Back

Distinct Evolutionary Trajectories in Early Lung Adenocarcinoma: Age Related Pathway with Epidermal Growth Factor Receptor with Genome Doubling and Smoking-Driven Pathway

Goto, A.; Nakaoka, H.; Yoshida, M.; Koyama, K.; Miyabe, k.; Zhou, J.; Umakoshi, M.; Takashima, S.; Imai, K.; Minamiya, Y.; Nishikawa, K.; Matsubara, D.; Inoue, I.; Sugimura, H.; Ishikawa, Y.

2026-07-23 pathology
10.64898/2026.07.21.26358543 medRxiv
Show abstract

Lung adenocarcinoma a progress from preinvasive lesions to invasive cancer; however, early evolutionary events in adenocarcinoma in situ (AIS) and minimally invasive adenocarcinoma (MIA) remain poorly defined, particularly in East Asian populations enriched for EGFR mutations. We performed whole-exome sequencing on 67 Japanese patients (38 AIS, 29 MIA), including multiregion sampling in 13 cases to identify EGFR as predominant driver (61.2%), followed by RBM10 (19.4%) and TP53 (9.0%). Two evolutionary trajectories emerged: age-related and smoking-driven pathways. In the former, EGFR-mutant tumors frequently exhibited early whole-genome doubling (WGD) (24.4%) with clock-like signature. The smoking-driven pathway, typically involving KRAS mutations, displayed a tobacco-associated signature. Multiregion sequencing revealed that driver mutations (EGFR, KRAS, and MET) were shared trunk events across in situ and invasive regions, while secondary alterations arose subclonally. This study defines the genomic evolution of early lung adenocarcinoma in Japanese patients, identifying two evolutionary trajectories: an age-related pathway with EGFR-linked genome doubling and a smoking-driven pathway involving KRAS mutations. These findings elucidate mechanisms underlying progression from preinvasive lesions to invasive cancer in Asian populations.

Matching journals

The top 9 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.