For those of us who work on drug-drug interactions, February 17, 2012 was a red-letter day: a new draft guidance on drug interaction studies¹ was posted that day on the FDA’s website. It was reported in the Federal Register on February 20, which started the clock on the 90-day period for posting public comments online.²

You’re probably asking yourself, “What’s the big deal about a new FDA guidance?” The answer is that this is a “new draft” guidance, following on the original draft guidance on the same topic, which was published in September 2006. In other words, it has taken five-and-a-half years not even to finalize the guidance, but to create a new draft. As incredible as that may appear, it actually reflects the rapid pace at which the field has advanced and matured, particularly in the area of drug transporters.

Through several workshops, numerous papers and oral presentations by FDA scientists, and the efforts of the International Transporter Consortium [ITC] (a group of drug transporter experts from industry, academia, and the FDA), including a highly influential white paper,³ it has been possible to track the gradual evolution of the FDA’s positions on drug interaction testing. As a result, there are no major surprises in the new guidance. Of course, the devil is in the details and many of the details have changed. Don’t worry; I don’t plan to discuss all of them, but I will say that, in contrast to the original draft guidance, the new version is explicit where it should be and appropriately vague in other places, allowing drug developers some latitude to use a variety of in vitro test systems (particularly for drug transporters, where the best approach is often unclear), as long as the choice can be justified on scientific grounds.

Maybe the biggest surprise is that the same seven transporters recommended for evaluation in the March 2010 ITC white paper and accepted by an FDA advisory panel in April 2010 are in the new guidance. The seven are P-glycoprotein (P-gp), breast cancer resistance protein (BCRP), organic anion transporting polypeptides 1B1 and 1B3 (OATP1B1 and OATP1B3), organic cation transporter 2 (OCT2), and organic anion transporters 1 and 3 (OAT1 and OAT3). It is significant that two others, bile salt export pump (BSEP) and OCT1, which are included in addition to the first seven in the April 2010 European Medicines Agency (EMA) Guideline on the Investigation of Drug Interactions,⁴ are not in the FDA guidance. That’s unfortunate; harmonization of requirements would make life simpler for drug developers that plan to request marketing approval in both the U.S. and Europe. In any case, it is significant that in the nearly six years since the original draft guidance the FDA has moved from recommending evaluation of only one transporter (P-gp) to seven.

Another significant change is that the original draft guidance relied heavily on the use of inhibitors to identify the transporter(s) involved in the active uptake or efflux of a new drug candidate. This made no sense to anyone active in the field — which was reflected in the comments posted online at the time — because it was well known even at that time that almost without exception, no truly selective pharmacologic reagents (probe substrates or inhibitors) were available for transporters. The new draft is much more scientifically sound in that regard, relying more on the specificity of a given in vitro test system than the available pharmacologic reagents, which are hardly more definitive than they were five-and-a-half years ago. It is remarkable that, given the amount of research activity in the transporter field, it is still true that there are practically no known specific probe substrates or inhibitors. On the contrary, we now know that most of the reagents that were previously thought to be specific are not.

OCT1 (required for evaluation by the EMA) appears to play a significant role in the disposition⁵ and efficacy⁶ of metformin, one of the most widely prescribed treatments for Type 2 diabetes. And that might be reason enough to screen new drug candidates as substrates of that uptake transporter. But it’s not. In fact, it’s not on the FDA’s list at all; not that the agency might not ask for it on a case-by-case basis. And the reason the EMA recommends evaluating new drug candidates as substrates of OCT1 is that it mediates the uptake of some anticancer tyrosine kinase inhibitors (TKIs) into tumors. Since OCT1 expression is down-regulated⁷ in some tumors, a higher dose (which might lead to additional side effects and/or toxicity) is required in such patients to achieve clinical efficacy for TKIs that are OCT1 substrates, such as imatinib.⁸ Some other TKIs, such as nilotinib, are not OCT1 substrates and do not require dose adjustment based on its expression.

One change from the ITC white paper is that all investigational drugs should be evaluated as inhibitors of the renal uptake transporters OCT2, OAT1, OAT3, without regard to likely co-medications; the white paper recommended evaluating new drugs as inhibitors of one of these transporters if they were likely to be co-administered with a substrate of the same transporter. The rationale for the lack of reference to co-medications is that the agency wants to evaluate the potential for any new chemical entity to alter the pharmacokinetics of endogenous uremic toxins. At the recent Second ITC Workshop, Les Benet of UC San Francisco took credit for identifying this previously unanticipated risk.

In the section on in vitro transporter studies, the guidance explicitly states that for drugs that are highly permeable and highly soluble (i.e., Biopharmaceutics Classification System [BCS] Class 1), intestinal absorption is not likely to be limited by P-gp or BCRP; thus, an in vivo study with a P-gp or BCRP inhibitor is unnecessary. This is not new or surprising, but it has not previously been stated explicitly in a guidance.

