Smoking Reduces Breast Tenderness During Oral Estrogen-Progestogen Therapy

ABSTRACT

Objective We wished to study the influence of smoking upon the occurrence of breast tenderness during oral estrogen-progestogen therapy (EPT).

Methods Data from 626 healthy postmenopausal women participating in three double-blind, randomized, controlled long-term trials of EPT versus placebo were included. The studies covered sequential, continuous and interrupted regimens of estradiol opposed by a selection of progestins. All studies were mono-center studies and performed in the period 1988-1997. Data on breast tenderness were collected from adverse event reporting and information on smoking status was obtained by interview.

Results Smoking was associated with an earlier age at menopause (difference: 1.3 years, p < 0.001) and a slightly lower body mass index (difference: 0.8 kg/m^sup 2^, p < 0.01). Smoking women in the EPT groups had significantly lower on-treatment estradiol levels compared to non-smoking women (p < 0.001), whereas no differences were observed in the placebo group. In parallel, the incidence of breast tenderness during EPT was reduced by about one-half (p < 0.001) in smokers compared to non-smokers, whereas no differences were seen on placebo.

Conclusion Current smoking reduces the incidence of breast tenderness in women receiving oral EPT. This may be caused by the increased degradation of estradiol during smoking.

Key words: BREAST TENDERNESS, BREAST PAIN, ESTROGEN, POSTMENOPAUSE, PROGESTOGEN, SMOKING

INTRODUCTION

Estrogen-progestogen therapy (EPT) is widely used in postmenopausal women for its effectiveness in the treatment of climacteric symptoms ‘ and in the prevention of bone loss2. Cigarette smoking occurs frequently and world-wide and, because the use in women is increasing, many women using EPT will also be smokers. We have previously shown that smoking reduces serum estradiol concentrations in healthy postmenopausal women on oral EPT3,4 and that this leads to a reduction in the gain of bone mass during low-dose EPT4. We also showed that smoking women pass menopause about 1 year earlier than non-smokers4,5. Smoking has been associated with a reduced level in markers of bone formation4; smokers have a lower bone mass than non-smokers4,6 and smoking has been found to increase the risk of hip fracture7. Thus, smoking also influences EPT metabolism as a target for the treatment of menopause, bone mass and fracture risk. At present, it is, however, not known whether smoking influences other estrogen-sensitive targets during EPT.

Standard HPT regimens induce side-effects from other estrogen- sensitive tissues, for example breast and endometrium. Although these side-effects may lead to discontinuation, factors that either predispose or protect against such side-effects have only rarely been studied.

By exploratory post-hoc analyses from three randomized, double- blind, placebo-controlled, mono-center studies of a broad selection of oral EPT regimens, we investigated the influence of smoking upon the occurrence of breast tenderness in healthy postmenopausal women.

MATERIAL AND METHODS

Design and subjects

Our criteria for study inclusion in our analyses aimed at ensuring:

(1) Double-blinded, randomization of healthy women to oral EPT or placebo;

(2) A similar collection method for breast symptom data without selection for – or against – breast symptoms;

(3) No selection for – or against – smoking.

We were able to use data from three studies for the present analysis8-10. All women were recruited by the same method: a random, blinded mailed invitation using social security numbers. This invitation consisted of a questionnaire eliciting anamnestic information with the purpose of inviting eligible women for screening. Study 1 was conducted from 1988 to 1991. Studies 2 and 3 were conducted simultaneously from 1993 to 1997 and the women in these studies were enrolled consecutively in one of the two studies from a common screened population. In total, 1099 women were screened in the three studies.

