The gender divide: how sex affects drug response

Matthieu L Kaltenbach
PharmD PhD
Professor of Pharmacology and Pharmacokinetics
Sylvain Dukic
MSc PhD
Assistant Professor of Pharmacology and Pharmacokinetics
Laboratoire de Pharmacologie et de Pharmacocinétique
Faculté de Pharmacie
Université de Reims Champagne Ardenne
France

Over the past decade, there has been a growing interest in obtaining adequate information about the effects of drugs in women. Indeed, although women are reported to use drugs more frequently than men and to have a higher incidence of adverse drug reactions, they have long been underrepresented as participants in clinical drug trials.(1,2)

After the tragedies caused by the use of thalidomide and diethylstilbœstrol (des) in pregnant women, the FDA issued a guidance in 1977 recommending the exclusion of women of childbearing potential in the early phases of drug development, except for life-threatening diseases. However, in later years that recommendation began to raise important ethical questions.

Acknowledging that excluding women from clinical trials was inadequate, the FDA revised its policy in 1993 and called for the study of both men and women in the evaluation of medicines.(3) This revised guideline, by lifting the restriction on women of childbearing potential, also emphasised the need to detect clinically significant sex-related differences. In 1998, the FDA went even further by requiring pharmaceutical companies to include and examine safety and efficacy data by sex, age and ethnicity in all investigational new drug (ind) applications.

In light of these recommendations, significant progress has been made, and numerous reports have now been published that shed some light on the potential existence and impact of sex-related differences in all aspects of clinical pharmacology.(4–10) As such, these differences in response can arise because of variation in a drug’s pharmacokinetics (ie, time course of drug concentrations in plasma or other tissues) or pharmacodynamics (ie, the body’s response to a given drug concentration). Consequently, the aim of this article is to briefly review the scope and clinical relevance of these findings and to explain how biological and molecular mechanisms can affect the kinetics and dynamics of therapeutic drugs.

The influence of menstrual cycle, pregnancy, menopause, oral contraceptives and supplementary oestrogens (as used in hormonal replacement therapy) on drug disposition and effects are not addressed here. The reader is referred to specific review articles as these issues deserve separate and more indepth attention.(11–15)

Sex-related differences in pharmacokinetics
Pharmacokinetics deals with the biological processes that affect the time course of drug concentrations in the body (see Figure 1). It is commonly separated into absorption, distribution, metabolism and excretion.

[[HPE06_fig1_80]]

Absorption
Several factors can contribute to sex-related differences in drug absorption. Women have lower gastric acid secretion (and thus a higher gastric pH), slower gastric emptying, altered bile composition and slower intestinal transit time than men.(16,17) These factors could in turn affect the rate and/or extent of absorption of oral medications. Yet few studies have thoroughly investigated sex differences in drug absorption, and clinically significant examples of such differences remain scarce. For example, aspirin administered orally as lysine salt was found to be absorbed more rapidly in women than in men, without any change in bioavailability. In contrast, another study found that the extent but not the rate of absorption of aspirin was higher in women than men, possibly due to a reduced first-pass effect.(8) Recently, mizolastine, a new nonsedating second-generation antihistamine, was reported to be absorbed about four times faster in women compared with men (mean absorption time of 0.67 hours versus 3.09 hours, respectively).(9)

Sex-related differences in first-pass metabolism in the gut wall or liver may also significantly affect drug bioavailability. Two well-documented examples concern ethanol and zolmitriptan.(9) Ethanol bioavailability after oral intake is significantly greater in women than in men due to a lower gastric alcohol dehydrogenase activity. Similarly, the bioavailability of zolmitriptan, a 5-HT(1D) agonist used in the treatment of acute migraine, has been reported to be much higher in women than in men, resulting in peak concentrations and AUC values approximately equal to twice those of men, although half-lives of elimination are similar. In the latter case, this sex difference was linked to a more extensive first-pass metabolism in men than in women. Its clinical relevance, however, is largely obscured by the fact that zolmitriptan is equally effective in both sexes.

Based on the limited amount of data available, it thus appears that, with regard to drug absorption, sex-related differences do not seem to be of major clinical importance.

