Clinical Cancer Research Versailles No Abst Frontiers in Basic Cancer Research
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hilsenbeck, S. G.
Right arrow Articles by Osborne, C. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hilsenbeck, S. G.
Right arrow Articles by Osborne, C. K.
Clinical Cancer Research Vol. 12, 1049s-1055s, February 2006
© 2006 American Association for Cancer Research

Recent Advances and Future Directions in Endocrine Manipulation of Breast Cancer

Is There a Role for Adjuvant Tamoxifen in Progesterone Receptor–Positive Breast Cancer? An In silico Clinical Trial

Susan G. Hilsenbeck and C. Kent Osborne

Authors' Affiliation: Breast Center and Department of Medicine, Baylor College of Medicine, Houston, Texas

Requests for reprints: Susan G. Hilsenbeck, Baylor College of Medicine, One Baylor Plaza BCM600, Houston, TX 77030. Phone: 713-798-1632; Fax: 713-798-1642; E-mail: sgh{at}breastcenter.tmc.edu.


   Abstract
Top
 Abstract
Materials and Methods
 Results
 Discussion
 Open Discussion
 References
 
Purpose: Subset analysis from the Arimidex, Tamoxifen, Alone or in Combination trial, a major adjuvant therapy trial, suggests that progesterone receptor–negative (PR–) cases may derive greater benefit from aromatase inhibitor compared with tamoxifen than PR+ cases. We postulated that estrogen receptor–positive (ER+)/PR+ patients might do as well or better by starting on tamoxifen and later switching to an aromatase inhibitor rather than by starting on an aromatase inhibitor as initial therapy.

Experimental Design: We constructed a computer model using retrospective data to approximate exponential failure rates for PR+ and PR– in tamoxifen-treated and tamoxifen-untreated patients, adding the assumptions that about half of patients are cured at surgery and that ~20% of postmenopausal ER+ early breast cancer cases are PR–. This model provided a very good approximation to the published overview data and to the clinical trials. We then used the failure rates to generate relapse times for a large number of cases (n = 50,000) for each treatment scenario.

Results: In PR– cases, initial therapy with an aromatase inhibitor is superior to tamoxifen and this advantage can never be made up by switching. In PR+ cases, tamoxifen is only modestly inferior to aromatase inhibitor at the outset, and after switching to an aromatase inhibitor at 3 or 5 years the tamoxifen relapse-free survival curve catches up and then begins to surpass the aromatase inhibitor curve at 7.5 or 12 years, respectively.

Discussion: Although our in silico trial is based on many assumptions, it closely approximates results of the published trials and, therefore, suggests that an in vivo comparison in ER+/PR+ patients of aromatase inhibitor versus tamoxifen followed by aromatase inhibitor may be worth considering.

Aromatase inhibitors are effective in the adjuvant treatment of breast cancer. Recently reported clinical trials have shown that for postmenopausal hormone receptor–positive early breast cancer, an aromatase inhibitor (anastrozole) is superior to tamoxifen with respect to relapse-free survival (1–4). At 6 years of follow-up in the Arimidex, Tamoxifen, Alone or in Combination (ATAC) trial, which compared 5 years of anastrozole to 5 years of tamoxifen, the hazard ratio for recurrence-free survival of anastrozole compared with tamoxifen was 0.74 (3). Adding an aromatase inhibitor after some period of tamoxifen treatment is superior to placebo or to finishing out the planned course of tamoxifen (5–8). In the MA.17 trial of letrozole or placebo after 5 years of tamoxifen, the hazard ratio for recurrence-free survival after the switch was 0.57 (6), and in the Intergroup Exemestane Study (IES) of 3 or 2 years exemestane after 2 to 3 years of tamoxifen compared with continuation to 5 years of tamoxifen, the hazard ratio was 0.64 (5). Data for the Italian Tamoxifen Anastrozole (7) and Austrian Breast and Colorectal Cancer Study Group 6a (8) trials are qualitatively similar, although follow-up is not yet mature.

