Does HER-2 status predict only a decreased response to hormone therapy in advanced breast cancer, or does it also predict the extent of metastatic disease?

Abstract

decreased response to hormone therapy in metastatic breast cancer. Their clinical study, the largest about the topic of HER-2 oncogene expression and hormone resistance, included 719 postmenopausal women with advanced breast cancer who participated in three randomized trials of second-line endocrine therapy. Two of the trials compared the aromatase inhibitor fadrozole with megestrol acetate, and one compared the aromatase inhibitor letrozole with megestrol acetate. 2,3 The efficacy end points evaluated in the analysis by Lipton et al were overall response, clinical benefit, time to treatment failure, and time to progression. All four end points were significantly worse in the patients overexpressing HER-2 when compared with patients not expressing HER2. Their data support the results of other retrospective studies, in which clinical lack of response to hormone therapy was also correlated with HER2 overexpression. 4,5 The article concludes that HER-2 is a molecular marker for predicting a decreased response to endocrine therapy in advanced breast cancer. However, this may not fully take into consideration some factors that are known to relate to prognosis in advanced breast cancer and that might also have an effect on HER-2 serum levels. These factors include site of metastases and metastatic tumor burden. In Lipton et al’s article, 1 the site of metastases was recorded. The proportion of a visceral dominant site of metastasis, a characteristic associated with poor prognosis and poorer response to therapy, was significantly larger in patients with elevated serum HER-2 levels than in patients with normal HER-2 levels (59% v 39%, respectively). Conversely, fewer patients with elevated HER-2 levels had soft tissue/bone metastasis, which is usually associated with a favorable prognosis (41% v 61%). 1 HER-2 status may therefore reflect the aggressiveness of advanced cancer, as suggested by others. 6-8 An association between HER-2 status and visceral disease has been observed by some authors, 4 although not by others. 9 Tumor burden, measured either with the number of metastatic sites or with the levels of a tumor marker such as CA 15-3, 10 is not reported in relation to HER-2 status in Lipton et al’s article. The addition in the analysis of the variable “tumor burden” might have shown that serum HER-2 is related to the amount of metastatic disease. In fact, we have shown previously that serum HER-2 in advanced breast cancer is related to HER-2 status in the primary breast carcinoma and also to metastatic tumor burden. 11 Therefore, part of the decreased response in HER-2‐positive patients may be attributed either to more aggressive tumor sites or to larger burden of metastatic disease. A multivariate analysis of prognostic factors, a statistical procedure that may correct the significance of competing variables, is not shown in Lipton et al’s report. However, in previous communications of the same material, the investigators did perform multivariate analyses, which included, in addition to HER-2 status, age, race, disease-free interval, performance status, visceral versus nonvisceral metastasis, and estrogen receptor and progesterone receptor status, although not tumor burden. 11,12 A multivariate analysis in the present report would have increased the confidence of investigators in the predictive value of serum HER2 and in its independence from the site of metastatic disease and tumor burden. Retrospective studies such as this report can provide predictive information, but the limitations are large. Prospective studies are needed that address the issue of HER2 expression and hormone resistance. Since the logistics of a randomized clinical trial may not be applicable here, we recommend the design of prospective cohort studies, in which patients are treated with a defined endocrine therapy, serum HER-2 levels are determined initially, and patients are followed up for relapse. The preliminary results of such a trial have been reported. 13 Until prospective data are available, it is not advisable to withhold endocrine therapy in patients with estrogen receptor‐positive, HER2-positive advanced breast cancer. Finally, although combinations of endocrine therapy plus anti-HER-2 antibodies are a logical approach, they should be used with caution, since some preclinical research has reported that such combinations may be less than additive. 14