Editor ’s Note

LESSONS FROM CLINICAL TRIALS ON LONG - TERM ADJUVANT ENDOCRINE THERAPY

Kathy Pritchard — a frequent guest on this series — often raises issues in breast cancer research that no one else seems to consider. It has been fascinating to observe the evolution of her thinking over the years in interpreting trial results from a practical clinical perspective. The last time we spoke, emerging data on third generation aromatase inhibitors suggested greater antitumor efficacy compared to tamoxifen. I was in the audience at the 1999 ECCO meeting in Vienna when the first data emerged for anastrozole,and Kathy was on her feet immediately, carefully querying the researchers — who were overturning the classic breast cancer paradigm that all endocrine therapies have the same efficacy.

Kathy prides herself on a conservative approach to applying research data to patient care,and one year later,when most US oncologists were routinely utilizing aromatase inhibitors as first-line therapy for postmenopausal women with metastases,she was still “thinking about it.” For our most recent interview — featured in this issue of Breast Cancer Update — Kathy indicated that she now is using AIs commonly as first-line therapy, but she also raised a fascinating question related to an important issue being addressed in a number of current randomized clinical trials.

While the aromatase inhibitors are rapidly moving into the first-line treatment of metastatic disease,many of us are also involved in adjuvant studies sequencing an aromatase inhibitor after two to five years of tamoxifen. However, the ATAC adjuvant trial is close to maturity and compares up-front Arimidex to tamoxifen to the combination. The large studies of post-tamoxifen AIs may provide interesting information that we may not use because we don’t give the drugs that way anymore.I question whether these studies of post-tamoxifen AIs are going to be relevant,because it may be that the trials using aromatase inhibitors up-front as adjuvant therapy will be the ones we ’re going to be interested in.

—Kathleen Pritchard, MD

Interviewing key breast cancer research figures for this series has been an adventure, and Kathy’s provocative but very salient comment reminded me of a series of interesting related concepts investigators have discussed over the years on the strategy of adjuvant endocrine therapy.

What has made this paradigm particularly interesting is the huge body of laboratory research that clinical investigators have turned to in seeking clues for trial strategies. At the center of this has frequently been Craig Jordan, whose animal studies in the ’70s and ’80s demonstrated the “cytostatic” effect of tamoxifen and led to studies of long-term adjuvant therapy in women with breast cancer.

A number of randomized trials and the International Overview confirmed that increasing the duration of tamoxifen up to five years was beneficial, but data on greater than five years suggested that — if anything — there was a deleterious effect of extending treatment.

As a result, in 1995, the NCI sent out an alert to physicians recommending that outside of a clinical trial setting, adjuvant tamoxifen should be given for five years.

National Cancer Institute clinical announcement:Adjuvant therapy of breast cancer — tamoxifen update. Bethesda (MD):National Institutes of Health;1995. Full-Text

The key NSABP trial was protocol B-14 in patients with ER+,node- negative tumors.A recent paper updated this experience.

“Through seven years after reassignment of tamoxifen-treated patients to either placebo or continued tamoxifen therapy,a slight advantage was observed in patients who discontinued tamoxifen relative to those who continued to receive it:DFS =82% versus 78% (P =.03), RFS =94% versus 92% (P =.13), and survival =94% versus 91% (P =.07), respectively.The lack of benefit from additional tamoxifen therapy was independent of age or other characteristics.”

Fisher B et al.Five versus more than five years of tamoxifen for lymph node-negative breast cancer:Updated findings from the National Surgical Adjuvant Breast and Bowel Project B-14 randomized trial. J Natl Cancer Inst 2001;93(9):684-90.Abstract

Interestingly, at about the same time that clinical trials were demonstrating that extending tamoxifen beyond five years was not beneficial, Craig Jordan was finding that cell lines exposed to tamoxifen for prolonged periods eventually became dependent on tamoxifen for growth. Subsequent research revealed an even more fascinating finding — namely, that tamoxifen-dependent cells were actually destroyed by exposure to physiologic levels of estrogen.

We took tamoxifen-stimulated tumors and retransplanted them into animals for five years.Without treatment, the tumors grew sluggishly. We then exposed these animals to either tamoxifen or physiological levels of estrogen as you would see in a normal woman. Tamoxifen stimulated the growth of the cells, but physiological levels of estrogen caused the tumors to melt away. I’ve never seen anything like this! This could explain how long-term tamoxifen works as an adjuvant therapy. Upon stopping after about five years, the woman’s own estrogen now comes back and seeks out all of these supersensitized micrometastatic breast cancer cells and initiates an apoptotic mechanism.

The woman’s own estrogen consolidates the beneficial effects of tamoxifen by causing a second antitumor action for months after the tamoxifen has stopped. So we have developed a five-year cycle within our model system where estrogen seems to attack the supersensitized tumors that have been exposed to five years of tamoxifen, creating apoptosis in breast cancer cells.

Another interesting laboratory observation is that if you continue treating these animals with estrogen after five years of tamoxifen, some of the tumors recur and start to grow again. If you then take these now estrogen-stimulated tumors and transplant them into a new generation, the tamoxifen now works as an antiestrogen again.

