Abstract
The aim of this study was to further characterise the clinical response to primary
systemic endocrine therapy in breast cancer and to determine whether it is possible to
identify biological markers of tumour phenotype that can be used to predict subsequent
clinical response to neoadjuvant treatment with letrozole for three months.
137 postmenopausal patients with locally advanced oestrogen receptor (ER) positive
breast cancer were treated with 2.5mg letrozole daily. Tumour samples were taken at
diagnosis, at three months and, in 62 patients, additionally at 10-14 days. Serial clinical
tumour measurements were made over the three month treatment period. Ultrasound
scanning (USS) was shown to be the most accurate method of assessing clinical
response to treatment and the modality that corresponded most closely with pathological
response. Patients with ER rich tumours were shown to be those most likely to derive
maximal benefit from neoadjuvant letrozole.
In this series, 67% of patients showed a clinical response to treatment (> 50% reduction
in tumour volume at three months on USS) and 63 % had their surgery down-staged
from mastectomy to breast conserving surgery. Of the 125 patients who completed the 3
month audit period, with a mean follow up period of 39 months (4-58), 42 patients had
died. Of these, 16 had evidence of recurrent breast cancer at the time of death. 7 local
recurrences have occurred in the series (5%).
75% of tumours displayed evidence of a pathological response (decreased cellularity/
increased fibrosis) at three months. Significant decreases in PgR expression were seen
after both 14 days and three months but this did not correlate with clinical or
pathological tumour response. Baseline proliferation, assessed using Ki67) was similar
in responders and non-responders whether assessed clinically or pathologically.
Treatment was associated with highly significant decreases in Ki67 in all tumour
subgroups (at least P<0.005 by paired Wilcoxon rank test) at 14 days. There was no
significant difference in Ki67 expression at 14 days between subsequent clinical
responders and non-responders. However, when correlating the decrease in proliferation
with pathological response, Ki67 expression at 14 days was significantly higher in
tumours which subsequently failed to show morphological evidence of response.
The percentage reduction in Ki67 over the three month treatment period showed a
significant correlation with cause specific survival (p=0.007). However, it was not
possible to use changes in proliferation to predict response for an individual patient. The
fresh tissue collected in parallel with the formalin fixed tissue in this study is currently
being analysed by microarray and will hopefully suggest possible avenues for other
markers which may prove more helpful in predicting response to neoadjuvant endocrine
treatment on an individual patient basis
