analytical methods used including validation of analytical parameters should be provided. An additional point is that, unlike the current FDA guidance, the new ICH guidance covers recommendations for in vitro DDI studies and clinical DDI studies in a single document.

In general, the new draft M12 provides a comprehensive viewpoint of the assays that need to be performed with guidance on specific assay design. It is clearly presented and, once finalised, should eliminate the need to repeat studies to meet the individual authority’s recommendations.

Cyprotex are able to perform full regulatory in vitro DDI studies which meet the current recommended guidance. If you are interested in discussing your study then please get in touch with our DDI experts.

Contact us to start planning your DDI study


Elsby R et al., (2012) Understanding the critical disposition pathways of statins to assess drug-drug interaction risk during drug development: it’s not just about OATP1B1.  Clin Pharmacol Ther 92(5); 584-598

Elsby R et al., (2017) Mechanistic in vitro studies confirm that inhibition of the renal apical efflux transporter multidrug and toxin extrusion (MATE) 1, and not altered absorption, underlies the increased metformin exposure observed in clinical interactions with cimetidine, trimethoprim or pyrimethamine.  Pharmacol Res Perspect 5(5); e00357,

Elsby R et al., (2019) Mechanistic in vitro studies indicate that the clinical drug-drug interaction between telithromycin and simvastatin acid is driven by time-dependent inhibition of CYP3A4 with minimal effect on OATP1B1.  Drug Metab Dispos 47(1); 1-8

This website is intended to assist industry participants, customers and employees to understand Cyprotex's global operations and ambitions.