An exploration of relapse data by hazard rate as a means of developing biological insights into the natural history and treatment of breast cancer: Data from the ’Arimidex’, Tamoxifen, Alone or in Combination (ATAC) trial and the Milan Institute database

Abstract

Listen

Translate article

612 Background The hazard rate (HR) for relapse following primary surgery for breast cancer demonstrates two peaks. The first, and steepest, peak occurs at approximately 18–24 months after surgery and may result from the activation of dormant/latent metastases by the initiation of angiogenesis, the expression of the genetic mechanisms responsible for wound healing, and the induction of proliferative activity in single quiescent tumor cells through growth factors. The second, flatter, peak occurs approximately 5–6 years after surgery and is thought to result from the stochastic transition of occult metastases from a latent to an active state, representing an “echo” of the natural history of the original disease. Method The ATAC trial compared 5 years of adjuvant tamoxifen (T) versus 5 years of adjuvant anastrozole (A), a third-generation aromatase inhibitor (AI), in women with early breast cancer, and 68-month follow-up data were reported in December 2004. Here, we compare this data with a matched series from the Milan Institute database of patients, who were treated before the introduction of systemic adjuvant therapy to produce a semi-quantitative comparison. Results The shape and timing of the signal for HR were identical for each of the three data sets (T, A and Milan), with the first peak at 18–24 months and the second peak beginning to appear at 4–5 years following primary therapy. However, in the T and A groups, the peaks were substantially flattened. The most striking aspect of our comparison was the near-obliteration of the first peak in the A group. Conclusion A possible explanation for the reduction in the first peak with anastrozole relates to the fact that COX2 expression, a natural response to wounding, leads indirectly to the activation of the aromatase enzyme. Therefore, an AI might in part suppress the activation of dormant metastases by the act of surgery. If this is the case, AIs should be initiated at (or even before) surgery in order to optimize this effect. Furthermore, this benefit might be enhanced by the concomitant use of a COX2 inhibitor. Author Disclosure Employment or Leadership Consultant or Advisory Role Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration AstraZeneca AstraZeneca, Novartis, Pfizer AstraZeneca AstraZeneca, Novartis, Pfizer