CHICAGO – according to a new analysis of the phase 3 LILAC trial.
The 725 women in the multinational trial received run-in, anthracycline-based chemotherapy and were then evenly randomized to receive ABP 980 or trastuzumab, each with paclitaxel, followed by surgery.
The difference in pathologic complete response (pCR) rate assessed by local pathologists has been previously reported (Lancet Oncol. 2018 Jun 4.); those findings established non-inferiority of the biosimilar but left the matter of non-superiority inconclusive. However, in the new analysis, reported in a poster session at the ASCO Annual Meeting, the difference in pCR rate when instead assessed by a central pathologist fell within all bounds for equivalence.
“This is part of the totality of evidence in the course of approval of ABP 980,” lead author Hans-Christian Kolberg, MD, head of the department of gynecology and obstetrics of the Breast Cancer Center of the Gynecologic Cancer Center at Marien Hospital Bottrop (Germany), commented in an interview.
The new data prompted European regulators to authorize marketing of the biosimilar (branded as Kanjinti) for HER2+ early breast cancer and metastatic breast cancer, as well as HER2+ metastatic gastric cancer. (In the United States, the Food and Drug Administration recently rejected the application for ABP 980 market approval.)
“Breast cancer therapy is getting more and more expensive, and we somehow have to raise the money to pay for it. If we have a chance to make an antibody that is 20%-30% cheaper, which is what we hope it will be in Europe, we have that money for other things,” Dr. Kolberg said, reflecting on the bigger picture.
“I am also a visiting professor at a university in China, where patients who are HER2+ don’t get Herceptin because they can’t afford it. We always have to remember that in Europe and the U.S., we are kind of living on an island. If you look at Africa, Asia, and South America, making things affordable is important,” he added. “I hope and believe that this is just the beginning of the price fight. I hope that the biosimilar companies really will fight to see who will have the lowest price because that will be good for the patients. The lower the price, the better for the patients.”
Research leading up to the LILAC trial established that ABP 980 had analytic characteristics, nonclinical attributes, and pharmacokinetics similar to those of trastuzumab. The trial, conducted in 97 centers in 20 countries in western Europe, eastern Europe, and other world regions, assessed clinical similarity.
“I think central review was done in the study because we had so many centers all over the world that it was questionable as to how we could monitor the quality in dozens and dozens of pathology labs,” Dr. Kolberg explained. “So the idea was that we make it a little bit more difficult, a little bit more expensive, but more reliable if we use one pathologist.”
The central review was not without logistical issues, he acknowledged. In particular, it was challenging to ensure that all centers – including some doing so for the first time – followed a standardized procedure for sending tissue to the central lab.
The previously reported locally assessed pCR rates in breast tissue and axillary lymph nodes were 48.0% with ABP 980 and 40.5% with trastuzumab. The risk difference was 7.3% (90% confidence interval, 1.2%-13.4%) and the risk ratio was 1.188 (90% CI, 1.033-1.366), with the upper bounds of the confidence intervals exceeding the predefined equivalence margins of 13% and 1.318, respectively.
The centrally assessed pCR rates were 47.8% with ABP 980 and 41.8% with trastuzumab. The risk difference was 5.8% (90% CI, –0.5% to 12.0%), and the risk ratio was 1.14 (90% CI, 0.993 to 1.312), with the upper bounds of the confidence intervals now falling within the equivalence margins.
“This is the first study ever that used central pathology review for pCR in a neoadjuvant breast cancer study. We were really skeptical at the beginning as to whether that would work because we had a lot of centers all over the world, from Russia, Brazil, the U.S., Germany,” Dr. Kolberg commented.
“It worked, and we were very lucky that it worked because in the local review, we did not reach our biosimilar margins, our equivalence margins. In the central review, we were well within the margins,” he said. “So if we had not in the beginning planned a coprimary endpoint with local and central pathology review, the medication would never have been approved.”
Dr. Kolberg disclosed that he is a consultant for Amgen, Carl Zeiss Meditec, Genomic Health, GlaxoSmithKline, Janssen, LIV Pharma, Novartis, Pfizer, Roche, SurgVision, Teva Pharmaceutical Industries, and Theraclion. The trial was sponsored by Amgen.
SOURCE: Kolberg HC et al. ASCO Annual Meeting, .