BATTLE-2: Next-Generation of Biomarker-Driven Trials

Roy S. Herbst, MD, PhD

Chief of Medical Oncology Associate Director, Translational Research, Yale Comprehensive Cancer Center Smilow Cancer Hospital at Yale-New Haven, Yale School of Medicine New Haven, CT

Ignacio I. Wistuba, MD

Jay and Lori Eissenberg Professor in Lung Cancer, Director of the Thoracic Molecular Pathology Lab, Departments of Pathology and Thoracic/Head & Neck Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX

1

What is BATTLE-2 and why is it important?

One of the themes at the 13th International Lung Cancer Congress in Huntington Beach, California, last July was the personalization of therapy based on rapidly increasing knowledge of the molecular profile of individual lung cancers. While the number of known mutations and matching targeted agents is relatively limited at present, clinical trials are being designed to identify effective therapies for specific mutations more efficiently. One such study is BATTLE-2, a biomarker-driven trial for patients with metastatic lung cancer. Roy S. Herbst, MD, PhD, and Ignacio I. Wistuba, MD, two of the principal investigators, discussed the study with OncologyLive.Herbst: The BATTLE-2 trial is a study that is designed to understand how tissue biomarkers predict who will respond to targeted therapy or targeted therapy combinations. The idea behind BATTLE is that every patient gets a biopsy at the time of their treatment so that it reflects the mutational status and biomarker status of their tumor, and then that information is used to put them onto one of four arms of the trial. In this case, the trial has an EGFR inhibitor (erlotinib); an inhibitor of Ras, Raf, and other [signaling molecules] (sorafenib); a combination of erlotinib plus an Akt inhibitor (MK-2206), which is thought to overcome EGFR resistance; and a combination to target KRAS,⎯a MEK inhibitor (AZD6244) plus an Akt inhibitor (MK-2206).

The way the BATTLE-2 trial works [is that] as soon as the patient has their new biopsy reflecting the tumor at the current time, it goes to Dr Wistuba’s lab. Within a week to 10 days his lab is able to provide a series of markers so we know what the profile is of that patient. Initially, the patients are equally randomized to each of the four arms, but as time goes by, as we learn how the marker profile predicts who does well, we start to adaptively randomize, meaning that they are more likely to get the arm that will benefit. So, this is an approach that the patients like because they’re not just being strictly randomized, but they’re being sorted based on their profile, and we’re learning. Dr Wistuba is also doing discovery work.

Wistuba: That’s the other important aspect—we use the information to assign the patients to the treatments, but we also use the tissue and the material that we have to potentially discover new things that could benefit future lung cancer patients. We are planning to take all this information and make clinical trials more specific based on what we are learning from these trials. So, it’s not just to assign patients based on the biomarker profile, but to discover new abnormalities in lung cancer that can help the entire field.

Herbst: Maybe 20%-30% of patients have a mutation or biomarker that can be used to give them a helpful therapy, but the other 70% don’t, so we’re going to learn by doing this work in a rigorous way with all the techniques that Dr Wistuba and his lab can employ. Dr Wistuba and I are doing this in collaboration with Dr Vassiliki Papadimitrakopoulou [University of Texas MD Anderson Cancer Center], and Dr J. Jack Lee [University of Texas MD Anderson Cancer Center]. We received an RO1 grant from the National Cancer Institute, so this is receiving peer-reviewed support.

Wistuba: We are also opening this at Yale, where Dr Herbst is now, so we’re going to show that this can be done in a multicenter way, because in the past it has been done at only one institution. By having this biopsy-driven trial in two places, I think this is also a step forward in the field.

2

Are clinical trials like this going to be the wave of the future for developing new cancer therapies?

Herbst: As we’ve heard at this meeting, some of these mutation and biomarker events are rare, 5%-10% of cases, and even a big center can’t do it by itself. We’ll need to collaborate, and we’re talking about SWOG or other collaborative groups running big studies. We’re piloting that this can be done at two centers. In the future, multicenter efforts such as these (perhaps through the cooperative groups) will be essential as we move forward with personalized medicine. We’re going to do some of the assays at Yale—initially, EGFR mutation, ALK translocation, and KRAS—and then the tissue for discovery is going to be sent to Dr Wistuba. So, a very nice collaboration.Herbst: Absolutely. I think it’s essential. If you don’t let the biology drive the treatment, everything is empiricism and it is going to have a very low chance of hitting the mark. So absolutely, as Dr Wistuba said, treat patients the best you can in real time, but also learn, discover, identify new markers. Patients get a drug and respond, but then they’re resistant; why are they resistant? That is, I believe, the fundamental question. Understanding why patients become resistant, and how we can understand what genes and proteins are involved so we can then target them.

Wistuba: One of the improvements in our clinical trial based on the BATTLE-1 trial result is that this approach of taking a biopsy from the patient at the time of the treatment has been safe clinically. There are not too many important adverse effects from having a biopsy at that time, and that has actually changed the paradigm of how these biomarker-driven clinical trials are going to be conducted. Instead of relying on a tissue specimen that was taken six months ago or a year ago, before several treatments have happened, this is real time and we’re learning about the genetic makeup of the tumor at that time.