Although the case for aromatase inhibitors is impressive, a closer look at the data suggests that some subgroups derive more benefit than others, and that clinical decision making may benefit from consideration of additional factors. Data presented by Dowsett (1) in San Antonio in 2003 indicate that, in the ATAC trial, there is a strong interaction with progesterone receptor (PR) status on the relative benefit of anastrozole versus tamoxifen, with PR-negative (PR–) cases deriving greater benefit from aromatase inhibitor compared with tamoxifen than PR-positive (PR+) cases (recurrence-free hazard ratio = 0.49 versus 0.81, respectively). Some of the advantage of anastrozole over tamoxifen in the overall results of the ATAC trial is likely due, therefore, to the large advantage of anastrozole over tamoxifen in PR– cases. Although the BIG 1-98 trial (9), which compares letrozole to tamoxifen, does not support a differential outcome for PR– and PR+ cases for either treatment group, a differential effect of PR on prognosis and treatment benefit is entirely consistent with historical clinical experience, which suggests that PR– cases have a worse prognosis (10) and respond less well to tamoxifen. It is also consistent with older clinical observations (11) and more recent laboratory studies by the Osborne-Schiff laboratory (12), which suggest that loss of PR is associated with acquired tamoxifen resistance. Both effects are likely a consequence of increased signaling from growth factor receptor kinases, which down-regulate transcription of the PR gene (13).

In view of the surprisingly large benefit from an aromatase inhibitor following tamoxifen and the modest advantage of an upfront aromatase inhibitor over tamoxifen in PR+ breast cancer, we wondered whether the group of ER+/PR+ patients might do as well or better by starting on tamoxifen and switching to an aromatase inhibitor after some period of time rather than by starting on an aromatase inhibitor as initial therapy. Because there are no clinical data addressing this question, we conducted the trial in silico.


   Materials and Methods
Top
 Abstract
 Materials and Methods
Results
 Discussion
 Open Discussion
 References
 
We investigated this question using a computer model that was constructed to account for breast cancer natural history and for the reported effects of each of the regimens. Then, a simulation was run to generate the expected relapse-free survival curves to compare the treatments of interest. The outcome of interest is time to recurrence, with nonbreast cancer deaths being censored. Published data from three clinical trials and a retrospective review were used (1, 3, 5, 6, 9, 14). Here, we assume all aromatase inhibitors are created equal and make no effort to attribute differential effects due to specific drugs.

Summary data for estrogen receptor (ER)–positive/unknown women reported in the Oxford Overview analysis (ref. 14; http://rum.ctsu.ox.ac.uk/~ebctcg/systemic2000/a8r.htm) for 5 years of tamoxifen versus placebo were used to construct the baseline model. We digitized "time to recurrence" curves from the Oxford Overview to use to compare to various parametric distributional models of time to recurrence to set baseline variables. As shown in Fig. 1, attempts to adequately mimic the Oxford Overview curves with standard parametric failure time distributions, such as exponential, Weibull, or lognormal, were unsatisfactory. However, adding the assumption that about half of patients are cured at surgery and are not at risk for relapse allowed us to use a mixture model. In this model, we gave no risk to the cured fraction and used an exponential model (or mixture of exponentials) for the uncured fraction.


Figure 1
View larger version (23K):
[in this window]
[in a new window]
  		Fig. 1. Relapse-free survival in ER+ cases observed in Oxford Overview (connected circles) and modeled (solid lines) by various probability distributions. Separate variable sets were used for tamoxifen- and placebo-treated cases. A, lognormal; B, Weibull; C, exponential; D, exponential with 47% of cases cured by surgery alone. Solid gray lines, exponential failure.

 
Relapse-free survival function was computed as follows:

Formula
where pc is the proportion cured by surgery, pPR– is proportion of cases that are PR–, pPR+ is the proportion PR+ (1 – pPR–), F is the cumulative distribution function for the exponential distribution, and rPR– and rPR+ are treatment-specific failure rates of noncured PR– and PR+ cases, respectively. Using retrospective data (10) to empirically approximate exponential failure rates for PR+ and PR– in tamoxifen-treated and tamoxifen-untreated patients separately, and assuming that ~30% of ER+ early breast cancer cases are PR–, this mixture model provided a very good approximation to the published Oxford Overview data (Fig. 2). To model results of the ATAC, MA.17, and IES studies, we reduced the proportion of PR– cases to 20% to account for the fact that all patients are postmenopausal and used the previously determined tamoxifen PR-specific failure rates. We then empirically determined failure rates in PR+ and PR– cases for an aromatase inhibitor by requiring that the resulting hazard ratios match those reported for the ATAC trial. Hazard ratios were computed by simulating data according to the selected variables for a large number of cases (n = 50,000) treated with either tamoxifen or aromatase inhibitor and computing a Cox regression for the simulated data. Maximum follow-up time was set at 10 years to simulate Oxford Overview and retrospective data, and 7 years for ATAC, MA.17, and IES because follow-up is shorter in these studies. Simulated data were generated, as shown in the flow diagram in Fig. 3, by censoring the cured cases at the maximum follow-up time and by generating relapse times from an exponential distribution with an appropriate failure rate and then censoring those cases with relapse times longer than the maximum follow-up time. Table 1 shows the variables used.


