Key Ideas

Resistance to third-generation EGFR inhibitors is far more heterogeneous than the resistance patterns seen after earlier-generation drugs, and in 40-50% of patients, no single resistance mechanism can even be identified. This heterogeneity is exactly why targeting one resistance pathway at a time, in the recurrent setting, is unlikely to be sufficient, and why broader strategies, along with routine tissue biopsy rather than liquid biopsy alone, matter more than ever.

A Different Resistance Landscape Than We're Used To

When first- and second-generation EGFR inhibitors (erlotinib, gefitinib) were standard, resistance was comparatively predictable: roughly 60-70% of patients developed a single point mutation, T790M, which directly led to the development of osimertinib. Resistance to osimertinib itself looks fundamentally different, and resistance to osimertinib used in the second line can look different still from resistance to osimertinib used upfront.

Today, no distinct resistance mechanism can be identified in 40-50% of patients progressing on osimertinib, a genuinely difficult population to treat systematically. Among identifiable mechanisms, EGFR-mediated resistance (most commonly C797S, in about 10% of patients) and MET amplification are the most common, but many additional resistance mechanisms occur, each in only the 1-2% range. Resistance can involve mutations in the drug target itself, "bypass track" alterations (such as ALK or RET fusions), mutations in downstream effectors, or histologic transformation (small cell or squamous transformation), each requiring a fundamentally different treatment approach. Without biopsy, these mechanisms often go undetected; liquid biopsy is useful, but when it's non-diagnostic, tissue biopsy remains necessary.

Does Frontline Regimen Choice Change the Resistance Pattern?

The MARIPOSA trial (amivantamab plus lazertinib versus osimertinib in the frontline setting) used a drug combination that is itself an EGFR-MET bispecific antibody paired with a TKI. Early data (worth interpreting cautiously) suggest this frontline strategy does shift the resistance pattern at progression, showing a reduction in both MET amplification and secondary EGFR resistance mechanisms such as C797S, compared with osimertinib alone.

The Long Search for Fourth-Generation EGFR Inhibitors

For years, the field has pursued fourth-generation EGFR inhibitors designed to target C797S-mediated resistance specifically, but many have failed, in part because patients frequently have other, "bypass track" resistance mechanisms operating simultaneously, a single-target drug can't address that heterogeneity alone.

Some history is worth remembering here. Older data (from Zosia Piotrowska's group, 2017) showed that identifying a specific bypass resistance mechanism (such as a RET fusion or, separately, a MET fusion) allowed off-label combination of a matched inhibitor (a RET inhibitor or crizotinib, respectively) with osimertinib, achieving real clinical and radiographic responses, though in my own experience these responses tend not to be durable, typically lasting only a few months.

More recently, for C797S-mediated resistance specifically, BLU-701 and BLU-945 achieved clinical and radiographic responses but did not advance further due to toxicity. Newer fourth-generation candidates show more promise: BDTX-1535 has shown activity against common EGFR resistance mechanisms in updated 2026 ASCO data, despite some toxicity. BH-30643 (from BlossomHill Therapeutics, which completed its IPO around the time of this data release), an oral "OMNI-EGFR" inhibitor, showed a 50% response rate in C797S-positive patients without prior chemotherapy and about 39-40% in those with prior chemotherapy, notably in a heavily pretreated population (66% with prior brain metastases, generally three or more prior lines). DZD6008 (Dizal Pharmaceutical, now partnered with AstraZeneca) showed response rates of 41.2% at a 40 mg dose and 43% at 60 mg in a roughly 40-patient C797S-mutated cohort, with relatively good tolerability.

An important strategic question running through all of this: should these newer fourth-generation agents be studied as monotherapy in later lines, or combined with third-generation EGFR inhibitors and moved into the frontline setting? Most companies appear to be pursuing the latter, though I think there remains real value in studying combinations specifically in the later-line, resistant setting as well.

Combination Strategies That Haven't Quite Panned Out Clinically

Several rational combination strategies (osimertinib plus a MET inhibitor such as savolitinib, in the ORCHARD study; osimertinib combined with a first-generation EGFR inhibitor like gefitinib; tepotinib plus osimertinib in MET-amplified, post-osimertinib disease, showing a 50% response rate) have produced scientifically interesting but, in most cases, clinically modest results, typically only a few additional months of benefit, and difficult-to-quantify true MET amplification when relying on plasma NGS alone.

Moving Toward Biomarker-Agnostic Strategies

A genuinely notable shift in the field: five years ago, suggesting a non-targeted approach for EGFR-mutant disease would have been considered heretical. That's increasingly no longer true. Immunotherapy carries a real, elevated risk of immune-related adverse events when used in sequence with third-generation EGFR inhibitors (Adam Schoenfeld's MSK-based publication showed grade 3 transaminitis rates of 35-40% in this setting), but newer biomarker-agnostic strategies are showing genuine promise:

  • COMPEL (continuing osimertinib alongside chemotherapy after progression) shows durable responses in a small but real dataset.

