Roundtable Discussion: Nadler Leads Debate on the Use of PD-L1 Expression in NSCLC Therapeutic Decisions

Eric Nadler, MD, MPP

Eric Nadler, MD, MPP, the medical director of US Oncology Health Informatics and Internet Technology at Texas Oncology-Baylor Charles A. Sammons Cancer Center, lead a Cased-Based Roundtable event during which participants discussed how PD-L1 expression in patient with non–small cell lung cancer impacts treatment decisions.

NADLER: What kind of testing do you want? Is it tissue? Is it plasma? Does anyone want to share what they’re doing?

SHIMKUS: We tend to do both tests up front with the concern that we may be missing something on solid tumor testing [only].

NADLER: Is turnaround time part of the reason for that, or for its comprehensiveness, or a little bit of both?

SHIMKUS: Maybe a little bit of both, but more comprehensiveness than timing.

CHALLAGALLA: Also, sometimes it’s inadequate tissue samples, so we don’t want to take a chance, so we do both.

AGARWAL: I like to do both at the same time, and it’s been more comprehensive. Like Dr Challagalla said, tissue is…always a problem. We can barely do PD-L1, and

then they say they can’t do anything else. So I do both at the same time.

NADLER: I agree. You’re never wrong testing both as fast as you can, if nothing else, because of turnaround time.

The National Comprehensive Cancer Network is mandating and recommending comprehensive panel-based testing, including for the more recent MET and RET mutations [added] last year.¹ HER2 and KRAS [mutations are] coming out as well. Next-generation sequencing is the only way of doing it. As you’ve all mentioned, there is really a move toward incorporating plasma within this.

NADLER: The question is, what do you do for a patient with pan-negative mutation, PD-L1 tumor proportional score 1% to 10%, intermediate-low [risk], and tumor mutational burden 6 mutations per megabase?

NADLER: There is no right answer here. All but 1 is an approved regimen to consider. One vote for immunotherapy [I/O] alone and 13 for chemotherapy and I/O with the KEYNOTE-189 [NCT02578680] approach.

Again, none of these are wrong; you could speak to each of these. Treatment patterns-wise, I find 1 probably for an intermediate-low [risk patient] is probably the most common. But I/O alone or combination I/O is certainly not wrong.

NADLER: How do you decide for a patient [who has] PD-L1 positive expression, 1% to 49%? How do you decide what to do and how do you start parsing through all the data?

SHIMKUS: I think it’s really unique in that people used to know about chemotherapy and maybe knew a couple of names, but [now] patients come in asking for [I/O]. It doesn’t matter what their disease is. They want immunotherapy; they don’t want chemotherapy. People are coming in now requesting it.

NADLER: Do you feel that patients with 1% to 49% [PD-L1 expression] are working more in the I/O alone, ipilimumab [Yervoy]/nivolumab [Opdivo], or single-agent pembrolizumab [Keytruda] direction? Or do you feel that they’re comfortable combining chemotherapy with immunotherapy?

SHIMKUS: I think they’re comfortable but what makes it difficult for me to make that decision is some of the data in melanoma and renal cell carcinoma—when rechallenging patients who have gone through ipilimumab/nivolumab, then later been on maintenance nivolumab, relapsed, and then received ipilimumab, [they] respond again.

You only get one shot at that. That’s where I tried to look at dual I/O therapy up front because you’re only going to get that one shot at the prolonged use of those drugs.

NADLER: It’s interesting, [and] I couldn’t agree more. You have that one shot. The one thing that’s just unusual—I’ve thought a lot about the distinction and the immunogenicity of melanoma and kidney cancer [compared with] head and neck cancer and lung cancer….In the days of interleukin 1 and 2 [IL-2], 6% of patients were cured [back] then. You could cure 6% of metastatic kidney cancer and melanoma 20 years ago with IL-2; those are the most immunogenic and then head and neck and lung cancer are the next [in line] after that.

NADLER: How important is the chemotherapy and how important is the I/O? Do most of you feel that the chemotherapy is the important secret sauce you need in those patients with PD-L1 expression 1% to 49%, or [more like] half your patients get I/O alone, half get chemotherapy with it?

SHIMKUS: I look at it as I could always go back to chemotherapy. I’m never going to have a problem getting that approved. But I can’t play with the I/Os and get a second I/O approved [later]. So I try to do [I/O and then] all the chemotherapy I want later.

SOLANKI: It’s going to come down to [insurance approval] or risk stratification. Right now, they are putting everything on PD-L1. We are not even sure if PD-L1 is what is determining [response]. The PD-L1 testing in, for example, the breast cancer [trial] IMpassion130 [NCT02425891]—they tested it differently.

They looked at the immune cell PD-L1 expression, not the tumor cell. So there would be a lot of differences, even in the PD-L1–expressing group. Eventually, it’ll come down to better risk stratification as we learn more so that you can parse out patients [and be able to know if they are] not going to benefit from chemotherapy at all or if I should give some chemotherapy along with it even though it will be harder.

