Pharmacokinetic study of different dosing regimens of levonorgestrel for emergency contraception in healthy women

Elof Johansson1, Vivian Brache2,6, Frank Alvarez2, Anibal Faundes2,3, Leila Cochon2, Sirpa Ranta4, Mark Lovern5 and Narender Kumar1

1 Center for Biomedical Research, The Population Council, New York, NY, USA, 2 PROFAMILIA, Santo Domingo, Dominican Republic, 3 CEMICAMP, Campinas, SP, Brasil, 4 Steroid Research Laboratory, Institute of Biomedicine, University of Helsinki, Helsinki, Finland and 5 Pharsight Corporation, Cary, NC, USA


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: Levonorgestrel (LNG) is a commonly used progestin for emergency contraception; however, little is known about its pharmacokinetics and optimal dose for use. METHODS: Serum levels of LNG and sex hormone-binding globulin (SHBG) were measured in five women who received three different regimens: A: 0.75 mg LNG twice with a 12 h interval; B: 0.75 mg twice with a 24 h interval; and C: 1.50 mg in a single dose, with a washout period of 28 days between each treatment. Blood samples were taken before pill intake and at 1, 2, 4, 8 and 12 h after each dose, every 12 h up to day 4 and every 24 h until day 10. LNG and SHBG were measured in all samples. RESULTS: Maximum LNG concentrations were of ~27 nmol/l for treatments A and B, and close to 40 nmol/l for treatment C. The area under the curve was significantly higher for treatment C during the first 12 h, and significantly lower for treatment B during the first 24 h. After 48 h and up to 9 days from onset of treatment, serum LNG levels were similar in all three regimens. SHBG levels remained stable for 24 h, decreasing to 60% of the initial value from day 5 until day 10, with no difference between regimens. CONCLUSIONS: The similarity of LNG serum levels obtained with one single dose of 1.5 mg or two doses of 0.75 mg with a 12 h interval justify a clinical comparison of these two regimes.

Key words: emergency contraception/levonorgestrel/pharmacokinetics/SHBG


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Emergency contraception (EC) is a woman's only reliable option for preventing pregnancy after unprotected sexual intercourse or contraceptive method failure. Although EC has been available for >20 years, it has been an underused modality among contraceptive methods. In recent years, it has regained relevance in the field of reproductive health, with the growing realization that EC could save millions of women (and health care money) from experiencing unplanned or unwanted pregnancies (Berer et al., 1995Go).

Although use of levonorgestrel (LNG) for EC is increasing, the knowledge of the pharmacokinetics of LNG when used for EC and the selection of the dose currently recommended is based on limited data. In the early 1970s, pioneering studies explored various doses (0.15, 0.25 and 0.4 mg) of LNG-only for post-coital contraception (Larranaga, 1971Go; Kesseru et al., 1974Go). Later studies evaluated the post-coital use of 0.75 mg of LNG (Seregely, 1982Go; Bhattacharjee et al., 1987) with a different concept of multiple use per cycle after each act of unprotected coitus. This concept was not pursued due to frequent cycle disruption and bleeding irregularities, as well as lower effectiveness than achieved with the combined pill (Larranaga, 1971Go; Kesseru et al., 1974Go; Seregely, 1982Go; World Health Organization, 1987Go).

A randomized comparative study of EC using the LNG (0.75 mg for two doses, 12 h apart) and Yuzpe regimes (two doses of 100 µg of ethinyl estradiol and 0.5 mg LNG, 12 h apart) showed equal effectiveness at 2.4 and 2.6% respectively (Ho and Kwan, 1993Go). However, fewer side-effects were observed among users of the LNG-only regimen. A large multicentre study reported higher efficacy and reaffirmed a lower incidence of side-effects with the LNG regimen (two doses of 0.75 mg, 12 h apart), compared with the Yuzpe method (World Health Organization, 1998Go). Based on this study, the LNG regimen would be the method of choice.