Figure 2
View larger version (12K):
[in this window]
[in a new window]
  		Fig. 2. Relapse-free survival in ER+ cases observed in Oxford Overview (connected circles) and modeled (solid lines) by cured fraction (47%) and mixture of exponential distributions for PR– and PR+, with 30% of cases being PR–.

 

Figure 3
View larger version (11K):
[in this window]
[in a new window]
  		Fig. 3. Schema for generating simulated relapse-free survival times in accordance with the models. The function rexp() generates a randomly selected time, t, to recurrence from the standard exponential distribution with the specified failure rate. If t is greater than Tc, the censor time, then t = tc and the observation is censored. Failure rates are a function of PR status and treatment.

 

View this table:
[in this window]
[in a new window]
  		Table 1. Parameters used in simulations

 
Next, we determined failure rates for an aromatase inhibitor after either 3 or 5 years of tamoxifen. We assumed the same continuing failure rate for tamoxifen-treated cases and used the IES and MA.17 data to approximate aromatase inhibitor–associated failure rates for cases still relapse-free after 3 or 5 years of tamoxifen, respectively. We required that the hazard ratios closely match those reported for the trials. Based on laboratory observations and limited clinical observations (11, 12), we assumed that 10% of PR+ cases will convert to PR– after 3 years of tamoxifen treatment and 30% will convert after 5 years. We believe that this conversion helps explain the magnitude of the effect seen for switching to an aromatase inhibitor. Without this conversion, the effect of an aromatase inhibitor after tamoxifen would be required to be unrealistically large—far larger than that of an aromatase inhibitor upfront. Having derived approximations for all of the relevant exponential failure rates, we used the rates in a final series of simulations to generate relapse times for 50,000 patients for each treatment scenario. Simulations of the Oxford Overview data were censored at 10 years of follow-up. Simulations of the ATAC, MA.17, and IES trials were censored at 7 years of follow-up. Cox regression was used to compute final hazard ratios for the relevant groups. Survival curves were plotted for the various scenarios using the model and extending results to 15 years. The carryover effect of an aromatase inhibitor after treatment stops is not yet known, and several plausible scenarios were explored to provide a range of possible outcomes, including no change in failure rate (indefinite carryover), modest increase in failure rate approximating that of tamoxifen (moderate carryover), or reversion of failure rate to that of no treatment (no carryover). All computations and simulations were carried out in S-PLUS (Version 6.2 for Windows, Insightful Corporation, Seattle, WA).


   Results
Top
 Abstract
 Materials and Methods
 Results
Discussion
 Open Discussion
 References
 
We used the variables in Table 1 to simulate 50,000 cases in each treatment group as described above to recreate the Oxford Overview and clinical trial data. The simulated hazard ratios closely approximate those reported in the literature (Table 2). We therefore used the model to explore the effects of various treatment strategies in biological subgroups of patients.


View this table:
[in this window]
[in a new window]
  		Table 2. Summary of modeled and observed hazard ratios

 
It is clear in the model that aromatase inhibitors are superior to tamoxifen in PR– cases and this advantage can never be made up by switching (Fig. 4). However, in PR+ cases, tamoxifen is only modestly inferior to aromatase inhibitor at the outset, and after switching to an aromatase inhibitor at 5 years the modeled curve for sequential therapy catches up and begins to pass the aromatase inhibitor curve at ~12 years, even if we assume that the benefit for aromatase inhibitor alone continues at a quantitatively similar level after the aromatase inhibitor is stopped at 5 years (Fig. 5A). If the aromatase inhibitor benefit declines or eventually disappears, then the long-term advantage for starting on tamoxifen and switching to an aromatase inhibitor may be greater (i.e., crossover happens earlier). The same analysis was also done for the scenario of switching to an aromatase inhibitor after 3 years of tamoxifen (Fig. 5B). The sequential curve catches up and begins to pass the curve for 5 years of an aromatase inhibitor at ~7.5 years. We did not compute hazard ratios for the aromatase inhibitor versus sequential therapy scenario because the curves cross, violating the assumption of proportional hazards. Indeed, the outcome of the comparison depends on the amount of follow-up, with short-term follow-up favoring aromatase inhibitor and very long-term follow-up favoring sequential therapy.


