Results of the randomized phase II trial of adjuvant anastrozole versus anastrozole plus fulvestrant in hormone-receptor-positive postmenopausal breast cancer patients with bone marrow micrometastasis (ABCSG trial 21).

Abstract

Abstract Abstract #4159 Purpose: Endocrine therapy is often the preferred therapeutic approach for patients (pts) with endocrine-responsive, invasive breast cancer. Bone marrow micrometastasis (BMM) is an established prognostic factor for early distant metastasis and cancer-related death. ABCSG-21 was designed to assess the predictive potential of BMM in an endocrine treatment setting.
 Methods: To be eligible for the study, pts had to be postmenopausal and to have hormone-receptor-positive invasive early breast cancer; it was also mandatory that they had completed their primary breast cancer treatment (neoadjuvant/adjuvant chemotherapy, definitive surgery). Pts with HER2 overexpressing tumors were ineligible. Pts found to have BMM were randomized to receive treatment with either anastrozole 1 mg/day (standard; N=6) or anastrozole 1 mg/day plus fulvestrant 500 mg every 28 days with a loading dose (500 mg) at day 14 (experimental; N=7). The primary objective was to compare the frequency of events (presence of BMM, clinical recurrence and/or death) after 12 months of randomized treatment.
 Results: The trial was closed after (i) 278 pts were screened, (ii) a BMM detection rate of 6.1% was found (17 of 278 pts), and (iii) 13 eligible BMM-positive pts (4.7%) had been randomized and treated (due to divergence of calculated [i.e. 15%] and actual [5%] risk ratio for the presence of BMM). Of the 17 BMM-positive pts, 65% had a T1 tumor, 60% had N0 disease, and 18% had G1 tumors. Despite small numbers, there were no significant differences in tumor size, frequency of lymph node metastasis, or tumor grading between pts with and without BMM at screening. A total of seven BMM positive pts have so far received 12 months of randomized treatment and also have had their scheduled second BM aspiration. All of these seven pts were BMM-negative at the second assessment, irrespective of study treatment received. No clinical recurrences and/or breast cancer-related deaths were observed.
 Conclusions: As stated in a previous pooled analysis (Braun et al 2006 NEJM), the presence of BMM is not individually predictable using established histopathological criteria. Thus, to select pts for treatment because of an individually increased risk of recurrence, additional markers (e.g. BMM) need to be tested. The low frequency of BMM found in our study may have several explanations but may also support the conclusion that the risk of distant metastasis and/or disease-related death among postmenopausal hormone-receptor-positive pts is lower than that reflected by the standard algorithm of decision-making with respect to endocrine breast cancer treatment. Both standard and experimental treatments appear to be effective in eliminating residual tumor cells, allowing to conclude that the trial concept remains valid. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 4159.