An inconvenience of the current LNG regimen is the required 12 h interval, which may be cumbersome for some women. This schedule of use was selected without a properly designed schedule-finding study; therefore, it is not known whether the same protection may be achieved with a 24 h interval between doses or with both pills taken together. Furthermore, there are no data available on the pharmacokinetics associated with the current recommended mode of administration of LNG for EC. Four studies have investigated different doses and regimens of LNG for EC: three studies involving a single administration of 0.75 mg LNG (Shi et al., 1988Go; Landgren et al., 1989Go; He et al., 1990Go) and one using a single dose of 1.0 mg LNG (Weiner et al., 1976aGo). These studies have demonstrated that LNG has a long half-life, which could be explained by the high affinity of LNG for sex hormone-binding globulin (SHBG) (Victor et al., 1976Go).

The objective of this study was to describe the pharmacokinetics of the currently accepted LNG EC regimen consisting of two doses of 0.75 mg LNG (Norlevo®; HRA Pharma, Paris, France) given 12h apart (treatment A), as well as the pharmacokinetics associated with two additional regimens: two doses of 0.75 mg LNG given 24 h apart (treatment B), and a single dose of 1.5 mg LNG (two 0.75 mg tablets; treatment C).


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Subjects
A total of five non-breastfeeding healthy women, attending the Reproductive Health Clinic at PROFAMILIA, Dominican Republic, were enrolled in the study. All participants read and signed a written informed consent before enrolment in the study. The inclusion criteria were: 18–45 years of age; haemoglobin levels >11 g/dl; body weight 55–80 kg; protection against pregnancy by use of barrier methods, abstinence or surgical sterilization; no use of hormonal contraceptives in the month before enrolment or of injectable contraceptives 4 months prior to enrolment; and normal liver function. Each subject included in the study was assessed by a medical history, a complete general and physical examination and determination of eligibility through pre-admission screening. Each subject provided a blood sample before the initiation of the study for haemoglobin measurement and for the standard assessment of liver function: bilirubin, alanine aminotransferase (ALT or SGPT), aspartate aminotransferase (AST or SGOT), alkaline phosphatase and albumin.

Design of the study
Each woman participated in the three arms of the study with a washout period between treatments of 27–28 days. The treatments were, A: two doses of 0.75 mg LNG given 12 h apart; treatment B: two doses of 0.75 mg LNG given 24 h apart; and treatment C: a single dose of 1.5 mg LNG (two 0.75 mg tablets).

Blood samples were taken just before pill intake and then serially at 1, 2, 4, 8 and 12 h after each dose; samples were then taken every 12 h on days 2, 3 and 4, and every 24 h on days 5, 6, 7, 8, 9 and 10. Clinical staff provided the LNG tablets at the time of intake. Subjects remained at the clinic for 12 h following each dose.

Assays
LNG and SHBG were assayed at the Steroid Research Laboratory, Helsinki, Finland. LNG was measured by a conventional radioimmunoassay as previously described (Weiner et al., 1976bGo). The steroid was extracted with diethyl ether and then measured by radioimmunoassay with a specific antibody and tritium-labelled LNG as a tracer, obtained from Schering AG (Berlin, Germany). The precision of the assay was evaluated by determining the intra- and inter-assay coefficients of variation (CV) in the optimal range of assay. The intra- and inter-assay CVs were 5.6–9.9% and 8.0–11.0% respectively. With the purpose of avoiding inter-assay bias, all samples of each subject (three arms) were analysed in the same radioimmunoassay run.

The concentration of SHBG in the sera was measured by a time-resolved fluoroimmunoassay, using a commercial kit (DELFIA) manufactured by Wallac Finland Oy, Turku, Finland. According to the manufacturer, the intra- and inter-assay CVs were 1.3–1.8% and 5.1–10.1% respectively. Free LNG index was calculated as the ratio between serum LNG (nmol/l) and SHBG (nmol/l) at 12, 24 and 24 h thereafter until day 9.

Comparison of pharmacokinetic parameters
Each individual concentration–time curve was fitted according to a two-compartment model using WIN NONLIN software (Pharsight Corporation, Cary, NC, USA). All treatments were compared in a single pharmacokinetic analysis. The area under the concentration–time curve, maximal concentration (Cmax), time to reach maximal concentration (Tmax) and biological half-life were obtained from software-calculated values for each treatment. Pharmacokinetic parameters are shown as geometric means with 95% confidence intervals and were compared using one-way analysis of variance (ANOVA).


