ORIGINAL ARTICLE

Comparative contraceptive efficacy and mechanism of action of the norgestimatecontaining triphasic oral contraceptive ROBERTS . LONDON,M.D.," ALCIDECHAPDELAINE, M.D.,b DAVIDUPMALIS,M.D.,' WILLIAMOLSON,PH.D.," AND JUDITHSMITH,PH.D" From Kaiser Permanente, Baltimore, Maryland,a the Hdpital Maisonneuve-Rosemont, Quebec, Canada,band The R W. Johnson Pharmaceutical Research Institute, Raritan, New JerseyC

Acta Obstei Gynecol Scanii 1992; Vol 71, Suppl 156: 9-14

Norgestimate (NGM), a derivative of 19-nortestosterone with very specific affinity for the progesterone receptor, has been used in combination with ethinyl estradiol (EE) at low doses in both monophasic and triphasic oral contraceptives (OCs). An open-label comparative clinical trial was conducted with 4,234 healthy women using triphasic levonorgestrel (LNG)/EE and NGM/EE through a total of 22.312 menstrual cycles. Contraceptive efficacy was excellent with both preparations, with no statistically significant between-regimen differences in pregnancy rates. The theoretical Pearl index was 0.12 for the NGMlEE triphasic, and 0.34 for the LNGlEE triphasic. Adverse experiences in both groups were typical of those that may occur among women taking low-dose OC agents. The percentage of subjects discontinuing treatment because of use-related adverse experiences was similar with the two preparations: 8.6% for the NGM/EE triphasic and 6.8% for the LNGiEE triphasic. In a separate mechanism of action study, specific endocrine parameters were investigated in 20 subjects using the NGM/EE triphasic for 4 cycles. Ovulation suppression was demonstrated in statistically significant decreases from pretreatment values in serum levels of luteinizing hormone, follicle-stimulating hormone, progesterone, and estradiol. Significant on-treatment increases in serum levels of sex hormone binding globulin evidenced minimal androgenicity. All hormonal values returned to or toward normal in the post-treatment cycle. The study results support those obtained in large noncornparative studies of the NGM/EE triphasic. This phased-dose combination suppresses ovulation and is a very effective, minimally androgenic contraceptive agent with a good safety profile. Key words: Norgestimate. ovulation suppression, SHBG, mechanism of action

Introduction The 1990s will be an exciting decade for clinicians using combined oral contraceptives (OCs) because of the introduction of pills containing previously unavailable progestins. Some of these new contraceptive options will be pills containing norgestimate (NGM), a 19-nortestosterone derivative with very specific affinity for the progestin receptor. However, there is limited clinical information available from

United States comparative trials using these new contraceptives. In U. S. multicenter trials, monophasic N G W ethinyl estradiol (EE) pills have proved to be safe and effective (1). To continue the trend of decreasing the doses of steroid hormones in OCs to minimize side effects, pharmacologists have developed a new phased-dose preparation containing EE and NGM (Ortho TriCyclena, Tri-CilestB). This regimen combines a Acta Obstet Gynecol Scand Sipp/ 156 (1992)

10

R. S. London

300

r

et

a/.

Norgestimate/EE Regimen 250

200

NGM EE

c19 100

-

n 0

7

Days

14

21

LevonorgestrelfEE Regimen 150

c1g

0 LNG

100

EE

parative multicenter study of the two preparations used through six menstrual cycles. The steroid content and the design of the two triphasic preparations are illustrated in Fig. 1. After informed consent was obtained, subjects who met the entrance critcria were assigned randomly to one regimen or the other. Contraceptive efficacy was assessed by evaluation of pregnancies and pregnancy rates. Case evaluation was blinded to the preparation being used. Safety was assessed by analysis of type and incidence of adverse effects and results of specialized laboratory tests, including an androgen profile in a subset o f women. The subjects were evaluated before treatment, near the end of treatment cycle 3 , and after treatment cycle 6.

