David P. Carbone, MD, PhD, discusses the evolution of immunotherapy and targeted therapy in non–small cell lung cancer and the importance of biomarker testing for personalized therapy.
The availability of chemoimmunotherapy regimens and dual checkpoint inhibition underscore the advances that have been made in immunotherapy options for patients with non–small cell lung cancer (NSCLC). Although these therapies have helped transform the treatment space, there is still potential for further improvement, according to David P. Carbone, MD, PhD.
“Immunotherapy has revolutionized the treatment of NSCLC, and, to a lesser degree, small cell lung cancer [SCLC]. We’re getting long-term survivors that we’ve never seen before. [However,] the therapy is still far from perfect,” Carbone said in an interview with OncLive® during an Institutional Perspectives in Cancer (IPC) webinar on lung cancer, which he chaired.
“There’s a lot of room for improvement, with better matching of therapies, selecting [the right amount of therapy], and deciding who should get immunotherapy alone, combination immunotherapy, and combination chemoimmunotherapy,” Carbone added.
In the interview, Carbone discussed the focus of each presentation, spanning the evolution of immunotherapy and targeted therapy in NSCLC and the importance of biomarker testing for personalized therapy. Carbone is a professor of internal medicine at The Ohio State University, the Barbara J. Bonner Chair in Lung Cancer Research, co-leader of the Translational Therapeutics Program, and director of the Thoracic Oncology Center at The Ohio State University Comprehensive Cancer Center–James (OSUCCC–James).
Carbone: [Immunotherapy] represents a new pillar of therapy available for treating patients with lung cancer. It started out in the second- and third-line [settings], and remarkably, it was better than available chemotherapies. It resulted in durable survival benefit, [extending several years].
These results appropriately moved [immunotherapy] to the frontline setting, in which it was shown to beat chemotherapy alone in many phase 3 trials. [Immunotherapy] is now the standard of care in the first line for non-driver mutant tumors. It is even being moved into early-stage disease in the perioperative setting.
[Pembrolizumab] totally transformed the landscape. Those were the definitive trials looking at immunotherapy in the first line, and they were all strongly positive studies.
I did point out that KEYNOTE-407 was less impressively positive, especially in [patients with] PD-L1–negative squamous [disease]. [However,] in the subset analysis, and in the overall analysis, [pembrolizumab] was still better than chemotherapy.
In adenocarcinomas, [PD-L1 status] may not be as important, in that chemoimmunotherapy is useful for all PD-L1 subsets. In my personal practice, in [patients with] high PD-L1, I use immunotherapy alone, without chemotherapy. In that respect, [PD-L1 status] important [when deciding to use pembrolizumab alone] in patients with adenocarcinoma and nonsquamous lung cancer.
In squamous [lung cancer], in KEYNOTE-407, [pembrolizumab plus chemotherapy] seemed to have an effect on patients with [high] PD-L1. In [patients who were] PD-L1 negative, the survival curves crossed, and there was no difference in the landmark survival past 2 years, which is disappointing because that’s about one-third of the squamous patients. Even though squamous cell lung cancer is a smaller subset than nonsquamous, it’s still a common diagnosis.
In a subset analysis, the combination of ipilimumab [Yervoy] and nivolumab [Opdivo] showed a respectable threefold improvement in landmark survivals in PD-L1–negative squamous cell lung cancer. Unfortunately, it’s not FDA approved in that setting. The same regimen with 2 cycles of chemotherapy is referred to as the [CheckMate 9LA (NCT03215706)] regimen, and that combination is approved for PD-L1–negative squamous cell lung cancer.
The benefit of the 2 cycles of chemotherapy is not entirely clear compared with the doublet immunotherapy regimen alone. The usage of that combination is not common. However, even with all those data, chemotherapy plus pembrolizumab is more frequently used.
[Initial data] from the phase 3 CheckMate 816 trial [NCT02998528] showed remarkably high complete pathologic response rates with nivolumab and chemotherapy compared with chemotherapy alone. Those data are likely to transform the management of early-stage lung cancer.
Now we have approvals in the adjuvant and neoadjuvant settings with immunotherapy. They haven’t been compared with each other [in a head-to-head trial], so we’re forced to make cross-trial comparisons. In theory, the neoadjuvant approach may be better than the adjuvant approach. We will have to see what the data pan out to be and what the overall survival [OS] outcomes are down the road, but immunotherapy is here to stay in early-stage lung cancer.
Brain metastasis is a huge problem in lung cancer. Most patients have brain metastases at some point in their [disease]. [Patients with brain metastases] have been a historically difficult-to-treat subset and have done worse than other subsets with chemotherapy. [Notably], the management with stereotactic radiation has aided that field.