Study 1

A total of 219 women were screened to include 148 women aged from 45 to 55 years. They had passed through a natural menopause 6 months to 3 years previously. None had illnesses or were taking medication known to influence bone or lipid metabolism. Further details of inclusion and exclusion criteria are described elsewhere8. The women were randomized to 2 years of oral therapy with either 2 mg estradiol valerate continuously combined with 1 mg cyproterone acetate (days 1-28 for both) (n = 25), 2 m g estradiol valerate sequentially combined with 75 g levonorgestrel (days 1-28 for estradiol valerate and days 17-28 for levonorgestrel) (n = 25), 2 mg estradiol valerate sequentially combined with 10 mg medroxyprogesterone acetate (days 1-21 for estradiol valerate, days 12-21 for medroxyprogesterone acetate and placebo for days 22-28) (n = 25), 1.5 mg 17β-estradiol sequentially combined with 150 g desogestrel (days 1-24 for 17β-estradiol, days 13-24 for desogestrel and placebo for days 25-28) (n = 25) or placebo (days 1- 28) (n = 48). The actively treated groups are abbreviated in the text as follows: 2 mg E2V/1 mg CPAc, 2 mg E2V/75 g LNGs, 1.5 mg E2/ 10 mg MPAs and 1.5 mg E2/150 g DGs, where ‘s’ indicates a sequential progestin administration principle and ‘c’ a continuous progestin administration principle.

Study 2

A total of 471 women were screened to include 278 women within 1- 6 years after menopause. In addition, they were healthy with an intact uterus, were not treated with medications known to influence bone metabolism and had no laboratory or clinical evidence for confounding conditions or diseases. Further details of recruitment, inclusion and exclusion criteria are described elsewhere9. The women were enrolled in the double-blind, mono-center 2-year study and randomized to daily oral therapy with either 2 mg 17β- estradiol sequentially combined with 25 g gestodene (n = 55), 2 mg 17β-estradiol sequentially combined with 50 g gestodene (n = 56), 1 mg 17β-estradiol sequentially combined with 25 g gestodene (n = 56), 1 mg 17β-estradiol continuously combined with 25 g gestodene (n = 55) or identically appearing placebo (n = 56). Study medication was administered in cycles of 28 days, during which gestodene was administered on days 17-28 in the sequential groups. The study was subsequently prolonged for a 3rd year (39 cycles in total) during which blinding was continued. The actively treated groups are abbreviated in the text as follows: 2 mg E2/25 g Gs, 2 mg E2/50 g Gs, 1 mg E2/25 g Gs and 1 mg E2/25 g Gc, where ‘s’ indicates a sequential progestin administration principle and ‘c’ a continuous progestin administration principle.

Study 3

A total of 409 women were screened to enroll 200 healthy women at least 1 year after menopause in the 2-year, mono-center study. Hysterectomy as well as diseases or therapies known to affect bone metabolism excluded participation. Further details of inclusion and exclusion criteria have previously been described10. The women were randomized to oral treatment with either placebo (n = 50), continuous piperazine estrone sulfate 1.5 mg combined with alternating 3-day cycles of 0.7 mg norethisterone, such that the norethisterone therapy was interrupted by 3 days after each 3-day intervention (n = 50), a similar group receiving medication comprising 0.75 mg continuous piperazine estrone sulfate combined with alternating 3-day cycles of 0.35 mg norethisterone (n = 50) or a continuous combination of 2 mg estradiol and 1 mg norethisterone acetate (n = 50). The placebo and piperazine interventions were double-blind and administered in blisters of 30 days; the continuous estradiol and norethisterone therapy (Kliogest) was open as it had not been feasible to produce a tablet appearing similar to those of the three other treatments. For the present analyses, we pooled the two groups receiving piperazine estrone sulfate combined with interrupted norethisterone. This was possible because the bone mass responses for these treatments were in the same order of magnitude10,11. The actively treated groups are abbreviated in the text as follows: 2 mg E2/1 mg NETAc and 1.5 mg or 0.75 mg PES/0.7 mg or 0.35 mg NETi, where ‘c’ indicates a continuous progestin administration principle and ‘i’ an interrupted progestin administration principle.

