Standard Biopsies Retain Significant Role in Lung Cancer Care

Philip D. Bonomi, MD, discusses some of the known biomarkers in non-small cell lung cancer (NSCLC), the utilization of PD-L1, and other testing modalities.

Philip D. Bonomi, MD

Liquid biopsies have emerged as an approach to better identify mutations in patients with lung cancer, allowing the unique benefit of allowing clinicians to follow their progress throughout treatment.

Tissue-based biopsies are still important, however, said Philip D. Bonomi, MD, and can provide information to guide treatment—as long as the physician collects enough tissue.

“Get as must tissue as you can,” said Bonomi, The Alice Pirie Wirtz Professor of Medical Oncology, Department of Internal Medicine, Rush Medical College. “You did a biopsy and made a diagnosis? That’s great, but it isn't enough. You are doing a disservice to the patient because they might miss out of a therapy that could extend their life for a number of years.”

OncLive: Can you discuss your talk on biomarkers in NSCLC?

During the 2017 OncLive® State of the Science SummitTM on Advanced Non—Small Cell Lung Cancer, Bonomi stressed the importance of good biopsy practices and communication within the oncology community. In an interview during the meeting, Bonomi discussed some of the known biomarkers in non-small cell lung cancer (NSCLC), the utilization of PD-L1, and other testing modalities.Bonomi: Probably the most important thing is that we have to make sure we do testing in patients. It’s true that some of the driver mutations are not so common; however, if you don't look for them, you’re definitely not going to find them. You have to get tissue. We have been talking about liquid biopsies, which have been helpful, too. They can't test for all of the mutations, but they can test for the ones that are targetable with approved therapies. A lot of times, people just do a needle aspirate and don’t get enough cells or tissue to do good assays.

What are the advantages and disadvantages of liquid biopsies?

Part of this is communication and education. You have to talk to your teammates—pulmonary, intervention radiology, and surgery. If somebody had a big lymph node, don’t just excise is—get some tissue so we can do the stuff that needs to be done for the patient. It is minimally invasive; you can do it on everybody. If people have bad performance status, or a difficult area to biopsy where you can't get enough issue, it is useful. Also, it can be done repeatedly so you can follow patients with it. There are many advantages.

When do you rebiopsy and what factors do you take into consideration to make that decision?

The biggest disadvantage is in patients who have a relatively low tumor burden, because you are not going to be able to find the EGFR mutation or other mutations. We even have patients with an initial high tumor burden that got treated with a first- or second-line EGFR inhibitor and went into remission. Now, they are starting to progress, but their tumor is way smaller. When you check the blood, you might not even find the primary mutation—so you aren't going to find a T790M [resistance mutation]. That is the type of patient for whom you need tissue. We rebiopsy in indications where there is a treatment that is available. The biggest example that we have is EGFR mutations. If you have exon 19 or exon 21—the most common primary mutations—you get treated with the first- or second-generation EGFR inhibitor. Sixty percent of those patients with T790M will progress, so you should really get tissue if you can.

PD-L1 as a biomarker is somewhat contested in lung cancer. What are your thoughts on the utility of PD-L1 testing?

I have to tell you: we go first to the liquid biopsy. If we don’t find it there, then we will go and try to get some tissue. Even [with] urine, you can sometimes find the EGFR mutation in that. Therefore, liquid biopsies are good to use in someone who has an EGFR mutation and is progressing on their first therapy. You want to make sure you check for T790M, because 60% to 65% will have it, and about 60% to 70% of them will respond to osimertinib (Tagrisso). PD-L1 is a player in the immunotherapy arena, but it is not a very good biomarker. The other thing we talked about at this meeting is how tumor mutational burden is probably going to be a factor, too. Neither of these are perfect biomarkers—we are going to need a panel of biomarkers. On the other hand, if we have things such as combining immunotherapy with chemotherapy showing excellent results, perhaps we won't even need a biomarker.

Initial results of immunotherapy following chemoradiation in stage III [disease] were exceptionally good—granted, it’s only progression-free survival data. Again, if we can find a biomarker that makes the combination of immunotherapy even better, that would be great. However, it is going to be interesting to see how this plays out.

What is the biggest take-home message from your talk?

Biomarkers are very important for single-agent immunotherapy in the first-line setting, but they are not in the second-line setting because it beat the chemotherapy in both side effects and efficacy, even if there was no PD-L1 [expression].First, community physicians have an incredibly hard job—they are trying to keep up with this avalanche of information. They are not just seeing lung cancer; they are seeing breast cancer, lymphoma, colon cancer, and pancreatic cancer. They have such a hard time keeping up. The question is, “how to we disseminate this information?” One of the things that is going to happen, and is starting to happen, is that there are more hospital-employed oncologists, and, in bigger oncology groups, they are starting to sub-specialize. There is so much information, and patients are savvy from studying on the web. The doctors have to keep up with that.

The key thing is that you need to be aware that molecular profiles are important—not just in lung cancer, but in all cancers. You must get enough tissue, and you need to know how to do liquid biopsies. You can't know every nuance, but you need to collect the specimens so you can do the molecular testing. Moreover, you especially need to keep up the best you can. As time goes on, community-based oncologists are going to be sub-specializing just like we have done in academic medicine for 2 and a half decades now.