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The weight range of the five participating women was 59–77 kg (mean ± SD: 67.6 ± 6.6). Mean serum LNG concentrations following the various treatments are shown in Figure 1Go. The pharmacokinetic parameters obtained following WIN NONLIN analysis are shown in Tables I and IIGoGo. The LNG Cmax after treatment C was ~50% higher than the Cmax obtained after treatments A and B (P = 0.03). Maximum serum concentrations of LNG were reached between 1.5 and 1.8 h after the administration of 0.75 mg and 2.6 h after the administration of 1.5 mg of LNG (Table IGo). The peak observed following the second dose, 12 or 24 h later, was slightly higher than the initial peak (32.8 versus 29.5 nmol/l and 30.0 versus 27.2 nmol/l respectively). By 48 h all three treatment arms had very similar LNG levels (A = 6.2, B = 7.4 and C = 6.3 nmol/l; Figure 1Go). It is interesting that due to the long biological half-life of LNG, serum LNG levels >1.3 nmol/l were observed 5 days after administration of the first dose, and levels near 0.6 nmol/l were present 1 week post-administration, with no discernible difference between the three regimens from days 3–9 after initiation of treatment.



View larger version (39K):
[in this window]
[in a new window]
 
Figure 1. Mean levonorgestrel (LNG) serum levels in women following oral administration of three different regimes of LNG for emergency contraception.

 

View this table:
[in this window]
[in a new window]
 
Table I. Geometric mean and 95% confidence interval (CI) of pharmacokinetic parameters of LNG in women
 

View this table:
[in this window]
[in a new window]
 
Table II. Pharmacokinetic parameters of LNG in women treated with LNG for emergency contraception (individual values)
 
The area under the curve (AUC) calculated for the first 12 h after LNG administration was very similar for treatments A and B and significantly higher for treatment C (P = 0.00014; Table IGo). The AUC for the 24 h period was lowest for treatment B (P = 0.0067), while there were no significant differences between treatments A and C (Table IGo). The total AUC (including 9 days of observation) for treatment C was significantly higher than for treatments A and B (P = 0.0003; Table IGo).

SHBG (mean ± SD) levels were 51.2 ± 21.7, 53.0 ± 15.1 and 53.4 ± 11.3 nmol/l, just before the administration of treatments A, B and C respectively. SHBG serum levels remained essentially unchanged during the first 24 h after drug intake (Figure 2Go). The first decrease was noted in the 48 h sample (~10%), followed by a continuous, regular decrease down to ~60% of baseline values at day 5 post-initiation of treatment. SHBG remained depressed at the same level through to day 9 after initiation of treatment, with no difference between the three regimens (Figure 2Go). There was a good correlation between baseline SHBG levels and LNG concentrations (Cmax) for treatments A and B (r = 0.79 and 0.86 respectively), but not for treatment C (r = 0.387) which corresponded to the highest dose and LNG serum levels. There was no correlation between body mass index of the five subjects and LNG levels (Cmax) achieved with any of the three regimes.



View larger version (17K):
[in this window]
[in a new window]
 
Figure 2. Serum SHBG levels (expressed as percentage of baseline) in women following oral administration of three different regimes of LNG for emergency contraception. Treatments A (0.75 mg twice with 12 h interval), B (0.75 mg twice with 24 h interval) and C (single administration of 1.5 mg).

 
The free LNG index curve followed a similar pattern as the LNG concentration curve, with higher levels for treatment C at 12 h and for treatment A at 24 h. The peak after the second dose with treatment B was missed, because SHBG was not measured in the 36 h sample. No differences between regimes were observed from 48 h to day 9 after treatment.

Adverse events were reported within 72 h after intake by one woman during treatment A, by two using treatment B and by one during use of treatment C. The reported side effects were nausea (three women), sleepiness (two women) and headache (one woman). No subject reported vomiting.


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The main purpose of this study was to compare the standard EC LNG regime with two alternative schedules: increasing the interval to 24 h between doses or a single administration of the total dose. The results of this study showed that the AUC after administration of one single dose of 1.5 mg was greater than after two doses of 0.75 mg with a 12 or 24 h interval between doses.

