In an interview, Martin F. Dietrich, MD, PhD, discussed his observations on the latest and upcoming trials that target biomarkers in earlier-stage non–small lung cancer.
Targeted Oncology: What do you feel are some of the biggest updates in precision medicine right now?
DIETRICH: We’ve seen 2 major developments. One is the broadening of molecular mutations in the metastatic setting. I think we’re refining mutations in the EGFR space. I believe we get a better understanding of prioritization of targeted mutations in the metastatic setting, like the newly published phase 3 LIBRETTO-431 trial [NCT04194944] that demonstrated superior survival with use of targeted therapy over chemoimmunotherapy.1
We’re also seeing a very exciting development in using targeted therapies in the curative intent setting. Post surgery, we’ve had the approval of the phase 3 ADAURA trial [NCT02511106] using osimertinib [Tagrisso] in the postresection space for EGFR-mutated disease.2 At the European Society for Medical Oncology [ESMO] Congress 2023, we’ve had an update on the phase 3 ALINA trial [NCT03456076] that demonstrated significant improvement in disease-free survival [DFS] with the use of adjuvant alectinib [Alecensa] following resection for the duration of 2 years.3
There’s a lot of refinement happening. To a certain extent, we’re repeating the lessons we learned from the metastatic space in biomarker settings. But I’m glad we’re bringing them into new and potentially more impactful indications.
How do you feel these updates have changed the way you sequence therapies for your patients?
The sequence of therapies overall is relatively similar. We do have some clinical trial designs that may or may not be as informative, but effective therapies sorted by their biomarkers are the most important, repetitive lesson that we’ve had. In the early stage, we’ve had a significant shift going from surgical resections to almost a universal leaning toward neoadjuvant approaches. We’ve had at least 5 perioperative trials and they demonstrated significant DFS or event-free survival in the pretreatment of surgical candidates. But these are equally biomarker sorted and it is very important that all the lessons we learned both about targeted therapies, the lack of effectiveness of immunotherapy, as well as the staggered benefit of immunotherapy with varying PD-L1 levels translate to a very similar extent into early-stage disease. We need to apply these biomarkers to the same stringency as we do in the metastatic setting.
Do you feel like this has put a large importance on testing with liquid biopsies?
We have a new challenge. We always were concerned about tissue availability and turnaround time in the metastatic setting. But now that we’re moving into the curative intent space, we have an additional challenge. In thinking about some of these options, we have to think about different concepts; I think we have to tailor this. Depending on the stage and early stage, the sensitivity of liquid biopsy may not be as useful, but…it’s certainly worth a try because we want to avoid any unnecessary delays. It’s a noninvasive test, often can be drawn based on the clinical diagnosis of lung cancer alone, and it can be complemented with a tissue biopsy. We still need histology for selection of chemotherapy as well as tailoring of the individual targeted therapies. If there’s a small specimen, I believe it would be reasonable to at least obtain the 3 most critical, actionable markers: EGFR, ALK, and PD-L1 by single-gene testing. However, ideally, you would get a broad molecular profile by next-generation sequencing [NGS] like we do in the metastatic disease setting to understand the full biology prior to mapping out a patient’s treatment path.
Do you run into any issues when you’re dealing with insurance when you’re sending that out?
I believe that many of the challenges we’re facing have shifted. I think administratively, reimbursement has gotten better. We are at the stage where lung cancer can be reflexively sent for NGS in all stages and all histologies. Our challenges mainly lie in the tissue utilization and in the distribution of the knowledge surrounding the need to wait for these molecular profiles, not only within medical oncology, but with our surgical and radiation colleagues as well because these therapies do influence the timing of their interventions just as much as they do ours. This is a new multidisciplinary approach more so than ever in lung cancer. Surgery has become much more involved in this multidisciplinary process, moving away from the typical sequence of surgery followed by adjuvant therapy. Now virtually every patient, [except for] those with maybe stage IA disease, receive a perioperative evaluation up front and likely qualifying for some degree of therapy. With the exception of EGFR and ALK disease, which are channeled down a different pathway of targeted therapy where we do not yet have neoadjuvant or perioperative, just adjuvant approaches with osimertinib and alectinib for EGFR and ALK alterations in early-stage lung cancer.
What would you recommend for physicians who are less familiar with lung cancer when they’re trying to help these patients?
We use very similar therapies in the metastatic setting and those we’ve moved into an earlier-line setting. It’s important to realize that the concepts are the same. We have to apply a sequence and encourage our colleagues to slow down on aggressive interventions like radiation and surgery until we have the overall picture understood well. I would read stage II and stage III disease...like a patient with stage IV. The reason is quite simple: Their biology is very similar and we’re dealing with widely systemic involvement of lung cancer. It’s a disease where we must assume high risk for metastatic manifestations; our survival rates would support that there’s no true lung cancer that does not require some of the degree of evaluation for systemic therapy. It’s very important to recognize that this has not just been a gradual but a tectonic shift to the use of perioperative therapies, and that the importance of biomarkers has now extended to every stage of lung cancer.
