The following questions were posed to the attendees of the 2001 Miami Breast Cancer Conference. The percent of attendees responding is shown along with edited comments from two faculty members, Dr Monica Morrow and Dr Nancy Davidson.

Section 1: Breast Cancer Clinical Trials Panel

1. About how many breast cancer patients (or high-risk women) have you entered on clinical trials in the last year?

If you look at these responses, though, the actual number of women whom physicians have put on trials is modest. Only six percent put more than 20 women on studies this past year, which reflects what is seen nationally — five percent or fewer of breast cancer patients enter clinical trials. Hopefully, the large number of physicians at this meeting who showed an interest in trials means that the total proportion of patients participating may go up.

2. In general, what would be your suggested nonprotocol approach to the axilla in case #1?

Large national patterns of care studies would be necessary to see if this is what is actually taking place across the country. It would take several years to collect and analyze the data, but we really don’t have an accurate idea of what’s going on right now with this procedure in practice. Probably most women are being offered sentinel node biopsy off protocol, in spite of the fact that we really don’t have good long-term follow-up on the outcome of the procedure.

3. How would you feel about offering this patient participation in the NSABP B-32 sentinel node trial? (randomizes to axillary dissection or no further surgery in sentinel node-negative patients)

Monica Morrow, MD

One issue with this study is, do you need a randomized trial if you already have data on many thousands of cases — as we do — documenting the false-negative rate of sentinel node biopsy? The NSABP already did trial B-04 that showed no difference in survival outcome between axillary dissection at the time of diagnosis and delayed axillary dissection if clinically positive nodes developed. So, if that trial didn’t show survival difference, is it reasonable to expect that NSABP B-32 would? I think that’s a very open question. I don’t believe that for all the time, effort and trouble that people are going to go through to randomize patients into this trial, that it’s going to provide a lot of new answers or different information than we have already.

Having said that, B-32 will tell us about the clinical false-negative rate when lots of surgeons do sentinel node biopsy, which is an important issue to inform patients about. That is, how many actual axillary failures are you going to see in the sentinel node-only arm versus the number of axillary failures that you’ll see in the dissection arm? For some patients, knowing their risk of coming back for surgery for recurrence is very, very important.

4. How would you feel about offering this patient participation in the American College of Surgeons (ACOS) sentinel node trial? (randomizes to axillary dissection or no further surgery in sentinel node-positive patients)

Monica Morrow, MD

The audience was a lot less enthusiastic about this trial, probably because it addresses a much more controversial question. It’s relatively easy to say, "Yes, I would enthusiastically randomize someone into a trial that I think I know the answer to," particularly when there’s not too much difference between the arms. However, this study addresses the critically important issue of whether there is a reason to remove axillary nodes after the patient has been staged as node-positive. It challenges the dogma that people have had for years. I would be willing to bet that the same people who told you for the last ten years that axillary dissection is a staging procedure are the same people who are uncomfortable with this trial — which intellectually doesn’t make a lot of sense.

I would be comfortable putting patients on either of these studies, but I actually am somewhat more comfortable with the ACOS study. I don’t think there’s a shred of data that says that removing normal nodes contributes to the cure of breast cancer.

5. The lesion in case #1 is excised and found to be 0.8 cm (ER+PR+). An axillary dissection is negative. Generally, what systemic therapy would you recommend in this situation?

I would certainly agree that this is an appropriate situation to discuss tamoxifen, but in this setting I would consider it to be optional. The primary reason you’re giving tamoxifen here is for prevention and for a very minimal reduction in risk of cancer recurrence. This wouldn’t be the case, for example, with a 1.5 centimeter tumor where I more strongly recommend tamoxifen. With this smaller tumor, I would say, "You have an excellent prognosis. You can make it even a little better, and you’ll have other benefits in terms of contralateral breast cancer risk reduction and potentially maintenance of bone density from using tamoxifen, and you should consider it."

Nancy Davidson, MD

In practice, most women in this situation would receive tamoxifen, and I would be comfortable with tamoxifen or with no systemic therapy. It would depend on what the patient perceives as important, and also her general health. The one thing that gives me pause is that her risk of systemic recurrence is reasonably low, and she’s in the age group where the side effects of tamoxifen — including clots and uterine cancer — are slightly higher. But I think I would be voting with the 95 percent of the audience who felt that tamoxifen or no therapy at all were reasonable choices for her. I would bring up chemotherapy if she were 40 rather than 67, but it would be hard for me to be very enthusiastic about it in either case. I would let these patients know about the NSABP trial 23 and the small margin of benefit that would be associated with chemotherapy.

