A. Travis et al., C-ERBB-3 IN HUMAN BREAST-CARCINOMA - EXPRESSION AND RELATION TO PROGNOSIS AND ESTABLISHED PROGNOSTIC INDICATORS, British Journal of Cancer, 74(2), 1996, pp. 229-233
A series of 346 patients with primary operable breast cancer and a ser
ies of 145 patients with advanced breast cancer were investigated for
c-erbB-3 protein expression using the monoclonal antibody RTJ1. Formal
in-fixed, paraffin-embedded tumour samples were stained using a standa
rd immunochemical method and staining was assessed on a four-point sca
le. The study aimed to observe the expression of the c-erbB-3 protein
and investigate any relationship between expression and established pr
ognostic indicators and prognosis. In both the primary and advanced se
ries breast tumour tissue was found to stain heterogeneously for c-erb
B-3. The staining was observed to be predominantly cytoplasmic and the
majority of tumours exhibited moderate positivity. However, 15% and 3
5% of cases in the primary operable and advanced series respectively d
isplayed strong positive staining. No significant difference was found
between the staining in the primary and advanced series. In the prima
ry operable breast cancers, no significant associations were demonstra
ted with overall survival, disease-free interval, regional recurrence,
the presence of distant metastases, age, menopausal status, oestrogen
receptor status, histological grade, lymph node stage, vascular invas
ion and c-erbB-2 protein expression; However, a significant associatio
n was seen between the degree of c-erbB-3 immunoreactivity and both tu
mour size (P < 0.01) and tumour type prognostic group (P = 0.05). No o
verall association with local recurrence was seen when the four groups
of c-erbB-3 expression were analysed (P = 0.12), but when those tumou
rs showing no or weak staining were compared with those showing modera
te and strong immunoreactivity it was seen that the latter were signif
icantly more likely to develop local recurrence (P = 0.03). In the ser
ies of patients with advanced disease, no significant associations wer
e demonstrated with survival, UICC criteria, age, menopausal status, o
estrogen receptor status, histological grade, c-erbB-2 status or the p
resence of vascular invasion. In conclusion this study found variable
expression of c-erbB-3 protein in human breast carcinoma and an associ
ation with some recognised prognostic factors in those patients with p
rimary operable breast carcinoma. It seems, however, unlikely that c-e
rbB-3 protein expression will emerge as a powerful enough prognostic f
actor to be of value in clinical practice.