Abstract:
Protein S100A6 is a member of the Ca2+-binding S100 family. Since it was proved, that in various malignant tumours (e.g. Melanoma and colorectal Adenocarcinoma) overexpression of this protein was related to a poorer prognosis, S100A6 has become highly interesting. The significance of S100A6 in breast cancer, however, has not yet been examined.
Methods: In a retrospective study we immunhistochemicaly analysed the expression of S100A6 in 266 primary breast cancer patients. Staining intensity (SI) and frequency of immunostained tumour cells (PP) were classified. From both figures an immunoreactive score (IRS) was calculated to evaluate the quantity of S100A6 in the tumour. Further, a threshold value for the IRS (cut off) was established, under which the survival curves, calculated according to the Kaplan-Meier method, most significantly differred. According to the threshold value an IRS of 0 or 1 was assessed with the status of "S100A6 - negative", an IRS of 2 up to 12 with the status of "S100A6 - positive".
The analysis was carried out in three turns of evaluation in order to determine the intraobserver- and interobserver variabilities. The author carried out two evaluation turns, a third was done by an expert (specialist in pathology). The results proved, that the determination of the S100A6 - status can be easily acquired and is reliably reproducible.
In the second turn of evaluation 78 patients (29,3%) were "S100A6 - negative" and 188 patients (70,7%) "S100A6 - positive". The results of this second analysis were used for further calculations.
Results: There was no correlation between S100A6 - status and the classical prognostic factors of tumour size, lymph node stage, histological grade or distant metastases. Tumours with a positive status of the oestrogene - or progesterone receptor and lobular carcinomas with a "S100A6 - negative" status were significantly more frequent.
In contrast to other carcinomas, in our study with breast cancer patients no prognostic relevance for the S100A6 - status could be determined, neither in our whole study population nor in subgroups. However, there was a non significantly poorer prognosis for overall survival and disease free survival in the "S100A6 - negative" group.
Surprisingly the S100A6 - status showed a significance in the prediction of the response to a CMF-chemotherapy (p=0.011). Patients with "S100A6 - negative" tumours had a clearly shorter disease free- or overall-survival than patients with "S100A6 - positive" tumours. On the other hand the S100A6 - status showed no prognostic relevance for a radiotherapy, therapy with tamoxifen or chemotherapy with anthracyclines.
In addition to that, there was a significantly more frequent occurence of bone metastases in the "S100A6 - negative" group.
A causative role of S100A6 during a CMF-chemotherapy or the formation of bone metastasis is presently not known, but new hypotheses were formulated. Their evaluation will lead to further studies and could play an important role in the individualization of breast cancer therapy.