50

n 0

6

11

21

Days

Fig 1 . Dose regimens of NGM/EE and LNG/EE triphasic formulations.

daily monophasic 35-pg dose of EE with three stepwise increasing dosages of NGM and was developed to minimize unwanted metabolic effects by reducing the total dose of progestin. Effects of the NGM/EE triphasic on lipid and carbohydrate metabolism were investigated over a total of 28,527 treatment cycles in 1,783 women (2). Changes in lipid/lipoprotein parameters and glucose and insulin levels were indicative of minimal androgenicity. To determine how the actual clinical performance of this new triphasic OC compares with that of the widely used triphasic formulation containing the gonane levonorgestrel (LNG), an open comparative clinical trial was performed at 29 centers throughout the United States. These findings will be reviewed. Additionally, because the mechanism of action of new progestins is of both clinical and research interest, a second investigation was undertaken to elucidate information concerning the effects of NGMiEE triphasic pills on specific endocrine parameters. The findings of this study are of interest not only to determine how pregnancy is averted using the new pills, but also to account for some of the protective effects associated with the use of combined OC agents, including the suppression of functional ovarian cysts (3). These findings will also be presented.

Methods The first investigation was designed to evaluate the safety and efficacy of triphasic NGM/EE when cornpared with a triphasic LNG/EE pill. This was a cornAcru O b ~ r e rGynecol Scund Sirppl 156 (19Y2)

Study population

The subjects were to be healthy women aged 18 t o 38 years with normal menstrual cycles, who were sexually active and not pregnant. No evidence o f cervical dysplasia was to be present on a Papanicolaou smear. Postpartum and post-termination subjects were to be admitted following the next spontaneous menses. All means of contraception other than the test preparations were to be discontinued before the study. If an intrauterine device had been used, at least one normal menstrual cycle was to follow its removal before entry o f the subject into the study. Among the exclusion criteria were a history o f vascular disorders, notably thrombophlebitis or thromboembolism, and a history of disorders possibly related to estrogen-containing products. Use o f an investigational drug or device within the previous 30 days was also an exclusion criterion. The subjects were required to record pill usage, any concomitant therapy, and any adverse experiences. Subjects who discontinued the study drug were to be reexamined as they were at entry. The second investigation was designed to evaluate the endocrine effects of the triphasic NGMiEE pill. This was a 6-cycle study comprising a pretreatment cycle, 4 treatment cycles, and a post-treatment cycle. Baseline studies taken in the control cycle included a Papanicolaou smear; hematology; blood chemistry: urinalysis; and physical, gynecologic, and ophthalmologic examinations. In each of the four treatment cycles, the NGM/EE triphasic was used by subjects for 21 days, followed by 7 days without treatment. Measurements were repeated at the end of treatment cycle 4 and at the end of the post-treatment cycle. In total, 20 healthy women aged 23 to 37 years were enrolled, and each had documented ovulation in the cycle preceding treatment. For the endocrine investigation, fasting blood

Efficacy and mechanism of action Tablc I . Comparison of demographic data

Characteristic Mean age (yr) Range ( y r ) Race, no. ('Y") White Black Oricntiil Othcr Total Smokers. no. ( % )

NGM/EE triphasic

LNG/EE triphasic

25.6 ? 4.7 15-41

2.5.5

2.014 (87.1) 220 (9.5) 73 ( I . O ) 55 (2.4) 7,312 617 (26.7)

2.040 (87.9) 218 (9.4) 18 (0.8) 45 (1.9) 2,321 612 (26.4)

* 4.5

1640

'' Of the 4,633 patients admitted to the study, 3,234 were evaluable for efficacy.

samples were drawn between 7 A M and 10 A M on multiple cycle days. Samples were frozen and assaycd later for estradiol (El), progesterone (P), folliclc-stimulating hormone (FSH), and luteinizing hormone (LH) by standard radioinimunoassay techniques. Data ana1y.si.r Contraceptive efficacy was assessed by the Pearl index. The Pearl index is defined as the number of pregnancies per 100 woman-years of exposure. The log rank test and the Gehan generalized Wilcoxon test were used to determine differences in pregnancy rates between regimens during the test period. Statistical tests were two-tailed and conducted at the 5%) significance lcvcl without adjustments for multiple comparisons.

