The accrual was very slow and there were many competing studies ongoing/initiated .

One of the basic principles of cancer chemotherapy is that these drugs act exclusively or mainly on cells in cycle. Estrogens have been shown to increase the fraction of breast cancer cells in cycle. Tamoxifen on the other hand, decreases the proliferative fraction and has been shown to negatively impact on the results of adjuvant chemotherapy in breast cancer when given concomitantly. A number of previous studies have attempted estrogenic recruitment of cancer cells (into cell cycle) to increase the efficacy of chemotherapy in locally advanced and metastatic breast cancer. Although some studies showed an increase in response rates in the recruitment arm, there was no benefit in time to progression or survival in any of the studies. These results may have been due to the inadequate sample size of the studies and advanced stage disease (with presumably higher fraction of inherently chemoresistant cells). The present study is designed to test the hypothesis that estrogenic recruitment of micrometastatic disease in operable breast cancer will increase the efficacy of standard adjuvant chemotherapy after surgery. The intervention arm of the study will involve administration of short duration estrogen prior to each cycle of adjuvant chemotherapy. The end-points are disease free and overall survival.

Results from a recent clinical study (1) show that tamoxifen administered concurrently with chemotherapy reduces the efficacy of the latter in patients with estrogen and/or progesterone receptor positive breast cancer. Tamoxifen exerts its anti-tumour efficacy in breast cancer primarily by its anti-estrogenic effect on the breast tissue. Anti-estrogenic effects of tamoxifen are mediated by competitive inhibition of the estrogen receptor, resulting in reduced transcription of estrogen-regulated genes (2). This results in blockade of cell cycle transit in G1 phase and inhibition of tumour growth. This mechanism of action might be the theoretical basis of the negative effect of concomitantly administered tamoxifen on the efficacy of adjuvant chemotherapy and could be explained thus: Most human solid tumours grow (and regress) following Gompertzian kinetics rather than the exponential one (3). The fundamental difference between Gompertzian and exponential models is that the growth fraction of the tumour (the fraction of cells in cell cycle) decreases with tumour growth in the former whereas in the latter it remains constant. Since many chemotherapeutic agents cause cell kill only in the fraction that is in cell cycle, this is used to explain (partly) the failure of chemotherapy in large tumours. Since tamoxifen also causes cell cycle arrest (in G1) and decreases the growth fraction, it could also impair the effects of chemotherapy in an analogous fashion. This is the result seen in this randomized trial.

The fraction of cells in cycle in breast cancer is low (5 to 10%) as determined by thymidine labeling index (4). Since most chemotherapeutic agents act preferentially or exclusively on cycling tumor cells, it is theoretically and intuitively appealing to increase the fraction of cycling cells to enhance the efficacy of chemotherapy. One way to do it in breast cancer would be to administer estrogen, which is known to enhance the proliferation of breast cancer cells. Weichselbaum et al (5) demonstrated that low concentrations of estradiol (10-9 M) increased the fraction of cells in S-phase and enhanced the rate of cell proliferation in estrogen receptor positive MCF-7 breast cancer cell line. The cell kill of this cell line on exposure to cytosine arabinoside was enhanced. Others have shown that even estrogen receptor negative tumors have increased cell proliferation in response to estrogenic stimuli (6). This has been explained partly as a result of modulation of the kinetic response of cancer cells to other growth factors (7,8). There have been a number of randomized studies in literature to test the concept of kinetic recruitment of breast cancer cells by estrogens to increase the efficacy of chemotherapy (9-14). All these studies have used diethylstilbesterol (DES) for few days before standard chemotherapy for breast cancer to recruit cells into cycle and all these studies have been in patients with locally advanced (LABC) or metastatic breast cancer (MBC). The results of these studies have largely been negative. In the trial by Baldine et al (14) in LABC patients there was no difference in the response rates between DES-CAF and CAF arms (56% Vs 63%) and no difference in the overall (47 Vs 49 months) and progression free (21 Vs 24 months) survival. DES-CAF was found to be more myelotoxic compared to CAF alone, which resulted in reduced dose intensity in the former. In the trial by Conte et al (13) patients of MBC were randomized to DES-CEF versus CEF alone. Again, there was no difference in the response rages (49% Vs 57%) and overall survival (20 Vs 17 months) in between DES-CEF and CEF, the former being more myelotoxic. In the trial by Ingle et al (12) in MBC patients, the response to DES-CMF was higher (39% Vs 25%, p=0.06) compared to CMF alone but there was no difference in time to disease progression or survival. In the study by Paridaens et al (11) in LABC and MBC patients, ethinyl-estradiol plus CAF was compared to CAF. There was no difference in response rates, time to progression or survival in the two groups. Toxicities were also similar.

To summarize, the results of estrogenic recruitment in patients with LABC or MBC have been negative with respect to survival but some studies have shown a trend towards higher response rates in the recruitment arm. There are two possible explanations for these negative results. All these studies have been in metastatic or locally advance breast cancer patients. It is possible that the fraction of cells with inherent chemoresistance is higher in these patients compared to early stage patients and this would negate any beneficial effect of cell recruitment into cycle. Secondly, the total trial size has been small (less than 260) in all these studies and therefore they were grossly underpowered to detect meaningful differences between the two groups.

As a direct corollary of the negative effect of tamoxifen administered concomitantly with chemotherapy and the proven ability of estrogens to increase the proliferating fraction in breast cancer cells, we hypothesize a beneficial effect of estrogen priming on the efficacy of standard adjuvant chemotherapy in operable breast cancer. Since the effect is hypothesized on micrometastasis, it is likely that chemoresistance will be a lesser or no impediment.

Thus in a manner that is inverse to tamoxifen, estradiol could 'prime' the tumour for subsequent chemotherapy. Since only operable breast cancer patients who have undergone surgery and need adjuvant chemotherapy will be the test population, the estradiol priming will theoretically act on micrometastases that chemotherapy seeks to eradicate.

Inclusion Criteria:

• 1. Patients with histological or cytologic proof of breast cancer. 2. Patients with operable breast cancer. 3. Patients who are candidates for adjuvant chemotherapy according to the standard policy. 4. Patients who have no contraindication to anthracycline based chemotherapy. 5. Patients who give informed consent to participate in the study. 6. Patients who can be followed up and can take all cycles of chemotherapy at the participating institution. 7. Patients should not have a known second cancer, present or past.

Exclusion Criteria:

-