Breast Cancer Update 2005 (8)

DR PETER M RAVDIN: I believe very few people are being started on tamoxifen with the intention of receiving five years of tamoxifen and then switching to an aromatase inhibitor.

The problem with initially starting on tamoxifen is that strategies that originally start with an aromatase inhibitor will have lower recurrence rates than those starting with tamoxifen.

If you start a patient on tamoxifen, you’re conceding that she is going to do worse initially than she would have done on an aromatase inhibitor. Then you have to feel that when you switch her to an AI, the curves will then recross.

In other words, the aromatase inhibitor will be so much more effective if delivered later that it will catch up and overtake the group that did receive the aromatase inhibitor from the beginning. That is possible, theoretically, because tamoxifen and the aromatase inhibitors have somewhat different mechanisms of action.

Therefore, a strategy that uses both agents might provide the most benefit. But that’s a theoretical consideration against the very real fact that we know if you start with an aromatase inhibitor, the patients do better.

Breast Cancer Update 2006 (2)

DR KATHLEEN I PRITCHARD: When you consider randomized studies of up-front aromatase inhibitors in which disease recurs more in patients on tamoxifen than in those on the aromatase inhibitor in the first two years, it’s difficult to suggest that you should begin with tamoxifen.

Until somebody shows in a randomized fashion that patients who begin with adjuvant tamoxifen are doing better at the end of five years than the patients who use an aromatase inhibitor initially, I will discuss with virtually all my postmenopausal patients the idea of beginning therapy with an aromatase inhibitor.

Breast Cancer Update 2006 (1)

DR PAUL E GOSS: Two points for upfront therapy with an aromatase inhibitor are valid and worth discussing. The first one is that a slight balloon of events appears in the first 24 months among patients with receptor-positive breast cancer. Distant metastases, which are potentially fatal, occur in the first 24 months at a slightly higher rate than later.

The second point is that women treated with tamoxifen face a higher risk of side effects that are important, particularly gynecologic intervention, most of which is inappropriate. Gynecologists inappropriately intervene in the uterus for ultrasound-detected thickness that should be left alone. Nevertheless, the argument that’s made is that an excess of events and toxicities makes the choice of tamoxifen in those initial two years inappropriate, and you should use an aromatase inhibitor up front.

The one thing that’s consistent is that lowering estrogen with letrozole or anastrozole causes an enhanced bone resorption. The body tries to prevent that by compensating with bone formation, but the net result is bone loss and a decrease in bone mineral density (BMD).

That is a fact now, but there are a few caveats. Number one, all of this happened in an era during which most of the trials did not specifically recommend the appropriate osteoporosis guidelines for calcium and vitamin D supplementation, which is now being done widely.

Number two, women on aromatase inhibitors were not adequately screened with serial BMD tests, which are now being conducted.

Number three, bone loss very much depends on whether the woman took tamoxifen before the aromatase inhibitor because if she took tamoxifen before, her bone density is built up above the population average first and then taken down by the aromatase inhibitors.

If you look at MA17, for example, at time zero with five years of tamoxifen, bone density beats a population age-matched control. Over the next five years, it goes below the population age-matched control, but the net result is likely to be a return to square one. So it’s not a massive clinical problem.

I don’t want to understate it; it’s a real issue, but we have tools to deal with it. In addition, now that we’re monitoring more and can introduce the effective bisphosphonates that we know will counteract the aromatase inhibitors, I think we have a tight handle on this.

Breast Cancer Update 2006 (2)

DR ANTHONY HOWELL: All three aromatase inhibitors are showing about two to three percent bone loss per year, and we need to do something about that. What’s interesting to us is that in the ATAC data, although the fracture rate was increased with anastrozole, it leveled off, and when the patient stopped treatment, the curves came right back together. If that’s true, that’s fantastic, but we need more data to confirm that.

I asked our bone specialists whether bone density can improve that quickly, and they pointed out that when steroids are stopped, bone reforms rapidly; they were not surprised by our findings.

Breast Cancer Update 2006 (2)

DR PRITCHARD: The absolute difference between the arthralgias in patients on aromatase inhibitors versus tamoxifen is about five percent. I believe the aches and pains that patients experience with aromatase inhibitors are real, but it’s such a peculiar phenomenon.

