Stephen V. Liu, MD, highlighted the importance of next-generation sequencing in lung cancer, key trials that have shifted the non–small cell lung cancer treatment landscape, and detailed ongoing research at Georgetown University Lombardi Comprehensive Cancer Center.
Immunotherapy may be the key to unlocking durable responses in patients with non–small cell lung cancer (NSCLC) regardless of whether the cancer is oncogenically driven, explained Stephen V. Liu, MD, who added that the sequence of treatment, particularly for patients with targetable alterations will be a critical element in delivering optimal care.
“Immunotherapy has to be our long-term goal because it's how we [can] achieve durable responses and can circumvent or avoid resistance,” Liu said in an interview with OncLive® following a State of the Science Summit™ on lung cancer, which he cochaired. “However, there are a lot of challenges in the way. Ongoing trials will take a big step in helping not just establish new options but establish a true understanding of what's happening at the cellular level.”
In the interview, Liu highlighted the importance of next-generation sequencing (NGS) in lung cancer, key trials that have shifted the NSCLC treatment landscape, and detailed ongoing research at Georgetown University Lombardi Comprehensive Cancer Center. Liu is an associate professor of medicine at Georgetown University, as well as the director of Thoracic Oncology and head of Developmental Therapeutics at Georgetown Lombardi Comprehensive Cancer Center in Washington, DC.
Liu: Dr Reuss discussed the evolving role of immunotherapy in resectable NSCLC. [Studies within the space include] the phase 3 IMpower010 trial [NCT02486718] and the phase 3 PEARLS/KEYNOTE-091 study [NCT02504372], which [evaluated] adjuvant immunotherapy in the form of either a PD-L1 or a PD-1 inhibitor after surgery. The phase 3 CheckMate 816 trial [NCT02998528] evaluated neoadjuvant chemoimmunotherapy. [Additionally], neoadjuvant [as well as] adjuvant [chemoimmunotherapy] were given in the phase 3 AEGEAN trial [NCT03800134] and the phase 3 KEYNOTE-671 trial [NCT03425643], among others. These approaches have produced tremendous improvement in outcomes with the addition of immunotherapy around the time of surgery. Clinicians see pathologic complete responses at the time of surgery. Furthermore, we see very durable responses and impressive event-free survival. Clearly, this is a major advance. However, as with all our treatments in lung cancer, the key is delivering it to the right patient population.
Investigators are looking at markers that predict responses, such as PD-L1 expression and different types of gene signatures, but just as important are markers that predict lack of response. For example, in patients whose tumor harbors a sensitizing EGFR mutation, we would not expect as much benefit from immunotherapy [as we would with targeted therapy], and we have better options [for them]. The phase 3 ADAURA study [NCT02511106] shows that giving adjuvant osimertinib [Tagrisso] to a patient with EGFR-mutant NSCLC improves disease-free survival [DFS] and prevents the spread of cancer to the brain. Moreover, we learned that the benefit extends to overall survival [OS]; those data [were] presented at the 2023 ASCO Annual Meeting.* Clearly, that [approach] is a good option. However, we need to identify those patients.
If a patient has already had surgery, although we're giving adjuvant chemotherapy, for example, we have [a lot of time] to do [genomic] sequencing to get full panels back. The challenge is in that neoadjuvant setting because we learned from our experience in the stage IV setting that we need the biomarker results before we start immunotherapy. If we're considering someone for perioperative immunotherapy, it's important to ensure that the patient does not have an EGFR-mutant lung cancer, because by giving neoadjuvant chemotherapy, we may be giving the wrong treatment, making it more challenging for us to deliver the right treatment. That extends not just to EGFR mutations, but to ALK, RET, and other actionable drivers. Therefore, we need NGS and that is challenging because tissue samples are smaller, and time is more precious as we're moving forward.
At our institution, we sometimes give 1 cycle of neoadjuvant chemotherapy alone as we're waiting for those results, and this is a space where liquid biopsy may not help too much because we often won't see tumor DNA being shed in the earlier stages. We need large samples from diagnostic biopsies, and we need to run NGS very quickly. Time is of the essence in this setting, more than any other.
