2 Clarke Drive
Suite 100
Cranbury, NJ 08512
© 2024 MJH Life Sciences™ and OncLive - Clinical Oncology News, Cancer Expert Insights. All rights reserved.
Hans-Christian Kolberg, MD, discusses ABP 980 and the future of biosimilars in oncology.
Hans-Christian Kolberg, MD
Trastuzumab (Herceptin) has proven to be a go-to target in the development of biosimilars. MYL-1401O (Ogivri; trastuzumab-dkst) was approved by the FDA in December 2017, and agents such as SB3 and ABP 980 have shown similar pathologic complete response (pCR) rates to the originator biologic.
In results from the phase III LILAC study presented at the 2018 ASCO Annual Meeting, ABP 980 showed a pCR equivalent to that of trastuzumab. This was the first study of a biosimilar in neoadjuvant breast cancer to include a central pathology review, as well as instances of switching patients from trastuzumab to ABP 980, according to lead author Hans-Christian Kolberg, MD.
The FDA issued Amgen, the developer of ABP 980, a complete response letter for the biosimilar in May 2018. However, Kolberg said that investigators are dedicated to working with the FDA to bring the biosimilar to the United States. Additionally, the European Medicines Agency’s Committee for Medicinal Products for Human Use has adopted a positive opinion for the marketing authorization of ABP 980.
In an interview with OncLive® during the 2018 ASCO Annual Meeting, Kolberg, head, Department of Obstetrics and Gynecology, Breast Cancer Center, and Gynecologic Cancer Center at Marienhospital Bottrop, Klinik für Gynäkologie und Geburtshilfe, Bottrop, Germany, discussed ABP 980 and the future of biosimilars in oncology.Kolberg: The LILAC study was the registrational study for ABP 980, a biosimilar for the monoclonal antibody trastuzumab, which is used in HER2-positive breast cancer. In HER2-positive breast cancer, it is used in the early and metastatic stages. Trastuzumab is also approved for [patients with] HER2-positive gastric cancer.
This study was in neoadjuvant breast cancer for HER2-positive patients who had tumors larger than 2 cm. These patients received chemotherapy, anthracycline, and taxane-containing chemotherapy in combination with either trastuzumab or ABP 980. After surgery, the group of patients receiving the originator trastuzumab was split again and randomized for a second time, because half of the patients received a biosimilar after receiving the originator in the adjuvant phase.
One thing that is important is that this is the first study of neoadjuvant breast cancer that included a central pathology review. This is always difficult because we are not only sending around biopsies, we had to send around specimens. This is the first time that a central pathologist decided if it was a pCR or not. We have centers all over the world, so there was a discussion of whether this was feasible or not. It did work; we had enough specimens brought to the central pathology. It was very important because in the end, the central pathology review showed the similarity of the biosimilar because the pCR ratios were in the predefined range. Whereas, in the local pathology, the results were so heterogeneous that we could not show the same results. It was very good that we included that before as a coprimary endpoint, and we met that endpoint.ABP 980 is a biosimilar, and the approval process and study program for a biosimilar is different from a study program for a new biologic. You must first do pharmacokinetic and pharmacodynamic studies when developing a biosimilar. In this case, the first study was in healthy individuals, just to show it was safe and to show pharmacokinetic and pharmacodynamic data.
This is the first study showing clinical data on the biosimilar. That is because for the approval of a biosimilar, you do not need the process of phase I, II, and III studies. What you need is 1 phase III study with an efficacy endpoint. That is adding to this concept of totality of evidence that we are using in biosimilars.
You can actually choose what population you want to treat, but it has to be 1 of the approved populations for the originator. You can only choose 1 population, as we utilize the concept of extrapolation. [A positive opinion has been adopted in] Europe for early breast cancer, metastatic breast cancer, and metastatic gastric cancer, as long as they are HER2-positive cancers. You do not necessarily get an approval for every indication that the originator is indicated in.The decision to use biosimilars is the payer’s decision, in the end. It is also a philosophical question. Cancer therapy is getting more and more expensive, and if we have the chance to save a little bit of money, then we can use that money elsewhere. There are countries where people cannot afford trastuzumab. If they can afford the biosimilar, then maybe it can grant access to patients who could have not afforded it before and maybe save lives. That is an ethical and philosophical question.
In many countries, the payers request the use of biosimilars. In Germany, we have a quota where if there is an approved biosimilar available, 30% of the time, we have to use the biosimilar. I spoke to a colleague from an English hospital and she told me that they will replace trastuzumab 100% by biosimilars as soon as they are available in the United Kingdom.
The next decision is which biosimilar to use. In some settings, the pharmacist in the hospital pharmacy may decide what to order. We have to think about what differentiates biosimilars from one another. The efficacy and safety of the biosimilar has to be the same, otherwise they would not be biosimilars. The question is, “What data do we have?” That is why we are proud of our study because we are the only ones who included a switch from the originator to the biosimilar. This situation will come up if you have a quota, because you will have metastatic patients who are on the originator that have to be switched to the biosimilar to reach the quota. We believe that the decision to include a switch may give people confidence to use that biosimilar.
In the end, it is a difficult decision, and I don't think it really is a decision—it is a question of confidence in the manufacturer, and the studies that were done. If you believe in the concept of biosimilars as the FDA and the European agencies do, you can do whatever you want.We try to set up studies that not only give confidence, but also bring new scientific data. Meaning, we will do studies that we would do with the originator, just replace it with the biosimilar, which makes both therapy and the study cheaper with the same effect. Combining anti-HER2 therapies with immunotherapies, perhaps. That is an idea being discussed at the moment. In the end, the biosimilar will develop as a study medication, but we will be testing it with new products rather than as a study medication.From the global perspective, many biosimilar studies are done in countries where trastuzumab is unaffordable for patients. Therefore, being a part of the study was a chance for these patients to get trastuzumab. The second thing is the fact that in many European countries, the reimbursement system is more centralized—it is easier for the payers to integrate biosimilars, because we do not have the system of hundreds of healthcare plans from different insurances.
Kolberg HC, Tomasevic Z, Demetriou G, et al. Efficacy analyses of central laboratory pCR results from the LILAC study comparing the biosimilar ABP 980 and trastuzumab. J Clin Oncol. 2018;36 (suppl; abstr 583).