Some of these women become miserable, and when you discontinue the drug, for many, the symptoms disappear. However, I’ve had some patients that I’ve put back on tamoxifen, and they still have the aches and pains.

I think this side effect is related to the lowering of estrogen. Aches and pains are reported as a menopausal symptom and are generally regarded as not all that common or serious, but maybe we don’t always listen to what women tell us about their menopausal symptoms.

With the aromatase inhibitors, we’re seeing more osteoporosis. In the MA17 data, Goss showed more fractures and osteoporosis in the patients on the placebo arm of the original trial who crossed over to letrozole in the last two years after unblinding compared to those who did not.

I think we will see some long-term complications from this unless these patients are properly treated for their osteoporosis. We have to consider how well we are prepared to either treat our patients or collaborate with primary caregivers to prevent osteoporosis, which I think is not well managed in the general population.

Duffy S et al. The ATAC (‘Arimidex’, Tamoxifen, Alone or in Combination) adjuvant breast cancer trial: First results of the endometrial sub-protocol following 2 years of treatment. Hum Reprod 2006;21(2):545-53.

In summary, this sub-protocol has found fewer endometrial abnormalities arising de novo during 2 years of treatment with anastrozole compared with tamoxifen. This study also found that endometrial thickness, as a surrogate marker of endometrial proliferation, remained consistently <5 mm in the anastrozole group. In addition, there was less need for medical intervention.

Finally, the majority of endometrial abnormalities occurred in the first year of treatment. These findings are consistent with the superior safety profile and the lower risk of endometrial cancer with anastrozole compared with tamoxifen as demonstrated in the main ATAC trial.

Coleman RE et al. Effect of anastrozole on bone mineral density: 5-year results from the ‘Arimidex’, Tamoxifen, Alone or in Combination (ATAC) trial. Proc ASCO 2006;Abstract 511.

Significant bone loss occurred throughout the five years in the anastrozole group, although there appeared to be a slowing down of the rate of bone loss in years two to five. Although no patients with normal BMD at baseline had become osteoporotic at five years, regular monitoring of BMD and bone protection strategies are likely to be required in patients receiving anastrozole in the presence of pre-existing osteopenia.

Breast Cancer Update 2006 (2)

DR HOWELL: The San Antonio data from Jakesz are important because they show that the effect of switching is not quite as big as we once thought. Whereas the hazard ratio is approximately a 40 percent reduction in the switching studies, when they took into account the first two years, the reduction in the hazard ratio was about 24 percent.

Another significant finding was the survival advantage seen in the metaanalysis of the ARNO 95, ABCSG- 8 and ITA trials. It was an important analysis because it showed, for the first time in an unselected population, the survival advantage of switching to anastrozole after two to three years of tamoxifen. Based on that, I feel we can use anastrozole in that clinical setting.

Paul Goss and Jim Ingle’s papers also presented some beautiful data — although some of that is selected — demonstrating the efficacy of letrozole for patients with hormone receptor-positive breast cancer.

Combined, I believe these data highlight the importance of the aromatase inhibitors therapeutically. We’ve also seen that apart from the bone events and aching joints, aromatase inhibitors are better than tamoxifen as far as toxicity is concerned.

Robert NJ et al. Updated analysis of NCIC CTG MA.17 (letrozole vs placebo to letrozole vs placebo) post unblinding. Proc ASCO 2006;Abstract 550.

With 54 months follow-up the HR for DFS was 0.31 (0.18, 0.55: p<0.0001) favoring patients who crossed over to letrozole compared to those who stayed on no treatment. The treatment switch was well tolerated with no significant difference in bone fractures or cardiovascular events. ... Women with hormone dependent breast cancer prescribed letrozole after a prolonged delay from completing tamoxifen experienced a significant improvement in outcome (DFS, DDFS, OS) and should be considered for this therapy.

Kaufmann M et al. Survival benefit of switching to anastrozole after 2 years’ treatment with tamoxifen versus continued tamoxifen therapy: The ARNO 95 study. Proc ASCO 2006;Abstract 547.

Median follow-up was 30.1 months. Switching to anastrozole significantly improved DFS and OS compared with continuing on tamoxifen. Fewer patients who switched to anastrozole reported serious adverse events (22.7%) compared with those who remained on tamoxifen (30.8%). ... Switching endocrine treatment improved DFS and OS in this well-defined population.

