Annals of Oncology Advance Access originally published online on July 27, 2006
Annals of Oncology 2007 18(1):198-199; doi:10.1093/annonc/mdl170
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© 2006 European Society for Medical Oncology
letters to the editor |
Reply to ‘Conclusions regarding relative cost–utility of alternative strategies for use of aromatase inhibitors in postmenopausal women with early breast cancer are premature’ by Delea et al.
Institute of Medicine Haukeland University Hospital, Department of Oncology, Bergen, Norway
* (E-mail: per.lonning{at}helse-bergen.no)
The letter by Delea et al. [1] raises several interesting questions with respect to cost–utility assessment. In principle, such analyses may be conducted in two different ways: we may calculate utilities based on actual data from the literature, which is the most common practice, or we may do what I did [2], i.e. assume a certain long-term effect and see whether that should provide a reasonable cost–utility estimate. For reasons to be explained below, I believe the second option is of particular value when assessing use of aromatase inhibitors at this stage.
In the first part of their letter, Delea et al. discuss the relapse rates detected in the different phase III studies. This is an interesting topic indeed; nevertheless, it is difficult to see the relevance to the discussion of the model applied in this study. Here, the assumption was that each approach (an aromatase inhibitor upfront or following 2–3 or 5 years on tamoxifen) would cause a 2% absolute improvement in long-term relapse-free survival and, if so, what would be the difference with respect to cost–utility? The reason for using such an approach is the uncertainty regarding long-term effects with aromatase inhibitors. Neither the data from the MA17 or any of the other phase III trials allow us to estimate long-term effects with respect to efficacy or toxicity. A few examples may illustrate this. Based on safety data, the argument has been raised that there could be a difference in cardiovascular morbidity between anastrozole and letrozole. If so, it may be related to a more potent aromatase inhibition of letrozole compared with anastrozole [3]. In case this has an impact on the side-effect profile, we could assume a difference with respect to antitumor efficacy as well. Similarly, the minor androgen-agonistic activity of exemestane [4] may impact not only on bone density, but could explain the lack of complete cross-resistance between the individual aromatase inhibitors [5], enhancing antitumor efficacy.
Thus, we are left with a number of open questions regarding potential differences between individual drugs, regimen of application (monotherapy versus sequential) as well as long-term effects. Using a model as applied—comparing cost–utility between different approaches based on assumption of similar efficacy as well as side-effects between compounds—may allow the reader to compare follow-up results to the assumptions included in the model. It is difficult to see why the details provided by Delea et al. with respect to short-time results should refute this approach. On the contrary, the fact that annual risk of breast cancer relapse is a non-linear process invalidates any form of indirect comparison of benefits between trials as indicated by Delea et al.
Next, they criticize the study for not including a potential impact on contralateral breast cancer and costs of recurrence. With due respect, these data are provided in Table 4 and Figure 4, respectively [2]. The reason for not including them in the initial analysis is the uncertainties with respect to these parameters. While preliminary results from the phase III studies indicate a dramatic reduction in contralateral breast cancer, we lack information as to whether this could be a short- or long-term effect. Considering costs of treatment of relapse, this scenario is rapidly changing due to implementation of novel, expensive targeted therapies as well as cytotoxic compounds. Also, there are reasons to believe that these costs may vary considerably between different countries. Accordingly, the result of this analysis is presented as a separate diagram. As pointed out by Delea et al., tamoxifen has a beneficial effect on BMD in postmenopausal women [6], although its effect with respect to preventing fractures is more difficult to estimate. Furthermore, it is not obvious that treating patients with tamoxifen for 5 years before implementing an aromatase inhibitor may prevent potential detrimental effects on bone metabolism. Considering data from the MA 17 [7], the difference in annual bone loss between the letrozole arm and the placebo arm was greater than that recorded in the 027 protocol comparing exemestane with placebo in tamoxifen-naive patients [8, 9]. Assessing the impact of an increased fracture risk on cost–utility, similar to that for survival, these are theoretical assumptions, illustrating different scenarios. Again, by providing these estimates, it allows future readers to assess cost–utility based on follow-up results.