Figure 4
View larger version (10K):
[in this window]
[in a new window]
  		Fig. 4. Simulation of ATAC trial, showing outcome for PR– (dotted) and PR+ (solid) for tamoxifen (lower lines), and an aromatase inhibitor (upper lines).

 

Figure 5
View larger version (18K):
[in this window]
[in a new window]
  		Fig. 5. Modeled effect of adding aromatase inhibitor after tamoxifen (dashed line), compared with aromatase inhibitor (dotted) or tamoxifen alone (solid) in ER+/PR+ early breast cancer. Shaded region shows varying carryover effects of aromatase inhibitor. A, 5 years of tamoxifen followed by 5 years of aromatase inhibitor. B, 3 years of tamoxifen followed by 2 years of aromatase inhibitor.

 

   Discussion
Top
 Abstract
 Materials and Methods
 Results
 Discussion
Open Discussion
 References
 
Based on the overall superiority of initial aromatase inhibitor compared with tamoxifen, many oncologists have chosen to adopt this approach for adjuvant endocrine therapy in postmenopausal patients, despite the lack of long-term efficacy or toxicity data. If the differential effects due to PR that have been reported from the ATAC and IES trials are confirmed, then PR– patients benefit from initial treatment with an aromatase inhibitor. Our analysis suggests that this strategy could be suboptimal in patients with PR+ tumors.

In our approach, we did not attempt to distinguish between different aromatase inhibitors or to match relapse-free survival distributions of all the individual studies in part because they are not reported in detail and in part because the underlying population mix probably differs. Instead, we developed a model of a baseline population based on the Oxford Overview data for tamoxifen-treated patients and applied treatment and biological effects to this population to mimic the relative effects reported by the clinical trials. This maintained consistency among groups and allowed us to compare groups that have not been compared clinically. We attempted to keep the model as simple as possible. An important feature of our model is accounting for the conversion over time of some tamoxifen-treated PR+ cases to PR–. The phenomenon, which is seen both in the clinic and in the laboratory, may help explain the surprisingly large beneficial effects of sequencing an aromatase inhibitor after tamoxifen. Although our in silico (i.e., "in or by means of a computer simulation") trial is based on a number of assumptions, simple models, and relatively little data, it does closely approximate results of the published trials and, therefore, suggests that an in vivo comparison in ER+/PR+ patients of aromatase inhibitor versus tamoxifen followed by aromatase inhibitor may be worth considering.

Using a different, decision analytic approach, Burstein et al. (15) also conclude that the optimal adjuvant therapy strategy may differ based on biological features of the tumor. Like our model, their analysis of long-term relapse-free survival favors "tamoxifen then aromatase inhibitor after 2 years" over aromatase inhibitor alone in PR+ cases, although some aspects of the analysis require further clarification. For example, contrary to clinical experience, their analysis predicts that ER+/PR+ cases have lower disease-free survival than ER+/PR–. Cuzick and Howell (16) have also derived models to compare the relative benefits of aromatase inhibitor as initial therapy versus tamoxifen followed by an aromatase inhibitor. They focus on years of life lost to recurrence and conclude that using an aromatase inhibitor as initial adjuvant treatment is a better option than switching patients to an aromatase inhibitor after ≥2 years of tamoxifen, but that this is most clear for PR– patients where the effects are large; for PR+ patients, the results are less clear and are model dependent. This also suggests a need to formally compare these treatments in the PR+ subgroup.

