DR TRACY: I saw a 79-year-old woman, very active skier, who discovered a right breast mass, which was four centimeters or greater in diameter.

Her mammogram and ultrasound were both pathologic. She had a mastectomy, which revealed a 7.5-centimeter, Grade II tumor with lymphovascular invasion. The tumor was ER/PR-positive and HER2-negative, and three of seven nodes were positive.

She had some mild hypothyroidism and absolutely no major comorbidities. She experienced some mild chronic pain as the result of a minor motor vehicle accident that caused a back injury. Her family was described as very long-lived.

Her physical exam was completely negative, with a well-healing chest wall scar. The staging studies included a bone scan, which was normal except for mild degenerative changes, and full CAT scans, because of the size of her tumor, which were negative.

DR LOVE: So, she’s a 79-year-old woman who sounds about as healthy as one can be at 79. Was she a proactive patient who wanted to do everything she could with regard to treatment?

DR TRACY: No. She was not.

DR LOVE: So she was concerned about the side effects of treatment?

DR TRACY: Absolutely.

DR LOVE: Peter, how do you approach the question of adjuvant chemotherapy in older patients?

DR RAVDIN: I think when we see an older patient, we view things a little differently. That isn’t necessarily a reflection of us being ageists but rather that we recognize, and patients themselves recognize, that they have less remaining life expectancy and, overall, a lot more competing problems. In fact, most people with node-negative disease, even if they’re fairly young, end up dying of something other than breast cancer. Sometimes looking at the competing mortality — for this patient it’s in the range of 30 to 40 percent, even though she is a skier — puts things in context, and many patients recognize that and are less enthusiastic about aggressive therapy.

The other thing to be said is that there are some uncertainties as to how effective adjuvant chemotherapy is in older patients, not because we have any deep biological rationale for why it shouldn’t work but because we have less data on that patient population.

There’s an excellent paper in JAMA by Dr Hy Muss that points out that older patients have about as much benefit as the younger patients, although they do have more toxicity (Muss 2005). So, for decisions where it’s difficult for us to know what to do, engaging the patient in the decision is reasonable because often, when they see the numbers and think about it, it actually helps them have a clearer view of what they want.

DR LOVE: How much additional survival benefit would this patient experience by receiving chemotherapy in addition to an aromatase inhibitor?

DR RAVDIN: The overview suggests that a chemotherapy regimen like CMF has little activity in postmenopausal women with ER-positive breast cancer (Early Breast Cancer Trialists’ Collaborative Group 1998), but other trials suggest adjuvant chemotherapy with anthracycline-based regimens actually has a 20 to 30 percent proportional risk reduction in this population. In terms of mortality reduction, chemotherapy — even that kind of chemotherapy — isn’t quite as good as hormonal therapy, but for a patient like this, it could mean approximately a five percent difference in mortality.

I think that it isn’t out of the question to treat healthy women in their seventies and eighties with adjuvant chemotherapy. I’d like to point out that this has been studied in other adjuvant scenarios such as colon cancer, and you can clearly see people older than 70 benefiting as much from adjuvant chemotherapy as younger patients.

DR OSBORNE: Peter, one of the problems with programs like your Adjuvant! Online model is that you can’t consider every single nuance of every single patient, so you have to generalize a bit. The program doesn’t consider PR, and I don’t think it takes into account quantitative ER, maybe because the data on those factors are still new, but I think both are becoming extremely important. Data from the SWOG-8814 study show that patients with high ER/PR-positive tumors receive no benefit from FAC (Albain 2004). Data from the Ludwig Breast Cancer Study Group similarly show that patients with highly endocrine-responsive disease receive little or no benefit from chemotherapy of any kind (Colleoni 2000), and there’s no benefit for dose-dense chemotherapy in the patients with strongly ER-positive disease.

In terms of endocrine therapy for this patient, the only prospective marker trial was conducted in the 1980s by Peter and showed that PR negativity does, at least in metastatic disease, predict for less response to tamoxifen (Ravdin 1992).

There are now three modeling studies of long-term endocrine therapy of postmenopausal patients. All three — Jack Cuzick’s, our own with statistician Sue Hilsenback, and Dana-Farber’s — have modeled whether it’s better to give an aromatase inhibitor or tamoxifen first. We can’t necessarily go by models, but I think they can tell us that jumping over to an aromatase inhibitor may not be, at 10 or 15 years, a better strategy.

DR LOVE: Aman, what are your thoughts regarding these modeling studies, and what do you think is the best long-term strategy?

DR BUZDAR: I think modeling studies are good to keep the biostatisticians busy until we generate the clinical data, but you have to go with the facts that are in front of you. The proper sequence is an interesting question, but it’s a research question, and until we have the answers, we have to use the data we have. We have data showing that in patients who are newly diagnosed, it’s better to start with an aromatase inhibitor rather than waiting two to three years. We don’t have data to show it’s better to start with tamoxifen and then switch. I think the major shortcoming of all three models is that they do not take into account the adverse effects in those first two to three years, which could be life changing for the patient.

DR LOVE: Kent, do these models bring in the toxicity issues with tamoxifen, including deep vein thrombosis, stroke and endometrial cancer?

