Skip directly to content

Pharmacokinetic (PK) Issues and Drug-Drug Interaction Studies, 2004

Debra B Birnkrant, MD
Center for Drug Evaluation and Research
Office of New Drugs
Division of Anti-Viral Products (HFD-530)
CRPII/Rm. N414
Food and Drug Administration
5630 Fishers Lane, Rm 1061
Rockville, MD 20852
Fax: 301.827.2523

Dear Dr Birnkrant,

I am writing to follow up on the productive meeting with the CDER staff and ATAC's Drug Development Committee (DDC) in November 2003. Many things were discussed during this meeting: (please see article by Bob Huff in Treatment Issues vol. 17 no. 11 at www.gmhc.org/health/treatment/ti/ti1711.html#3, that reviews specifically

  • PK data
  • Drug-drug interaction studies
  • 'Real life' study populations
  • Post-marketing research
  • Long-term side effects—à la sentinel cohorts).


I am writing solely as a member of the Treatment Action Group to address the first three points—what data needs to be seen with new drugs before approval decisions are made.

This letter will focus on pharmacokinetic (PK) issues as they apply to specific populations and drug-drug interaction studies (1), to wit, on outlining possible changes regarding required or desirable pharmacokinetic and drug-drug interaction data. This data should be available from sponsors of investigational new drugs (INDs) for treatment of HIV disease and associated complications, either at the time the new drug application (NDA) is submitted, or as part of the suite of required post-marketing studies.

As discussed a year ago, the community is suggesting that certain data be routinely included in every approval package and be required—or strongly encouraged—by the FDA of sponsors. In some cases, e.g., accelerated approval, the FDA may have more leverage than in others. However, we encourage the FDA to think creatively about strategies for obtaining and requiring the capture of data necessary to better guide clinical care for people living with HIV. Can full approval be deterred until the data submitted?

The interaction studies most needed that would help in better defining the toxicity and safety of new drugs, are:

By approval:

  • All currently approved ARVs (including AZT, ddI, d4T, 3TC, ABC, efavirenz, nevirapine, tenofovir, indinavir, nelfinavir, fosamprenavir, saquinavir, atazanavir with and without ritonavir, lopinavir/r. They should be conducted in small numbers of HIV+ patients before efficacy data is collected. For example, would the Boehringer Ingelheim BI 1182.51 study have shown the same results in a shorter sub-study—to see that there were some serious interaction issues between tipranavir/r and (all) other PIs? Remember, these were 296 highly treatment experienced patients who at week 8 were experiencing treatment failure in all double PI arms.)
  • ARVs in the pipeline, as soon as dosing is determined.
  • Methadone
  • Hormonal Contraceptives—oral, patch, and topical delivery of progestional and combined estrogen/progestional agents)
  • Lipid regulators (statins, fibrates)
  • Food and liquid—what kind of and how much food, water intake


Within 6 months of approval:

  • Ribavirin
  • Ulcer drugs (H2 receptor antagonists, Proton Pump Inhibitors)
  • Certain herbal medications & supplements (see Appendix 1. Because the list is extensive, we might suggest starting with the most commonly used according to FDA's Center for Food Safety and Applied Nutrition (www.cfsan.fda.gov) and have a multi-tiered approach, the most common at 6 months, the rest within a year.)
  • Antipsychotics, ie, chloroprothixen, zuclopenthixol, haloperidol
  • Seizure drugs, ie, anti-epileptics, anti-convulsants
  • Erectile dysfunction drugs, ie, sildenafil, tadalafil, vardenafil


Within 12 months of approval:

  • Street drugs (ecstasy, methamphetamine, heroin)—these studies can be undertaken safely and legally in The Netherlands, Switzerland, Spain
  • TB drugs (rifampin, rifabutin)
  • Pegylated interferon
  • Cardiac Drugs, ie, amiodaron, disulfiram, verapamil, beta blockers
  • Antibacterials, ie, roxythomicin, clindamycin
  • Anti-depressants, ie, buproprion, SSRIs, trazodone, TCAs
  • Hypnotics, ie, the non-benzodiazepine selective agonists of the GABA-A receptor complex, as well as benzodiazepines, flurazepam, zaleplon, diazepam
  • Antihistamines, ie, levocetirizine, loratadine, cetirizine, promethazine
  • Anti-fungals, ie, fluconazole, itraconazole, ketoconazole, voriconazole
  • Broncho-dilatators
  • Immunosuppresants, ie, cyclosporin A
  • Antimalarials
  • Uricostatics, ie, allopurinol
  • Cytotoxic drugs, ie, tamoxifen, paxlitacel, vincristine
  • Complementary and alternative medicines.


Not enough information is generated on specific sub-populations of the HIV+ patients that may use the drug at time of approval. We reiterate some common points here below, compliance to which is needed more so now than ever before:

Specific Populations

  • Gender percentages that more realistically reflect the population in which the drug is studied, ie, in the US there should be at least 30% women in all trials, including phase I and II pharmacokinetic trials.
  • Ethnic and racial differences, again to reflect the epidemic itself.
  • People with hepatic and renal insufficiency, especially those with HCV and/or HBV co-infection. For patients with mild liver impairment, are dose reductions needed/recommended? Ditto for those with more serious liver problems? Should more visits and laboratory analyses be performed on people with abnormal LFTs at baseline, including those with chronic hepatitis?
  • Pediatrics/Pregnant women data is rarely generated. Bioequivalence is often not demonstrated for a powder/liquid formulation, if there is one. Pediatric formulations need to be devised and studied in both newborns and infants. Most ARVs are defined as Class C for pregnancy; they have unknown placental passage (newborn:mother drug ratio), and often much after full FDA approval, long-term animal carcinogenicity studies are still not done; along with the post-marketing pregnancy registry, ARVs need to be studied and categorized for use in pregnant women.
  • Hyperbilirubinemia and other hepatic abnormalities need to be better characterised in drugs to be approved before approval.


