Exploring adjuvant endocrine therapy use with letrozole in postmenopausal women with hormone receptor-positive breast cancer

Dr Terry Mamounas - Orlando Health UF Health Cancer Center, Orlando, USA

Tomorrow we’ll be presenting an update on the NSABP B-42 trial. This is a ten year update. This is a study evaluating the effect of extended aromatase inhibitor therapy with letrozole for patients that have completed five years of endocrine therapy, primarily with an aromatase inhibitor in the first five years although we also allowed patients to receive up to three years of tamoxifen and then the remaining the aromatase inhibitor. At that point if patients were disease free they were randomised to five years of letrozole versus five years of placebo, again evaluating the extended adjuvant endocrine therapy.

We presented the results here about three years ago with a median follow-up of six years and it was the seven year data which showed a not statistically significant but a trend towards a significant benefit in disease free survival with about a 15% reduction in recurrence in favour of extended letrozole. Technically it did not meet statistical significance due to the interim analysis where the p-value had to be 0.041 and it was 0.048.

At that time we had also shown a significant reduction in the distant recurrence and also a significant reduction in breast cancer recurrence overall. The adverse events, particularly in terms of osteoporotic fractures or arterial thrombotic events, were not significantly different. So there wasn’t much in terms of those events.

In this meeting we’re presenting now the ten year data with almost a 9.3 year median follow up which now shows a statistically significant improvement in favour of letrozole, no surprise. The p-value is now 0.01 and the absolute benefit is about 4% in favour of letrozole versus placebo. The relative reduction is about 16% reduction so the hazard ratio is 0.84. Similarly we see a statistically significant improvement in terms of reduction in distant recurrence and in terms of reduction in breast cancer recurrence overall. We don’t see any more safety signals, still no significant difference in the osteoporotic fractures or arterial thrombotic events. We were a little concerned initially with the arterial thrombotic events because in the first 2½ years of follow-up letrozole had less than placebo but then the curves flipped so we thought that perhaps there could be an increase with letrozole over time. But it certainly hasn’t happened so the safety has been demonstrated.

What we take out of this study, as we already presented before and published, is that we need to have a careful assessment of risks versus benefits in order to recommend extended endocrine therapy for these patients. We have to look at the absolute risk of recurrence which is a factor sometimes of the age of the patient, nodal status, tumour size. Also we need to look at bone mineral density because that’s one of the potential adverse events from aromatase inhibitors and also how they tolerate the aromatase inhibitor in the first five years.

As far as the future goes we hope that maybe integrating genomic classifiers into this algorithm may help us actually select more appropriate candidates for extended letrozole therapy or extended aromatase inhibitor therapy in general. As you may know, there have been several other studies that have looked at this question, most of which show a trend towards benefit with extended letrozole therapy. Most of the studies show that at least eight years of total therapy is beneficial, maybe you don’t necessarily need up to ten years. Our study first compared five to ten so we said ten is better but maybe close to eight years may be an optimal duration.

How are the adverse events managed?

Sometimes there could be constitutional side effects such as hot flushes, menopausal symptoms, with aromatase inhibitors, vaginal dryness, vaginal discharge, in which case we can manage the symptoms in terms of lubricants, in terms of medications that counteract the effects of hot flushes such as, for example, some of the antidepressant medications. Also another, of course, major effect is osteoporosis which, of course, we monitor. Luckily it doesn’t come abruptly, it comes over time so we can intervene with bisphosphonates if we need to. As far as, obviously, the arterial thrombotic events are something that you cannot anticipate but the other issue with aromatase inhibitors are arthralgias and myalgias. There are multiple ways of dealing with that including non-steroidal medications, including sometimes acupuncture. Some people have shown that you can improve on that. Also, of course, exercise because most of this is essentially morning stiffness type of thing so if you get going it improves.

Sometimes you have to interrupt therapy and hopefully restart if the patient can be re-challenged. There are different aromatase inhibitors so sometimes you can switch from one to another, although there is usually cross-reaction in those adverse events. Sometimes you may find a patient that tolerates one versus another. So these are all methods that we use to alleviate adverse events.

Why might there have been no improvement in overall survival with letrozole?

That’s primarily, we think, because most of these patients if they get a distant recurrence event or a recurrence they go on endocrine therapy for metastatic disease, they do well and it takes many years before these patients succumb to the disease. Perhaps it’s too early to actually even see that kind of difference.

The other speculation is that the effect was modest in terms of disease free survival and we may not have enough power to detect an overall survival benefit. So we still need to do further analysis, potentially up to fifteen years or even longer because, as you know, the ER positive disease is a disease that tends to recur over time. Up to 15-20 years later you may still have recurrences. So we cannot exclude something may become significant later on but certainly at this point we don’t see it.

Interestingly, when we looked at the seven year data the hazard ratio is actually favouring the placebo versus the letrozole, it was 1.15, it was the other way. The hazard ratio now is 0.97 so in other words it favours a little bit letrozole although not significant. But over time the two lines come together and so it’s possible that with additional follow-up, because of the difference in distant recurrence, we may see some survival benefit emerging. Whether it will be significant or not remains to be seen.