Timing is Everything: Breast Cancer Surgery in Premenopausal Women
abstract & commentary
Synopsis: Controversy exists regarding the relationship between survival and the phase of the menstrual cycle during which breast cancer surgery is performed in premenopausal women. This retrospective analysis included 112 patients for whom the date of the last menstrual period was known prior to surgery. Surgery performed during the follicular phase was associated with a poorer prognosis (10-year survival rate of 42%) for patients with either ER-positive or ER-negative tumors. On the other hand, surgery performed during the luteal phase was associated with an improved survival. Unlike the follicular phase, a relationship between survival and ER status of a tumor operated on during the luteal phase emerged (10-year overall survival of 80% for ER-positive tumors and 60% if ER-negative).
Source: Cooper L, et al. Cancer 1999;86:2053-2058.
The 112 patients in this study were treated at the Guy’s Hospital in London from 1975-1985. Receptor status was determined by immunohistochemistry from formalin-fixed, paraffin-embedded specimens. The percentage of cells with positive staining was converted into a scale of 0 to 4. The intensity of staining was scored on a scale of 0 to 3. The sum of these two scales was used to divide tumors into receptor-positive (sum > 3) or receptor-negative (sum < 3). ER-positive and PR-positive tumors comprised 63% and 61%, respectively, of the cases and both types were found without significant variation throughout the menstrual cycle. ER and PR analysis of adjacent normal tissue showed no variation between the phases.
Overall survival in the study was approximately 50%, with a slight advantage for ER-positive patients compared with ER-negative patients that did not reach statistical significance. However, the 10-year survival rate plummeted from 75% if surgery was performed during the luteal phase (days 0-2 and 13-32) to 45% if performed during the follicular phase (days 3-12). During the follicular phase, patients with either ER-positive or ER-negative tumors did equally poorly. On the other hand, patients with ER-positive tumors had superior survival (80%) compared with ER-negative tumors (60%) if surgery was performed during the luteal phase. Results for progesterone receptor status were similiar.
COMMENT By Kenneth W. Kotz, MD
The "normal 28-day" menstrual cycle can be divided into a follicular (proliferative) phase and a luteal (secretory) phase. The follicular phase is characterized by a rising FSH (and estrogen levels) peaking at the time of ovulation (with the LH surge). This is followed by the luteal phase, with rising progesterone levels that peak on approximately the 21st day. The onset of menstruation is considered day 1 of the new cycle. The 14 days preceding the onset of menses is by definition the luteal phase and women with longer cycles have longer follicular phases. The presence of the luteal phase can be verified by checking a basal body temperature that will be more than 98 degrees in the vast majority of patients.1
There is now a large body of data regarding how the timing of surgery during the menstrual cycle influences survival. A recent critical review, although not a formal meta-analysis, of 32 retrospective studies encompassing almost 10,000 patients concluded that it was "likely" that the phase of the menstrual cycle is relevant to outcome but did not recommend any "immediate change in practice."2 We now have another retrospective analysis that also supports the hypothesis that surgery should be performed during the luteal phase. Not all studies have demonstrated this effect, but in this study the improvement in survival was dramatic.
It has been hypothesized that the unopposed estrogen in the follicular phase accounts for the poorer outcome, perhaps due to increased viability of micrometastases. However, contributions from other mechanisms may play a role too. Serum levels of vascular endothelial growth factor (VEGF), a potent angiogenic cytokine, have been shown to be significantly lower during the luteal phase.3 Additionally, the expression of genes that can affect the proliferative capacity and metastatic potential of a tumor, such as those for cathepsin L, matrix metalloproteinases, and TP53, may occur in a cyclical fashion that mirrors the phases of the menstrual cycle.4
Estrogen receptor status is a known prognostic factor for breast cancer. One would have predicted that ER-positive tumors operated on in the luteal phase would have fared better than the ER-negative tumors. What is interesting is that ER-negative tumors operated on in the luteal phase did better than ER-positive tumors operated on in the follicular phase. Furthermore, ER status did not appear to predict for survival if surgery was performed during the follicular phase.
Although controversial, this article supports the notion that "optimally timed" or "suboptimally timed" surgery should be included in descriptions of premenopausal patients’ tumors along with tumor size, number of involved lymph nodes, receptor status, grade, S-phase, Her-2-neu score, etc. How oncologists should incorporate this into the decision-making process before recommending adjuvant therapy is unknown. However, if premenopausal patients are willing to take four cycles of paclitaxel after four cycles of AC for a small absolute improvement in survival,5 then they will certainly be willing to delay definitive surgery for up to two weeks, even if the benefit is modest. Considering the lack of harm from a short-term delay, oncologists seeing patients preoperatively should consider recommending breast cancer surgery be timed in relation to the menstrual cycle. COLOR=black>
1. Kotz H. J Ob Gyn Nursing 1973;2:43-48.
2. Hagen A, et al. Am J Surg 1998;175:245-261.
3. Heer K, et al. Br J Cancer 1998;78:1203-1207.
4. Saad Z, et al. Lancet 1998;351:1170-1173.
5. Henderson I, et al. PASCO 1998;17:101a.
Retrospective data support the hypothesis that survival in premenopausal patients may be improved if breast cancer surgery is performed:
a. as soon as breast cancer is suspected.
b. during the follicular phase of the menstrual cycle.
c. during the luteal phase of the menstrual cycle.
d. after waiting for one full menstrual cycle.