DR OSBORNE: No. They’re not built into the models, but the mortality from those events is, so death from them would be an issue. We have data on tamoxifen followed by an aromatase inhibitor and we see a certain benefit there. We also have data comparing an aromatase inhibitor to tamoxifen in the first five years. So we do have clinical data. What we don’t have is what happens, side effect- and benefit-wise, to the patient after being treated for five years with an aromatase inhibitor and then off. We have no data for that, not even modeling data.

DR BUZDAR: We pick the numbers that we like to believe. The first publication on the ATAC trial reported a proportional reduction in hazard rate at year one, year two and year three. In year two, the reduction in the risk of events was about 39 percent. When you look at the curves, you are looking at the overall effect, and when you look at hazard rates, you look at a certain point.

DR GRADISHAR: In terms of how I would approach this 79-year-old patient, I would probably utilize the data that we have on up-front therapy — the ATAC data set and more modest follow-up in the BIG 1-98 trial — and begin the patient on anastrozole or letrozole (1.1).

DR LOVE: Do you prefer one or the other — anastrozole or letrozole — or do you consider them to be equal?

DR GRADISHAR: The long-term data at this point — the more mature data — are with anastrozole, looking at follow-up of patients from the ATAC data set.

DR LOVE: What about the issue of chemotherapy in this patient, Bill?

DR GRADISHAR: The obvious caveat is that the patient has to be engaged in the discussion. This patient has been described as the super-athlete 79-year-old with a clear mindset about what she wants to do in her life, not only in sports but probably day to day in terms of tolerating certain side effects that are going to inhibit her activities.

So I would feel comfortable discussing chemotherapy with her, but she sounds like somebody who’s going to think about the numbers carefully and realize that the added contribution of chemotherapy, despite her relatively poor prognosis, is still going to be modest, and she may elect to get endocrine therapy alone.

DR LOVE: Would you be comfortable giving this woman chemotherapy?

DR GRADISHAR: I would, based on how she was described in terms of performance status.

DR LOVE: What if she were 85 years old?

DR GRADISHAR: I think you have to look at the competing morbidities and mortalities as a patient ages. Peter’s Adjuvant! Online model takes you out to 10 years, so when we are talking about the added contribution of chemotherapy in an 85-year-old patient, we have to think about her odds of being alive at 95. Then I think it becomes a different issue.

DR LOVE: Peter, there was a paper in the JCO that evaluated the validity of the Adjuvant! Online program (Olivotto 2005). I understand that not only were the numbers related to breast cancer validated, but your nonbreast cancer mortality numbers were also validated.

DR RAVDIN: We had an interest in validating the model, which is derived largely from US data. The group in British Columbia has, perhaps, one of the larger North American databases, with very complete clinical data, so they sent us blinded information about the patients and their tumors. We then projected outcomes for those patients, sent the data back and they compared that to the actual outcomes of those patients.

Overall, it was a very tight fit. There were some groups where we were a little bit off. For instance, in younger women it appears that perhaps the classic pathologic variables don’t tell us everything that’s different about that population in that they had a somewhat worse prognosis than we had projected.

DR LOVE: What specifically would this 79-year-old woman most likely be treated with in your clinic, Peter?

DR RAVDIN: Hormonal therapy with an aromatase inhibitor. This is also a very active woman who’s probably going to be taking a few falls, and the one thing worth emphasizing is that she’d have to be told about the osteoporosis risk and actually evaluated for that.

DR LOVE: Which aromatase inhibitor?

DR RAVDIN: I’d probably treat her with anastrozole, although there are substantial data for letrozole also. However, the anastrozole data in up-front therapy is actually more robust, because it’s based on almost six years of follow-up now, so that’s encouraging.

DR LOVE: Aman, there is no survival difference at this point in the ATAC trial. How do you think the risk-benefit stacks up without survival differences, and what are your thoughts about cost?

DR BUZDAR: At this point, none of these studies show any statistically significant survival advantage. However, there is a trend in the same direction in every study: With aromatase inhibitors there are fewer breast cancer deaths, whether they are used up front, in the middle, or down the line. Why hasn’t the ATAC trial, which has the longest follow-up, shown a statistically significant effect on survival?

Because the patients in that trial are the most favorable subset of women in that the majority are node-negative, and it takes a much longer time to see a survival effect. Even the NSABP study, which consisted of patients with node-negative disease treated with tamoxifen, took more than seven or eight years to show a survival advantage (Fisher 1989). I think if the therapy is effective and it prevents recurrence, eventually it will manifest a reduction in mortality down the line.

DR LOVE: Dr Tracy, what happened with your patient?

DR TRACY: We discussed these issues, although a little less scientifically, and she did not want chemotherapy. We discussed the slight potential benefit and the side effects of the different categories of drugs. Her two sons were in favor of chemotherapy, but she chose no chemotherapy. She did receive radiation therapy to her chest wall and her draining lymphatics and has not had any sign of local recurrence.

We put her on anastrozole, and she’s tolerated it beautifully with no musculoskeletal side effects. She’s been on it for more than two and a half years and has no evidence of metastatic disease.

We did do a bone mineral density scan initially, and she had very mild osteopenia and worked with her primary care physician on that. She did not, however, choose to take a bisphosphonate. She and her family are very happy with her decisions.

Select publications