6. How would you feel about offering this patient participation in the NSABP B-34 study (adjuvant clodronate vs placebo)?

Monica Morrow, MD

This is a very interesting research question, well worth studying. There is no reason that this patient — an otherwise healthy woman — shouldn’t be offered participation in a well-designed, prospective randomized trial.

One of the interesting things that we’ve seen is that many older women don’t even get asked to participate in trials, and that’s part of the reason we don’t have great data on older women. It may be presumed that they won’t want to enter the study, or they may have visual difficulties which complicate reading these incredibly long consent forms, or they may make up their mind a little more slowly about things. All of these factors discourage physicians from inviting them to participate.

But, it would be really useful to get information from a trial like this to define the optimal treatment for older women. Quality-of-life issues and competing causes of mortality are much more important in this population. So, you might be not willing to trade off a one or two percent improvement in survival associated with substantial toxicity.

Nancy Davidson, MD

I chair the ECOG breast committee, and we would very much like to have an adjuvant bisphosphonate trial. This NSABP study is asking a very practical clinical question, and the clinical trials that we have available give conflicting results.

There are three randomized trials using clodronate in the adjuvant setting. One is the German trial, which was pretty positive. The second was a Finnish trial and was pretty negative — in fact, it suggested that there was a detriment to the use of clodronate. The third trial is from the U.K. and hasn’t been written up formally, but it suggests that there are fewer bone metastases.

So, I’m kind of neutral right now. Bisphosphonates have anti-osteoporosis qualities that would be beneficial, but the anti-breast cancer benefits are uncertain.

7. If clodronate were available in the U.S., would you be inclined to use it as adjuvant therapy outside a protocol setting?

8. How would you feel about offering this patient participation in the proposed NSABP adjuvant trial of node-negative, ER-positive patients — Faslodex® (fulvestrant) vs tamoxifen?

Nancy Davidson, MD

I’m glad to see that the audience was largely favorable, because I would be, too. It seems to me that the only difficulty in this particular trial is the monthly IM injection versus a pill. Sometimes randomized trials are tougher if the modes of treatment administration are a little bit different. But there’s a lot of excitement about Faslodex as a drug and a lot of information to lead us to think that this is a very rational comparison. So, I’d be very supportive.

The one thing that gave me pause when I looked at this proposed trial is that we don’t have a lot of lengthy follow-up information on Faslodex. That reflects our wish to develop things quickly, and also, perhaps a little more sophisticated drug development, because this agent has been extensively tested pre-clinically, and we have a fair idea of what we might expect in terms of the clinical side effects.

9. What would your usual pharmacologic management be of case #2?

Monica Morrow, MD

Generally, I use the five-year Gail risk estimate, because it’s been validated. With lifetime risk, you get into competing causes of morbidity. I inform the patient about her risk level, tell her that tamoxifen has been shown in clinical trials to reduce that risk and, in the case of atypical hyperplasia, the risk reduction may be as great as 86 percent, which is quite substantial. I talk in real terms about what that means, and many patients will opt to take tamoxifen. Some feel that a five-year risk of only 2.3 percent is not overwhelmingly alarming in that they still have a 97 plus percent chance of not developing breast cancer. These women may opt not to be treated. I do not offer raloxifene for breast cancer risk reduction off protocol.

10. Which of the two following trials would you be comfortable with case #2 entering?

Monica Morrow, MD

It’s bit surprising that so many people would be interested in putting patients on both studies. I would have a good deal of trouble with the placebo arm in IBIS 2. I think the STAR trial is a great study — it’s asking an important question to define agents for prevention, and we enter women on that study.

Unlike our colleagues in Great Britain, I believe that the data from the NSABP P-1 trial quite clearly demonstrates that tamoxifen is effective at reducing risk in high-risk women. We have that same data from the overview analysis with 15 years of follow-up in terms of contralateral breast cancer. We have that same data from the NSABP DCIS trial. I don’t know how many times we need to see it.

I think there are very good explanations for why the Italian tamoxifen prevention study and the Royal Marsden study didn’t show benefits. Therefore, I would have a great deal of difficulty asking a high-risk woman to enter a study where she might receive a placebo, when there is an agent that’s already of proven benefit.

