Resistance Profiling Changes the Lung Pathway

Osimertinib plus savolitinib improved progression-free and overall survival over chemotherapy after MET-driven resistance emerged during osimertinib treatment in EGFR-mutant NSCLC. Progression is becoming a reason to profile the tumour again, not simply a reason to change drugs.

November 20, 2026
Editorial
MET-driven resistance after osimertinib can be targeted directly, strengthening the case for molecular reassessment at progression.[New Africa] / Shutterstock.com

IPM Take

Progression is not simply the end of one treatment line. It is a new biological state. SAFFRON shows why precision pathways need to profile resistance rather than automatically move patients from targeted therapy to chemotherapy.

Executive Summary

The international Phase III SAFFRON trial randomised 338 patients with EGFR-mutant advanced NSCLC that had progressed during osimertinib and showed MET overexpression, amplification or both. Osimertinib plus savolitinib significantly improved progression-free survival compared with platinum-pemetrexed chemotherapy, with a hazard ratio of 0.55, and significantly improved overall survival, with a hazard ratio of 0.68. Twelve-month overall survival was 70% versus 53%.

Why it matters

  • Patients / advocates: Progression does not automatically mean the end of targeted treatment.
  • Diagnostics / pathology: Re-biopsy and resistance profiling can now identify a treatment-relevant escape mechanism.
  • Clinicians: MET-driven resistance can be attacked while continued EGFR inhibition maintains pressure on the original driver.
  • Payers: Repeat molecular testing may prevent default chemotherapy in an actionable subgroup.

The peer-reviewed SAFFRON Phase III paper makes an important conceptual point: “progressed on osimertinib” describes what happened clinically, but it does not explain what changed biologically. SAFFRON enrolled patients whose EGFR-mutant cancers had developed MET overexpression, amplification or both and showed that continuing osimertinib while adding the selective MET inhibitor savolitinib produced better progression-free and overall survival than switching to platinum-pemetrexed chemotherapy. The trial therefore turns a recognised resistance mechanism into a therapeutic pathway rather than treating every post-osimertinib progression as the same event.

That approach depends on something many health systems still do inconsistently: molecular reassessment at progression. Tumours evolve under selective pressure, and a biopsy or liquid biopsy taken after targeted therapy may reveal an actionable state that was not present or not detectable at diagnosis. Yet repeat profiling remains uneven because tissue re-biopsy is invasive, ctDNA reimbursement varies, turnaround times can be slow and some pathways still treat molecular testing as a one-time event performed only at initial diagnosis.

SAFFRON therefore strengthens an implementation argument that extends well beyond MET. Modern targeted oncology should not ask only what mutation started the cancer; it should ask what mechanism is sustaining the cancer now. When resistance biology changes the next treatment decision, repeat profiling stops being an optional academic exercise and becomes part of standard pathway design. The clinical trial has shown what happens when the resistance mechanism is found. The next test is whether routine care can find it reliably.

Source & Evidence