The effect of duration on cardiac events was a key question for the PHARE and HERA trials to answer. In general, longer duration of treatment was associated with more cardiac events, especially declines in left-ventricular function, but overall the risks were low.

Cardiac Endpoints

In the PHARE trial, the 12-month regimen conferred greater cardiac risk, with events occurring in 5.7% of this arm, vs 1.9% of the 6-month treatment arm (P < .0001). The main drivers of this difference were the declines of left-ventricular ejection fraction to < 50%, observed in 6.3% of the 12-month arm and 4.7% of the 6-month arm (P = .04).

 In HERA, secondary cardiac endpoints and other adverse events were increased in the 2-year arm, but overall they were rare. Dr. Gelber noted that to be eligible, patients had to have an ejection fraction ≥ 55% at baseline, after treatment with chemotherapy.

A primary cardiac event (class III or IV congestive heart failure or left-ventricular ejection fraction < 50% and ≥ 10% below baseline, or cardiac death) occurred in 0.8% of the 1-year group, 1.0% of the 2-year group, and 0.1% of the observation arm. A secondary cardiac event occurred in 4.1%, 7.2%, and 0.9%, respectively.

HERA Data Considerations

Invited discussant Sandra M. Swain, MD, of Washington Hospital Center’s Washington Cancer Institute, said she would like to see a more specific analysis of the cardiac endpoints in PHARE before commenting on the findings, but she explored the findings from the HERA trial.

“In HERA, severe heart failure or cardiac deaths were uncommon (≤ 1.0%), and remember that patients were required to have a baseline ejection fraction ≥ 55%. This was after chemotherapy, which could include anthracyclines,” she noted. When asymptomatic declines were included, 2 years of treatment was more cardiotoxic than 1 year.

She pointed out that these outcomes are much better than those seen in the recent 7-year follow-up of National Surgical Adjuvant Breast and Bowel Project (NSABP) B-31, which found a 4% risk of heart failure and cardiac deaths, compared with 1.3% among patients not receiving trastuzumab (though the cardiac toxicity definitions were not exactly the same). But in B-31 and in previous analysis of HERA, approximately 80% of asymptomatic patients with declines in ejection fraction eventually normalized.

“This is good news, and it hopefully means that these events are not clinically meaningful to our patients,” she said.

A combined dataset will evaluate the cardiac risk from all the adjuvant trastuzumab trials and possibly identify patients likely to develop heart failure or serious declines in ejection fraction, she added. ■

Disclosure: Dr. Swain receives research support from and is on advisory boards (uncompensated) for Genentech/Roche.