There is an overdue role for well-planned and rigorous population PK in Phase III studies. Would the use of therapeutic drug monitoring (TDM) be helpful in managing different patients and patient populations? Studies to look at re-dosing via TDM in overdosed/underdosed patients need to be addressed. Because HIV+ patients are usually taking multiple interacting drugs, many factors may complicate the "normal" concentration-response relationships expected in the plasma (or intracellularly with NRTIs). Might TDM also be a useful tool in finding a dose that balances efficacy and toxicity (to maintain the targeted level of drug exposure, maintaining the therapeutic success window open without too much toxicity on one side or therapeutic failure on the other)? (2, 3)

Studies in HIV-negative volunteers may provide a more rapid and efficient way of obtaining the in vivo data for two- and three-way drug interactions. [It must be said that there are safety and ethical issues here as well as possible differences between HIV-negative and HIV-infected volunteers.] For antiretrovirals likely to be in wide use, we recommend that as many ARV three-way interactions as possible be done before approval. (4)

We consider that many of these studies mentioned above are important enough to somehow limit the marketing rights of a sponsor who does not carry them out. We are interested in further exploring what to do when a company does not offer all the data considered "minimal", without in any way keeping that drug from someone in a life or death situation. For example, were FDA to demand 30% women in all HIV trials, and another 20% of people with hepatic failure level C, what would happen if that were not achieved? Why would it not be achieved if FDA mandated it?

There needs to be a regulatory requirement for drug interaction data not to withheld but released early as an issue of public safety. Transparency and better definition need to be the hallmarks of all studies. [We refer to our support of the AMA and the growing association of scientific journals to publish all studies, both positive and negative, and new Open Access initiatives, like the Public Library of Science (www.plos.org).]

Thank you for your anticipated assistance. The community has been making the same PK/population requests of Industry for many, many years and greatly appreciate the agency's concern in providing Industry with direction in this regard. Please do not hesitate to contact me with questions or feedback.

Very truly yours,

Rob Camp
Antiviral Project Director
Treatment Action Group

cc: Deputy Director Jeffrey Murray, M.D. HFD-530 301-827-2338 CRPII/Rm. N413


Appendix 1: Complementary and alternative medicines (CAM) (vitamins/minerals, homeopathy, herbal, other) commonly used need to be looked at with ARV therapy. Although we realize that FDA does not regulate CAMs, the interactions of these with ARVs would be very helpful and possibly preventative of complications. A large percentage of people living with HIV use many of the products listed below. A good example of why these products need to be studied is the indinavir/hypericin interaction, where the metabolism of indinavir is speeded up and lower blood levels are the result, with the raised risk of treatment failure and resistance:

Echinacea, Ginseng, Evening primrose oil, Brewer's yeast, Cannabis, Soy lecithin, Hepatoprotective products, Cat's claw, Aloe vera, Propolis, Pollen, Bach's flowers, St John's wort (hypericin), Garlic supplements, Algae, Kombucha, Supplements such as zinc, selenium, vitamins, melatonin, carnitine, Co-Q, etc (5)

Refs:
1. Identifying metabolic differences in patient groups based on genetic polymorphism, or on other readily identifiable factors, such as age, race, and gender, can aid in interpreting results.

The effects of an investigational drug on the metabolism of other drugs and the effects of other drugs on an investigational drug's metabolism should be assessed relatively early in drug development so that the clinical implications of interactions can be assessed as fully as possible in later clinical studies.

– from FDA Guidance for Industry Population Pharmacokinetics, CDER and CBER, Feb 1999


Identifying metabolic differences in patient groups based on genetic polymorphisms, or on other readily identifiable factors such as age, race, and gender, could help guide the design of dosimetry studies for such populations groups. This kind of information also will provide improved dosing recommendations in product labeling, facilitating the safe and effective use of a drug by allowing prescribers to anticipate necessary dose adjustments. Indeed, in some cases, understanding how to adjust dose to avoid toxicity may allow the marketing of a drug that would have an unacceptable level of toxicity were its toxicity unpredictable and unpreventable.

Two of the major clinical reasons, as previously mentioned, are (1) to identify all of the major metabolic pathways that affect the drug and its metabolites and (2) to anticipate the effects of the drug on the metabolism of other drugs. With these objectives in mind, an understanding of the metabolic profile of a drug in vitro would be useful prior to the initiation of phase 2 studies and is especially important before phase 3 trials, when a broader population will be studied. This knowledge would permit the efficient design of clinical dose/response, interaction, and special population studies and also would enable adequate attention to be given to patient variability and potential interactions in phase 2 and 3 studies.

– from FDA Guidance for Industry Drug Metabolism/Drug Interaction Studies in the Drug Development Process: Studies In Vitro, CDER and CBER, April 1997


2. Acosta EP, Gerber J and the Adult Pharmacology Committee of the ACTG, Position Paper on Therapeutic Drug Monitoring of Antiretroviral Agents, AIDS research and Human Retroviruses, vol. 18, n. 12, 2002, pp 825-834.

3. FDA Guidance for Industry In Vivo Drug Metabolism/Drug Interaction Studies—Study Design, Data Analysis, and Recommendations for Dosind and Labeling, CDER and CBER, November 1999.

4. Jonathan Schapiro and Terrence Blaschke, personal communications.

5. Partial list compiled from two posters at the 5th International Workshop on Clinical Pharmacology of HIV Therapy, 1-3 March 2004, Rome, I.