Distribution
Sex-related differences in drug distribution mostly result from differences in physical constitution between men and women (eg, body weight, proportions of muscular and adipose tissue, body water spaces, organ size and blood flow). As such, women generally weigh less than men and have a higher proportion of body fat, two factors likely to limit the distribution of hydrophilic drugs. For example, several fluoroquinolones (ofloxacin, fleroxacin and grepafloxacin) exhibit smaller volumes of distribution, as well as lower oral clearances, in women than in men. Similarly, women have a lower volume of distribution for ethanol than men since this substance is known to distribute evenly into total body water. The volume of distribution of theophylline is also 10–22% smaller in women than in men. These differences, however, are small and can often be adjusted for by normalising pharmacokinetic parameters for lean or total body weight.(9)

Sex may also represent a source of variability in drug binding to plasma and tissue proteins. Although the concentration of albumin, the major drug binding protein in plasma, is largely unaffected by sex, the concentration of alpha(1)-acid glycoprotein is slightly lower in women due to the presence of endogenous oestrogens. Differences in binding might thus be expected for basic drugs that mostly bind to alpha(1)-acid glycoprotein. Lastly, differences in plasma concentrations of lipoproteins and specific transport proteins might modulate drug binding as well.

Reports of sex-related differences in drug binding, however, are scarce and often of limited clinical relevance. For example, although the free (pharmacologically active) fractions of diazepam, chlordiazepoxide, nitrazepam and imipramine have been reported to be higher in women than in men, the clinical significance of these findings remains unclear.(8,9)

Metabolism
Sex-related differences have also been demonstrated in hepatic phase I and phase II metabolism, suggesting that the activity of some cytochrome P450 isozymes and of several isoforms of glucuronosyl- and sulfo-transferases might be greater in men than in women.(18–20)

The cytochrome P450 system is a group of related enzymes or isozymes found primarily in the liver, but also in the gastrointestinal tract, lungs and kidneys. It is responsible for the oxidative metabolism of numerous drugs and endogenous substrates.(19) Individual isozymes are grouped into three major families (cyp1, cyp2, and cyp3), with seven primary isozymes each involved in the metabolism of a different set of drugs: cyp1a2, cyp2a6, cyp2c9, cyp2c19, cyp2d6, cyp2e1, and cyp3a4. It should, however, be noted that more than one isozyme can be involved in the metabolism of any given drug. The activity of these enzymes is dependent on genetic, physiological and environmental factors.(20) Over the past decade, it has also become increasingly apparent that the activity of specific cyp isozymes may be influenced by sex, most notably through endogenous hormonal influences (see Table 1).

[[HPE06_table1_82]]

cyp3a4 is one of the most abundant cyp isoforms in the human liver, accounting for approximately 30% of total hepatic cyp. It displays no apparent genetic polymorphism, has the broadest substrate specificity, and could be involved in the metabolism of more than 50% of all medications.

Although numerous studies have reported evidence of increased cyp3a4 activity in women compared with men (20–40% increase in clearance for drugs such as erythromycin, nifedipine, verapamil, diazepam, methylprednisolone and cyclosporin), the current thinking among pharmaceutical researchers is that drug transporters are in fact responsible for these sex-related differences in metabolism.(10) Indeed, drug transporters such as P-glycoprotein shunt drugs out of certain cells, and thus ultimately control access to the liver and gut enzymes. Since women have only 30–50% of the hepatic P-glycoprotein level of men, intracellular substrate availability is increased, resulting in a greater metabolic clearance for drugs that are cosubstrates of both cyp3a4 and P-glycoprotein. Thus, previously reported sex-related differences in cyp3a4 activity might just be the result of lower P-glycoprotein expression in women relative to men.

cyp2d6 (2% of total hepatic cyp) is the second most important cyp isoform in the metabolism of drugs. It displays a genetic polymorphism and is responsible for the oxidation of more than 40 drugs, including many antidepressants, antiarrhythmics, analgesics, and beta-blockers. Yet few studies have examined the influence of sex on cyp2d6-mediated metabolism. Slightly higher clearances have been reported in women for some cyp2d6 substrates, such as clomipramine, desipramine, ondansetron or propranolol. However, the relevance of these findings remains to be confirmed, since most studies were weakened by many confounders such as not normalising for body weight, concomitant use of oral contraceptives, and presence of other metabolic pathways. Therefore, at this point in time, there is only limited evidence to support the fact that the activity of cyp2d6 is dependent on sex.(9,10,20)

cyp1a2 (13% of total hepatic cyp) plays a minor role in drug metabolism as far as the number of drugs is concerned, but it is involved in the oxidation of theophylline, clozapine and ondansetron. It is the only cyp inducible by cigarette smoking. To date, fairly consistent data have demonstrated a lower cyp1a2 activity in women compared with men.(10,20)