Total study population

In total, 626 individuals were included in the three studies. This cohort represented a broad range of healthy women having past menopause between 6 months and 18 years earlier and being treated with a selection of optimized EPT regimens. Neither smoking nor breast tenderness was mentioned in any inclusion or exclusion criteria in the studies. In addition, the women had no history of breast cancer; baseline physical examination gave no indication of breast abnormalities and mammographies in studies 2 and 3 were without cancer suspicion.

Methods

Breast tenderness

Information about breast tenderness was obtained as part of the adverse event data collection. For the collection of data on breast tenderness, we included adverse events with the following descriptions: breast tenderness, breast pain and mastalgia. The information was collected at the clinic visits. Degree of severity was not captured. In addition, the localization o\f symptoms within each cycle was not recorded.

Demographic data and vital signs

Pretreatment, weight was measured on an electronic scale and height on a Harpenden Stadiometer. Both measurements were performed wearing indoor clothes and no shoes. Information about number of years since menopause and smoking status was elicited by interview. For the present analyses, smoking refers to current smoking whereas non-smoking refers to non-current smoking, i.e. including both never- smokers and previous smokers. The number of cigarettes was not assessed. The type of smoking consisted of cigarettes in all cases.

Serum hormones

Scrum estradiol was analyzed in studies 1 and 2(3,4). Both studies used radioimmunoassay, but the assays in the two studies were different. The average on-treatment estradiol was calculated as the time-averaged value after randomization.

Statistics

All analyses were conducted on the total group of randomized women. A χ^sup 2^ test was applied to analyze breast tenderness frequency in smokers versus non-smokers and also to compare baseline smoking frequency. ANOVA was used to compare baseline continuous data and average on-treatment serum estradiol levels. Significance was accepted at the 5% level. The statistical analyses were performed using SPSS (SPSS Inc., Chicago, IL, USA).

RESULTS

Table 1 gives the characteristics of the cohort separated by smokers and non-smokers. The smoking women passed through menopause 1.3 years earlier than the non-smoking women (p < 0.0001), and smoking women had a slightly lower body mass index (BMI) than non- smoking women (0.8 kg/m^sup 2^, p < 0.01). Otherwise, the smoking and non-smoking women were comparable. Tables 2-4 demonstrate the baseline characteristics after randomization in the three studies. The groups were comparable in the studies apart from a small difference in BMI in study 2 (Table 3).

Table 1 Baseline characteristics of the total cohort separated by smokers and non-smokers. Data are expressed as mean standard deviation

During EPT, breast tenderness was overall reduced by about one- half in smoking compared to non-smoking women, whereas no differences were seen in the placebo groups (Figure 1). Comparing the total group of women on EPT in each study, there was a borderline significance in study 1 (p < 0.2) and a significant difference in studies 2 and 3 (p < 0.01 for both). In studies 1 and 2, where serum estradiol had been measured, there was a comparable and constant low level of average on-treatment serum estradiol in the postmenopausal range in both smoking and non-smoking women on placebo. In contrast, during EPT, average on-treatment serum estradiol was significantly reduced in smoking compared to non- smoking women (p < 0.001) (Figure 2).

Figure 1 The frequency of breast tenderness in smokers (solid bars) versus non-smokers (hollow bars) treated with sequential, continuous or interrupted regimens of oral estrogen-progestogen therapy (EPT0) compared to placebo in studies 1-3. The χ^sup 2^ test comparing smoking and non-smoking women in the total EPT- treated group in each study lead to p < 0.2 for study 1, and p < 0.01 for studies 2 and 3, respectively. EPT1, 2 mg E2V/1 mg CPAc; EPT2, 2 mg E2V/75 g LNGs; EPT3, 1.5 mg E2/10 mg MPAs; EPT4, 1.5 mg E2/150 g DGs; EPT5, 2 mg E2/25 g Gs; EPT6, 2 mg E2/50 g Gs; EPT7, 1 mg E2/25 g Gs; EPT8, 1 mg E2/25 g Gc; EPT9, 2 mg E2/1 mg NETAc; EPT10, 1.5 mg or 0.75 mg PES/0.7 mg or 0.35 mg NETi; E2V, estradiol valerate; E2, estradiol; CPA, cyproterone acetate; LNG, levonorgestrel; MPA, medroxyprogesterone acetate; DG, desogestrel; G, gestodene; NETA, norethisterone acetate; PLS, piperazine estrone sulfate; NET, norethisterone. 's' indicates a sequential progestin administration principle, 'c' a continuous progestogen administration principle and T an interrupted progestin administration principle

