Cancer Care · global
Targeting MET After Resistance to Lung Cancer Targeted Therapy: SAFFRON Trial Shows Two-Drug Combination Prolongs Survival
In a group of patients selected through tumor testing, osimertinib plus savolitinib achieved a median overall survival of 17.8 months, compared with 12.9 months for chemotherapy. However, fatal adverse events and immature long-term data add considerations that require careful assessment of the treatment's benefits.
When lung cancer progresses again after targeted therapy, whether another precision treatment is possible often depends on new tumor testing results. The phase 3 SAFFRON trial, presented at the 2026 European Society for Medical Oncology (ESMO) Congress, showed that some patients with EGFR mutations whose disease progressed after osimertinib (Tagrisso) could survive longer than with chemotherapy by adding the MET inhibitor savolitinib, provided their tumors also met criteria for MET overexpression or gene amplification.
The study randomized 338 patients to receive either osimertinib plus savolitinib or chemotherapy with a platinum agent plus pemetrexed. According to OncLive and Targeted Oncology reports from the same meeting, median overall survival was 17.8 months in the combination group and 12.9 months in the chemotherapy group. The hazard ratio for death was 0.68, and the difference was statistically significant. The proportions of patients still alive at 12 months after treatment were 70% and 53%, respectively.
Disease control showed a consistent pattern. By blinded independent central review, median progression-free survival was 7.6 months in the combination group and 5.4 months in the chemotherapy group, with a hazard ratio for disease progression or death of 0.55. The proportion of patients with confirmed tumor shrinkage meeting objective response criteria increased from 27% in the chemotherapy group to 51% in the combination group. Among patients who responded, median duration of response was 9.7 months and 5.7 months, respectively.
However, these results have clearly defined testing criteria. Patients had to undergo a new tumor tissue biopsy after disease progression, with central laboratory confirmation of either MET immunohistochemistry staining of 3+ in at least 90% of tumor cells or at least 10 MET gene copies detected by fluorescence in situ hybridization. This evidence applies to a specific biomarker-defined population and cannot be directly generalized to all patients whose osimertinib treatment has failed.
The safety findings were more complex. The proportions of patients experiencing grade 3 or higher adverse events were similar in the combination and chemotherapy groups, at 64% and 63%, respectively. When only events judged to be treatment-related were counted, the proportions were 40% and 52%. However, discontinuation of any study treatment because of adverse events was more frequent in the combination group, at 14%, compared with 8% in the chemotherapy group. The most common adverse event in the combination group was peripheral edema.
Fatal events require separate interpretation: 12 patients in the combination group experienced fatal adverse events, compared with none in the chemotherapy group. One case of pneumonia was judged to be treatment-related. The reports also noted that nine of these 12 patients had documented disease progression before death or had disease progression listed as a cause of death. Therefore, not all fatal events can be attributed to the drugs, and the lower rate of severe treatment-related events alone does not establish that the combination is safer overall. Median treatment exposure also differed between the groups, at 8.5 months and 3.5 months, respectively.
SAFFRON provides survival evidence from a randomized phase 3 trial for this treatment approach after resistance develops. However, data maturity for the current overall survival analysis is 53%, and estimates of long-term benefit may still change with further follow-up. For clinical practice, the significance of these results extends beyond adding another drug combination: they also highlight the importance of obtaining new samples and identifying tumor characteristics. Efficacy, safety, and testing thresholds must all be considered in subsequent assessments.