Postmenopausal women with hormone-sensitive EBC who have already received 2 years’ adjuvant tamoxifen therapy should be switched to anastrozole.

Coombes RC et al. First mature analysis of the Intergroup Exemestane Study. Proc ASCO 2006;Abstract LBA527.

Switching postmenopausal patients with receptor-positive or unknown disease who remain disease free after two to three years of tamoxifen does appear to reduce the risk of dying — about a 15 to 17 percent reduction in the risk of death…serious side effects are rare, and many may in fact be attributable to tamoxifen withdrawal.

The switching strategy appears to minimize the adverse risks of both agents. Lastly, we conclude that with two to three years post treatment follow-up, the only disease related benefits previously reported appear to be maintained. We can conclude that exemestane is safe and well tolerated.

Breast Cancer Update 2006 (2)

DR RAVDIN: Paul Goss presented follow-up data on patients who participated in the Canadian trial comparing letrozole versus a placebo after completing five years of adjuvant tamoxifen. When they broke the code at two and a half years, some of the patients taking placebo decided to switch to letrozole and a few chose no further therapy. The patients who went on to take letrozole had much lower recurrence rates, although some of them had been off any endocrine therapy for four years.

This analysis suggests that even years after stopping tamoxifen, patients can gain benefit from an aromatase inhibitor. In my practice, that means some of my patients, particularly the patients at high risk who have already been off tamoxifen for a year, should consider taking an aromatase inhibitor, specifically letrozole, because it’s the only one that has been tested in this context.

Breast Cancer Update 2006 (1)

DR PRITCHARD: We’re just starting to see women who have either received five years of an aromatase inhibitor, or who switched to an AI after two to three years of tamoxifen, and we don’t know what to do in terms of continuing or stopping. I told the last patient I saw to continue her aromatase inhibitor and come back in six months because we would have a clinical trial for her. Both Jim Ingle and Paul Goss have presented data from the MA17 trial that suggest, year upon year, letrozole continues to add benefit.

However, until we see randomized studies, we’re not going to know the best way to manage these cases. I think it’s great that the NSABP is launching a study to evaluate patients who have had five years of any aromatase inhibitor or two or three years of tamoxifen followed by an aromatase inhibitor.

These patients will or will not then be randomly assigned to an additional five years of an aromatase inhibitor.

Breast Cancer Update 2006 (2)

DR RAVDIN: Aromatase inhibition probably should be continued indefinitely, and my opinion is based on two factors. One is that if a patient stops an aromatase inhibitor, her hormone levels will, of course, recover.

Second, it may be more difficult to develop resistance to estrogen deprivation than it is to develop resistance to tamoxifen. Tamoxifen is an agonist/antagonist, and preclinical work has shown that it can be reinterpreted as an estrogen by cancer cells, but I can’t conceive of a pathway that would reinterpret no estrogen as an estrogen.

Breast Cancer Update for Surgeons 2006 (1)

DR J MICHAEL DIXON: In our preoperative study, we found the aromatase inhibitors were as effective at reducing proliferation in patients with HER2-positive disease as in those with HER2- negative disease. The degree of reduction was identical in patients with HER2-positive and HER2-negative disease. It’s as though HER2 isn’t important in relation to the likelihood of responding to an aromatase inhibitor.

Breast Cancer Update — Think Tank Issue 1, 2006

DR C KENT OSBORNE: One important issue is whether HER2 overexpression and PR loss predict for less benefit from tamoxifen than from an aromatase inhibitor. To me, the data are overwhelming that PR status predicts for response to tamoxifen.

In a prospectively designed SWOG trial published by Peter Ravdin, patients with metastatic disease were treated with tamoxifen. The trial was designed to address the value of PR status. On multivariate analysis, PR status was found to be an independent predictor. That was the first prospective trial following another five or 10 studies published in the early 1980s and late 1970s suggesting that patients with PR-negative disease responded less well to tamoxifen.

What about HER2 overexpression and tamoxifen? Most, but not all, studies show less benefit if HER2 is overexpressed. Preclinical studies strongly support the clinical data. So I tend to believe the majority of the clinical data, along with the biology, that HER2 does predict for less responsiveness to tamoxifen.

