COMMENTARY

Lung Cancer Advances: A Year Like No Other

Mark G. Kris, MD; Pasi A. Jänne, MD, PhD; Christian Rolfo, MD, PhD, MBA

Disclosures

November 16, 2018

Mark G. Kris, MD: Hello. I'm Mark Kris, professor of medicine at Memorial Sloan Kettering Cancer Center in New York City. Welcome to Medscape Oncology Insights. Today we'll be discussing recent clinically relevant advances in lung cancers. Joining me is Professor Christian Rolfo, director of thoracic medical oncology and early clinical trials at the University of Maryland Greenebaum Comprehensive Cancer Center in Baltimore; and Professor Pasi Jänne, director of the Lowe Center for Thoracic Oncology at the Dana-Farber Cancer Institute in Boston, Massachusetts.

This has been an amazing year for the treatment of lung cancers. It's hard to know which topics are the most important. Pasi, let's talk first about the developments in targeted therapies. What would you highlight? What should oncologists know that would be practice-changing?

Pasi A. Jänne, MD, PhD: Certainly, the use of our next generation of targeted therapies for epidermal growth factor receptor (EGFR)-mutated lung cancers and ALK-rearranged cancers is important. Osimertinib is emerging as the standard-of-care, first-line EGFR inhibitor for EGFR-mutated lung cancer. Alectinib is emerging as the first-line ALK inhibitor in that space, although we recently had data showing that brigatinib was better than crizotinib.[1] Then we have many other treatment subsets—for example, for RET-rearranged lung cancer data on both LOXO-292 and BLU-667 showing great responses,[2] and entrectinib for ROS-rearranged lung cancer.[3] So I believe that in the targeted therapy space, we're finding the best single therapies for these oncogenic alterations. It's been a great advance.

Kris: The question that comes up is, what to do next? What can you do when these new target therapies stop controlling cancer?

Jänne: I believe that this is still in evolution. The most important thing is to try to understand why the therapy stops working. So for example, if you have an ALK-rearranged lung cancer, are there any ALK secondary mutations? If so, some of the other inhibitors may work or may not work in the context of that specific mutation. If the patient develops resistance to osimertinib, is it because the cancer has another EGFR mutation? Is it because of MET amplification? Then it may be a candidate for combination therapy with an EGFR and a MET inhibitor. Do you have small cell transformation and need chemotherapy? We need to continue to study how the tumors are evolving in response to these therapies because that may give us clues as to what the next likely therapy is.

Kris: I'm getting the impression from what you say, though, that people should have a biopsy when they progress, despite their initial targeted therapy.

Jänne: Absolutely. You can do a liquid biopsy, but I believe that may be limited because you don't know if you'll get small cell transformation from a liquid biopsy. You also don't know if the tumor is shedding. For the moment, a tumor biopsy would be preferable, but if that can't be done, you can certainly get some information from a blood-based biopsy.

Kris: Christian, would you agree?

Christian Rolfo, MD, PhD, MBA: Absolutely. More and more, we are using next-generation sequencing not only in tissue but also in these liquid biopsies. Several things still need to be clarified in the area to make it possible to apply in clinical practice. We just published [a workflow],[4] and I also see the International Association for the Study of Lung Cancer guidelines, for liquid biopsies[5] that will help to perform these kinds of tests in a practical. In addition to the excellent results that we have with the new drugs, it is most important that we test our patients for the mutations. There is a tendency to implement treatments on the basis of a small panel or even single determinations of mutations; we quickly go for PD-L1 status and treat patients in this way.

We need to be very careful because some of the subsets of the biomarker driver mutations are not working properly with immunotherapy. So I believe that the message is to test your patients, to identify not only the patients who are able to respond to immunotherapy but also the patients who are able to respond, even in a small, particular subgroup population, to different target therapies.

Kris: Are the blood-based tests good enough for fusions?

