The Cancer Journal - Volume 8, Number 1 (January-February 1995)
The menstrual cycle and timing of breast cancer surgery - a "post-modernistic" vision of an advance that remains at arm's length
A strange controversy that deserves attention has surfaced over the last five years. It arises from a seminal paper by Hrushesky et al. (1) who showed in studies on breast cancer in women, which were based on prior investigations on mammary tumors in mice, that timing of breast surgery is a major and independent risk factor with regard to recurrence.
This finding was immediately contested by other investigators (2) and, since then, several papers addressing the issue have seen the light of day. One of the most recent is by Veronesi et al. (3) who have compared 525 women undergoing surgery during the follicular phase of the cycle (the first 14 days) with 650 women undergoing luteal phase surgery (the following 14 days). Veronesi and colleagues noted a highly significant difference in relapse rate between the two groups: 36.6 % (192/525) in the follicular phase group, 29.6 % (192/650) in the luteal phase group, which was even more marked when the comparison was confined to node-positive women who had undergone axillary dissection. Unfortunately, the results of Hrushesky's and Veronesi's teams cannot be compared directly because applicable to different days of the menstrual cycle but there is no reason why the discrepancy cannot be resolved by considering both time-schedules in these retrospective studies.
A more recent meta-analysis of available studies by Fentiman et al. (4) has added further support to the notion that a premenopausal woman with breast cancer is liable to gain from deliberately choosing her operation date in relation to her menstrual period.
What evidence could be simpler to act upon ! At first sight, at least, as there are questions we should nevertheless ask ourselves.
1. What are the odds regarding survival and death ?
2. Is there a risk to patients if the study is pursued ?
3. What will it cost; will expenditure increase ?
4. How long will it take to obtain a more convincing answer ?
5. What are the biological mechanisms involved and are they amenable to manipulation to improve the results ?
6. Can the chronology of other therapies for breast cancer or of the management of other cancers be considered in relation to the menstrual cycle ?
Satisfactory answers to the first three questions are already available. (1) The incidence of node-positive breast cancer in premenopausal women is known from epidemiological studies. Veronesi's paper gives a 12% difference for disease-free interval rates at 5 years allowing the expected gain to be calculated, should these promising results be confirmed. (2) Even the most conservative estimates do not reveal any increase in risk in performing luteal phase surgery or in avoiding operations during the perimenstrual period (days 1 to 6 and 21 to 36). At the very worst, the relapse rate remains the same (3). Suitable operation dates would be spread out over time and surgical practice remain unperturbed because of the lack of synchrony among menstrual cycles. Consequently, neither our present state of knowledge nor practical considerations provide any obstacle to positive action that, needless to say, can be undertaken after the customary consultation of ethics committees.
A conclusive answer to the issue would depend upon the numbers of patients that could be enrolled and surgeons implicated in a study. Both disease-free interval and survival would have to be evaluated. Doubtless, a randomized prospective study is called for but a rigourous follow-up indicating an early frankly positive response might render the continuation of randomization unethical and lead to its interruption.
Five years ago in this journal (5), we stressed how simple are the studies required for a quick answer but too few surgeons took up the challenge. S. Epistème (6) set forth the rationale of the steps to be taken on the basis of the following propositions that are still applicable today :
- if the positive result is false, there will be no difference;
- if it is true, there will be a positive difference that will advantage patients who underwent surgery at the right time and, on the contrary, a negative difference to the disadvantage of the patients whose operation was at the wrong time.
In the light of the above, ignoring the significance of these findings or failing to seek to confirm or refute them, becomes contrary to the precepts of medicine. No considerations of economy are involved; both rich and poor countries can perform this type of study. The only true obstacle to the undertaking is the firm and widespread conviction that a simple procedure cannot help medicine progress at a time when sophisticated techniques and studies are unable to reduce breast cancer mortality..
1. Hrushesky WJM, Gruber SA, Bluming AZ et al. Menstrual influence on surgical cure of breast cancer. Lancet ii, 949-952, 1989.
2. Powles TJ, Ashley SE, Nash AG et al. Timing of surgery in breast cancer (letter). Lancet 337, 1604, 1991.
3. Veronesi U, Luini A, Mariani L et al. Effect of menstrual phase on surgical treatment of breast cancer. Lancet 343, 1545-1547, 1994.
4. Fentiman IS, Gregory WM, Richards MA. Effect of menstrual phase on surgical treatment of breast cancer. Lancet 344, 402, 1994.
5. Salomon JC. Incredibly simple, effective and probably true. Cancer J 2, 410, 1989.
6. Epistème S. Menstrual cycle and breast cancer surgery. Cancer J 3, 7-8, 1990.