Dr. Paul E. Goss
Paul E. Goss, MD, PhD, of Massachusetts General Hospital Cancer Center and Harvard Medical School, presented results of the Canadian Cancer Trials Group’s MA.17R trial during the Plenary Session on June 5. He explained that 5 years of AI therapy with or without prior tamoxifen treatment has become standard practice for postmenopausal women with HR-positive early breast cancer, but the effect of extending that treatment was not previously known. Dr. Goss noted that HR-positive breast cancer is a chronic relapsing form of cancer, suggesting that extending therapy could provide some benefit.
The trial included 1,918 postmenopausal women with early-stage breast cancer, randomly assigned to receive either extended letrozole or placebo. The trial included three cohorts: two of which had received some duration of prior tamoxifen, and one, representing only about 20% of the full trial, that had no prior tamoxifen. All patients had recently received 4.5 to 6 years of adjuvant AI therapy, and all patients were required to have a minimum life expectancy of 5 years. Patients were permitted to enroll up to 2 years after completing previous AI therapy, but approximately 90% began receiving letrozole or placebo within 6 months of that prior therapy.
The patients were followed for a median of 6.3 years, and during that period there were a total of 165 DFS events, 67 of them in the letrozole arm and 98 in the placebo arm.
Of those DFS events, 42 and 53 were distant recurrences in the letrozole and placebo groups, respectively. There were 19 locoregional recurrences in the letrozole group and 30 in the placebo group. Contralateral breast cancer occurred among 13 patients receiving letrozole (1.4%) and among 31 patients receiving placebo (3.2%). The 5-year DFS rate was 95% for the letrozole arm compared with 91% for the placebo arm (hazard ratio [HR] 0.66, 95% CI [0.48, 0.91]; p = 0.01). The improvement in DFS was significant among patients with node-positive disease, but not for those with node-negative disease.
Overall survival was no different between the two groups, and 100 patients in each group died during follow-up. The 5-year overall survival rate was 93% with letrozole and 94% with placebo (HR 0.97; p = not significant).
The annual incidence rate of contralateral breast cancer, however, was significantly better in the letrozole group at 0.21%, compared with 0.49% with placebo (HR 0.42; p = 0.007).
There was no difference in quality of life scores based on several measures. But some adverse events occurred more frequently with letrozole, including bone pain, elevation of alkaline phosphatase, and elevation of alanine transaminase. More patients suffered a fracture in the letrozole arm (14%) than in the placebo arm (9%; p = 0.001), and there was also a greater incidence of new-onset osteoporosis with letrozole than with placebo (11% compared with 6%; p < 0.0001).
Dr. Ian E. Smith
“We make a big fuss about this, but the figures suggest it is not a big deal,” Dr. Smith said.
ASCO breast cancer expert Harold J. Burstein, MD, PhD, of the Dana-Farber Cancer Institute, said during a press briefing that “there will be tremendous interest in longer duration of therapy with an AI,” but he noted that therapy will likely be tailored based on specific patients’ risk factors. The study, he said, shows that longer duration of AI treatment is clinically valuable, but exactly how long is not yet clear.
The study also carries weight given the wide availability of the drugs in question. “Unlike many anticancer therapies, AIs are readily accessible around the world, and therefore our results will improve the outcome of many women with breast cancer,” Dr. Goss said.
Dr. Smith noted that the difference in distant recurrence between the treatment groups was only 1.1%, with contralateral breast cancer contributing significantly to the DFS improvement. With no overall survival benefit so far, he said, “If this were a novel adjuvant therapy, these data would not sustain.”
Still, from the point of view of “pragmatic clinicians,” Dr. Smith said, if patients have managed to stay on AI therapy beyond early unpleasant toxicities, he might tell them they could carry on with the treatment for longer based on these data.
The value discussant for the Plenary Session, Deborah Schrag, MD, MPH, of the Dana-Farber Cancer Institute, said that based on both the ASCO and ESMO scoring systems, this intervention has low value from a health system perspective, in part due to the excess costs associated with the increased risk of osteoporosis and fracture.“It would be a priority to focus on adherence to the first 5 years of AI therapy,” she said.
– Dave Levitan