Dr Brahmer discussed rising immunotherapy agents, and we've had several new additions to the field. For years, we've been comfortable using PD-1 inhibitors like pembrolizumab [Keytruda] and nivolumab [Opdivo], and PD-L1 inhibitors, like atezolizumab [Tecentriq]. [Additionally], we've recently seen multiple ways to deliver cemiplimab-rwlc [Libtayo], a very potent PD-1 inhibitor with data that seemed very compelling. Most recently, the combination of durvalumab [Imfinzi] and tremelimumab-actl [Imjudo] in the phase 3 POSEIDON [NCT03164616] regimen. We're left with multiple different regimens, and the agents all seem active, and they do seem comparable. The key is always going to be patient selection, and when to deliver chemotherapy. Similarly, it is important to consider when it is advantageous to add CTLA-4 inhibition.
We saw some data from the POSEIDON study looking at STK11 mutations, and some [data within] the context of KRAS mutations. [The trial addresses whether the] addition of a CTLA-4 inhibitor will convert more patients to long-term survivors, [which is our goal]. Getting an initial response is nice, but what we want is long-term, durable, meaningful survival. We achieve that in a subset of patients, but we want that subset to be the majority. We're not there yet. Are we going to get there by inventing new and more potent immunotherapy agents? Maybe. A closer more immediate step is better patient selection, understanding which patients are getting that maximal benefit, and using the drugs we have in the most optimal way.
Dr Scott discussed the rapid evolution seen in KRAS. The targeted drugs we have approved, such as sotorasib [Lumakras] and adagrasib [Krazati], are specifically approved for KRAS G12C. The G12C is critical, that cysteine is necessary for the binding of these agents. KRAS G12D, KRAS G12V are important subsets but those are areas where sotorasib and adagrasib are not appropriate. Newer drugs are coming along to fill those needs. However, the big question that remains is how to manage this subset in the first line. The current standard of care is immunotherapy, and we [administer] immunotherapy in KRAS G12C differently than in EGFR and ALK due to the response to immunotherapy. Some patients will have long-term durable responses to immunotherapy and that's not something we want to gamble with. We don't want to have someone miss out on the opportunity for long-term control with immunotherapy. All patients deserve the opportunity for immunotherapy.
This raises the question of whether there is a subset that we can identify up front where targeted therapy may be better, or whether there is a way we can combine targeted therapy with immunotherapy. Early studies looking at sotorasib with checkpoint inhibition seemed too toxic. There's some question as to whether it is the class or whether it is the molecule. From the phase 2/3 KRYSTAL-7 [NCT04613596] data, there was a suggestion that it may be better tolerated with adagrasib and that's still largely unknown. These are small studies but we're trying to understand if we can deliver a combination of a KRAS G12C inhibitor and immunotherapy in the frontline to make immunotherapy work better. Will it be a way to extend that long-term effort to more people? This is a big unanswered question.
The next unanswered question is how we move our response rates from the 30% to 40% range in the second-line setting up to the majority of patients. KRAS G12C is extremely heterogeneous. The rate of comutations is very high and previous work shows us that those comutations are very important. We must identify who is going to respond to a KRAS G12C inhibitor alone, and who is likely to need a combination. Many groups are putting us well on the path to getting closer to achieving that goal.
Dr Kim discussed the evolving role of EGFR inhibition in earlier stage lung cancer. We're familiar with the phase 3 ADAURA study [NCT02511106], which demonstrated that treatment with adjuvant osimertinib improves DFS and OS. It is important that we identify patients with a resected, EGFR-mutant NSCLC and deliver osimertinib. However, there's still a lot of questions. Although we see that the DFS benefit is profound, the [optimal] duration of treatment is largely unknown. The study used 3 years, and other trials show that 2 years [of adjuvant therapy] was not effective. Is 3 years the optimal duration or do we need longer? A better approach [is] identifying who needs treatment at all because some people will be cured with surgery and chemotherapy. Some patients are not, and those who are not may need a longer course of osimertinib. Those are big questions that we don't have the answer to.
What we really are waiting for is a minimal residual disease marker that will be different from the currently available commercial ctDNA platforms, something a little more specific. There are a lot of candidate platforms under investigation. This is a setting where we may have to wait for the technology to catch up. In the meantime, it's important that we test our patients with resected lung cancer for EGFR mutations to see if they're candidates for this very powerful therapy and keep an eye out for upcoming studies looking at ALK or RET inhibition in similar settings. Will we see the same types of benefits? I suspect so, but we're waiting for those data.