Do you ever see conflicting results when you send out for those biomarkers? If a patient has an EGFR mutation and PD-L1 expression, how do you pick between those?
That’s an additional layer of challenge. Sometimes we run into the problem of overlapping biomarkers that are seemingly actionable. Then we have to rely on our experience and our molecular genetics colleagues to help us prioritize a hierarchy of disease. You can have EGFR-or ALK-positive disease, which is by no means mutually exclusive from having PD-L1 positivity. Further, the labels for perioperative therapy do not even require some of these biomarker tests, but just because it’s not required doesn’t mean it’s not necessary. So, when we have PD-L1 positivity, especially in high levels, we still must recognize that EGFR and ALK biology predict for a very poor response to immunotherapy. In addition to a lack of efficacy, we also risk an exacerbation of toxicity in the sequence of immunotherapy prior to targeted therapy. The very firm recommendation here is to hold off on immunotherapy until at the very least EGFR and ALK mutations have been excluded. We do need PD-L1 at least for prognostication; there’s certainly an exponential benefit in the neoadjuvant or perioperative space with higher PD-L1 levels. Fifty percent of patients with a PD-L1 level greater than 50% have pathologic complete responses [pCRs]. That’s certainly a staggering number and the benefit is certainly very different than patients who are PD-L1 negative where we see limited to no benefit. It’s very important to establish the hierarchy of biomarkers and understand this from the very get-go, which is why PD-L1 stain shouldn’t be acted on. They come in quickly, often done in house. They truly tempt us to engage in a therapy that is only advisable once the complete biomarker setting is available.
Is there something that you would counsel others on from recent trends? Is there anything from ESMO and other conferences you think is standing out in terms of what’s coming for the future?
The trend toward perioperative therapies is clear. At least the neoadjuvant part is very clear; the benefit is massive. Sometimes when patients are eager to get started on treatment—we’ve done this in the metastatic setting as well— we give the first cycle of chemotherapy, and most patients with systemic disease concern will receive chemotherapy, get the first cycle of chemotherapy without immunotherapy, and then add it on once biomarker testing is completed. Even regarding EGFR- and ALK-positive disease, I believe chemotherapy will still play a role with a benefit in survival—a modest benefit, but certainly not discarded.
The trends are going to be shifting our practices. There has been an avalanche of data coming on since the introduction of CheckMate 816 [NCT02998528] with the neoadjuvant part. Moving our treatment pattern, practice pattern, and referral pattern into a multidisciplinary up-front evaluation similar to breast or rectal cancer—I believe that’s going to take some time for full implementation.
The other question is: What do I do with patients who do not achieve pCRs? We now have the option to continue with adjuvant immunotherapy. But I believe that more of the same is less likely to benefit a majority of these patients. And we have to think about alternative strategies of escalation. These could be switched immunotherapies, or combinations of immunotherapy plus different angles of chemotherapy. Similar like we’ve seen, for example, in the phase 3 KATHERINE trial [NCT01772472] in breast cancer, where the pCR as a validated surrogate end point aided us in selecting patients for escalation of therapy. I believe this will follow in lung cancer in very similar ways, that we truly need to be looking out for options for patients who did not achieve a pCR in their neoadjuvant setting to enhance the therapies in the adjuvant setting to optimize outcomes.
How do you know which patients are right for that kind of treatment?
What we’ve learned, and this was new in lung cancer and probably unexpected to most, is that the pCR rates are somewhere around 12-fold enhanced by the addition of immunotherapy. I would stratify patients by the depths of pathologic responses, with the pCRs being a positive outlier where I would not escalate therapy. There [are] currently no further data available to suggest augmentation, but certainly we see the higher risk for patients who have had lesser pathologic responses or even no pathologic responses. Then I believe a switch mechanism in the curative intent setting is critical. There will be a little bit of gray zoning, going from major pCRs then to the lesser responses.
Another focus that’s been a learning lesson consistently across the different trials is the PD-L1–negative population continues to pose a particular challenge for which we will need…,like we do in the metastatic setting, combination treatments, and the CTLA-4 mechanism with ipilimumab [Yervoy] and tremelimumab [Imjudo] have shown some sensitization in the metastatic setting. I would expect to see a similar picture in the neoadjuvant setting. There’s no biological difference in the tumor microenvironment. I believe this might be the first approach to enhance neoadjuvant therapy based on selection, but those trial data results are not ready yet for implementation in the clinical setting.
What is coming down the pipeline for patients with NSCLC?
