SABCS 2008: AIs "Somewhat Better" Than Tamoxifen as Adjuvant Therapy for ER-Positive Breast Cancer

Zosia Chustecka

December 17, 2008

December 17, 2008 (San Antonio, Texas) — More evidence suggesting that aromatase inhibitors (AIs) have better efficacy than tamoxifen as adjuvant therapy for postmenopausal women with estrogen-receptor-positive breast cancer was presented here at the 31st Annual San Antonio Breast Cancer Symposium (SABCS). However, although the survival data are very similar, the toxicity profiles are quite distinct, and there is an enormous difference in the cost of the 2 approaches, experts pointed out.

"Tamoxifen is a very good drug, it just appears that the aromatase inhibitors are somewhat better," commented James Ingle, MD, director of the Breast Cancer Specialized Program of Research Excellence at the Mayo Clinic Cancer Center, in Rochester, Minnesota. He was presenting data from a new meta-analysis involving nearly 20,000 patients on behalf of the Aromatase Inhibitors Overview Group.

Both drug approaches reduce estrogen, but in different ways. Tamoxifen is a selective estrogen-receptor modulator that prevents estrogen from interacting with its receptor, whereas AIs act by inhibiting an enzyme involved in the synthesis of estrogen, and hence reduce levels of the hormone. Three AIs are currently marketed — anastrozole (Arimidex, AstraZeneca), exemestane (Aromasin, Pfizer), and letrozole (Femara, Novartis).

Both approaches have been shown to significantly reduce the rate of relapse in women with estrogen-receptor-positive breast cancer when used as adjuvant therapy, but evidence suggesting that AIs may be more effective than tamoxifen is mounting, as previously reported by Medscape Oncology.

"The aromatase inhibitors are usually a few percent better, especially for preventing relapse," when you consider all the data, Hyman Muss, MD, from the Vermont Cancer Center, in Burlington, who cochaired the session, commented to Medscape Oncology. This includes data both from head-on comparator trials and studies in which patients were switched over from one drug to another, he added.

"But when you look at the survival data, even in this huge meta-analysis, then the difference is very small," Dr. Muss pointed out. There are some concerns about the interpretation of the mortality data, he said, adding: "We don't know enough yet about mortality with the aromatase inhibitors."

Significant Reduction in Relapse of Breast Cancer

The meta-analysis that Dr. Ingle presented at the meeting was split into 2 parts:

  • Cohort 1 consisted of data from monotherapy clinical trials that directly compared an AI to tamoxifen (9586 patients with a median follow-up of 5.9 years).

  • Cohort 2 consisted of data from "switching" trials in which the patients on tamoxifen were switched to an AI after 2 or 3 years (9015 patients with a median follow-up of 3.9 years).

AIs produced significantly lower recurrence rates than tamoxifen in both cohorts, Dr. Ingle reported. The proportional reduction was 23% in cohort 1 and 29% in cohort 2. The absolute gains were 2.9% at 5 years and 3.9% at 8 years in cohort 1, and 3.1% at 3 years and 3.5% at 6 years in cohort 2.

In cohort 1, the proportional reductions were similar in the first 2 years and the last 3 years of treatment; there were further reductions beyond 5 years, but these were not significant, Dr. Ingle noted. In cohort 2, there were significantly greater reductions during the first 3 years of treatment than there were beyond the 3 years, he added.

"These findings raise the issue of value for longer AI therapy," he commented

There was no significant difference in breast cancer mortality between the 2 therapies in cohort 1 (there was a 1.1% benefit for AIs at 5 years, and 0.5% at 8 years). There was a significant reduction in breast cancer mortality for the AIs in cohort 2 (0.7% at 3 years and 1.6% at 6 years from treatment divergence), but as Dr. Ingle and several physicians in the audience pointed out, the gains were very small.

The fact that nonbreast cancer mortality and overall mortality were not higher with the AIs in either cohort is reassuring regarding their overall safety.

Dr. Ingle said the fact that nonbreast cancer mortality and overall mortality were not higher with the AIs in either cohort is "reassuring regarding their overall safety."

He also said that any efficacy gain had to be balanced against tolerability in each individual patient.

