Conclusions regarding relative cost-utility of alternative strategies for use of aromatase inhibitors in postmenopausal women with early breast cancer are premature.
T. Delea,J. Karnon,P. Goss
DOI: https://doi.org/10.1093/ANNONC/MDL171
IF: 51.769
2006-07-27
Annals of Oncology
Abstract:We read with interest the study by Lønning [1] examining the cost–utility of aromatase inhibitors as adjuvant therapy in postmenopausal women with early breast cancer. We are concerned, however, with the validity of the analysis, its results and conclusions. The fundamental assumption underlying the analysis is that each strategy results in the same 2% absolute reduction in the risk of relapse on an intent-to-treat basis. Given the assumption of a similar 2% benefit for all strategies regardless of timing or duration of therapy, it is not surprising that the authors conclude that switching to an aromatase inhibitor after 2–3 years treatment with tamoxifen is more cost-effective than 5 years of treatment with an aromatase inhibitor (either up front or after 5 years of tamoxifen), as the latter will obviously be associated with higher costs than the former. Unfortunately, this fundamental assumption is unsupported by data from the controlled trials that are the basis of the comparison. In the MA-17 study, the absolute reduction in Kaplan–Meier estimated disease-free survival (DFS), defined as the time from randomization to the recurrence of the primary disease or development of a new primary breast cancer in the contralateral breast, at 4 years was 4.6% with extended adjuvant letrozole versus placebo (94.4% versus 89.8%) [2]. In the IES study, there were 144 breast cancer events in the exemestane group (n = 2362) versus 227 in the tamoxifen group (n = 2380), representing an absolute risk reduction of 3.4% over median follow-up of 30.6 months [3]. In ATAC, there was a 2.4% absolute difference in Kaplan–Meier estimated DFS after 5 years for patients receiving up-front anastrozole versus tamoxifen [4]. While we recognize that differences in study populations, trial designs and definitions of study outcomes make comparisons across trials difficult and potentially misleading, these findings clearly suggest that there may be important differences between sequencing strategies in the absolute magnitude of risk reduction. The magnitude of these differences, which will determine the relative cost-effectiveness of different strategies, are unknown, however, and will remain so pending results of randomized controlled trials which explicitly compare them, such as the ongoing BIG 1–98 trial [5]. The study has a number of other important limitations. The author fails to consider in the primary analysis the effects of aromatase inhibitors on incidence of contralateral tumors, an event that has important clinical and economic consequences for breast cancer patients. Also, the costs of recurrences were not considered in the primary analysis, an omission that is inconsistent with good modeling practices [6]. Although the author considered the effects of aromatase inhibitors on fractures, other adverse events that might be less frequent with aromatase inhibitors (e.g. endometrial cancer, thromboembolism) were not considered. Furthermore, in estimating the effects of different strategies on fracture risk, Lønning failed to consider the potential beneficial effects of tamoxifen. As these benefits are probably greatest with the ‘tamoxifen’ and ‘letrozole’ strategies, the analysis may be biased in favor of the other strategies. Also, the author assumed that the effects of aromatase inhibitors on risk would persist after therapy discontinuation, whereas data from ATAC suggest that fracture risk is similar after discontinuation of anastrozole versus tamoxifen [7]. Finally, it should also be noted that there exists a large group of women who are currently receiving or have recently completed initial adjuvant therapy with tamoxifen, and for whom the question is not what initial therapy to receive, but whether to initiate extended adjuvant therapy with an aromatase inhibitor. In these women, the conclusions of the MA-17 trial are clear, that extended advjuvant letrozole significantly improves survival and has a favorable cost–effectiveness ratio in such patients [8–10]. Cost–utility analysis is an important tool for evaluating treatment strategies for women with early breast cancer. The usefulness of such analyses depends not only upon the appropriateness of the assumptions employed, but also the extent to which potential users of the information can understand and interpret the results [6]. Unfortunately, the study by Lønning fails on both counts. Further research is needed before conclusions can be drawn regarding the relative cost–effectiveness of alternative sequencing strategies of adjuvant hormonal therapy in patients with early breast cancer.