If IBIS 1 shows an advantage to tamoxifen, the investigators will drop the placebo arm, which would leave Arimidex® (anastrozole) versus tamoxifen. I think that is a very reasonable trial and a good question to ask. The fact that they’re considering this approach affirms the data from the advanced disease setting on the safety and efficacy of Arimidex.
I would also definitely be supportive of a trial comparing an aromatase inhibitor to tamoxifen in postmenopausal women with DCIS, like the trial that the NSABP is considering. We know how effective tamoxifen is, and we have a very good idea of the side effects profile. If we can use what we’ve learned from that experience to find better agents that either are more effective or less toxic, it would be an advance.

The standard therapy — if you will — in this country would be tamoxifen, but many of these folks don’t take anything, and that’s okay, too. It depends on the patient’s perception of her risk-benefit ratio. But I would be supportive of either STAR — where we think we’re going to have preserved efficacy and hopefully less toxicity — or IBIS 2, if they remove the placebo arm — where I think we’re all hoping that maybe Arimidex will turn out to be better than tamoxifen.

Aromatase inhibitors have been moved into a high-risk setting before there’s data available in the adjuvant setting, and this is pretty common now — this kind of leap-frogging based on what you hope are going to be positive results. Sometimes that’s very successful, and sometimes it’s a failure. A clear example of that would be the NSABP B-22 and B-25 trials of adjuvant chemotherapy, where if they had known the results about high dose in the B-22 trial, I’m not sure they would have designed the even higher dose trial in quite the same way. So, sometimes a good scientific hypothesis doesn’t pan out. On the other hand, you can spend an awful lot of time waiting, if you wait for positive results before you move on.

As far as toxicities with aromatase inhibitors, we don’t have long-term monitoring on bones and lipids, particularly in large populations of women.

Monica Morrow, MD

I agree that the standard duration of tamoxifen therapy is five years. Having said that, the supportive data comes from node-negative patients, where the risk is certainly lower than in a patient with three positive nodes. But, since we know that there is an issue of tamoxifen-stimulated tumor growth related to duration of exposure, outside of a trial, stopping at five years is appropriate.

Nancy Davidson, MD

I would support the majority who said that their standard approach would be to stop the therapy. While there’s uncertainty about the duration of tamoxifen administration in node-positive patients, we don’t have reliable information right now to suggest that continuing is a good idea. So, until the results come from those other randomized trials, I would agree with the NIH consensus conference that stated that five years of tamoxifen would be regarded as standard. I can’t imagine giving raloxifene or an aromatase inhibitor to this patient right now outside of a trial.

12. Which of the two major trial types would you be comfortable with the patient in case #3 entering?

Nancy Davidson, MD

I’m surprised that this audience is as supportive of clinical trials as they say they are. That’s great to see. The concept behind the ATLAS/ATTOM and aromatase inhibitor trials is rational for this woman who is node-positive. I’d have a little more reservation about ATLAS or ATTOM in a patient who is truly node-negative — given the trial information we have available — but not enough to keep someone off the trial. These kinds of trials are very appealing, as the patients have already completed their standard therapy and are interested in whether there is something else that they might be able to do in addition. This offers a pretty painless way of trying to do that.

Monica Morrow, MD

Both of these are important research questions, because high-risk women, like this case — even with chemotherapy and five years of tamoxifen — still have a substantial risk of relapse. If you could lower that either by following witharomatase inhibitor or continuing tamoxifen, it would be an important step forward. In terms of continuing tamoxifen, historically, we thought we knew that this agent was cytostatic and that the benefits would go away when the drug was stopped. This clearly is not the case, so there is plenty of reason to ask this research question.

It’s interesting that if you look back over all of the audience responses to these clinical trial questions, the overwhelming majority of the physicians state that they would offer patients clinical trials, yet not that many are actually entering patients. What that says is that theoretically physicians would offer the trials, but the practical impediments of entering patients may be excessive. The numbers don’t really match. It suggests that there is a practical problem, not a theoretical or randomization-ethics problem.

Clearly some physicians are uncomfortable saying, "We don’t know the best treatment and if you enter this trial, you will be randomized." For most of the studies presented in this panel, it’s fairly easy to get beyond that by saying, "We’re offering you the standard of care or something that may be better or less toxic or, in some way, represents an advance." This type of trial presents very little problem with the ethical issues of randomization.

My impression is that if you look back over recent clinical trials comparing a known therapy to a newer one, while the trials may show equivalence or non-statistical benefit, I can’t recall very many that have shown that the new agent being tested is significantly worse than the standard therapy. These Phase 3 national clinical trials are so expensive to mount that we have to have a great deal of supporting data and evidence to proceed, and there is a very high likelihood of benefit. So, the patient is protected in that way, and that’s something that many patients don’t realize. This relates to this idea of "I don’t want to be experimented upon. I want the best therapy that you already know about."