—Craig Jordan, PhD DSc

This increasingly accepted practice of limiting adjuvant tamoxifen therapy to five years was reinforced in the 2000 NIH Consensus Conference,but during that same meeting,I interviewed Sir Richard Peto, who told me that the existing data on tamoxifen duration was inadequate to determine whether extending treatment beyond five years would provide additional benefit.

We have really good evidence that five years of tamoxifen is better than two years of tamoxifen. It’s not a big difference, but it’s real. We don’t have comparable data in terms of whether ten years is better than five years.A few thousand women have been randomized, but most of them have been randomized so recently that we have minimal follow-up. In terms of recurrences, we’ve got a few hundred recurrences, whereas in the trials of five years versus two years, we’ve got a few thousand recurrences. So, we don’t have reliable information on whether ten years is better than five years. And even worse, the recurrences that have been observed are nearly all in the first few years after randomization, where the carry-over benefit from that first five years of tamoxifen is still going to be providing a lot of protection in the control group.

Breast cancer is a disease with at least a 20-year natural history, and the question is, “Would 10 years of tamoxifen be better than five years of tamoxifen in providing protection against recurrence in the second decade after diagnosis?” For a woman in her 50s or 60s, 20-year outcome is very relevant. There are almost as many breast cancer deaths in the second decade after diagnosis as in the first decade after diagnosis, so we’ve really got to think on a long-term scale. When you do that,then the idea that ten years might be better than five years actually becomes quite interesting.

—Richard Peto, FRS

Peto and his colleague, Christina Davies, are coordinating the massive Adjuvant Tamoxifen Longer Against Shorter (ATLAS) trial to attempt to address this question by recruiting 20,000 patients internationally, who are willing to be randomized to either continue tamoxifen or stop five or more years after initiating therapy. A similar study is the Adjuvant Tamoxifen Treatment, Offer More? (aTTom) trial, in which patients also are randomly assigned either to discontinue the drug or to receive tamoxifen for at least five more years. It is not likely that the major findings from these trials will be available for about 10 years. However, when one considers Kathy Pritchard’s comment, this question of tamoxifen duration may in the long run become a moot point if trials like ATAC demonstrate that the initial adjuvant therapy should include an aromatase inhibitor.

The ebb and flow of breast cancer clinical research is very evident in the issue of the optimal strategy of long-term endocrine therapy, but there are other important lessons here that relate to all of cancer research.

One key factor is the value of the translational model. Craig Jordan and many other basic researchers have worked closely with the cooperative breast cancer clinical trials groups, participating in a two-way partnership that has resulted in randomized trials that have tested key basic paradigms and provided tissue and serum samples leading to new investigation in the laboratory.

Endocrine therapy has been viewed as the first targeted breast cancer treatment that fully capitalized on the translational model, but many others have followed, including trastuzumab and other agents targeting EGFR, and tissue microarray studies evaluating the impact of cytotoxic treatments.

Another important research concept that is evident in the example of long-term endocrine therapy is the need for randomized trials with sufficient statistical power to identify modest improvements in outcome. Richard Peto has carried the torch for this issue and over the last 15 years has taught us to critically evaluate findings like the five-versus ten-year tamoxifen data.

The result has been the implementation of massive trials like ATAC — which has more than ten times as many accrued patients as the largest original adjuvant systemic studies reported in the late ’70s and ’80s.

To understand the clinical implications of this research database, the Breast Cancer Update series has had as its focal point interviews with key breast cancer laboratory and clinical investigators. We have also focused a great deal of attention over the years on education about the background, rationale and design of current randomized clinical trials. Recently, our team put together an educational exhibit and book on this topic for the Third Annual Lynn Sage Breast Cancer Symposium, sponsored by Northwestern University. Bill Gradishar, the meeting chair, and co-chairs, Monica Morrow and Craig Jordan, frequent guests on our series, assisted in developing this “clinician-friendly” first attempt at summarizing the key issues and questions being asked by cooperative research groups.

The entire clinical trials book and exhibit are posted on BreastCancerUpdate.com,along with hundreds of relevant links to journal articles, protocols and cooperative group sites. The special feature in this issue of Breast Cancer Update includes the first chapter of the educational supplement, which lists major cooperative groups and their current phase III randomized breast cancer clinical trials.

The rapid evolution of useful knowledge on breast cancer treatment directly reflects the efficacy of these trial mechanisms in identifying important questions and obtaining salient answers. Putting aside Kathy Pritchard’s thought about the possible obsolescence of post-tamoxifen aromatase inhibitor studies, it is interesting to consider that it was only 15 years ago that trials clarified that tamoxifen significantly made an impact on mortality in the adjuvant setting, and third-generation aromatase inhibitors have only been available to clinicians for about five years.

Because of an integrated and surprisingly efficient cooperative clinical trials system, we now are exploring issues that are likely to pay dividends to patients much more rapidly than anyone would have imagined a decade ago. Even practicing clinicians who are not accruing patients on trials will benefit from being informed about where we’ve been, where we are and where we’re headed in this fascinating journey.

—Neil Love, MD

Select Publications

Home · Search