More mature results of the ATAC (2), MA.17 (6), Italian (7), Austrian (8), and BIG 1-98 (9) trials with central laboratory determinations of ER, PR, Her2, and a proliferation marker will also be informative. In the absence of data on the long-term carryover effects and toxicity of aromatase inhibitor treatment, the long-term outcomes of aromatase inhibitor alone versus tamoxifen followed by aromatase inhibitor may be very similar and a noninferiority design might be the most appropriate choice for comparison. Nonbreast cancer events, such as fracture, cardiovascular events, or endometrial cancer, and quality of life might then become the determining factors. Our model suggests, however, that it is very possible that tamoxifen followed by an aromatase inhibitor may produce superior long-term results in patients with ER+/PR+ tumors and the widespread use of initial aromatase inhibitor therapy in this group could be a mistake.


   Open Discussion
Top
 Abstract
 Materials and Methods
 Results
 Discussion
 Open Discussion
References
 
Dr. Myles Brown: Twenty years ago, people were saying patients should stay on tamoxifen indefinitely. That was the initial assumption because we didn't understand the mechanism fully; we thought that "tamoxifen forever" was better, but in fact stopping was better.

Dr. Hilsenbeck: At this point, there are no data on continuing an aromatase inhibitor longer than 5 years, and there are no data on what happens to people 5 or 10 years after you stop the aromatase inhibitor. We don't have any data for the model on which to set changing failure rates. The question is, do you assume complete carryover effects? Or do you assume that after the aromatase inhibitor stops that the failure rate increases a little bit, or it increases a lot? I tried those what-ifs and that shaded area in the figure was basically the result of those simulations. It's essentially just guesswork at this point.

Dr. James Ingle: In following ATAC, it will be important to watch if the survival curves start to come back together. If they start changing slope, then we have to worry that maybe we ought to be continuing an aromatase inhibitor for longer than 5 years, if one can do that.

Dr. Steven Come: What happens in your model if you change the frequency of conversions in PR status? I always wonder what it is about initial tamoxifen exposure that would allow a subsequent aromatase inhibitor to be better and, therefore, for the curves to cross? Are you postulating that it is setting a tumor up biologically for conversion to PR negative?

Dr. Hilsenbeck: Yes, exactly. If you make the conversion rate zero, then tamoxifen upfront doesn't look as good and it is hard to make the model fit the reported data. If you increase the conversion rate, tamoxifen gets better. So I think you're right, even if a relapse hasn't been observed, tamoxifen is setting the tumor up for being more sensitive to the aromatase inhibitor.

Dr. Rakesh Kumar: If initial exposure of tamoxifen is priming this patient for a better response by aromatase inhibitor, perhaps there are tamoxifen-responsive genes that are part of the biosynthesis of estrogen pathway. This is a testable possibility.

Dr. Mitch Dowsett: In terms of published data, there is Dr. Johnston's report (Cancer Res 1995;55:3331–8) showing the drop-off in PR in the tamoxifen-treated patients who had relapsed. The numbers weren't enormous, ~20 or 30 patients, but there was a highly significant decrease in PR. Some of them were becoming frankly PR negative.

Dr. Brown: If I understand the model correctly, that conversion number is the critical determinant. That is an experimentally determinable thing, by designing clinical trials so that at first relapse the patient will have a biopsy and the PR status will be ascertained.

Dr. Hilsenbeck: The people who relapse are not necessarily a random sample of the converted but are probably an oversampling of the PR negative.

Dr. Aman Buzdar: From the clinical point of view, the issues are as follows: are there patients for whom tamoxifen should be initial therapy?; what is the evidence to support its use?; and which are the patients for whom an aromatase inhibitor should be the initial therapy?

Dr. Come: There are no comparative data. All you can say is that if you start tamoxifen, you probably shouldn't continue that by itself forever. That we know. That's inferior. But we don't know what's superior to that. If you are focusing on the first 2 or 3 years after diagnosis, your best strategy to be disease-free is to start with an aromatase inhibitor. But we don't know whether the long-term outcome is going to be best with that strategy.

Dr. Eric Winer: We can say that adding an aromatase inhibitor at some point in time lowers recurrence rates compared with taking tamoxifen alone. We know that starting with anastrazole versus tamoxifen does not lead to any difference in the proportion of women who are alive at 5 years. The only study to date that has even suggested a survival advantage for this particular strategy is the MA.17 trial.

Dr. Ingle: It comes down to what the primary end point of these studies has been. It has been disease-free survival. Nobody has required that you have to show a survival advantage to identify an optimal strategy. Based on the available data, you are better off starting with an aromatase inhibitor.