Rolfo: Yes. I believe that we have technologies that are able to catch fusions. We need to be aware of which kinds of panels we are using. Some of the panels are not covering, for example, TRK [tropomyosin receptor kinase]. They cover only TRAK1 fusion, and thus they are not capturing all of the fusions that we are able to target with compounds such as larotrectinib and entrectinib. It is important to know the coverage analysis that the lab is doing and how deep they are able to go—so, the allelic fraction variation as well. It can be a bit confusing when we see liquid biopsies with a single determination. So we need to go for panels to give us more complete information.

Kris: And when you decide to do a biopsy at the time of progressive disease, if you need to change therapy, do you have a preference for the liquid or solid biopsies?

Rolfo: It depends. I believe that the liquid biopsies are preferable for the patients because they are less invasive. But if this approach is not getting all of the information, we must not forget the tissue, and we can use the tissue for some of the determinations. Liquid and tissue biopsies are complementary at this moment. We cannot substitute one for the other.

Jänne: I would agree. You learn something from the tissue-based analyses, as I mentioned. Histologic transformation—if you need to do any additional protein-based test, you do need the tissue. Sometimes clinical trial enrollment requires tissue and blood isn't an acceptable substitute. So they do provide complementary information, and I certainly prefer tissue at the moment, if it's feasible.

Kris: Let's move on to immunotherapeutics. It seems to me, in 2018, that it has become much easier, because except for some small groups of patients, everyone gets chemotherapy and a checkpoint blocker. And the type of chemotherapy is determined by the cell type. So for small cell, you would use etoposide and a platinum; for squamous cell, you would use a taxane and a platinum or gemcitabine and a platinum; and for adenocarcinomas, you would use pemetrexed and a platinum. Do you agree with that?

Rolfo: In the past 2 years in immunotherapy, we are like kids in the candy shop. We are overcome with all of the information that is coming out. We also have new data coming for small cell lung cancer, as you say—atezolizumab in combination with chemotherapy—that demonstrates some positive results. That is very important information.

But we need to be careful with the information we are processing in these times. We need to be aware of several things—for example, the PD-L1 status. We know that PD-L1 is an imperfect biomarker, but this is the only one we have at the moment in clinical practice. Tumor mutational burden is another biomarker that we are seeing included more and more in clinical trials, and the subpopulation analyses are also being used to say that certain results are positive or negative according to these subanalyses. In my opinion, we need to be careful when we apply tumor mutational burden in clinical practice. We don't have a lot of standardization of the methods we are using. We don't have a cutoff that we can rely on. We still don't have prospective data on overall survival. Most important, we don't know the quantity or the quality of the mutations. The main clue here is to find mutations that are able to present antigens and can produce an immunoreaction in this setting. There are several other factors—the microbiome, for example—that may come in the future. We also need to be careful in selection of the patients, specifically for the combinations in first line with chemotherapy. We need to consider the side effects, and the selection of the patient is crucial.

Kris: Pasi, did you have a PD-L1 cut-off where you would use single-agent pembrolizumab versus a combination chemotherapy plus pembrolizumab?

Jänne: I believe that in the front-line setting, for greater than 50% [PD-L1 expression] patients, you have the option of using single-agent pembrolizumab or chemotherapy plus pembrolizumab. Certain considerations will go into that determination: How bulky is the patient's disease? How symptomatic is the patient? Age may be a factor; as opposed to being able to give triple therapy, you would give single therapy. Thus, some of those considerations go into that decision.

I like being able to treat patients with single agents, especially if they respond. But I also know that if I start out with single-agent pembrolizumab and they don't respond, I can potentially then use chemotherapy.

Kris: Regarding the comments on tumor mutational burden [as a biomarker], is there enough rationale today to recommend treatment with nivolumab-ipilimumab or are more data needed?