Considering potential effects of tamoxifen on cardiovascular events and death, this is a complex issue. Looking at the most recent Oxford meta-analysis [10], the number of trials by which the exact cause of CV mortality was reported is low with a limited number of observations. Looking at the risk of CV death, there seems to be a small benefit in favor of tamoxifen in comparison to placebo, although the statistical significance is border-line. While recent evidence suggests potential benefits may appear several years after terminating tamoxifen [11], hence further complicating this issue, evidence from the WHI study [12] indirectly suggests hormonal manipulation may have minor effects on CV death. I concur with the statement by Delea et al. that we need more information, in particular with respect to long-term data to learn all the details with respect to effectiveness. I do not, however, concur with their negative and unsubstantiated statement claiming this analysis fails with respect to assumptions and interpretation of results.
 |
References |
|---|
 Top References |
|---|
1. Delea TE, Karnon J, Goss PE. Conclusions regarding relative cost-utility of alternative strategies for use of aromatase inhibitors in postmenopausal women with early breast cancer are premature. Ann Oncol doi:10.1093/annonc/mdl171.
2. Lønning PE. (2006) Comparing cost/utility of giving an aromatase inhibitor as monotherapy for 5 years versus sequential administration following 2–3 or 5 years of tamoxifen as adjuvant treatment for postmenopausal breast cancer. Ann Oncol 17:217–225.
3. Geisler J, Haynes B, Anker G, Dowsett M, Lønning PE. (2002) Influence of letrozole (Femara) and anastrozole (Arimidex) on total body aromatization and plasma estrogen levels in postmenopausal breast cancer patients evaluated in a randomized, cross-over-designed study. J Clin Oncol 20:751–757.
4. Johannessen DC, Engan T, Salle ED, et al. (1997) Endocrine and clinical effects of exemestane (PNU 155971), a novel steroidal aromatase inhibitor, in postmenopausal breast cancer patients: a phase I study. Clin Cancer Res 3:1101–1108.[Abstract]
5. Lønning PE, Bajetta E, Murray R, et al. (2000) Activity of exemestane in metastatic breast cancer after failure of nonsteroidal aromatase inhibitors: a phase II trial. J Clin Oncol 18:2234–2244.
6. Powles TJ, Hickish T, Kanis JA, Tidy A, Ashley S. (1996) Effect of tamoxifen on bone mineral density measured by dual-energy X-ray absorptiometry in healthy premenopausal and postmenopausal women. J Clin Oncol 14:78–84.[Abstract]
7. Perez EA, Josse RG, Pritchard KI, et al. (2004) Effect of letrozole versus placebo on bone mineral density in women completing >5 years (yrs) of adjuvant tamoxifen: ncic ctg ma.17b. Breast Cancer Res Treat 88:Suppl 1, Abstr 404,s36.
8. Lønning PE, Geisler J, Krag LE, et al. (2005) Effects of exemestane administered for 2 years versus placebo on bone mineral density, bone biomarkers, and plasma lipids in patients with surgically resected early breast cancer. J Clin Oncol 23:5126–5137.
9. Geisler J and Lønning PE. (2006) Aromatase inhibitors as adjuvant treatment of breast cancer. Crit Rev Oncol Hematol 57:53–61.[Web of Science][Medline]
10. Group EBCTC. (2005) Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials. Lancet 365:1687–1717.[CrossRef][Web of Science][Medline]
11. Nordenskjold B, Rosell J, Rutqvist LE, et al. (2005) Coronary heart disease mortality after 5 years of adjuvant tamoxifen therapy: Results from a randomized trial. Journal of the National Cancer Institute 97:1609–1610.
12. Anderson GL, Limacher M, Assaf AR, et al. (2004) Effects of conjugated, equine estrogen in postmenopausal women with hysterectomy. The women's health initiative randomized controlled trial. JAMA 291:1701–1712.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||