Abstract 2434: Prognostic and predictive effects of TP53 co-mutations and RET fusion partners in RET -rearranged advanced NSCLC

Abstract Background: RET fusions occur in ∼1-2% of advanced NSCLC (aNSCLC). Selective RET inhibitors (SRIs) significantly improve outcomes, yet ∼10% of patients progress by 6 months and ∼30% by 12 months. Predictors of response or early progression remain poorly defined. Methods: A multicenter retrospective analysis (RET-MAP) of RET+ aNSCLC from 47 international centers evaluated clinical/genomic correlates of SRI outcomes. Multivariable Cox models estimated progression-free (PFS) and overall survival (OS) and tested interactions with first-line (1L) SRI versus chemotherapy ± immune checkpoint inhibitor (CH±ICI). In parallel, genomic/transcriptomic features were characterized in an external Caris Life Sciences (CLS) RET+ aNSCLC cohort. Results: Among 510 RET-MAP patients (median age 63; 59% female; 36% ever-smokers; 92% adenocarcinoma), 401 received an SRI (1L n=151; later lines n=250). Median PFS on SRI (any line) was 17.1 months (95% CI, 14.5-21.1), median OS 30.4 months (95% CI, 27.0-41.1). TP53 co-mutation was present in 27% (78/292 tested) and was associated with shorter PFS (8.9 vs 19.2 months; p<0.001) and OS (22.7 vs 30.9 months; p=0.002), without a significant treatment-by-TP53 interaction for 1L SRI versus CH±ICI (p=0.22). Fusion partner contributed with additional signal: KIF5B (73%, 274/373) had inferior PFS (13.4 vs 60.2 months; p<0.001) and OS (26.6 vs 73.0 months; p=0.002) compared with CCDC6 (17%, 65/373), and a significant treatment-by-fusion interaction favored SRI over CH±ICI in CCDC6 versus KIF5B (p=0.007). On multivariate analysis, shorter PFS and OS were independently associated with TP53 mutation (HR 1.74, p<0.001; HR 1.77, p=0.001), KIF5B fusion (HR 1.98, p=0.002; HR 1.64, p=0.040), ECOG ≥2 (HR 2.20, p<0.001; HR 3.06, p<0.001), and brain metastases (HR 1.59, p=0.003; HR 1.88, p<0.001); worse OS was associated with non-adenocarcinoma histology (HR 1.74, p=0.049) and smoking (HR 1.42, p=0.034). Notably, KIF5B fusions were enriched for TP53 mutations versus CCDC6 (40% vs 19.6%, p=0.011), suggesting partially overlapping biology. In the CLS cohort (N=211), TP53-mutant tumors (74/211) were enriched for RB1 mutations, had higher PD-L1 expression and tumor mutational burden, and demonstrated increased M1 macrophage/B-cell infiltration with reduced neutrophils; by fusion partner, CCDC6 showed higher PD-L1, while global transcriptomes were otherwise similar to KIF5B. Conclusions: In RET+ aNSCLC, TP53 mutation is prognostic for inferior outcomes on SRIs but not predictive of differential benefit versus CH±ICI. Fusion partner carries both prognostic and predictive relevance; CCDC6 is associated with a more indolent course and greater relative benefit from SRI. External profiling supports an inflammatory microenvironment in TP53-mutant disease, reinforcing biologically distinct—and clinically meaningful—subsets within RET-driven lung cancer. Citation Format: Daniela Miliziano, Julia K. Rotow, Meghanne Lomibao, Tolulope Adeyelu, Arianna Marinello, Helena Bote-de Cabo, Jamie Feng, Andrea De Giglio, Mariana Brandão, Florian Guisier, Michael Duruisseaux, Christina Falcon, Massimiliano Cani, Francesca Colamartini, Barliz Waissengrin, Isabelle Monnet, Anna Eisert, Emilio Bria, Amin H. Nassar, Ayesha Aijaz, Patricia Iranzo, Colin R. Lindsay, Elizabeth Fabre, Vladmir Cordeiro de Lima, Judith Raimbourg, Laura Mezquita, Nicolas Minatta, Sophie Cousin, Katarzyna Szymczak, Vincent Fallet, Clarisse Audigier-Valette, Helene Doubre, Philippe Rochigneux, Annarita Avanzo, Antonio Calles, Marco Tagliamento, Diego Cortinovis, Balazs Halmos, Nicholas Girard, Andrew Elliott, Jair Bar, Alessio Cortellini, Diana N. Ionescu, Frances A. Shepherd, Fabrice Barlesi, Karen L. Reckamp, David Planchard, Benjamin Besse, Alexander Drilon, Mihaela Aldea. Prognostic and predictive effects of TP53 co-mutations and RET fusion partners in RET-rearranged advanced NSCLC [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 2434.