DR GRADISHAR: ATLAS probably got more traction with patients than anything else that came out at San Antonio. And I’m confident that we were all inundated with questions about whether that data pertains to me as a patient, whether you’re node-positive, negative, DCIS or whatever, as I suggested earlier.
The ATLAS trial has been long awaited. It’s a trial that really was meant to address the issue of longer durations of tamoxifen compared to 5 years of therapy, so 10 versus 5. We had some data from smaller trials in the past and, in fact, the NSABP-B-14 trial really, in a sense, set practice for many, many years, suggesting that longer durations of therapy did not result in a better outcome. In fact, it sort of veered towards a worse outcome. Most people adopted 5 years of tamoxifen as the standard.
Then, of course, the whole era of AIs came into play and investigation over the last decade. This was sort of an old story that sat in the background for some time. Now, with pretty good follow-up, thousands of patients that were enrolled in this study, we have data that would suggest that 10 years is better than 5 years. And there’s some nuances to it that are interesting, too, and that is that the impact of that additional therapy is seen or observed most in the second decade. We have this notion of a carryover effect that we’ve been fully aware of for a long, long time about once you stop tamoxifen after 5 years, the lines don’t come together. There’s that carryover effect.
If you continue tamoxifen beyond 5 years, where you’re seeing the impact is not necessarily in years 6 through 10 or 5 through 9, however you want to phrase it, but rather in the second decade. I think it’s worth revisiting this issue in certain subsets of patients where the use of tamoxifen for longer durations may be applicable.
I think where that would occur is in patients who are still premenopausal, where you deem them at high risk, which is, of course — now we’re going to discuss what “high risk” means, but I would submit that it’s probably patients with multiple positive lymph nodes, where an AI short of rendering them postmenopausal wouldn’t be viable. You could consider tamoxifen.
Certainly in women who if you made the switch to an AI and they did not tolerate an AI, you would have some argument and basis for either continuing tamoxifen for a longer duration of time. And then there’s some theoretical reasons where you could add tamoxifen after an AI is complete, but there’s no data to support that yet.
DR LOVE: Any comments in terms of complications, endometrial cancer, second breast cancers, et cetera?
DR GRADISHAR: They did take into account in the analysis that there are finite risks associated with tamoxifen, even at 5 years. As you go beyond that, you start to accumulate an increased number of such events. But when you take into account the improvement in breast cancer-related mortality, it more than offsets the increase in those other side effects. There’s a net plus for the use of longer durations of tamoxifen.
DR LOVE: How did seeing these data affect your practice, if at all, Edith?
DR PEREZ: Actually very little because at the same time they presented the data in San Antonio, the manuscript came out with the toxicities in a separate forum. It was very interesting that the toxicities were not highlighted as much in the presentation at San Antonio. And remembering the NCIC CTG MA.17 trial where patients received 5 years of tamoxifen followed by 5 years of AI has driven our practice. In the postmenopausal setting, it’s pretty easy. I would not use tamoxifen for 10 years. I would use AI first and continue the AI, actually, for 10 years, depending on risk, although we’re waiting for some definite data.
The issue is in the premenopausal setting. And the challenge that we have is that there are no good biochemical tests that we can do in a premenopausal woman who’s on tamoxifen to determine whether she’s still premenopausal or not. But after 5 years, that’s an issue that would need to be considered. And the question is: Should that patient then have an ovarian ablation instead of going to an agent that may have significant cumulative toxicities?
And last but not least important is, there was a subset analysis of MA.17 that showed that the patients who benefited the most from the switch of tamoxifen to 5 years of AI were the patients who started being premenopausal, where they were enrolled in MA.17.
DR LOVE: I translate that to: You don’t use more than 5 years of tamoxifen?
DR PEREZ: No.
DR LOVE: Adam?
DR BRUFSKY: The way we’ve decided to do this is that, again, the major toxicity that’s different between 5 and 10 is endometrial cancer. It’s doubled. It’s 1.6% to 3.1%. And most of it is in the postmenopausal women. And most of those postmenopausal women after 5 years of tamoxifen will likely be changed to an AI anyway, based on MA.17. What we’ve decided to do is use 10 years of tamoxifen in the premenopausal women, not so common in women who still remain premenopausal after 5 years after therapy. We treat younger and younger women with breast cancer. We’re having women who are like that, but that’s generally what I’m applying this to.
Now, again, ovarian ablation has toxicity, as well.
DR PEREZ: Correct.
DR BRUFSKY: We complain about the toxicity of tamoxifen, but if you put someone into ovarian failure too early, that has problems too. You’ve got to really balance the 2 risks.
DR LOVE: Monica, what are your thoughts about this? I also noticed that the second breast cancer rate was slightly lower in the people who continued tamoxifen. How are you and the people at Memorial thinking this thing through?
DR MORROW: I think based on what we already knew about tamoxifen and second breast cancers, it’s not entirely surprising that you would continue to see that benefit of risk reduction. And I think that what we’re doing, actually, complies with what a lot of people have said — namely, that for postmenopausal women, they’re getting AIs, and it’s not really very much of an issue, except in a small subset who don’t tolerate AIs. But for premenopausal women, especially higher-risk premenopausal women, this offers a benefit. And it’s being offered to patients on an individual basis.
I think if you look at the patient who has a T1A node-negative breast cancer, that nobody’s really flogging them to take tamoxifen for 10 years because their risk of relapse is so small to start with.