Predicting Prognosis and Tamoxifen Response in Breast Cancer. With a special focus on contralateral breast cancer

Sammanfattning: One of the great challenges in breast cancer treatment today is to customize adjuvant treatment to each patient’s individual needs. To do this it is necessary to learn more about the prognostic and treatment predictive factors that determine the risk of relapse and response to a certain mode of treatment. This thesis describes studies on the effect of amplified in breast cancer 1 (AIB1), a coactivator of the oestrogen receptor, on prognosis and tamoxifen response through a controlled trial on premenopausal patients randomized to tamoxifen or a control group. AIB1 was found to be a negative prognostic factor, although patients with high AIB1 responded very well to tamoxifen. The findings were validated in two independent cohorts, one consisting of premenopausal patients not receiving tamoxifen, and the other of pre- and postmenopausal patients receiving tamoxifen.
It has recently been suggested that the effect of AIB1 is modified by paired box 2 gene product (PAX2). PAX2 is a transcription factor important during embryogenesis, and may also play a role in carcinogenesis. This is the first time PAX2 has been investigated in well-defined cohorts of patients receiving or not receiving tamoxifen. PAX2 was not found to affect prognosis on its own, or to modify the effect of AIB1.
The second part of this thesis focuses on contralateral breast cancer (CBC). Within their lifetime, previous breast cancer patients have a 2-20% risk of developing a second tumour in the contralateral breast. From the trial on premenopausal patients randomized to tamoxifen or control, it was found that without tamoxifen 12% developed CBC within a median follow-up period of 14 years. This risk was even higher in the youngest women (<40 years), in which 20% developed CBC. Treatment with tamoxifen reduced the risk by 50% in all patients, and by 90% in the youngest women.
Since CBC is still a rather rare event, previous studies are often small or based only on register data. Detailed patient, tumour and treatment information has been collected for a large cohort (>700) of patients with CBC in the Southern Healthcare Region of Sweden. From these data it was found that a short time interval between tumours was associated with a poorer prognosis, especially in young patients. This could indicate that some of these CBCs are in fact metastases of the first tumour, and would thus require different treatment. It could also be that tumours that develop soon after previous treatment have developed resistance to treatment and are of a more aggressive phenotype.
Finally, it was found that patients who first noticed symptoms of their CBC themselves had a higher risk of developing metastases than patients diagnosed by mammography or clinical examination in a follow-up programme. The difference in prognosis in relation to mode of detection remained even when the time interval between tumours was ≥10 years, indicating that a long follow-up period is valuable.