Much Ado About Small Differences
CR Magazine: Collaberation – Results


THURSDAY, DEC. 11, 2008

Much Ado About Small Differences:
Are we asking the right questions about hormonal therapy for primary breast cancer?

Posted by Musa Mayer

The first morning of the San Antonio Breast Cancer Symposium is usually reserved for the most newsworthy research presented at the conference. On Thursday, the focus was on hormonal therapies for breast cancer prevention and treatment—especially for adjuvant treatment of early disease. And for me at least, what I heard was less than impressive.

In the majority of newly diagnosed primary breast cancers, the estrogen receptor is overexpressed—meaning that the cancer cells are overproducing these receptors, which bind to estrogen. These cancers may be stimulated to grow in the presence of estrogen, so treatments that prevent estrogen from fueling tumor growth could offer an important clinical benefit.

These mostly oral drugs work in various ways. Tamoxifen, arguably the first real targeted therapy for breast cancer, selectively competes with estrogen, preventing it from entering the cell and activating the receptor. In postmenopausal women, whose ovaries are no longer producing estrogen, the three approved aromatase inhibitors—Femara (letrozole), Arimidex (anastrazole) and Aromasin (exemestane)—work by blocking an enzyme called aromatase from converting androgens (male hormones) to estrogen. Vanishingly low levels of the hormone, which are the result of aromatase inhibitor treatment, are intended to “starve” the tumor. The newer drug Faslodex (fulvestrant) works by permanently down-regulating the estrogen receptor. In younger women, the ovaries may be removed to rid the body of its most plentiful source of estrogen. Or more commonly, a drug known as an LHRH agonist, like Lupron (leuprolide) or Zoladex (goserelin), may be used to shut down ovarian function. A few other, older off-patent approaches tend to have more side effects and have fallen out of favor.

James Ingle of the Mayo Clinic, in Rochester, Minn., presented a meta-analysis of the results of multiple adjuvant trials in postmenopausal, hormone-sensitive breast cancer that have compared two of the three aromatase inhibitors with tamoxifen, either as monotherapy (use of a single drug for five years) or in the setting of switching from one class of hormonal drug to another—from tamoxifen to an aromatase inhibitor or vice versa. A meta-analysis was needed because so many similar studies have been done, in tens of thousands of patients. In the hopes of cornering a sizable market, each company that makes one of the three aromatase inhibitors—all still under patent protection—has invested in these large clinical trials to compare its drug with the previous standard of care, tamoxifen, a drug that has been off-patent for a number of years.

One of the studies in the meta-analysis, BIG 1-98, was also presented during this session. This randomized European trial of more than 8,000 patients compared four treatment approaches:

  • five years of tamoxifen
  • five years of letrozole
  • two years of tamoxifen followed by three years of letrozole
  • two years of letrozole followed by three years of tamoxifen

After nearly six years of follow-up, this study suggests letrozole (Femara) is slightly better than tamoxifen in reducing recurrence, closely replicating similar findings related to the other two aromatase inhibitors. The headline in the Novartis press release trumpets Femara as the “first aromatase inhibitor to indicate survival benefit versus tamoxifen when taken for five years after breast cancer surgery.” Only in the small print do we read that the survival benefit was actually not statistically significant on the basis of the standard, accepted “intention to treat” analysis that includes all patients as randomized. But because nearly 25 percent of patients in the control arm crossed over (in other words, they later took Femara), the investigators justified performing a non-standard “censored” analysis instead. I wonder what the statisticians will have to say about this?

Completing the third piece of the triumvirate of studies on aromatase inhibitors, Stephen Jones of US Oncology presented the TEAM trial, of 10,000 post-menopausal women, which tested Aromasin versus tamoxifen and in sequence with it. Unsurprisingly, this trial came up with similar findings, meaning that all three aromatase inhibitors show a slight benefit in reducing recurrences, especially local recurrence that is not life-threatening.

One caveat, mentioned by Jones as an aside in his report, is that patient compliance with adjuvant hormonal treatment was poor in his study: 29 percent of patients on tamoxifen and 19 percent of patients on an aromatase inhibitor failed to complete their treatment. This is far from a unique finding, but as far as I know, no one is asking about the impact on trial results of poor compliance.

But what’s particularly striking to me is that not a single one of all of these trials, with their enormous costs, and the participation of thousands of patients, has explored a head-to-head comparison between these three aromatase inhibitors.

So at the end of the day, what more do we really know from all this research? As an advocate, I have to wonder: Did these studies really ask the most important research questions?