Figure 3 shows the average frequency of breast tenderness in smoking and non-smoking women in all EPT groups compared to all placebo groups. Breast tenderness is reduced by about one-half in smoking women versus non-smoking women. The difference between smoking and non-smoking women in the EPT group is highly statistically significant (p < 0.001), whereas no difference is detected in the pooled placebo group.

DISCUSSION

This large cohort of representative women consisted of three mono- center, long-term, double-blinded, placebo-controlled, randomized studies. The blinded assessments supported the validity of our results.

We found that current smoking reduces the occurrence of breast tenderness in healthy post-menopausal women by about one-half during oral EPT. The difference was significant in two studies, borderline in the third, but highly statistically significant in the pooled population of 626 women. Our collection method for breast tenderness results was based on adverse event data, which leads to a credible and conservative estimate for the occurrence of breast tenderness.

Figure 2 The average on-treatment serum estradiol in the combined estrogen-progestogen therapy (EPT) groups and the placebo groups separated by smoking (S) and non-smoking (NS) women. The top panel represents data from study 1 and the bottom panel from study 2. The difference between smoking and non-smoking women on EPT is highly significant (p < 0.001 for both studies). In contrast, there is no significant difference between smokers and non-smokers on placebo in either study. The y-axis unit is nmol/l

Figure 3 The frequency of breast tenderness in smokers (solid bars) versus non-smokers (hollow bars) treated with oral estrogen- progestogen therapy (EPT) or placebo. The difference between smoking and non-smoking women on EPT is significant with p < 0.001 in the γ^sup 2^ test

The influence of smoking on breast tenderness is in parallel with the influence of smoking on serum estradiol concentrations. This may indicate that the adverse influence of estradiol leading to breast tenderness is induced by estrogen. That the influence of smoking upon breast tenderness is seen on both high- and low-dose estradiol therapies suggests that the influence of estradiol on the breast tenderness risk is fully expressed at estrogen doses lower than those studied. We have previously shown that, during therapy with 1 mg estradiol, the gain in bone mass over 3 years is halved in smokers as compared to non-smokers4. Thus, that a reduction in serum concentrations of estradiol due to smoking is reflected in a response tissue has now been found both for bone and breast. During therapy with 2 mg estradiol, there was no difference between smoking and non-smoking women in the gain of bone mass4. This is in contrast to the results on breast tenderness and suggests that the dose- concentration response curves of estradiol on the breast and the bone are different such that the influence on bone is fully expressed at estradiol doses lower than those that would lead to a fully expressed effect on breast tenderness. Our results thus indicate that a desirable effect is reached sooner than an undesirable effect, but that there is an overlap of the doses leading to an effect on bone and the doses leading to adverse effects on the breast. This suggests that the optimum regimen may benefit from individualization and current smoking may be a factor to be considered when tailoring EPT regimens. That smoking reduces the occurrence of breast tenderness suggests that smokers may have a reduced risk of discontinuation due to this adverse effect.

Because of the heterogeneity of the progestins studied, it was not possible for us to evaluate whether the progestin effect is influenced by smoking or whether there are differences between sequential and continuous regimens. These important questions need further assessment. Another limitation of our study was the lack of a specific pain assessment in the collection of breast tenderness data. For future studies of smoking during EPT, it will be important to include more specific evaluations, including a specific pain scale, degree of severity and association with the cycle.