Regarding other cyp isozymes, the bulk of the data suggests no significant sex difference in clearance of cyp2c9 and cyp2c19 substrates. In contrast, there is some evidence that the activity of cyp2e1 may be slightly higher in men than in women. However, cyp2e1 (7% of total hepatic cyp), an ethanol-inducible form of cyp, plays only a minor role in drug metabolism, and few clinically used drugs are affected: ethanol, chlorzoxazone, halothane, isoflurane and isoniazid. The clinical significance of this sex difference in drug metabolism thus appears very limited.(8,9,10,20)

Besides cyp isozymes, significant sex-related differences have also been identified in phase II conjugation reactions. For example, UDP-glucuronosyl-transferase activity has been suggested to be lower in women than in men, as demonstrated for five drugs predominantly metabolised by glucuronidation (temazepam, oxazepam, paracetamol, clofibrate, diflunisal). Similarly, there is some evidence that the activity of some sulfo-transferase isoenzymes is slightly lower in women than in men. In contrast, sex does not appear to affect the elimination of clofibrate or ibuprofen, and the conjugation of fenofibrate appears to be more important in women than in men. These somewhat conflicting results seem to indicate that sex-related differences in drug conjugation result only from sex differences in some glucuronosyl- and sulfo-transferase isoenzymes, but not others.(9,20) On average, and in spite of the limited data available, it appears that UDP- glucuronosyl- as well as sulfo-transferase activity is greater in men than in women.

In summary, sex-related differences in drug metabolism seem to exist in both phase I and phase II reactions. While data are still limited and not in complete agreement, the majority of studies seem to indicate that cyp2d6 activity is higher in women, whereas cyp1a2, cyp2e1, glucuronosyl- and sulfo-transferase activity is probably greater in men. cyp3a4, cyp2c9 and cyp2c19 activity is most likely not sex-dependent. These differences, however, are small and often do not warrant any adjustment in drug dosage.

Excretion
The influence of sex on the renal excretion of drugs remains poorly evaluated despite well-known differences in renal physiology between males and females. Women have a 10% lower glomerular filtration rate than men when normalised for body surface area. This might explain the lower digoxin (12–14% reduction) and cefotaxime clearances in women compared with men, since both drugs are primarily eliminated via renal excretion. Since the glomerular filtration rate is also known to be directly proportional to body weight, some apparent sex-related differences might simply be a weight effect.(8–10) Lastly, differences in active secretion and tubular reabsorption might also occur (as seen with amantadine), but remain to be fully characterised. However, published data to date seem to indicate that sex-related differences in renal excretion might have clinical significance only for drugs that are primarily excreted unchanged via the renal route.

Sex-related differences in pharmacodynamics
Although pharmacodynamic variability is often much larger than its pharmacokinetic counterpart, there is now a growing body of evidence that drug effects in women are sometimes different from those observed in men. As such, therapeutically relevant sex effects have been clearly identified regarding the clinical efficacy and adverse effects of many antipsychotic, antidepressant, analgesic, antihypertensive and antiarrhythmic drugs.(1,8,9,21,22)

For example, women show greater improvement and more severe adverse effects than men in response to antipsychotic medications, such as chlorpromazine.(16,17) This sex difference, commonly attributed to the antidopaminergic properties of oestrogens, may also be partly explained by a pharmacokinetic (higher blood levels) rather than a pharmacodynamic mechanism.(8) A number of studies have also shown that women respond differently from men to antidepressants. Men suffering from panic attacks are better treated with tricyclic antidepressants, whereas women respond better to monoamine oxidase inhibitors.(16,23) Women also seem to have a greater magnitude of response to serotonin agonists and selective serotonin reuptake inhibitors (SSRIs) than men.(8)