Dr. Winer: For the PR-negative, HER2-positive patients, an upfront aromatase inhibitor would be preferred, albeit not based on the firmest evidence. But for the other patients, I don't know that one can say there is a preferred strategy.

Dr. Dowsett: You are basing these comments on these switching data. How concerned are you that not all the patients have got into that switching because we have already lost many of the patients to tamoxifen relapse at 2 or 5 years.

Dr. Winer: Dr. Hilsenbeck said it perfectly. It is all about how much of a difference do you have to make up and how fast do you make it up. Clearly, at 2 years, the tamoxifen bunny is running behind the aromatase inhibitor bunny. But the question is whether crossing over then allows you to get ahead.

Dr. Stephen Johnston: After all, you still lose patients on aromatase inhibitors. So we want to know the optimal long-term strategy for a population of 100 patients over 10 years: Will there be a net gain by a switch approach? We have argued enough over the years about the mechanisms for fooling resistance and we have talked about trying to do switch strategies in advanced disease. Here is the scenario where we actually may be sitting on the data that tell us the net long-term gain is the switch approach, for a proportion of patients. Some, I agree, we can already identify who should be getting an aromatase inhibitor upfront, but I don't know that we can say that everyone should have an aromatase inhibitor upfront.

Dr. Buzdar: Yes, the lines may come together or the lines may crossover, but the life-years lost with one strategy versus another is an issue that we have to consider. If a patient relapses at year 9 and dies at year 10, 1 year is lost. If a significant fraction of patients relapse after years 1, 2, or 3, then when you calculate the time lost for over the 10-year period, that is a huge time-loss in years of life lost.

Dr. Brown: Would you have that one person recur at year 1 or 10 people recur at year 9? It is still 10 years.

Dr. Pritchard: But we don't know that is happening yet.

Dr. Brown: Let's wait and see what the crossover will be—that is what the model shows.

Dr. Ingle: If you're a doctor sitting in the room with Mrs. Smith, is there a profile that woman could have so that you'd say, "You really need tamoxifen"? Tamoxifen would be the preferred initial therapy?

Dr. Arteaga: The woman with ER-positive, PR-positive, HER2-negative disease.

Dr. Ingle: So the phenotype is ER-positive, PR-positive, HER2-negative? The reason I couldn't say, "You absolutely should take tamoxifen," is that I know that the hazard ratios for adverse effects, such as the thromboembolic events, are higher. I think you can tell a patient, "Tamoxifen is almost as good for reducing risk of recurrence as anastrozole." But I don't see that is where you say, "You absolutely have to take tamoxifen."

Dr. Winer: I worry about P values of 0.08 around non–breast cancer deaths, and so particularly for the patient who is at lower risk, I am concerned that we may, for some number of women, be causing more harm than good. I would really like to have more data, particularly in those situations where at the moment the two choices are relatively equivalent.

Dr. Ingle: Okay, so you're in equipoise. It sounds like you have said there is not a population where you would say, "You ought to take tamoxifen, Mrs. Smith."

Dr. Winer: I am in equipoise for patients with ER-positive, PR-positive, HER2-negative disease, in terms of whether to use tamoxifen followed by an aromatase inhibitor, or an aromatase inhibitor upfront. I would be happy to put any patient in that situation on a randomized trial to compare those two strategies.


   Footnotes
 
Grant support: Grants 5P50CA058183 and 5P01CA030195.

Presented at the Fifth International Conference on Recent Advances and Future Directions in Endocrine Manipulation of Breast Cancer, June 13-14, 2005, Cambridge, Massachusetts.

Received 9/28/05; accepted 10/11/05.


   References
Top
 Abstract
 Materials and Methods
 Results
 Discussion
 Open Discussion
 References
 