We have very little data with the aromatase inhibitors. We have three separate neoadjuvant trials and a fourth from Mike Dixon’s group in Edinburgh that show very similar results. Whether it is letrozole or anastrozole, the responses are really quite good for patients with HER2-positive disease.

Breast Cancer Update — Think Tank Issue 1, 2006

DR GEORGE W SLEDGE JR: I find the ER-PR data interesting biologically. Having said that, I don’t know how much real-world relevance it has because I can’t pick out any population of patients in whom tamoxifen does better than an aromatase inhibitor.

Because of that, my default — unless it’s going to be the oddball patient who can’t tolerate an aromatase inhibitor for some reason — will be to use an aromatase inhibitor.

Breast Cancer Update 2006 (2)

DR MARTINE J PICCART-GEBHART: I tend to look at the profile of the tumor. If I’m dealing with a highly endocrine-responsive tumor with little worry about early relapse on therapy — a situation in which both ER and PR are very high, the proliferation genes are very low, the tumor is Grade I, and there is no HER2 overexpression — I believe there is a very low risk that the patient will relapse if you put her on tamoxifen for two years.

Breast Cancer Update 2006 (2)

DR RAVDIN: The ATAC trial found that the extra advantage of an aromatase inhibitor — in this case, anastrozole — was seen strongly only in the patients with ER-positive, PR-negative disease. This finding was based on 6,000 patients, and it had a very large p-value.

However, the BIG FEMTA study, which compared letrozole to tamoxifen as up-front therapy, did not find a significant difference in the efficacy of these agents relative to the status of the progesterone receptor.

The BIG investigators also evaluated the HER2 status of roughly 4,000 patients because data suggest that aromatase inhibitors may be more effective in tumors that are HER2-positive and ER-positive. They conducted a well-controlled study and found no significant difference on the basis of HER2, either.

The relationship between hormone receptor status and the impact of endocrine therapy was also examined in the NCIC-CTG MA17 trial. This trial randomly assigned patients who had taken five years of adjuvant tamoxifen to five years of letrozole versus a placebo. Patients with ER- and PR-positive disease particularly benefited from letrozole.

However, patients with ER-positive but PR-negative disease received no additional benefit from letrozole compared to tamoxifen. Interestingly enough, that observation is exactly opposite to that in the ATAC trial.

At this point, we have no way in clinical practice to specifically select patients, and in this state of uncertainty, an aromatase inhibitor is probably the better adjuvant endocrine therapy for postmenopausal patients with ER-positive breast cancer.

Dowsett M et al. Retrospective analysis of time to recurrence in the ATAC trial according to hormone receptor status: An hypothesis-generating study. J Clin Oncol 2005;23(30):7512-7.

Retrospective subgroup analyses reported here showed that TTR was longer for anastrozole- than tamoxifen-treated patients in both the ER-positive/PgR-positive and the ER-positive/PgR-negative subgroups of patients, but the differential benefit was greater in ER-positive/ PgR-negative tumors. These data are exploratory, should be considered hypothesis generating, and should be confirmed prospectively in other trials comparing the adjuvant use of an aromatase inhibitor with tamoxifen.

Breast Cancer Update 2006 (3)

DR KEVIN R FOX: If a premenopausal woman is treated with chemotherapy and becomes amenorrheic, it is inappropriate to assume that she will remain in a state of real menopause. Based on the natural history data, it appears to take two years to establish with some certainty that a woman will remain in a state of menopause.

If a 45-year-old woman — five years from the mean age of menopause — receives chemotherapy and becomes amenorrheic, I do not believe we can be assured that her ovaries will remain in a state of menopause until we’ve followed her for two years.

Breast Cancer Update 2006 (3)

DR FOX: The most significant challenge in developing new adjuvant strategies for premenopausal women with hormone receptor-positive breast cancers is the issue of ovarian suppression. We are participating in one of the two largest clinical trials addressing this issue: the SOFT trial, which randomly assigns premenopausal women with receptor-positive cancer to receive tamoxifen alone, ovarian suppression for five years with tamoxifen or ovarian suppression for five years with exemestane.

Breast Cancer Update 2006 (3)

DR FOX: If our patients report two years of amenorrhea following adjuvant chemotherapy and we are considering switching them to an aromatase inhibitor, we always try to corroborate that information with an estradiol and an FSH level, recognizing the occasional shortcomings of either of those measurements. In my own practice, I require that a patient have nonmeasurable levels of estrogen and an elevated FSH level in the postmenopausal range before prescribing an aromatase inhibitor.