Results In the investigation for safety and efficacy, a total of 2.1 10 women aged 16 to 41 years took the NGM/EE triphasic for at least one cycle; 1,597 completed 6 cycles of treatment. A total of 2,124 women aged 16 to 30 years took the LNG/EE triphasic for at least one cycle; 1,667 women completed 6 cycles. There were 11,006 cycles of treatment with NGM/EE and 11,306 cycles with LNG/EE. The two groups were similar demographically at the initiation of the investigation (Table I). The demographic data for the two treatment groups were comparable at baseline. Contraceptive efficacy Excellent contraceptive efficacy was demonstrated with both regimens during the study. There were no statistically significant differences between the prep-

11

arations in pregnancy rates. Pregnant patients were assessed with regard to compliance in the cycle of conception. Seven of 8 pregnancies occurred in the NGM/EE triphasic group hecause patients failed to take the preparation daily as prescribed. There were 7 pregnancies with the LNG/EE triphasic, 4 involving noncompliance. and 3 product failures. The theoretical Pearl index, which excludes from calculation those pregnancies where noncompliance was a factor, were 0.12 and 0.34 with the NGM/EE triphasic and the LNG/EE triphasic, respectively; the overall Pearl indexes, which include pregnancies in noncompliant patients, were 0.94 and 0.80, respectively. Safety and adverse effects A review of all adverse experiences revealed that they are characteristic of those that may occur among women using low-dose OCs (Table 11). These included headache, nausea, mastodynia, and dysmenorrhea. Adverse experiences that resulted in the discontinuation of OC use occurred primarily in the first three cycles of the study. Among those with adverse experience information, the number of subjects who discontinued treatment because of userelated adverse experiences was similar with the two preparations: NGM/EE triphasic, 8.6'X (I82 of 2,115 subjects); LNGlEE triphasic, 6.8% (135 of 2,132 subjects). Androgen effects A subset of 45 subjects was evaluated for SHBG changes. Values for SHBG increased from baseline to cycle 6 with triphasic NGM/EE (56.7 nmol/L), but decreased 6.0 nmol/L with triphasic LNG/EE, a net difference of 62.7 nmol/L. This represents a 68.6% increase with triphasic NGM/EE vs a 6.1"An decrease with triphasic LNG/EE, a between-regimen difference significant at the 5% level. Free testosterone levels dropped 18.2% from baseline to cyclc 6 in the Table 11. Adverse experiences*

Adverse experience Headache Upper respiratory infection Pharyngitis Nausea Mastodynia Vaginitis Dysmenorrhea *Incidence >5%

NGM/EE triphasic (N=2,1 IS)

LNGlEE triphasic (N=2,132)

29.31

29 .so

12.15 501 10.87

I I ')I

5.86 7.14 11.63

S.86 10.32 3.97 7.97 11.87

12

R . S. London et af. 60

-

50

MlUlml

40

-

-

30 20

Pre-therapy

-

' & o

-.

10

Post-therapy

Cycle 4

%----+----m

0

I

I

I

4

8

11

I

d

I

a

,

14 16 18

Fig 2. Mean levels of luteinizing hormone. Blood samples for luteinizing hormone, follicle-stimulating hormone, progesterone, and estradiol were drawn on multiple days. Each data point represents 2 or 3 days.

21

24

27

Days

NGMlEE group, but increased 30.8% in the LNG/EE group. This difference was not statistically significant.

Hormonal function In the investigation of the N G W E E triphasic effects on ovarian, pituitary, and adrenal functions, the new OC formulation suppressed ovulation in all 20 women, as evidenced by on-treatment decreases in serum levels of L H , FSH, P, and E, (Figs 2-5). Changes in serum levels of the other endocrine analytes (eg, increased growth hormone and thyroidstimulating hormone concentrations, and decreased dehydroepiandrosterone sulfate) were statistically significant but were not considered clinically important. The exception was a significant increase in sex hormone binding globulin (SHBG), a manifestation of low androgenicity. At cycle 4, SHBG levels showed a change from pretreatment of 204%. Values for all the analytes returned to or toward normal in the post-treatment cycle, during which ovulation was again demonstrable in 19 of the 20 subjects. The mean post-treatment concentrations of FSH, L H , P, and Ez were similar to pretreatment concentrations, indicating a rapid return to ovulatory status.

Over the course of the study, there were n o clinically meaningful changes in vital signs o r in the results of the examinations or laboratory tests.

Discussion In the near future, practitioners are going to be faced with clinical choices about using OCs containing the new progestins in their daily practices. These two investigations were designed to provide a body of clinical and pharmacologic information about the effects of one of these new pills, a new triphasic NGM-containing OC, on a population of healthy women. When evaluating the efficacy of a hormonal OC, however, the effectiveness of any agent should be determined as a direct function of the reduction in the probability of conception, but fertility and fecundity are not readily quantified. Contraceptive efficacy is therefore usually assessed from the number of unplanned or accidental pregnancies that occur during the time that a contraceptive agent is used. Ideally, direct comparisons of products form the basis for efficacy assessments. Analysis of the comparison data in the large study reported here shows that excellent contraceptive efficacy was

20 16

MlUlrnl '2

:

*-.