  1. Dowsett M; ATAC Trialists. Analysis of time to recurrence in the ATAC (Arimidex, Tamoxifen, Alone or in Combination) trial according to estrogen receptor and progesterone receptor status. In: San Antonio Breast Cancer Symposium, San Antonio, Texas; 2003.
  2. Baum M, Budzar AU, Cuzick J, et al. Anastrozole alone or in combination with tamoxifen versus tamoxifen alone for adjuvant treatment of postmenopausal women with early breast cancer: first results of the ATAC randomised trial. Lancet 2002;359:2131–9.[CrossRef][Medline]
  3. Howell A, Cuzick J, Baum M, et al. Results of the ATAC (Arimidex, Tamoxifen, Alone or in Combination) trial after completion of 5 years adjuvant treatment for breast cancer. Lancet 2005;365:60–2.[CrossRef][Medline]
  4. Dowsett M, Cuzick J, Wale C, et al. Retrospective analysis of time recurrence in the ATAC trial according to hormone receptor status: an hypothesis-generating study. J Clin Oncol 2005;23:7512–7.[Abstract/Free Full Text]
  5. Coombes RC, Hall E, Gibson LJ, et al. A randomized trial of exemestane after two to three years of tamoxifen therapy in postmenopausal women with primary breast cancer. N Engl J Med 2004;350:1081–92.[Abstract/Free Full Text]
  6. Goss PE, Ingle JN, Martino S, et al. A randomized trial of letrozole in postmenopausal women after five years of tamoxifen therapy for early-stage breast cancer. N Engl J Med 2003;349:1793–802.[Abstract/Free Full Text]
  7. Boccardo FM, Rubagotti A, Puntoni M, et al. Switching to anastrozole (ANA) vs continued tamoxifen (TAM) treatment of early breast cancer (EBC). Updated results of the Italian tamoxifen anastrazole (ITA) trial. 2005 ASCO annual meeting proceedings. J Clin Oncol (Suppl) 2005;23:10s.
  8. Jakesz R, Samonigg H, Greil R, et al. Extended adjuvant treatment with anastrozole: Results from the Austrian Breast and Colorectal Cancer Study Group Trial 6a (ABCSG-6a). 2005 ASCO annual meeting proceedings. J Clin Oncol (Suppl) 2005;23:10s.
  9. Thurlimann BJ, Keshaviah A, Mouridsen H, et al. BIG 1-98: randomized double-blind phase III study to evaluate letrozole (L) vs tamoxifen (T) as adjuvant endocrine therapy for post-menopausal women with receptor-positive breast cancer. 2005 ASCO Annual Meeting Proceedings. J Clin Oncol (Suppl) 2005;23:6s.
  10. Bardou VJ, Arpino G, Elledge RM, Osborne CK, Clark GM. Progesterone receptor status significantly improves outcome prediction over estrogen receptor status alone for adjuvant endocrine therapy in two large breast cancer databases. J Clin Oncol 2003;21:1973–9.[Abstract/Free Full Text]
  11. Gross GE, Clark GM, Chamness GC, McGuire WL. Multiple progesterone receptor assays in human breast cancer. Cancer Res 1984;44:836–40.[Abstract/Free Full Text]
  12. Massarweh S, Osborne CK, Wakeling AE, Schiff R. Tamoxifen resistance in a breast cancer xenograft model coincides with a switch from an ER+/PgR+ to an ER+/PgR– phenotype accompanied by EGFR/HER2 activation. In: San Antonio Breast Cancer Symposium, San Antonio; 2004.
  13. Cui X, Zhang P, Deng W, Oesterreich S, Lu Y, Mills GB, Lee AV. Insulin-like growth factor-I inhibits progesterone receptor expression in breast cancer cells via the phosphatidylinositol 3-kinase/Akt/mammalian target of rapamycin pathway: progesterone receptor as a potential indicator of growth factor activity in breast cancer. Mol Endocrinol 2003;17:575–88.[Abstract/Free Full Text]
  14. Early Breast Cancer Trialists' Collaborative Group. Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials. Lancet 2005;365:1687–717.[CrossRef][Medline]
  15. Burstein HJ, Winer EP, Kuntz KM, Weeks JC, Punglia RS. Optimizing endocrine therapy in postmenopausal women with early stage breast cancer: a decision analysis for biological subsets of tumors. 2005 ASCO Annual Meeting Proceedings. J Clin Oncol (Suppl) 2005;23:11s.
  16. Cuzick J, Howell A. Optimal timing of the use of an aromatase inhibitor in the adjuvant treatment of postmenopausal hormone receptor-positive breast cancer. 2005 ASCO Annual Meeting Proceedings. J Clin Oncol (Suppl) 2005;23:43S.




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hilsenbeck, S. G.
Right arrow Articles by Osborne, C. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hilsenbeck, S. G.
Right arrow Articles by Osborne, C. K.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online