C-erb-B2 gene amplification and serum level of c-erb-B2 oncoprotein at primary breast cancer diagnosis.

We studied c-erb-B2 gene amplification of DNA of primary breast tumours without distant metastasis from 236 women admitted to our institute during 1992. For 125 of them, we had a serum sample at diagnosis, before any treatment. C-erb-B2 gene amplification (> or = 2 copies) was observed in 26% (62/236) of the cases. There was a correlation with higher histological grades (p < 0.03) and with absence of hormone receptors: ER-(p < 0.0001). PgR-(p < 0.0001), association ER- and PgR-(p < 0.0000). Large tumours T3 and T4 taken together tended to present more c-erb B2 gene amplifications (p < 0.08). There was no correlation with age, histological type or nodal status. At diagnosis, mean concentration of serum c-erb-B2 oncoprotein was 8.5 +/- 18 U/ml with a median of 4 U/ml (4-150). Choosing a cut-off value of 8 U/ml gave a sensitivity of 21% (26/125). Serum levels of c-erb-B2 oncoprotein were correlated with tumour spread: large tumours T3-T4 (p < 0.001), nodal involvement (N+) (p < 0.01), association T3-T4 and N+(p < 0.0005), high levels of CA 15:3 (normal value < 25 IU/ml) (p < 0.05). There was no other correlation, particularly with age, histological type, hormone receptors or c-erb-B2 gene amplification. c-erb-B2 oncoprotein serum levels could be helpful to detect recurrences. Assessment of c-erb-B2 oncoprotein serum concentration, before treatment, as an independent prognostic factor is necessary.