*June 2025*
Key Takeaways
- In recent research, a range of therapies showed favorable efficacy and safety results in patients with epidermal growth factor receptor (EGFR)-mutant advanced non-small cell lung cancer (NSCLC) after disease progression on EGFR tyrosine kinase inhibitors (TKIs) and/or chemotherapy.
- Among patients with EGFR-mutant and mesenchymal epithelial transition factor (MET)-amplification advanced NSCLC after disease progression on EGFR TKIs, savolitinib plus osimertinib was associated with a greater improvement in progression-free survival (PFS) compared to chemotherapy in the phase 3 SACHI trial.
- In a subset of patients from the SAVANNAH trial with EGFR-mutated advanced NSCLC with MET overexpression and/or amplification after disease progression on first-line osimertinib, savolitinib plus osimertinib showed greater efficacy and central nervous system (CNS) benefit vs savolitinib plus placebo.
- In the phase 3 MARIPOSA-2 trial, a longer median PFS was observed with amivantamab plus platinum-based chemotherapy vs chemotherapy alone in patients with EGFR-mutant advanced NSCLC with disease progression after osimertinib, across baseline resistance subgroups.
- The phase 2b REZILIENT1 study demonstrated promising activity and manageable safety with zipalertinib in patients with EGFR exon 20 insertion-mutant NSCLC previously treated with platinum-based chemotherapy, including those with prior amivantamab exposure.
The 2025 American Society of Clinical Oncology (ASCO) Annual Meeting, which took place from May 30 to June 3 in Chicago, Illinois, featured numerous studies focused on treatment advances for patients with NSCLC. Among the research presented on therapies specifically for EGFR-mutant advanced NSCLC, findings highlighted emerging strategies, confirmed current standards of care, and showed promise for novel therapies.
Julia Rotow, MD, is the clinical research director of the Lowe Center for Thoracic Oncology at Dana-Farber Cancer Institute in Boston, Massachusetts, and assistant professor of medicine at Harvard Medical School. As a clinician and researcher, her main area of interest is the treatment of oncogene-driven lung cancer, particularly in patients with non-smoking associated driver mutations such as EGFR and human epidermal growth factor receptor 2 (HER2)-activating mutations. In this article, Dr Rotow discusses the results of multiple studies from ASCO 2025 that focused on treatment strategies for patients with EGFR-mutant advanced NSCLC.
Various studies presented at ASCO 2025 showed promising efficacy and safety results with a range of therapies in patients with EGFR-mutant advanced NSCLC after disease progression on EGFR TKIs and/or chemotherapy.1-3 Can you broadly discuss this area of significant clinical need and the current treatment strategies applied in such cases?
This certainly is an area of unmet need for patients with EGFR-activating mutations with NSCLC. Patients can do extremely well on initial EGFR-directed TKI therapy — sometimes the response is measured in years. Patients usually receive such therapy alone or in a combination regimen in the frontline setting. But we know that resistance inevitably develops. Historically, when patients develop resistance, particularly after having a TKI and after having our initial standard platinum-based doublet chemotherapy, treatment options become more limited and are often limited to later-line, single-agent chemotherapy. Unfortunately, such treatment has modest clinical activity and does not reach the kinds of efficacy we want for this often very young and very healthy patient population.
Fortunately, in the past year or so, and especially at ASCO 2025, we have seen the emergence of not just multiple options in the frontline setting, but also the emergence of multiple treatment strategies we hope will offer efficacy and improve outcomes for these patients once they have acquired resistance to their TKI or TKI plus chemotherapy.
Current themes of emerging strategies include all-comer treatment strategies for all patients with treatment-resistant EGFR-mutant lung cancer and biomarker-directed strategies. The latter have been oriented around other subsets of antibody drug conjugates (ADCs) or other targeted therapies for patients with specific acquired resistance mutations.
In the randomized phase 3 SACHI study (ClinicalTrials.gov identifier: NCT05015608), savolitinib plus osimertinib demonstrated a significant PFS improvement vs chemotherapy in patients with EGFR-mutant and MET-amplification advanced NSCLC after disease progression on EGFR TKIs, and the combination was safe and well-tolerated.1 Can you discuss the efficacy and safety results observed in this study and the anticipated next steps regarding this potential new treatment combination?
A subset of patients who develop resistance to EGFR inhibitors will have identifiable mechanisms of resistance driven by MET activation — either MET amplification or MET overexpression — and an active area of study has focused on identifying the best strategies to target this acquired resistance. One of the strategies is the addition of a MET TKI to the EGFR TKI backbone.
The SACHI study evaluated the combination of the MET TKI savolitinib with continued osimertinib in patients who had previously progressed on a third-generation EGFR TKI, such as osimertinib. This was a randomized phase 3 study that compared this combination regimen to standard platinum-based doublet chemotherapy alone,1 which would otherwise, in many global settings, be the standard of care for these patients. Standard platinum-based doublet chemotherapy is very much a non-biomarker directed, all-comer strategy.