The new guidance recognizes the possibility of pharmacokinetic interactions between therapeutic proteins (TPs) and small molecule drugs, a phenomenon that has garnered increasing attention in recent years as the pharmaceutical industry has invested more and more heavily in the development of biologics. A typical interaction results from suppression by a TP of the expression of a drug-metabolizing enzyme for which a drug is a substrate. One needs to consider the disease state as well; for example, in chronic inflammatory diseases such as rheumatoid arthritis, cytochrome P450 enzymes (CYPs) are generally down-regulated. Administering a therapeutic antibody targeting an inflammatory cytokine could indirectly result in a generalized increase in CYP expression, thereby affecting the PK of a co-administered drug. The novelty of therapeutic protein-drug interactions (TPDIs) is reflected in the fact that in vitro models for their identification are not even mentioned in the guidance.

The criteria driving metabolism-mediated DDIs are now more conservative than in the past. For example, regardless whether in vitro studies are negative for any DDI risk, if an investigational drug is a substrate of an enzyme responsible for ≥ 25% of its systemic clearance you must conduct in vivo DDI studies with a strong inhibitor and inducer, and/or in subjects with different genotypes for the enzyme. In addition, if several enzymes are together responsible for ≥ 25% of its systemic clearance you must conduct in vivo studies to evaluate the potential for complex DDIs, i.e., those involving inhibition of a drug-metabolizing enzyme and a transporter or more than two co-administered drugs. This is undoubtedly driven by cases such as repaglinide as the victim of an interaction with a metabolite of gemfibrozil, magnified greatly by the addition of itraconazole.9 In general, more attention is given to such complex DDIs, which can lead to surprisingly large increases in peak plasma concentrations and/or exposure of the victim drug. “Surprising,” because one could not have predicted the magnitude of the clinical effect based on the available in vitro data. In some cases, even “significant” metabolites should be evaluated for DDI risk, although the guidance is vague as to when this would be appropriate; it’s probably one of those things that the agency would help you decide during one of the frequent meetings you should be having with them as your new drug candidate proceeds through development. (Hint, hint.)

The conservative nature of the new guidance is also evident from the recommendation that if an investigational drug is not metabolized by any of the major CYPs (1A2, 2B6, 2C8, 2C9, 2C19, 2D6, and 3A), other CYPs (e.g., 2A6, 2J2, 4F2, 2E1) or non-CYP enzymes should be evaluated. For the first time, evaluation of specific members of the UGT family of Phase II enzymes (i.e., those expressed in the liver) is recommended, if an appropriate in vitro system indicates that glucuronidation is responsible for at least 25% of the total metabolism of a drug.

With regard to CYP induction as a mechanism of metabolism-mediated drug interactions, it’s interesting to note that the only in vitro endpoint specified is mRNA; enzyme activity is not even mentioned. This is one point where I have to disagree: measuring both mRNA and enzyme activity provides valuable additional information. For many compounds, there is a disconnect between the two; a large increase in mRNA for the CYP of interest with little or no change in enzyme activity suggests that the compound is both an inducer and a time-dependent inhibitor of the enzyme. Although no criterion is specified for a “positive” in vitro CYP induction result, the guidance mandates much more reliance on complex, mechanistic computer models to translate in vitro CYP induction or inhibition results into expected clinical outcomes.

In summary, the FDA has done a commendable job of capturing the current state of the art and science of predicting the risk of clinical DDIs, not an easy task given the rapid pace of change, particularly with regard to drug transporters. Experts in the field breathed a collective sigh of relief when it was published, and will do so again if, following the usual 90-day period for public comments, this time a final guidance on drug interaction studies comes out in relatively short order.  

References

1. www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/UCM292362.pdf
2. www.regulations.gov/#!submitComment;D=FDA-2006-D-0036-0031
3. Giacomini KM, et al., Membrane transporters in drug development. Nature Rev Drug Discov. 2010 Mar;9(3):215-36
4. www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2010/05/WC500090112.pdf
5. Wang DS, et al., Involvement of organic cation transporter 1 in hepatic and intestinal distribution of metformin. J Pharmacol Exp Ther. 2002 Aug;302(2):510-5
6. Shu Y, et al., Effect of genetic variation in the organic cation transporter 1 (OCT1) on metformin action. J Clin Invest. 2007 May;117(5):1422-31
7. Heise M, et al., Downregulation of organic cation transporters OCT1 (SLC22A1) and OCT3 (SLC22A3) in human hepatocellular carcinoma and their prognostic significance. BMC Cancer. 2012 Mar 22;12(1):109 [Epub ahead of print]
8. White DL, et al., OCT-1-mediated influx is a key determinant of the intracellular uptake of imatinib but not nilotinib (AMN107): reduced OCT-1 activity is the cause of low in vitro sensitivity to imatinib. Blood. 2006 Jul 15;108(2):697-704
9. Bode C, The nasty surprise of a complex drug-drug interaction. Drug Disc Today. 2010 May;15(9-10):391-5