We did a subset analysis of CheckMate 9LA looking at the outcomes of patients with and without brain metastases. Remarkably, the chemotherapy control arm did far worse in patients with baseline brain metastases than patients without brain metastases, [meaning the addition of nivolumab and ipilimumab generated significant activity in these patients]. In the CheckMate 9LA treatment arm, the patients did about the same, whether they had brain metastases or not, in terms of OS. That is remarkable and implies significant brain activity [with nivolumab and ipilimumab plus chemotherapy].
Other parameters were impressive, [including] a doubling of the time to central nervous system [CNS] progression and the improvement in intracranial response rate with [nivolumab and ipilimumab plus chemotherapy]. There are signals that the combination may be active in the CNS. That has also been shown in melanoma, as well, so the results are consistent.
Biomarker testing, specifically testing for genetic drivers, is a crucial part of the workup for patients, especially those with nonsquamous lung cancer. The treatment and the optimal sequencing of therapies are completely different. The efficacy of immunotherapy in [patients with nonsquamous lung cancer] is low, except for the KRAS population, which seems to be about the same. The subset of BRAF-mutant tumors in patients who have smoked may respond to immunotherapy as well.
For the common drivers—EGFR, ALK, RET—those patients do better with first-line targeted therapies and may have more [adverse] effects [AEs] if [treatment] starts with immunotherapy and switches to a targeted therapy.
Many of the targeted therapies get into the CNS well. If you have a patient who has multiple asymptomatic brain metastases, especially in a non-smoker with a high likelihood of a driver mutation, we should do everything we can to avoid whole brain radiation until the genetics come back, because these patients are often effectively treated with systemic oral therapy and may completely avoid whole brain radiation. [These patients] may live for years on these targeted therapies, long enough to manifest the AEs and the long-term consequences of whole brain radiation. Doing reflexive whole brain radiation with multiple brain metastases in a [patient who has not undergone genotyping] is a huge mistake.
Some of the early drugs in that space, [including] gefitinib [Iressa], erlotinib [Tarceva], and afatinib [Gilotrif], had dramatic efficacy, but they were hampered by AEs, especially afatinib. Most patients couldn’t tolerate the recommended dose of afatinib, and none of them got into the CNS very well, [which was problematic because] EGFR-mutant lung cancer often goes to the brain. Osimertinib is a selective inhibitor that had little EGFR wild-type effect. The toxicity was much less [than earlier-generation agents], and it got into the brain much better. It has [become] our standard of care in this space.
In a similar way to EGFR, ALK fusion–positive disease is effectively treated with the newer-generation drugs. Again, some of the older drugs, [such as] crizotinib [Xalkori] and ceritinib [Zykadia], had some activity. However, they were toxic, and the activity didn’t last very long and didn’t get into the brain well. Alectinib [Alecensa] gets into the brain well, has low toxicity, and recent data show a 65% 5-year OS rate. This is remarkable in metastatic disease when you consider that, historically, patients with metastatic disease had a 4- to 6-month survival. Now we’re talking about more than half of patients [being] alive at 5 years. That’s a major difference.
The activity [of alectinib] in the brain, similar to osimertinib, was a game changer. Half of patients treated with crizotinib would relapse in the brain within 2 years, whereas it’s only 9% with alectinib. Some of the other agents in the field, [such as] brigatinib [Alunbrig] and lorlatinib [Lorbrena], have activity against some of the acquired resistance mutations to alectinib. We are doing resistance genotyping on these patients’ tumors and picking second-line agents by virtue of the type of resistance mutation. These are patients where you want to avoid up-front whole brain radiation because of the efficacy of the newer-generation drugs in the brain.
When I was in training, SCLC was [previously] the success story for lung cancer. We had an active regimen that resulted in responses in most patients, and survival went from 6 weeks to 12 months, and we thought that was great. Now, SCLC is the stepchild in lung cancer. NSCLC advances have outpaced those in SCLC.
We have some intriguing new therapies on the horizon. The addition of immunotherapy to chemotherapy in extensive-stage disease has made a definite advancement in the outcomes of patients. Better radiation therapy in limited-stage disease is improving outcomes in those patients.
We used to have common high-grade toxicities from radiation that now are uncommon because of modern treatment-planning approaches. There are also new agents targeting various small-cell markers. For example, [the phase 2 TRINITY trial (NCT02674568)] used an antibody-drug conjugate [ADC] targeting DLL3, rovalpituzumab tesirine, that was not a huge success, mostly because of toxicity. There was a hint of activity, and now there are new ADCs and bispecific antibodies targeting DLL3 that are showing significant efficacy and may be part of our standard of care down the road.
There are other targets in SCLC that are being investigated. We are making progress in SCLC, but there’s room for improvement.
Lurbinectedin is a chemotherapy, and it was found to be active in the second-line [setting] in SCLC therapy and represents a treatment option for our patients. We are now routinely using it in the clinic, and we’re seeing significant activity.
It’s not ideal. It has some toxicity, and the patients often progress after a few months. But in this population, any advance is welcome. Lurbinectedin is clearly an advance in chemotherapy for SCLC.