Background
Rearranged during transfection (RET) rearrangement has been identified as one of the crucial oncogenic drivers in non-small cell lung cancer (NSCLC). Recently, two highly selective RET inhibitors have been approved by the US Food and Drug Administration and demonstrated remarkable responses. However, the clinical characteristics, outcomes and optimal diagnostic method of RET-rearrangements are not well understood. This study sought to evaluate the prevalence and characteristics of RET rearrangement, identify an effective diagnostic method for it, and correlate its presence with outcomes.
Conclusions
RET rearrangement occurs in 1.52% of Chinese NSCLCs and has identifiable clinicopathologic characteristics. RET IHC has a low sensitivity, disavowing its use in routine practice. While NGS and FISH has good performance in identifying RET rearrangement. Both IHC and FISH demonstrated lower sensitivity for NCOA4-/others-RET fusions. Clinical benefit with chemotherapy is different between CCDC6-RET and KIF5B-RET fusion patients, optimal treatment should be considered when selecting therapies for patients with RET-rearranged lung cancers.
Methods
A total of 9,431 Chinese NSCLCs from two cancer centers who have undertaken targeted DNA-NGS were enrolled and 167 RET-positive cases were screened. Non-canonical RET rearrangements were confirmed by targeted RNA-NGS. If material was sufficient, positive cases were analyzed by fluorescence in situ hybridization (FISH) (n=30) and immunohistochemistry (IHC) (n=57). Clinicopathologic characteristics, molecular profiling and treatment outcomes of RET rearrangement were evaluated.
Results
The prevalence of RET rearrangement was 1.52% (138/9,101) in unfiltered cases and 8.79% (29/330) in EGFR/KRAS/BRAF/ALK-negative cases. RET rearrangement was common in females, never smokers, and lung adenocarcinoma patients. Additionally, 40.3% of stage IV RET-rearranged NSCLC patients developed brain metastases. TP53 was the most common concurrent mutation, and 8 patients harbored concurrent driver oncogenic alterations, including EGFR (N=5), KRAS (N=2), and ALK (N=1). Non-canonical fusion partners were identified in 13.8% (23/167) of cases by DNA-based NGS, and RNA-based NGS identified 3 new partners (EPS8, GOLGA5, and TNIP1). The concordance of FISH and NGS was 83.3% (25/30), while the concordance of IHC and NGS was only 28.1% (16/57). Both IHC and FISH demonstrated lower sensitivity for NCOA4-/other-RET fusions. The CCDC6-RET subgroup had significantly longer progression-free survival than the KIF5B-RET subgroup, both after chemotherapy (23 vs. 9.7 months; P=0.014). Conclusions: RET rearrangement occurs in 1.52% of Chinese NSCLCs and has identifiable clinicopathologic characteristics. RET IHC has a low sensitivity, disavowing its use in routine practice. While NGS and FISH has good performance in identifying RET rearrangement. Both IHC and FISH demonstrated lower sensitivity for NCOA4-/others-RET fusions. Clinical benefit with chemotherapy is different between CCDC6-RET and KIF5B-RET fusion patients, optimal treatment should be considered when selecting therapies for patients with RET-rearranged lung cancers.