The SACHI study was an international study that did not have enrollment from the United States, and it used MET amplification as its biomarker. MET amplification is a gene-based biomarker, not an immunohistochemistry (IHC)-based biomarker. What researchers found was that the EGFR TKI and MET TKI combination was significantly more effective than standard platinum doublet chemotherapy for this biomarker-selected population, with a PFS of 8.2 months vs 4.5 months.1 This is meaningful because it gives patients an oral TKI option potentially at acquired resistance, and it is an example of using a biomarker match strategy with a meaningful durability of response. A lot of concerns and questions have arisen over the years with biomarker-directed treatments. For example, is this tumor heterogeneous at this point, and are you going to pick off 1 subclone but not control everything? Here, there was a clinically meaningful duration of effect.
Another important point that came out of ASCO and that we have seen reported before, is that it is important to do the EGFR TKI plus MET TKI combination, not MET TKI alone. You need to cover both the original EGFR clone and the acquired MET-resistant clone. That is consistent with prior published data in the field looking at a different MET TKI, tepotinib, so it is consistent with our understanding of the biology of the disease.4
Because the SACHI study had only international enrollment, a more global study incorporating US patient populations is needed. Another ongoing phase 3 study called the SAFFRON study (ClinicalTrials.gov identifier: NCT0526139) is comparing the same combination — osimertinib plus savolitinib — against platinum-based doublet chemotherapy using a slightly different but similar MET biomarker in a more global population.2 We are awaiting the results of that randomized study in hopes for broader availability of this combination in the US, but in the meantime, tepotinib also has published data — a different MET TKI that is indeed available right now for other indications — and we have used it off-label in these sorts of combinations.4
Levy et al analyzed data from a subset of patients from the SAVANNAH trial (ClinicalTrials.gov identifier: NCT03778229) with EGFR-mutated advanced NSCLC with MET overexpression and/or amplification after disease progression on first-line osimertinib, including patients with asymptomatic stable brain metastases. According to their findings, savolitinib plus osimertinib demonstrated greater efficacy and more favorable CNS activity compared to savolitinib plus placebo.2 Based on these results, what are the implications regarding strategies to overcome treatment resistance and reduce the risk for new brain lesions in this patient population?
The SAVANNAH study was the other savolitinib plus osimertinib study reported at ASCO. This was a slightly smaller phase 2 study, and it studied a similar patient population — people with acquired resistance to the EGFR TKI osimertinib and EGFR-mutant lung cancer with MET-driven resistance identifiable on repeat biomarker testing. However, there were some key differences. The SAVANNAH study used a slightly different biomarker for MET-driven resistance. It used MET amplification similar to the SACHI study, but it also used MET overexpression, so it is protein level overexpression as an acceptable biomarker, and it was also a different control arm. The study looked at the osimertinib plus savolitinib combination, but compared it to savolitinib (a MET TKI) alone plus placebo.2
Here, the results were more distinct, which is not unexpected. The study showed a greater difference in response rate (58% vs 16%), a slightly larger magnitude gain in PFS (8.3 vs 3.6 months), and a slightly lower hazard ratio.2 That is not surprising because, unfortunately, the savolitinib plus placebo arm was likely only controlling the MET-driven subclones but not the overall driver clone of EGFR-driven disease that may not have MET-driven resistance. So the study showed the superiority of continuing the EGFR TKI backbone post-progression when we are doing these sorts of resistance-targeted strategies.2 Again, that is consistent with data for tepotinib, where response rates were better if you continued the TKI backbone with osimertinib.4 This is consistent with our knowledge of the field with these resistance-targeted strategies.
The SAVANNAH study also presented data looking at CNS response in a small subset of patients, and once again, there was a better CNS response rate in those patients who had the continued osimertinib post-progression vs the MET TKI alone.2 This is also not surprising because we do not know whether those CNS clones were MET-driven resistant. I think 1 of the strongest arguments for the EGFR backbone post-progression has been continuing to maintain control of known or occult CNS disease in this setting. That is also being shown, although in small numbers, in the SAVANNAH study, and it is consistent with our understanding of the field.
In the MARIPOSA-2 trial (ClinicalTrials.gov identifier: NCT04988295), the addition of amivantamab to platinum-based chemotherapy significantly improved median PFS compared to chemotherapy alone in patients with EGFR-mutant advanced NSCLC whose disease progressed on osimertinib, regardless of the underlying resistance mechanism.3 Given these results, what is the clinical impact of combining amivantamab with chemotherapy in the post-osimertinib setting, and how might this strategy influence treatment sequencing for patients with EGFR-mutant NSCLC?
This has been our standard for a while now, but I think it is relevant to mention in the context of next-line options. MARIPOSA-2 evaluated amivantamab (which is an EGFR–MET bispecific antibody) plus chemotherapy vs chemotherapy alone at acquired resistance to first-line EGFR TKI therapy with osimertinib.5 The study showed that adding amivantamab was more effective than standard chemotherapy alone, both as measured by PFS and response rates to therapy.3
This strategy is our established standard of care now for patients who have had progression on first-line EGFR TKI therapy alone and are going on to a next-line treatment option when they have not previously had platinum doublet chemotherapy. The strategy reflects the most established all-comer treatment strategy where it is not being identified as a regimen based on resistance profiling.
I do consider amivantamab plus chemotherapy to be our continued standard of care post-monotherapy for patients who are going on to a next regimen without an identifiable mechanism of prior resistance.