The pharmacokinetic parameters observed in this study, with the administration of 0.75 mg of LNG, fit within those reported in China by Shi et al. and also by He et al. after administration of Postinor (He et al., 1990Go; Shi et al., 1998). On the other hand, Landgren et al. found peak LNG concentrations much lower than ours (16 nmol/l versus 29 or 27 nmol/l, at 2 h after administration) (Landgren et al., 1989Go). In addition, these last authors found that LNG levels were undetectable 72 h after one single administration, while we found mean levels of 0.6 pmol/l 1 week after two successive doses with 12 or 24 h intervals. Pharmacokinetic studies of the Chinese-manufactured pill of 0.75 mg (He et al., 1990Go) showed lower concentrations of LNG and a delay in reaching peak levels. While the Tmax observed with European-manufactured pills ranged between 1.5–2.0 h, the Tmax observed with the Chinese pill ranged between 3–4 h. The authors attributed these differences to the lesser degree of micronization of LNG in the Chinese formulation (He et al., 1990Go).

The comparison of the data from all these studies underlines the importance of repeating these analyses in different settings, given the variability between sites, subjects and formulations.

SHBG did not seem to influence LNG serum levels during the initial 2 days after treatment, since SHBG was unchanged during the first 48 h after initiation of treatment. The decrease in SHBG was observed at a time when the expected contraceptive effect already should have occurred.

An earlier study on the effect of the Yuzpe regime upon ovarian function (Croxatto et al., 2002Go) showed that the effect of hormone administration on gonadotrophin levels was already observed 24 h after the first dose. Therefore, there are good reasons to believe that the bioavailability of LNG during the first 12 h after administration is critical to achieve the expected effect on ovulatory function and, possibly, a local effect on sperm penetration. If this is the case, the administration of a single larger dose would be more effective than the same amount of LNG divided into two doses administered 12 h apart.

We do not know, however, if there is a real need for such high LNG plasma levels as those observed after administration of the 1.5 mg dose (~40 nmol/l). It may well be that a single dose of 0.75 mg would be sufficient to achieve maximal biological effect, and there would be no advantage with the administration of a higher dose. In fact, earlier reports suggest that even a lower single dose, 0.4 mg of LNG, could be sufficient to cause the desired contraceptive effect (Kesseru et al., 1974Go). Furthermore, in a comparative study of Postinor and the Chinese-manufactured pill with 0.75 mg of LNG, the same clinical effectiveness with both formulations was observed, in spite of considerably lower LNG serum levels attained with the Chinese formulation (Cmax 18.9 nmol/l) than after the administration of Postinor (Cmax 33.9 nmol/l) (He et al., 1990Go). In addition, when Norplant® contraceptive implants were inserted during the advanced follicular phase (days 8–13), ovulation inhibition was achieved in 60% of users in the first cycle of use, even though peak levels reached only 3 nmol/l at 24–72 h post-insertion. This level is well below the peak LNG concentrations observed with the three regimens described above (Brache et al., 1999Go).

There are obvious practical advantages of administering a single dose over two doses given 12 h apart; a single dose would improve compliance and eliminate the need to disrupt sleep for drug intake. A possible disadvantage of the higher single dose could be greater intolerance with more side-effects, including the possibility of vomiting. In this small sample no evidence of greater intolerance and no episodes of vomiting were observed.

On the other hand, administration of the second dose 24 h after the first one (treatment B) was associated with a lower AUC during the first 24 h, as should be expected. It is doubtful that the rise in plasma level after the second administration, 24 h later, would make a significant contribution to the mechanism of action of EC. Thus, although the 24 h interval is programmatically convenient, it may not offer advantages over one single administration of 0.75 mg of LNG.

The overall results of this study suggest that the clinical comparison of the standard LNG regimen with single administration of the entire dose is fully justified. In addition, it is worth considering a single dose of 0.75 mg as a potentially equally effective alternative to the standard LNG regimen. Moreover, the effects of lower doses of LNG on gonadotrophin levels and the ovulatory process should be explored, in order to identify the minimal dose of LNG that could be effective as EC.


    Acknowledgements
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The authors gratefully acknowledge Ms Margaret Small for editorial assistance.