We’ll see a refinement by biomarkers. I’ll say we’ll see a relatively quick arrival of some of our new chemotherapies via antibody-drug conjugates in the curative intent setting. I think this is going to be interesting for 2 reasons. One is the density of drug delivery, as well as the previously unrecognized mechanism of action. I’m excited about seeing these trials coming to clinic and patients actively participating. It’s a big knowledge gap that we need to close for treatment. The core message is to catch up clinical practice to the status of clinical evidence right now, which would include universal testing of all patients with lung cancer independent of stage, independent of histology, and then implementing multidisciplinary [standards], whether it is formalized in a tumor board setting or through careful care coordination into clinical practice of all settings.
Just because a tumor is resectable doesn’t mean that it should be resected without the consideration for up-front treatment with a systemic immunotherapy. The reason is quite clear. It is simply a matter of biology that the interaction between tumor cell and tumor resident lymphocytes is one that requires a tumor microenvironment. If you resect this tumor microenvironment, you probably lose the sparring ground. We’ve seen this in other tumor types like melanoma, that the neoadjuvant approaches vs adjuvant approaches demonstrate impressive HRs of 0.5. We see in our cross-trial comparisons very similar discrepancies between the neoadjuvant approaches independent of agent and the adjuvant approaches in the phase 3 PEARL [NCT03003962] and IMpower010 [NCT02486718] trials.
There are still patients who were thought to have early-stage disease, and they were discovered to be higher stage or higher risk during initial therapy where the adjuvant paradigm still applies. But I believe we’ll see a massive shift toward neoadjuvant therapy. I think this will facilitate surgery; we do not see any major delays.
We want to make sure that patients are selected properly, and then to keep in mind also that for stage III disease, radiation therapy still has a massive overall survival benefit with the use of immunotherapy and hopefully soon with the adjuvant use of targeted therapy as well. There are still plenty of options to consider. But it’s getting more patient-specific care for each treatment approach. Are there any tools that you use when you’re making these decisions or when you’re deciding to enroll a patient in a clinical trial?
The most pertinent clinical question up front is the selection by biomarkers. We know what the subgroups [are] who have a major benefit and we know those that are particularly challenging, and I believe those are where we need to focus our clinical trial efforts.
In the adjuvant setting, I believe it’s a risk-adaptive strategy. A patient who has had no significant pathologic responses should be treated with additional angles of therapy. That’s where clinical trials are critically important. But I believe if we utilize EGFR, ALK, and PD-L1, and now maybe by extension other biomarkers like RET as good guidance for treatment selection, we’ll be able to enhance outcomes in early-stage lung cancer and shift our survival curves in a favorable direction.
What are those targets that you feel need some more attention or work?
We have basically 2 columns of biomarkers. One is the EGFR/ALK column that we know is associated with less smoking, less mutational burden, less inflammatory activity, and therefore not good candidates for immunotherapy. Then we have the KRAS column where we know we see the biggest benefit of therapy, typically associated with smoking, higher inflammatory activity, and tumor mutational burden.
But we have a lot of biomarkers in between, and we don’t know what to do with them yet. RET [inhibition] is one that now closely aligns with EGFR and ALK, even though we don’t have those data in the early-line setting. And then there are many others about which we still need more information and more work. For some of them, like the HER2 mutations, for example, we don’t have any targeted therapy options yet. We have some agents like [zongertinib] in development that are showing promise to look very much like the EGFR and ALK inhibitor experience.
But there is still need for agents and this is why targeted therapy in the adjuvant setting is so attractive. They’re not only highly effective, but they’re also very tolerable and provide a very good risk-benefit ratio for the patient. My impression is that we need to move on the tailoring of immunotherapy. PD-L1 serves some patients really well, and many patients not well at all.
Then there are the improvements in the targeted therapy options that are available for our actionable mutations that would help us and make decisions in a similar direction like we do with adjuvant osimertinib and alectinib, where we clearly see good tolerance for many years and patients without any additional long-term toxicities. This is going to be the main part; matching what we know to what we’re doing is the next quest of this wonderful development of perioperative therapies.
1. Zhou C, Solomon B, Loong HH, et al; LIBRETTO-431 Trial Investigators. First-line selpercatinib or chemotherapy and pembrolizumab in RET fusion-positive NSCLC. N Engl J Med. 2023;389(20):1839-1850. doi:10.1056/NEJMoa2309457
2. Herbst RS, Wu YL, John T, et al. Adjuvant osimertinib for resected EGFR-mutated stage IB-IIIA non–small cell lung cancer: updated results from the phase III randomized ADAURA trial. J Clin Oncol. 2023;41(10):1830-1840. doi:10.1200/JCO.22.02186
3. Solomon BJ, Ahn JS, Dziadziuszko R, et al. LBA2 – ALINA: Efficacy and safety of adjuvant alectinib versus chemotherapy in patients with early-stage ALK+ non–small cell lung cancer (NSCLC). Ann Oncol. 2023;34(suppl 2):S1254-S1335. doi:10.1016/annonc/annonc1358