  • MARIPOSA-2 (amivantamab plus chemotherapy) remains, to me, the current standard of care in this setting, with a 63% response rate and progression-free survival nearly double that of chemotherapy alone. It's worth highlighting that amivantamab shows genuinely durable CNS activity (intracranial PFS hazard ratio of 0.55, roughly a four-fold improvement over chemotherapy alone), a meaningful point since not all TKIs have equally good CNS penetration despite general assumptions to the contrary.

  • Antibody-drug conjugates, including a genuine failure worth learning from: patritumab deruxtecan (a HER3-directed ADC), which initially showed a 30% response rate but ultimately failed against chemotherapy in the confirmatory HERTHENA-Lung01/02 program and is not an approved agent. By contrast, based on the (negative, ultimately) TROPION-Lung05 trial in which we participated, in combination with TROPION-Lung01 data, datopotamab deruxtecan is now, in my view, a standard of care option in the third-line setting (after osimertinib and chemotherapy, and, I'd add, after amivantamab as well). Also genuinely exciting: MK-2870 (sacituzumab tirumotecan), a Trop-2-directed ADC now a Merck asset, showing response rates around 60% early on and improving in larger, mostly China-based randomized phase 3 studies; we look forward to seeing this data replicated in US populations.

Conclusions

Acquired resistance to EGFR-directed therapy is inevitable, and we see both on-target and off-target mechanisms driving it. We've historically spent a great deal of effort trying to overcome on-target resistance specifically; a broader strategic approach may now be more important. Biomarker profiling remains essential: roughly 15% of patients will show histologic transformation (squamous or small cell), and relying on liquid biopsy alone is insufficient in 2026, tissue biopsy still matters. We need therapeutics with genuinely novel mechanisms of action and broader therapeutic strategies overall. Given the heterogeneity of acquired resistance, MARIPOSA-2, COMPEL, and the TROPION-Lung program currently represent our standard of care in the second- and third-line settings.

For Patients

Resistance to EGFR-targeted therapy is more varied than once thought: in nearly half of patients who progress on osimertinib, no single cause can be found. A repeat tissue biopsy at progression, not just a blood test, matters, since it can catch a transformation to a different, more aggressive cancer type that liquid biopsy alone would miss. Ask your care team about repeat tissue biopsy at progression and whether a clinical trial testing a broader combination strategy might apply to your case.

Key Takeaways

  • No single resistance mechanism can be identified in 40-50% of patients progressing on osimertinib, a substantial shift from the more predictable T790M-driven resistance seen with earlier-generation EGFR inhibitors.

  • Resistance mechanisms include on-target EGFR mutations (most commonly C797S), MET amplification, bypass-track alterations (ALK or RET fusion), downstream mutations, and histologic transformation (squamous or small cell), each requiring a different treatment approach.

  • Newer fourth-generation EGFR inhibitors (BH-30643, DZD6008, BDTX-1535) show encouraging activity against C797S-mediated resistance, though optimal positioning (later-line monotherapy versus frontline combination) remains unresolved.

  • Patritumab deruxtecan (a HER3-directed ADC) failed its confirmatory trial and is not approved, a useful reminder that early single-arm response data doesn't always translate to confirmed clinical benefit.

  • MARIPOSA-2, COMPEL, and the TROPION-Lung program (datopotamab deruxtecan) currently represent standard-of-care options in the second- and third-line settings for EGFR-mutant NSCLC.

  • Tissue biopsy remains essential at progression; liquid biopsy alone is insufficient, particularly for detecting histologic transformation, which occurs in roughly 15% of patients.

References

  1. Piotrowska Z, Isozaki H, Lennerz JK, et al. Landscape of acquired resistance to osimertinib in EGFR-mutant NSCLC and clinical validation of combined EGFR and RET inhibition with osimertinib and BLU-667 in osimertinib-resistant RET fusion-positive lung cancer. Cancer Discov. 2018.

  2. Cho BC, et al. Amivantamab plus lazertinib versus osimertinib in EGFR-mutated NSCLC (MARIPOSA).

  3. BlossomHill Therapeutics. Phase 1 dose escalation/backfill data for BH-30643 (SOLARA trial) in advanced EGFR-mutant NSCLC. Presented at 2026 ASCO Annual Meeting.

  4. Schoenfeld AJ, Arbour KC, Rizvi H, et al. Severe immune-related adverse events are common with sequential PD-(L)1 blockade and osimertinib. Ann Oncol. 2019.

  5. Yu HA, et al. Patritumab deruxtecan in EGFR-mutated NSCLC (HERTHENA-Lung01/02).

  6. Ahn MJ, et al. Datopotamab deruxtecan in previously treated EGFR-mutated NSCLC (TROPION-Lung01, TROPION-Lung05).