In brief
- In a multicentre retrospective study of 111 patients with advanced EGFR exon 19 deletion-positive NSCLC treated with osimertinib, 86.5% received the drug as first-line therapy.
- Patients with deletions other than E746_A750del (n=25) had a shorter median progression-free survival (PFS) than those with E746_A750del (n=86): 14.3 vs 20.6 months, p<0.05.
- The four patients with L747_A750delinsP had a median PFS of 3.5 months and a median overall survival (OS) of 11.8 months, against 20.6 and 48.5 months with E746_A750del (both p<0.001).
- In public genomic databases, non-E746_A750del deletions carried more RBM10 co-mutations.
- L747_A750delinsP was characterised by frequent CDKN2A/B homozygous deletions and MYC amplifications.
Study at a glance
- Design: multicentre retrospective cohort, with genomic analysis of AACR Project GENIE and MSK-CHORD data
- Population: 111 patients with advanced EGFR exon 19 deletion-positive NSCLC
- Treatment: osimertinib, first line in 86.5%
- Groups: E746_A750del (n=86) and other exon 19 deletions (n=25), including L747_A750delinsP (n=4)
- Median PFS: 20.6 months (E746_A750del) vs 14.3 months (other deletions), p<0.05
- L747_A750delinsP (n=4): median PFS 3.5 months, median OS 11.8 vs 48.5 months with E746_A750del (both p<0.001)
- Co-alterations: RBM10 in non-E746_A750del; CDKN2A/B homozygous deletion and MYC amplification in L747_A750delinsP
- Not reported: hazard ratios, adjustment for other prognostic factors, and OS for the non-E746_A750del group as a whole (in the abstract)
Exon 19 deletions differ, and osimertinib data by subtype are scarce
E746_A750del is the most frequent of the EGFR exon 19 deletions, and the authors set out to test whether the rarer variants behave differently on osimertinib, for which subtype-level evidence has been limited.
Differences in survival between exon 19 deletion subtypes have been suggested after treatment with EGFR tyrosine kinase inhibitors. The authors note that the biological mechanisms behind any such difference are poorly understood, which is why the study pairs its clinical cohort with a genomic analysis.
A 6-month shorter median PFS with the less common deletions
Median PFS was 14.3 months in the 25 patients whose tumours carried a deletion other than E746_A750del, compared with 20.6 months in the 86 patients with E746_A750del (p<0.05).
Most of the cohort (86.5%) received osimertinib as first-line therapy, so the comparison largely reflects the setting in which the drug is now standard. The design is retrospective, the smaller group has 25 patients, and the abstract gives the result as p<0.05 without an exact value.
L747_A750delinsP: a median PFS of 3.5 months in four patients
Compared with E746_A750del, the delinsP subtype had a median PFS of 3.5 versus 20.6 months and a median OS of 11.8 versus 48.5 months, both with p<0.001.
The size of the gap is large, and it rests on four patients. The authors describe L747_A750delinsP as potentially a high-risk subtype, and that qualification matches the evidence: a finding of this size in so small a group needs confirmation in a larger series before it could inform treatment decisions.
RBM10 co-mutations in the uncommon deletions, CDKN2A/B loss and MYC gain in delinsP
These co-alterations come from whole-exome sequencing data in the AACR Project GENIE registry and the MSK-CHORD real-world dataset, not from the treated cohort itself.
Non-E746_A750del deletions showed a higher rate of RBM10 co-mutations. L747_A750delinsP was characterised by frequent CDKN2A/B homozygous deletions and MYC amplifications. The authors conclude that differences in co-occurring genomic alterations may contribute to the prognostic heterogeneity among exon 19 deletion subtypes. Because the genomic findings come from public databases rather than the treated cohort, that link remains a hypothesis.
Sources
- Lung Cancer. Clinical outcomes and genomic features of uncommon EGFR exon 19 deletion subtypes in osimertinib-treated non-small cell lung cancer (2026-08-31). doi.org
Featuring Lung Summit faculty
This study was co-authored by Lung Summit faculty Hidehito Horinouchi.
This article was produced independently by the Lung Summit editorial team, without industry funding or input.