During the discussion period, a British doctor questioned the cost-effectiveness of using AIs instead of tamoxifen in light of the very small gain in mortality that was seen. Tamoxifen is now available as a generic, so is much cheaper than the branded AIs. In the United States, it's the difference between $5 and $300, Dr. Muss pointed out to Medscape Oncology.

Concerns Over Survival Data

In an interview, Dr. Muss elaborated on some of the concerns surrounding the mortality data, which he said were "not clean."

First, there is CYP2D6, Dr. Muss said, the enzyme that metabolizes tamoxifen to its active form. Between 5% and 10% of white women are deficient in this enzyme and are poor metabolizers of tamoxifen, and there is research to suggest that the drug is not effective in this patient population, as reported by Medscape Oncology.

So, in all the clinical trials, it could be that between 5% and 10% of women taking tamoxifen were not metabolizing it and the drug was ineffective, Dr. Muss said. Conversely, some of the women who were high metabolizers and who were responding to tamoxifen might have come off the drug early because of adverse effects. Both of these effects could introduce bias into the results, he pointed out.

There is the question of patient compliance with the 2 different approaches, Dr. Muss continued. He pointed at that in another presentation at the same session, results from the Tamoxifen Exemestane Adjuvant Multinational (TEAM) study reported noncompliance in 29% of women taking tamoxifen and in 19% of women taking AIs. This trial was just being reported, so data from it were not included in the meta-analysis, but there may be a similar issue with compliance in the other trials. "When you are looking at a difference of 10% between compliance with these drugs, and a difference of only 1% in survival, that may be due to chance," he commented. An intention-to-treat analysis would not take into account the patients who dropped out, he said.

AIs May Be a Better Choice

Overall, Dr. Muss said that he feels that AIs "may be a better choice," than tamoxifen, particularly for older patients, but he would be influenced more by toxicity than by efficacy.

Tamoxifen has 2 major adverse effects — blood clots and uterine cancer, Dr. Muss said. Thromboembolism is a major concern, because most women with breast cancer are older and have comorbidities, such as hypertension and vascular problems. Neither of these adverse effects are seen with AIs, although these drugs have their own problems, he commented. In particular, AIs are associated with aches and pains in the joints, which can be severe and pose major problems. "Some women say they are unable to walk," he noted, "and there is not much that can be done, [other than] maybe change to another aromatase inhibitor."

AIs also carry the risk for osteoporosis, but this can be monitored and countered with bisphosphonates, he pointed out. "Oncologists have become very good bone doctors . . . since these drugs have been around," Dr. Muss commented. Also, there is now some research suggesting that these drugs also have an effect on the breast cancer itself, he noted. (New data suggesting the bisphosphonate zoledronic acid has a direct anti-tumor effect was presented at the SABCS meeting, as recently reported by Medscape Oncology.)

AIs should be a part of the treatment for every postmenopausal woman, Dr. Muss said, as laid out in the American Society of Clinical Oncology guidelines. "But from where we sit today, it is not, to me, totally convincing that you have to start every patient on an aromatase inhibitor as initial endocrine adjuvant therapy."

"However, I would say that at some stage in the course of treatment, maybe 2 years later, and there are data for 5 years later, an aromatase inhibitor should be used as part of the patient's treatment," he said. The switching data show that there may be some survival advantage, he added.

An aromatase inhibitor should be used as part of the patient's treatment.

A major question than remains, however, is how long such treatment should be continued, Dr. Muss said. "The overview analysis shows that most women who are treated with tamoxifen relapse after 5 years, so it may be that you have to continue women on endocrine therapy for many, many years to do the best you can." These studies are currently in progress, comparing 5 years of therapy with 10, and 10 years of therapy with 15, he noted.

Dr. Ingle has disclosed no relevant financial relationships. Dr. Muss reported acting as a consultant for Pfizer, Genentech, Amgen, Roche, Bristol-Myers Squibb, and Tragera; and receiving research funding from AstraZeneca, Pfizer, GlaxoSmithKline, Genentec,h and Novartis.

31st Annual San Antonio Breast Cancer Symposium (SABCS): Abstract 12. Presented December 11, 2008.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: