Breast Cancer Update 2

You are here: Home: BCU 2 | 2008: Sir Richard Peto, FRS

Sir Richard Peto, FRS

Tracks 1-10

Track 1	Early Breast Cancer Trialists’ Collaborative Group (EBCTCG)
Track 2	EBCTCG Overview: Impact of postmastectomy radiation therapy
Track 3	Influence of ER and PR status on efficacy of adjuvant tamoxifen in the EBCTCG Overview
Track 4	Time course of recurrence for ER-positive versus ER-negative breast cancer
Track 5	ATLAS (Adjuvant Tamoxifen, Longer Against Shorter) trial

Track 6	Estimates of recurrence risk after five years of adjuvant tamoxifen
Track 7	EBCTCG: Overview of the benefits of adjuvant chemotherapy
Track 8	EBCTCG: Influence of ER status on the efficacy of adjuvant chemotherapy
Track 9	Trends toward overall reduction in breast cancer mortality
Track 10	Commentary on randomized trials as prerequisites for adjuvant therapy guidelines

Select Excerpts from the Interview

Track 3

DR LOVE: Would you talk about the data you presented related to ER and PR status and responsiveness to endocrine therapy?

DR PETO: It’s been known for some time that tamoxifen works for ER-positive disease and has little or no effect on ER-negative disease. It had been suggested that tamoxifen might be of some value for patients with ER-poor but PR-positive disease. That turns out not to be true. Tamoxifen is an antiestrogen that doesn’t do any good for ER-poor disease (Peto 2007b; [2.1]).

Conversely, it had been suggested that if a patient had ER-positive disease and didn’t have a functioning progesterone receptor, you wouldn’t gain much from treating with an antiestrogen. That, again, turns out not to be true. In patients with ER-positive and PR-poor disease, tamoxifen has a substantial effect on the long-term risk of recurrence. In fact, it’s just as effective as in patients with ER-positive, PR-positive disease (Peto 2007b; [2.1]).

If you want to know whether to use tamoxifen — or, I suspect, any other hormonal treatment — all you need to do is obtain a reliable ER measurement. Measuring PR does not provide any further guidance as to whether to use endocrine treatment.

2.1

Track 5

DR LOVE: Can you discuss the initial findings from the ATLAS trial?

DR PETO: This study compares 10 years of tamoxifen to five years in terms of the 15-year outcome. It involves a large international group with about 400 centers in 38 countries and is run by Christina Davies. They’ve randomly assigned 11,500 women who have completed five years of adjuvant tamoxifen. About half of the women had ER-positive disease, and half didn’t have their ER status tested, but most of them would have been ER-positive (Peto 2007a).

She followed these patients for an average of four years, so these are preliminary results. But it’s clear that further reduction in recurrence is achieved by continuing tamoxifen beyond five years. Continuing tamoxifen beyond five years doesn’t increase the recurrence rate of breast cancer. It decreases it further by about 15 percent. This 15 percent decrease is in addition to the decrease from the carryover effect of the first five years. In terms of preventing recurrence, 10 years of adjuvant tamoxifen is better than five years (Peto 2007a).

It’s too early to determine how 10 years of tamoxifen will affect breast cancer mortality. Also, longer treatment will increase the incidence of the significant side effect of uterine cancer. There are disadvantages in continuing tamoxifen, but certainly the myth that tamoxifen beyond five years will start stimulating the breast cancer is wrong. This was a mistake that emerged in the mid-90s.

Unfortunately, the NCI decided to issue a Clinical Alert stating that nobody should go beyond five years on tamoxifen. It was an overreaction to a small preliminary result. The data from the ATLAS trial are about 10 times as extensive as the data on which that NCI Clinical Alert was based (Fisher 2001; NCI 1995). I expect that when the ATLAS data are published, they’ll revise that Clinical Alert.

Tracks 7-8

DR LOVE: Would you discuss the data you presented from the EBCTCG Overview on adjuvant chemotherapy?

DR PETO: We don’t have trials of taxane-based regimens versus no adjuvant therapy, but we have trials of taxane-based regimens versus anthracycline-based regimens. These regimens will not be so different, but the taxanes are better. Although I must add that a fair number of trials are still not available, and the follow-up isn’t long enough. We need the 2010 cycle to include all the taxane trials and data that will be out toward five to 10 years.

From what we have at the moment, taxane-based regimens appear to involve about a 15 percent lower recurrence rate and lower breast cancer mortality rate versus anthracycline-based regimens. With anthracycline-based regimens versus CMF, about a 15 percent lower recurrence rate is evident. I don’t mean a 15 percent absolute reduction: These are proportional risk reductions. In the old trials of CMF versus no adjuvant therapy, CMF looked good for young women, but for older women the relative risk reduction was not large (Peto 2007b).

To evaluate chemotherapy, we have to put various trials together. CMF versus nothing — 1970s chemotherapy versus no treatment — provided a moderate gain. Anthracycline-based therapy versus CMF — 1980s chemotherapy versus 1970s chemotherapy — provided another moderate gain. Taxane-based regimens versus anthracycline-based regimens — 1990s chemotherapy versus 1980s chemotherapy — provided yet another moderate gain (Peto 2007b).

If you combine all of these, then you conclude that if we had been comparing taxane-based regimens to no adjuvant therapy, for older women in their fifties and sixties we’d probably be reducing breast cancer mortality by about one third and recurrence rates by about half. For the younger women, the effects are even greater. Taxane-based regimens would be reducing breast cancer mortality by about half and recurrence rates by more than half (Peto 2007b; [2.2]).

A related issue is that when you make this indirect comparison of taxane-based regimens to no chemotherapy separately for ER-poor and ER-positive disease, the proportional risk reduction is the same (Peto 2007b; [2.3]). The idea that ER determines the proportional reduction in risk that chemotherapy can produce is not true. It’s widely believed, but the evidence doesn’t support it.

If you combine the CMF trials, the anthracycline trials and the taxane trials as I’ve described and then ask, “What do we see in terms of proportional risk reduction produced by a modern taxane-based regimen?” it’s the same for ER-negative and ER-positive disease. There’s no difference (Peto 2007b; [2.3]).

2.2