Breast Cancer Update — Think Tank Issue 1, 2006

DR ERIC P WINER: In premenopausal women with ER-positive disease, the issue of ovarian suppression with an aromatase inhibitor is being addressed in the SOFT and TEXT trials. At least some reason exists to be concerned that this could possibly be an inferior strategy.

In a woman who has a high level of estrogen in the premenopausal state, the estrogen levels go down after she receives ovarian suppression. Then adding an aromatase inhibitor and taking a woman down to extremely low levels of estrogen may add benefit.

It’s also possible that taking those two steps down is, in fact, no better than a single step.

Of course, from a toxicity standpoint — as I think we’re learning from both TEXT and SOFT — that deep plunge into not only menopause but menopause and an aromatase inhibitor is a pretty tough maneuver for most of these patients.

So for premenopausal women, I would strongly argue against using ovarian suppression and an aromatase inhibitor as an up-front strategy outside of a clinical trial.

What about the use of an aromatase inhibitor for a woman who is premenopausal at diagnosis, stops cycling soon after diagnosis and is now on tamoxifen for two years? This situation is much more analogous to the postmenopausal woman. She has now been without premenopausal levels of estrogen for two years. It is more likely that substituting an aromatase inhibitor for tamoxifen after two years could be of additional benefit.

We don’t know that from any of the clinical trials that have been performed, but it seems more rational. However, we’ve all seen in practice — and Hal Burstein actually has a whole series of these women — patients who have been without menstrual cycles for a couple of years go off tamoxifen and start cycling again.

Breast Cancer Update — Think Tank Issue 1, 2006

DR HAROLD J BURSTEIN: The point is made that amenorrhea is menopause, but that’s not a very good definition for treating patients with aromatase inhibitors. We began to notice some patients — all of whom were women in their forties who had chemotherapy-induced amenorrhea — who were thought biochemically or on strong clinical grounds to be truly menopausal and were put on an aromatase inhibitor.

Usually, within six to 18 months they began to have menstruation again or had biochemical evidence of residual ovarian function, suggesting that they were not obtaining a therapeutic gain from an aromatase inhibitor.

Smith IE et al. Adjuvant aromatase inhibitors for early breast cancer after chemotherapy- induced amenorrhoea: Caution and suggested guidelines. J Clin Oncol 2006;24(16):2444-7.

Most women older than age 40 treated with chemotherapy will develop permanent amenorrhea. However, in a small minority, reported as 0% to 11%, ovarian suppression may be temporary, and they may renew menses over time. Our clinical observations suggest that the incidence of recovery is probably increased by the use of AIs (27% in our audit, compared with 0% to 11% spontaneously in women older than age 40).

Breast Cancer Update CME 2005

DR ROWAN T CHLEBOWSKI: The data you have to support using an LHRH agonist and an aromatase inhibitor for a younger woman at high risk with a HER2-positive tumor are a couple of Phase II trials in metastatic disease with fewer than 100 patients. So, in a certain sense, you have very limited information. Alternatively, you could administer ovarian suppression and tamoxifen.

That’s what we’ve been doing. You deviate from these protocols in various ways. For women under 40 with hormone receptor-positive disease, we’re routinely doing ovarian suppression with tamoxifen. We haven’t utilized the combination of aromatase inhibitors with ovarian suppression yet. But I can see how it may not be an unreasonable extrapolation to do so.

Breast Cancer Update CME 2005

DR G THOMAS BUDD: Off protocol, in general, I use tamoxifen for premenopausal women. Whether ovarian ablation adds — in terms of efficacy — to chemotherapy or tamoxifen, I don’t believe we know. We do know it adds toxicity.

That’s what ECOG-E3193 in node-negative breast cancer showed. It’s quite possible that it will end up adding efficacy, if we can select the right patient population, unconfounded by early menopause from chemotherapy.

Breast Cancer Update CME 2005

DR WILLIAM J GRADISHAR: The data we have available on ovarian suppression with tamoxifen or an aromatase inhibitor are still relatively limited. I would still view tamoxifen as the optimal therapy.

Select publications