From our results on the influence of smoking on response to oral EPT in the bone and the breast, it is clear that smoking and non- smoking women represent two different populations, which need different therapies. The health authorities, when developing guidelines for the required documentation in the registration process, should consider this.

In conclusion, we have demonstrated that current smoking reduces the occurrence of breast tenderness by one-half during oral EPT. Parental regimens remain to be investigated.

ACKNOWLEDGEMENTS

The authors thank Schering AG, Berlin, Germany as well as NV Organon, Oss, The Netherlands and R.W. Johnson Research Institute, Raritan, New Jersey, USA and Zrich, Switzerland for supplying study medication and placebo preparations.

Conflict of interest Nil.

Source of funding Schering AG, Berlin, Germany; NV Organon, Oss, The Netherlands; RW Johnson Research Institute, Raritan, New Jersey, USA and Zrich, Switzerland.

References

1. Notelovitz M, Lenihan JP, McDermott M, Kerber IJ, Nanavati N, Arce J. Initial 17betaestradiol dose for treating vasomotor symptoms. Obstet Gynecol 2000;95:726-31

2. The Writing Group for the PEPI trial. Effects of hormone therapy on bone mineral density: results from the postmenopausal estrogen/progestin interventions (PEPI) trial. J Am Med Assoc 1996;276:1389-96

3. Jensen J, Christiansen C, Rdbro P. Cigarette smoking, serum estrogens and bone loss during hormone replacement therapy early after menopause. N Engl J Med 1985;313:973-5

4. Bjarnason NH, Christiansen C. The influence of thinness and smoking on bone loss and response to HRT in early postmenopausal women. J Clin Endocrinol Metab 2000;85:590-6

5. Andersen FS, Transbol I, Christiansen C. Is cigarette smoking a promotor of the menopause? Acta Med Scand 1982;212:137-9

6. Ward KD, Klesges RC. A meta-analysis o\f the effects of cigarette smoking on bone mineral density. Calcif Tissue Int 2001;68:259-70

7. Cornuz J, Feskanich D, Willett WC, Colditz GA. Smoking, smoking cessation and risk of hip fracture in women. Am J Med 1999;106: 311-4

8. Haarbo J, Hassager C, Jensen SB, Riis BJ, Christiansen C. Serum lipids, lipoproteins and apolipoproteins during postmenopausal estrogen replacement therapy combined with either 19- nortestosterone derivates or 17-hydroxyprogesterone derivates. Am J Med 1991;90: 584-9

9. Bjarnason NH, Byrjalsen I, Hassager C, Haarbo J, Christiansen C. Low dose estradiol in combination with gestodene to prevent early postmenopausal bone loss. Am J Obstet Gynecol 2000;183:550-60

10. Alexandersen P, Byrjalsen I, Christiansen C. Piperazine oestrone sulphate and interrupted norethisterone in postmenopausal women: effects on bone mass, lipoprotein metabolism, climacteric symptoms, and adverse effects. Br J Obstet Gynaecol 2000;107: 356- 64

11. Bjarnason NH, Alexandersen P, Christiansen C. Number of years since menopause: spontaneous bone loss is dependent, but response to HRT is independent. Bone 2002;30:637-42

N. H. Bjarnason, C. Jrgensen, H. Kremmer, P. Alexandersen and C. Christiansen

Center for Clinical & Basic Research, Ballerup, Denmark

Correspondence: Dr N. H. Bjarnason, Center for Clinical and Basic Research, Ballerup Byvej 222, 2750 Ballerup, Denmark

ORIGINAL ARTICLE

2004 International Menopause Society

DOI: 10.1080/136971 30400012171

Received 18-08-03

Revised 24-04-04

Accepted 26-04-04

Copyright CRC Press Dec 2004