    Notes
 
6 To whom correspondence should be addressed at: PROFAMILIA, Biomedical Research Department, P.O. Box 1053, Santo Domingo, Dominican Republic. E-mail: biomedica{at}codetel.net.do Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Berer, M., Coutinho, E., Delano, G., Ellertson, C., Garza-Flores, J., Glasier, A., Greenslade, F., von Hertzen, H., Huezo, C. and Kapoor, I. (1995) Consensus statement on emergency contraception. Contraception, 52, 211–213.[ISI][Medline]

Brache, V., Blumenthal, P.D., Alvarez, F., Dunson, T.R., Cochon, L. and Faundes, A. (1999) Timing of onset of contraceptive effectiveness in Norplant® implant users II. Effect on the ovarian function in the first cycle of use. Contraception, 59, 245–251.[ISI][Medline]

Croxatto, H.B., Fuentealba, B., Brache, V, Salvatierra, A.M., Alvarez, F., Massai, R., Cochon, L., Faundes, A. (2002) Effects of the Yuzpe regimen, given during the follicular phase, on ovarian function. Contraception, 65, 121–128.[ISI][Medline]

He, C.-H., Shi, Y.-E., Liao, D.-L., Zhu, Y.-H., Xu, J.-Q., Matlin, S.A., Vince, P.M., Fotherby, K. and Van Look, P.F.A. (1990) Comparative cross-over pharmacokinetic study on two types of postcoital contraceptive tablets containing levonorgestrel. Contraception, 41, 557–567.[ISI][Medline]

Ho, P.C. and Kwan, M.S. (1993) A prospective randomized comparison of levonorgestrel with the Yuzpe regimen in post-coital contraception. Hum. Reprod., 8, 389–392.[Abstract]

Kesseru, E., Garmendia, F., Westphal, N. and Parada, J. (1974) The hormonal and peripheral effects of d-norgestrel in postcoital contraception. Contraception, 10, 411–424.[ISI][Medline]

Landgren, B.M., Johannisson, E., Aedo, A.R., Kumar, A. and Shi, Y.-E. (1989) The effect of levonorgestrel administered in large doses at different stages of the cycle on ovarian function and endometrial morphology. Contraception, 39, 275–289.[ISI][Medline]

Larranaga, A. (1971) d-Norgestrel in postcoital administration. Preliminary report (abstract only). In Kleinman, R. and Pickles, V.R. (eds) Seventh World Congress, Fertility and Sterility, October 1971, Amsterdam, Netherlands, Excerpta Medica. (International Congress Series No. 234a). pp. 87–88.

Seregely, G. (1982) Results of a multicentre trial of Postinor. Therapia Hungarica, 30, 72–78.[Medline]

Shi, Y.-E., Zheng, S.-H., Zhu, Y.-H., He, C.-H., Yu, P.-P. and Fotherby, K. (1988) Pharmacokinetic study of levonorgestrel used as a postcoital agent. Contraception, 37, 359–369.[ISI][Medline]

Victor, A., Weiner, E. and Johansson, E.D.B. (1976) Sex hormone binding globulin: the carrier protein for d-norgestrel. J. Clin. Endocrinol. Metab., 43, 244–247.[Abstract]

Weiner, E., Victor, A. and Johansson, E.D.B. (1976a) Plasma levels of d-norgestrel after oral administration. Contraception, 14, 563–570.[ISI][Medline]

Weiner, E., Victor, A. and Johansson, E.D.B. (1976b) Plasma levels of d-norgestrel, estradiol and progesterone during treatment with Silastic implants containing d-norgestrel. Contraception, 14, 81–92.[ISI][Medline]

World Health Organization (1987) Special programme of Research Training in Human Reproduction. Postcoital contraception with levonorgestrel during the peri-ovulatory phase of the menstrual cycle. Task Force on Post-Ovulatory Methods for Fertility Regulation. Contraception, 36, 275–286.[ISI][Medline]

World Health Organization (1998) Special Programme of Research, Development, and Research Training in Human Reproduction. Task Force on Postovulatory Methods of Fertility Regulation. Randomized controlled trial of levonorgestrel versus the Yuzpe regimen of combined oral contraceptives for emergency contraception. Lancet, 352, 428–433.[ISI][Medline]

Submitted on November 1, 2001; accepted on February 12, 2002.