CHICAGO – Topoisomerase II-alpha protein, which is frequently overexpressed in breast cancers, appears to be an independent predictor of response to anthracycline-based chemotherapy and a marker for rapidly proliferating disease, investigators reported at the annual meeting of the American Society of Clinical Oncology.
By testing 171 samples from a series of primary breast tumors and validating the results in an additional 2,630 samples from women with both primary and locally advanced cancers, Dr. Stephen Chan and his colleagues from the University of Nottingham (England) found that topoisomerase II-alpha (TOP2A) protein expression predicted pathologically confirmed complete responses, independent of HER2 status.
The TOP2A gene is coamplified in about a third of breast cancers in which the HER2 gene is amplified, leading some researchers to propose coamplification of the genes as a possible explanation for the relationship between TOP2A levels and breast cancer. Others, however, have suggested that levels of TOP2A protein, the expression product, may be a better marker than gene amplification for tumor aggressiveness and/or sensitivity to anthracyclines, the investigators noted.
To test the latter hypothesis, the authors looked for changes in the number of gene copies, and measured TOP2A messenger RNA (mRNA) levels and protein expression in a series of 171 unselected breast cancers.
They then sought to validate their results in samples from 2,000 consecutive cases of primary breast cancer that was treated with adjuvant cyclophosphamide, methotrexate, and fluorouracil and/or hormonal therapies; 240 locally advanced primary breast cancers treated with neoadjuvant anthracycline-based regimens with or without a taxane; 245 breast cancers treated with adjuvant anthracycline-based chemotherapy; and 145 primary breast tumors overexpressing HER2 that were treated with sequential adjuvant anthracycline-based therapy (the FEC regimen, consisting of 5-fluorouracil, epirubicin, and cyclophosphamide) plus trastuzumab (Herceptin).
They found a statistically significant correlation between TOP2A mRNA levels and gene copy number changes (Pearson correlation = .6817; adjusted P = .00001). Additionally, TOP2A mRNA was significantly overexpressed when the gene was amplified (adjusted P = .002). Protein overexpression poorly correlated with copy number changes, however, suggesting that overexpression is independent of gene amplification.
Looking at the relationship between TOP2A protein expression and clinical outcomes, they found that among patients with tumors that were positive for TOP2A protein expression, the pathological complete response rate was 27%-30%, regardless of whether TOP2A or HER2 genes were amplified. In contrast, when TOP2A protein expression was absent, only 2%-10% of patients had a pathological complete response.
"The authors have done a nice job to correlate [TOP2A] mRNA with amplification and mitotic index, but not protein expression, which brings into play [the question,] Is it an analytical variable?" commented Dr. Stephen K.L. Chia of the British Columbia Cancer Agency and the University of British Columbia in Vancouver. Dr. Chia was the invited discussant.
He said that the results could not be used to assess for predictive effect because the study lacked adequate power and an interaction test, and that the authors did not have data from randomized, controlled trials.
Dr. Chia noted that in a recent Canadian trial in which 477 of 710 patients had tumor samples tested for TOP2A and other biomarkers, TOP2A was – as the U.K. investigators also found – predictive of a benefit from anthracyclines, "but so was gene amplification and HER2 over-expression," which calls into question the value of TOP2A as a stand-alone marker, he said.
The study was supported by Breakthrough Breast Cancer, a U.K. charity. Neither the authors nor Dr. Chia had relevant financial disclosures.