8

Pre-therapy

\

-..

*m-

-

.-.=----.----* \--0

4.

Post-therapy Cycle 4

4

01

I

I

I

4

8

11

1

1

Days Actu Ohstrt Gynecol Scmd Sicppl 156 (IY92)

1

t 4 16 18

I

I

21

24

I

27

Fig 3 . Mean levels of follicle stimulating hormone.

Efficacy and mechanism of crction

13

ported here, the significant on-treatment decreases of the pituitary gonadotrophins and P support an antiovulatory mechanism of action with the NGMiEE triphasic formulation. The rapid return to normal gonadotropic function portends an excellent return to fertility after discontinuation of the pills, a very desirable effect of reversible contraception. Finally, the on-treatment increases in serum levels of SHBG reflect selectivity and low androgenicity, characteristics of NGM that, in laboratory tests, have demonstrated its similarity to natural progesterone (5), and that in clinical use may result in favorable effects on the lipid profile. Increased serum levels of high-density lipoprotein cholesterol have been reported with NGM/EE (6).

mglml

Days

Conclusion

Fig 4. Mean levels of progesterone

The findings of this large study comparing NGM/EE and LNG/EE triphasic formulations support those obtained in earlier noncomparative studies demonstrating excellent contraceptive action and selectivity. The NGM/EE triphasic is a minimally androgenic and highly effective ovulation-suppressing OC agent with a good safety profile.

achieved with both the NGM/EE and LNG/EE triphasic formulations. Additionally, no matter how effectively a contraceptive method prevents pregnancy, if the profile of real or perceived side effects is unacceptable to individuals requiring the method, it will not be utilized. In the case of the NGM/EE triphasic pill used in this investigation, the profile of minor side effects as well as discontinuations compares quite favorably with a widely used norgestrel-containing pill. The lack of effect on vital signs, hematologic and biochemical parameters bodes well for the safety profile of this new pill.

References 1 . Carson SL. Efficacy and clinical profile of a new oral contraceptive containing norgestimate: U . S . clinical trials. Acta Obstet Gynecol Scand Suppl 1990; 152: 25-3 1. 2. Burkman RT, Kafrissen ME. Olson W , Osterman J . Lipid and carbohydrate effects of a new triphasic oral contraceptive containing norgestimate. Acta Obstet Gynecol Scand Suppl 1992; 71: 5-8. 3. Vessey M, Metcalfe A, Wells C. et al. Ovarian neoplasms, functional ovarian cysts, and oral contraceptives. Br Med J 1987; 294: 1518-20.

Hortvionul uctions

in a double-blind, crossover study of 16 healthy women in the United Kingdom, Eyong et a1 demonstrated that both 180-yg and 250-~(gdoses of NGM suppressed the hypothalamic-pituitary axis, leading to a very low ovulation rate (4). In the study re220

140 Pglml

I8O 100

r

Pretherapy

.---.----

0

-.*----a

Post-therapy

Cycle 4

20

Fig 5 . Mean levels of estrddiol.

#

I

1

4

8

11

I

1

1

14 16 18 Days

1

1

I

21

24

27

14

R. S. London et al.

4. Eyong E, Buchi K, Elstein M. Effects of 180 pg and 250 pg norgestimate on pituitary-ovarian function and cervicat mucus. Fertil Steril 1988; 50: 756-60. 5 . Phillips A, Demarest K, Hahn DW, et al. Progestationa1 and androgenic receptor binding affinities and in vivo activities of norgestimate and other progestins. Contraception 1990; 41 (4): 399410. 6. Chapdelaine A, Desmarais J-L, Derman RJ. Clinical evidence of the minimal androgenic activity of. norgestimate. Int J Fertil 1989; 34: 347-52.

Aita Obstet Gynecol Scand Suppl156 (1992)

Address for correspondence: Robert s, London, M , ~ , Chief of Obstetrics and Gynecology ~~i~~~Permanente 7141 security ~ l ~ d ~ ~ l ~ Maryland 21207

i

~

~

~