NADLER: I think it’s a fantastic question. So 93% of us said we wanted to start carboplatin, pemetrexed [Alimta], and pembrolizumab. If you gave it 4 to 6 times and it’s working, what’s the next step and why? Do you do [maintenance with] pemetrexed, do you do pembrolizumab, or do you give them both?

AGARWAL: I do pembrolizumab with pemetrexed as maintenance.

NADLER: At the 14-month mark, do you drop something?

AGARWAL: Yes, at that time, I drop the pemetrexed. I just continue pembrolizumab until progression.

NADLER: Some people are just using I/O at that point and some people are using both. If you start both, which the majority of us do, do you ever stop one or the other? And if you do, when is that time?

CHALLAGALLA: A lot of times it’s the myelosuppression toxicity that makes us drop pemetrexed. [Timing] varies patient to patient...In my patients, I have to drop pemetrexed at some point.

AGARWAL: I just have 1 patient who is reaching almost a year now, so I’m thinking of dropping pemetrexed.

NADLER: Does anybody buy into the whole if it isn’t broken, why fix it, and continue them both, or does everybody feel compelled at some point stop 1, or the other?

SHIMKUS: I continue until toxicity.

NADLER: Anyone have final comments on this idea of what the role of chemotherapy in maintenance is?

SOLANKI: The problem we have had with combining a molecular therapy or targeted therapy with chemotherapy is that there was no effort made to see how long you need to give, for example, bevacizumab [Avastin] in lung cancer. In colon cancer, for example, it took another study that took another 10 years before we realized that Avastin maintenance in metastatic colon cancer is useless. But until then, because of the original study, no effort was made.

So what is going to happen is, all of us are going to have few patients who did wonderfully well and feel like that is what we should do for everybody. That’s because we have lack of data to help us.

NADLER: Certainly, [we] don’t know what you’re supposed to continue and the ultimate duration of what that is. There are no answers to that and there won’t ever be. That isn’t actively being investigated. There was a trial looking at pembrolizumab, pemetrexed, or both, but even that trial and the results of it may not settle this debate. So it’s a challenge to get great answers in lung cancer to these challenging questions.

XIONG: I think it is a challenging issue facing us, but in a good way. In the past we didn’t have that many challenges. We really question how long we want to continue. Now I have patients with lung cancer that always have something you can see on the scan but have been stable for 2 years. It gives you some excuse that something is still there but deep in your mind, [you think it] might not even be cancer. But if something is there, you continue.

I have another patient that [has] nothing you can see [on scans]. It’s a complete remission for a year or 2 years. So we continue I/O and, hopefully, it won't come back, but for how long? It is question that we don’t really have an answer to, but the mentality tends to [be to] continue something that’s still working.

NADLER: It’s interesting. We have data in melanoma. I’m not a melanoma doctor, but we do have some. In lung, we don’t.

The trials were all 2-year trials but at the end of it, patients came off trial and if the doctor wanted to continue to give the drug, they were allowed to do that.

So when you look at the survival data…the trial stopped at 2 years, but the doctors kept giving it because patients were doing so well. So we don’t really have the answer when you can stop and what do you do.

AGARWAL: That’s interesting because I have a patient experience where I continued a patient with metastatic disease for 3 years, who went into complete remission. After 3 years, the patient could not afford the co-pay part of it, so she stopped on her own. She continues to be in remission 2 years later.

AGARWAL: I biopsy these patients to rule out a small percentage; 5% may have small cell and things like that.

NADLER: Probably not here [because] the small cell transformations were seen predominantly in the EGFR-mutated space. In a patient who is mutation negative, with PD-L1 15%, never had a driver mutation, and they progressed on chemotherapy and I/O, what are our second-line options?

TIJANI: There’s a role for a docetaxel in these patients. Sometimes I continue with the monotherapy with taxane.

NADLER: That is the only [option]. They’ve seen platinum pemetrexed but haven’t seen a taxane.

SOLANKI: I would just give docetaxel at 75 mg/m2.

DAVE: [I have used] docetaxel with ramucirumab [Cyramza].

NADLER: Have you done that frequently?

DAVE: I have done it just once.

NADLER: I think a taxane-based therapy, whether it be on trial or with Risk Evaluation and Mitigation Strategies or as a single agent, is the next logical thing to give a person who hasn’t seen a taxane yet if they have no other mutational data and progressed on [chemotherapy and] immunotherapy.

Adding ramucirumab to docetaxel is not an unreasonable thing to do whatsoever but it isn’t something on our pathway.

_REFERENCE

_{NCCN. Clinical practice guidelines in oncology. Non–small cell lung cancer; version 4.2021. Accessed May 24, 2021. https://bit.ly/3oIujwf}