Preventive Effects of Tamoxifen Vary With HER2 Level

LOS ANGELES--Tamoxifen (Nol-vadex) may not prevent breast cancer recurrence in women whose tumors over-express the HER2 oncogene (also known as c-erbB-2), and such overexpression may be a predictor of responsiveness to chemotherapy, researchers said at the ASCO integrated symposium on HER2 in breast cancer.

The tamoxifen/HER2 interaction was reported in a 20-year update of the GUN (Group of the University of Naples) trial of adjuvant tamoxifen. Angelo R. Bianco, MD, reporting for the Naples investigators, noted that in vitro studies have shown that overexpression of HER2 induces resistance to tamoxifen.

This long-term clinical trial found that "tamoxifen improved disease-free survival and overall survival in c-erbB-2-negative patients only." The proportion of breast cancers that overexpress HER2 appeared to be about 26% in Dr. Bianco’s sample.

The GUN trial randomized 308 patients to receive either tamoxifen, 30 mg/d for 2 years, or no hormonal therapy. HER2 expression was evaluated on 245 paraffin-embedded specimens by immunohistochemistry. A sample was considered positive if at least 10% of the cells stained with antibody for HER2.

At 20 years, when the median follow-up was 13.1 years, both disease-free survival and overall survival were significantly longer for patients treated with tamoxifen than for controls, Dr. Bianco reported. This improvement occurred regardless of menopausal status or lymph node status.

"More interestingly," he said, "when we evaluated the interaction between c-erbB-2 and tamoxifen, we found that tamoxifen improved disease-free and overall survival only in c-erbB2-negative patients, while showing a paradoxical detrimental effect in c-erbB-2-positive patients (Table 1)."

A multivariate test for interaction adjusting for lymph node status, menopausal status, estrogen receptor (ER) status, and ER/tamoxifen interaction confirmed the predictive value of HER2 expression (overall survival, P = .006; disease-free survival, P = .03).

A significant association was observed between HER2 expression and ER expression, with ER-positive tumors more likely than ER-negative tumors to be HER2 negative (Table 2).

A multivariate analysis showed no significant interaction between the effects of treatment with tamoxifen and ER status, but the analysis did demonstrate a first-order interaction between expression of HER2 and treatment with tamoxifen. This interaction, Dr. Bianco said, is a significant prognostic factor for disease-free survival and overall survival, "suggesting that total expression of c-erbB-2 affects the outcome of adjuvant therapy with tamoxifen."

He noted that the results of this study need to be confirmed "before the routine evaluation of c-erbB-2 expression can be recommended for the selection of patients who are likely to benefit from tamoxifen."

A related role for HER2 expression was reported by Peter M. Ravdin, MD, and his colleagues from the Southwest Oncology Group (SWOG).

He said that an Intergroup trial of 1,470 ER-positive, node-positive, postmenopausal breast cancer patients found a modest but statistically significant disease-free survival advantage for patients who received adjuvant cyclophosphamide-doxorubicin-fluorouracil-tam-oxifen (CAFT), compared with tamoxifen alone. "This raised the question of whether c-erbB-2 overexpression could identify a subset of patients who would benefit from the addition of CAF to tamoxifen," he said.

He noted that a study by the Cancer and Leukemia Group B (CALGB-8541) had earlier suggested that patients with overexpression of HER2 are more likely to benefit from high doses of CAF, compared with low or intermediate doses.

In the Biological Correlative study reported by Dr. Ravdin, HER2 was retrospectively measured in 595 samples from the Intergroup trial. The results showed that CAFT appeared to be superior to tamoxifen alone in patients whose tumors overexpressed HER2. However, a formal statistical analysis found only a trend toward an interaction between HER2 overexpression and additional benefit of CAF, in part, he said, "because of the low number of c-erbB-2-positive patients randomized to receive tamoxifen alone (21 patients)."

Thus, Dr. Ravdin said, "we are unable to state with statistical confidence whether the addition of CAF was more effective in c-erbB-2 positive patients. This study has low statistical power to answer predictive questions, given the short follow-up, the 10:3 randomization of CAFT vs T, and the 16% rate of overexpression of c-erbB-2." He added that the researchers are currently collecting and analyzing additional cases for HER2 overexpression.

Dr. Ravdin concluded that "this study, like several others, provides tantalizing but as yet inconclusive evidence that c-erbB-2 expression may be a predictor of particular chemoresponsiveness to anthracycline-containing combination chemotherapy."

In his discussion of the papers, Edison Liu, MD, director of the NCI’s Division of Clinical Sciences, said that the two studies considered together "raise a very important concept: That tumors that overexpress HER2 do poorly with tamoxifen as the sole adjuvant therapy, and that this bad outcome may be overcome by use of adequate doses of adjuvant chemotherapy and especially by the use of agents such as doxorubicin."