Experts Review NCI Recommendation to Limit Tamoxifen Duration to Five Years

An expert panel of seven cancer researchers and a representativeof the National Alliance of Breast Cancer Organizations (NABCO)came together at the San Antonio Breast Cancer Symposium for aroundtable discussion of the use of tamoxifen (Nolvadex), sponsoredby PRR, Inc., publisher of Oncology News International and thejournal ONCOLOGY.

Last month, an article based on the roundtable focused on thepossible risks of secondary cancers with tamoxifen use. This reportcovers the appropriate duration of tamoxifen therapy. Future articlesin this series will include the panel's discussion of possiblenoncancer benefits of tamoxifen and how to deal with side effects.

SAN ANTONIO--The National Cancer Institute, in a clinical announcement,has recommended that physicians limit tamoxifen (Nolvadex) usein the treatment of early breast cancer to 5 years in clinicalpractice.

"But nobody really has done a valid comparison to determinean optimal duration," said V. Craig Jordan, PhD, DSc, ofNorthwestern University, who chaired the tamoxifen roundtable.He believes that the question of 5 vs 10 years is "almosta null hypothesis. You won't be able to detect any differencewith clinical trials as they are set up because they are not largeenough."

The NCI announcement came on the heels of the decision by theNational Surgical Adjuvant Breast and Bowel Project (NSABP) tostop a clinical trial (B-14) comparing 5 and 10 years of tamoxifenuse after surgery in women with node-negative, estrogen-receptor(ER)-positive breast cancer.

In this trial, 1,172 women who had been on tamoxifen for 5 yearsand had not relapsed were randomized a second time to either 5more years of tamoxifen, 20 mg/day, or placebo. After 4 yearsof follow-up, 92% of the placebo group were alive and free ofdisease, compared with 86% of the group scheduled to receive 10years of tamoxifen.

C. Kent Osborne, MD, of the University of Texas Health ScienceCenter, San Antonio, said that ideas on tamoxifen duration havechanged over the last 10 years. Initially, he said, it was feltthat since tamoxifen seemed to be predominantly a cytostatic agent,it might be best to continue the drug for a longer period of time,maybe indefinitely.

"Now," he said, "additional data, both clinicaland laboratory, are accumulating to suggest that we may have beenwrong in our initial hypothesis." Molecular biology studiessuggest that, in some cases, tamoxifen, an antiestrogen, can becomea stimulatory agent in the breast.

"It raises the possibility that there may be an optimal durationfor tamoxifen in the adjuvant situation, and it may be shorterthan we might have imagined before," Dr. Osborne said.

Node-Positive Patients

Dr. Jordan called the NSABP study "the first piece of evidenceas to what might be the optimal duration for the pre- and postmenopausal,ER-positive, node-negative woman. And it looks like 5 years maybe the right way to go." He added that the optimal durationmight be different in node-positive women.

For Dr. Osborne, however, there is "no reason at the momentto think that the situation would be different in node-positivebreast cancer," although he believes it is a worthwhile questionto ask.

"We would like to think that because node-positive patientshave a slightly worse prognosis that giving more tamoxifen isbetter," he said, "but that may well not be the case.And I think outside of a clinical research trial, I wouldn't givethe agent for more than 5 years, even in node-positive patients."

Dr. Osborne added that the NSABP trial is not alone in showingthat extending tamoxifen duration beyond 5 years adds no furtherbenefit. "There is a smaller trial from Scotland that showedsimilar results," he said, "and the Scottish trial includedboth node-negative and node-positive patients."

Amy Langer, executive director, National Alliance of Breast CancerOrganizations (NABCO), preferred to emphasize the positive datafrom the B-14 trial--that tamoxifen in node-negative breast canceris clearly beneficial compared with no tamoxifen, and that 5 years'duration is "now going to be more or less the standard."

Norman Wolmark, MD, of the Medical College of Pennsylvania/HahnemannUniversity, and head of the NSABP, agreed with Dr. Osborne thatthe issue of 5 vs 10 years in node-positive patients remains avalid research question.

"Having said that," he continued, "you should beaware that we at the NSABP have stopped tamoxifen at 5 years forour node-positive breast cancer protocols, based on the data fromthe node-negative population. I think this is the best informationwe have, and this decision has to be made by individual patientstogether with their physicians."

Dr. Jordon reiterated his belief that a clinical trial the sizeof the NSABP study would probably show no difference in benefitbetween 5 and 10 years. "My early concern with the NSABPstudy was that it would show a significant detrimental effectafter 10 years, with a large increase in recurrent disease, butthat has not turned out to be the case," he said.

He believes we have entered an age of uncertainty about how longto give tamoxifen. On the one hand, he pointed out, evidence fromclinical trials has led to a re-evaluation of previous hypothesesabout tamoxifen duration. On the other hand, it may be that thefinding of no additional benefit beyond 5 years represents a balancebetween suppression of tumor growth in some patients and tamoxifen-stimulatedgrowth in others.

"Patients are not all the same," Dr. Jordan said, "andthis is why we must have huge clinical trials to be able to answerthe question."

As a final note, Dr. Wolmark said that "just because we'rerecommending that tamoxifen be stopped at 5 years, it doesn'tmean that stopping tamoxifen will cause the tumor to start growingagain. That's not what's happening. Whatever benefit is accumulatedover that 5 years continues through 10 years of follow-up."

What the NSABP study has shown is that adding 5 extra years oftamoxifen therapy does not increase the benefit, he said, andalso that "there's no rebound effect, no increased incidenceof recurrence after tamoxifen is stopped."

Why Tamoxifen Keeps Working

Just 1 year of tamoxifen therapy has been shown to reduce therisk of breast cancer recurrence 10 years later. "So thereseems to be some permanent effect," Dr. C. Kent Osborne saidat the tamoxifen roundtable (see story above).

He explained that "you don't have to kill tumor cells toeradicate the tumor if you can stop tumor cell proliferation."

Tumor cells continue to die off naturally, but the number of replacementcells is so large that the tumor grows bigger over time. "Ifyou can shut off that proliferation, which is what we think tamoxifendoes, then you could eradicate all the tumor cells without directlykilling any of them," he said.

Laboratory evidence suggests that, in addition to its cytostaticeffects, tamoxifen also has a cytocidal effect, "killingoff breast cancer cells," V. Craig Jordan, PhD, DSc, said.But tamoxifen may help control breast cancer in a variety of ways.

It appears, for example, that tamoxifen may have a profound effecton angiogenesis, eradicating tumor cells by denying them a bloodsupply, Dr. Jordan said.