For cardiovascular drugs, there is now conclusive evidence that men and women respond differently to antihypertensive agents, and women appear to experience more severe adverse effects from these drugs than men.(8,9) Recently, a meta-analysis found that women are at greater risk than men of developing torsades de pointes from cardiovascular drugs that prolong repolarisation, such as quinidine, amiodarone, procainamide and disopyramide.(24) In fact, women have a longer QT interval (time period needed for the heart to recharge between beats) than men, and this difference in sensitivity to drugs that lengthen the QT interval could be due to the effect of sex hormones on the activity of potassium channels, which in turn govern repolarisation of the heart.(25) Lastly, men and women also appear to respond differently to antithrombotic therapy.(8)

Unfortunately, and except for a few well-designed studies, the clinical relevance of these findings remains to be demonstrated. Most of these results have been obtained with a limited number of subjects, without adequate control of the many confounding factors (age, ethnicity, menstrual cycle, comedications) known to influence drug effects. More research is definitely needed to assess the exact role of sex in drug pharmacodynamics.

Conclusion
Currently, scientific evidence of sex-related differences in pharmacokinetics and pharmacodynamics abounds. The mechanisms and origins of these differences are complex, not easily summarised and often poorly understood. Fortunately their clinical significance remains limited, and most drugs probably do not present a relevant sex difference in their pharmacokinetics and/or pharmacodynamics.

Except for a few select cases where sex-specific dosage recommendations should be provided, dosing based on body weight may well be considered as an effective approach to eliminate most sex differences in drug disposition.

References

1. American College of Clinical Pharmacy. Pharmacotherapy 1993;13:534-42.
2. Schmucker DL, Vesell ES. Clin Pharmacol Ther 1993;54:11-5.
3. Food and Drug Administration. Fed Regist 1993;58: 39406-16.
4. Institute of Medicine. Wizemann TM, Pardue ML, editors. Exploring the biological contributions to human health: Does sex matter? Washington DC: National Academy Press; 2001.
5. Berg MJ. J Am Pharm Assoc 1997; NS37(1):43–56.
6. Berg MJ. Pharmacological differences between men and women. In: Atkinson AJ, et al, editors. Principles of clinical pharmacology. New York: Academic Press; 2001.
7. Fletcher CV, et al. J Adolesc Health 1994;15:619-29.
8. Harris RZ, et al. Drugs 1995;50: 222-39.
9. Beierle I, Meibohm B, Derendorf H. Int J Clin Pharmacol Ther 1999;37:529-47.
10. Meibohm B, Beierle I, Derendorf H. Clin Pharmacokinet 2002;41:329-42.
11. Back DJ, Orme ML. Clin Pharmacokinet 1990;18:472-84.
12. D’Arcy PF. Drug Intell Clin Pharm 1986;20:353-62.
13. Kashuba AD, Nafziger AN. Clin Pharmacokinet 1998;34:203-18.
14. Gleiter CH, Gundert-Remy U. Eur J Drug Metab Pharmacokinet 1996;21:123-8.
15. Loebstein R, et al. Clin Pharmacokinet 1997;33:328-43.
16. Yonkers KA, et al. Am J Psychiatry 1992;149:587-95.
17. Kando JC, et al. Drugs 1995;50:1-6.
18. Meyer UA. J Pharmacokinet Biopharm 1996;24:449-59.
19. Wrighton SA, et al. J Pharmacokinet Biopharm 1996;24:461-73.
20. Anderson GD. J Gend Specif Med 2002;5:25-33.
21. Pollock BG. Psychopharmacol Bull 1997;33:235-41.
22. Schwartz JB. J Gend Specif Med 1999;2:28-30.
23. Thurmann PA, Hompesch BC. Int J Clin Pharmacol Ther 1998;36:586-90.
24. Makkar RR, Fromm BS, Steinman RT, et al. JAMA 1993;270:2590-7.
25. Ebert SN, et al. J Women Health 1998;7:547-57.

Resources
Journal of Gender Specific Medicine
W:www.mmhc.com/jgsm
Institute of Medicine report. Exploring the biological contributions to human health: does sex matter?
W:www.nap.edu/books/0309072816/html/index.html
NIH Office of Research on Women’s Health
W:www4.od.nih.gov/orwh/
Society for Women’s Health Research
W:www.womens-health.org