Jänne: I think more data are needed. Unlike PD-L1 results, which most of us can get back in 24 hours, tumor mutational burden sequencing takes 10-14 days. So the practicalities of that as a biomarker, especially in diseases like small cell lung cancer, are not quite there yet. And so I believe that we need more data to see whether we can ultimately use it as a biomarker. The other issue is that we don't quite have standardization of tumor mutational burden. We spent a long time looking at different antibodies on PD-L1 detection and looking at how good they are among each other. I believe that we need to think about that same sort of analogy for tumor mutational burden if that is going to emerge as a clinically useful biomarker.

Kris: This year, more data are out about using immune checkpoint blockade with surgery and with the combination chemotherapy and radiation. Christian, would you comment on that?

Rolfo: During the 2018 European Society for Medical Oncology (ESMO) meeting, some data on neoadjuvant treatment were presented, and it seems to produce a good response—even a pathologic complete response. Several studies are being conducted. We have one study of neoadjuvant with atezolizumab,[6] and another combination with chemoradiation. This is an important topic, but we need more data before we apply this in clinical practice.

Kris: Those of us who have toiled in this mine for many years are pretty amazed at how high the rates of pathologic and major complete response are with just single-agent checkpoint blockers. Is there something about the intact biology of the tumor, and the nodes surrounding it, that it's a somewhat better milieu for these drugs to work?

Jänne: The early data are amazing, and I believe that we are fortunate in this situation. We now are able to apply and study immunotherapy in the earlier stages due to a sufficient number of samples, sample sizes, etc., allowing us to ask these questions.

Unfortunately, lung cancer, even if it's surgically resected, has a high degree of relapse. So, incorporating the therapies that are effective in the advanced disease setting into the local advanced disease setting, and into the earlier disease setting, is great. We hope that this will lead to even better outcomes and higher rates of cures.

Kris: And we do have a unique situation in lung cancer. When the primary tumor is resected, so are the regional lymph nodes. It's not a sampling; it can be a very comprehensive dissection. So there is sufficient material that doctors can use to answer some of these questions. I believe that in addition to giving another opportunity to our patients, it gives us a chance to understand how these treatments work.

After decades of setbacks in trying to improve the odds for people with local regional disease that cannot be resected, suddenly we have data that show improvements in progression-free survival, and now overall survival, with the use of durvalumab after conclusion of chemotherapy and radiation.[7,8] Christian, has that trial and paradigm changed your care of these patients?

Rolfo: Absolutely. I would recommend that for our patients who have stage III disease. This is a very important development in this population because we know that these patients often have a recurrence in a short time. It is important that we protect them from that.

We can have an impact on progression-free survival and overall survival. An analysis of that was presented at ESMO this year, speaking about PD-L1 status, showing some data that PD-L1 status is positive—patients have an extra benefit with this. This is a very nice scenario. These patients will be pioneers in this kind of investigation. We need to follow these patients very well in order to understand whether this impact on overall survival is maintained over several years and we can protect them from the recurrence.

Kris: I'm not surprised to see that they say that the higher your PD-L1 expression, the more benefit you will get. When you look at all of the data together, that is the trend.

Rolfo: As I said before, this is a very imperfect biomarker; we are working with a biomarker that has intrapatient heterogeneity. We need to think carefully. But the most important message of this trial was that the entire population had a benefit.

Kris: How has it changed the way you approach these patients?

Jänne: We recommend it now for everyone [in this situation]. We always knew that we could cure some fraction of individuals with chemotherapy radiation. But as Christian mentioned, the relapse rates are high, and finally to be able to have a therapy that improves not only progression-free survival but overall survival is a real advance. I believe that our opportunities moving forward are to think about which patients may need more therapy, which patients are most likely to benefit. There will be a lot of science involved in that moving forward. It is a great advance.

Kris: I can't think of another year in my career where there have been more changes. When you think back a year ago, what you may have told someone, that's not right anymore. And it's in so many areas. We have greater opportunities for our patients and also [more options] for us to offer patients. Christian, Pasi, thank you for joining me today. This has been a great discussion. We are fortunate to have two true experts here with me.

This is Mark Kris for Medscape Oncology.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....