Dr Benyounes gave a wonderful talk on EGFR exon 20 insertion mutations, pointing out that if we're using polymerase chain reaction–based testing for EGFR mutations, we're going to miss a lot of EGFR exon 20 insertions. These are important not to miss, as they generally don't respond well to immunotherapy. We know patients can respond to some of the newer targeted agents. The historic EGFR inhibitors, such as erlotinib [Tarceva], gefitinib [Iressa], or osimertinib, don't seem to work as well in an EGFR exon 20 insertion because of where that C-helix is, blocking access to the pocket. However, newer drugs including mobocertinib [Exkivity], an oral EGFR kinase inhibitor, and amivantamab [Rybrevant], a bispecific antibody targeting MET and EGFR have shown activity and are now approved [in that population].
We have studies now looking to move these [agents] into the frontline space and that's a big question. Currently, they're FDA approved for previously treated [patients], but it is not known if we should use these approaches up front. Our response rates are around 40% and we'd like that number to be higher. Part of it may be specific mutations. As we know, EGFR exon 20 insertions are not 1 mutation, but rather a family. We are waiting for the results from ongoing studies looking at more potent and novel agents, which hopefully will help advance the field. In the meantime, amivantamab and mobocertinib are 2 competing standards of care for EGFR exon 20 insertion NSCLC. Both present with relative pros and cons, both are active in this space, and important drugs to be sure our patients have access to. Again, the key here is to test if we're going to treat [patients] properly.
Dr Choudhury provided a nice review on SCLC, focusing on the extensive-stage setting where our frontline standard of care is chemoimmunotherapy with carboplatin, etoposide, and atezolizumab, based on the phase 3 IMpower133 [NCT02763579] regimen, or platinum, etoposide, and durvalumab, based on data from the phase 3 CASPIAN trial [NCT03043872]. Both [regimens] have been approved in the frontline setting for several years now. We do know these [agents] have improved outcomes. We do have a subset of patients that have long-term survival, but most patients unfortunately will relapse and need other therapies.
Looking to the horizon, lurbinectedin [Zepzelca] has FDA accelerated approval in the second-line space with activity in both patients with [platinum-]sensitive and [platinum-]refractory relapse. An important study, the phase 3 IMforte trial [NCT05091567], is looking at moving lurbinectedin up into the maintenance setting. Is it better to prevent than to react? We await results from that important phase 3 trial.
Looking ahead at newer active agents, Dr Choudhury highlighted tarlatamab [AMG 757], a bispecific T-cell engager that brings together DLL3 and CD3, trying to induce immune responses. Regarding toxicity, there are [incidences] of cytokine release syndrome, and the in-patient administration is a barrier to delivery. However, ongoing studies hopefully will show the true benefit of that [agent] in a large population. DS-7300, an antibody-drug conjugate [ADC] has shown a lot of activity. This is a drug targeting B7-H3, which is highly expressed on neuroendocrine tumors, including SCLC. It's shown impressive activity, though in a small data set.
We're looking for larger studies and there are other very interesting ADCs also being delivered in this space. We are still trying to figure out how to leverage the different transcriptional subsets of SCLC and we'll stay tuned for some data looking at the delivery of PARP inhibitors. For now, the best option in the relapsed setting is a clinical trial. It’s important that we refer patients to get access to these new and exciting drugs.
One of the studies ongoing at Georgetown Lombardi I'd like to highlight is an investigator-initiated trial written by Joshua Reuss, MD, looking at bringing immunotherapy into subsets where it hasn't been effective. This is a single-arm prospective study that's looking at the combination of atezolizumab, a PD-L1 inhibitor, bevacizumab [Avastin], a VEGF antibody, and tiragolumab, an anti-TIGIT antibody, in patients with EGFR-mutant NSCLC. We know immunotherapy has been disappointing in EGFR-mutant lung cancer, but we do see responses. Perhaps we haven't unlocked the proper combination. There have been multiple signals, both with atezolizumab and sintilimab that VEGF inhibition might be important in inducing an immune response in EGFR-mutant lung cancer. Will tiragolumab be that extra push to get more durable responses? This trial includes patients with EGFR-mutant lung cancer, who have already exhausted tyrosine kinase inhibitor [TKI] options. We know that [treatment] sequencing is important, and we would never recommend immunotherapy as an initial treatment [for this population]. However, after TKIs have stopped being effective, this combination will hopefully induce responses.
Importantly, biopsies done before and after starting the immunotherapy regimen will demonstrate the impact of this treatment on the microenvironment within the tumor, highlighting what those barriers are, and how we can overcome them.
Editor’s Note: This interview took place prior to the 2023 ASCO Annual Meeting. To read more about the OS results from the ADAURA trial, visit: https://www.onclive.com/view/adjuvant-osimertinib-provides-os-benefit-in-egfr-stage-ib-iiia-nsclc