Screening of FGFR patients for FGFR directed clinical trials in Network Genomic Medicine (NGM): Real-world data

e21013

Background: The fibroblast growth factor receptor (FGFR) 1-4 genes show a heterogenic landscape of alterations in non-small cell lung cancer (NSCLC) whereas only a small amount is yet considered to have oncogenic potential. The frequency of activating FGFR alterations is low, counting for approximately 2% of NSCLC. We have screened NSCLC patients (pts) for FGFR translocations/mutations within NGM and analysed them on FGFR alteration frequency, patient characteristics and outcome. Methods: From 04/2019 to 01/2020 we screened 472 squamous NSCLC for FGFR gene alterations and from 02/2020 to 12/2021 an additional 5286 patients including all NSCLC cases. Of these 5286 pts, 1097 pts were analysed for FGFR fusions. We used DNA-NGS for FGFR-mutations and RNA-NGS for FGFR–translocations. Activating mutations were defined according to the publicly available molecular data bases and published data. Results: Within the cohort of 5758 NSCLC patients, we found 316 (5.5%) patients with FGFR alterations. Sixty-six (20.9% of FGFR, 1.1% of NSCLC) patients had alterations classified as activating, of whom 39 had FGFR point mutations and 27 FGFR translocations. Concerning the patients with activating alterations, they had UICC stage III or IV at time of diagnosis; 22 were females; 58 patients had squamous cell carcinoma, 6 patients had adenocarcinoma and 2 had large cell neuroendocrine carcinoma. Fifty-three patients (80.3%) with activating FGFR alteration had a co-mutation: TP53 (inactivating) co-mutation was seen in 41 cases (62.1%) and 19 cases had either PTEN (7 pts), KRAS (4), EGFR (3), PIK3CA (2), ROS1 (1), ALK (1) or BRAF (1) mutations. Ten patients were included in a FGFR-targeted trial. Sixty patients were available for follow-up. The median overall survival (mOS) was 21.4 month (95%CI: 16.8–25.9) for all patients with activating FGFR alteration, whereas mOS was 18.5 month (95%CI: 13.2-23.9) for FGFR mutation and 25.3 months (95%CI: 17.8-32.9) for FGFR fusions. Conclusions: FGFR 1-4 gene alterations are rare. Large molecular and clinical networks are necessary to identify these pts. Prognostic factors of FGFR patients are currently not defined. Further assessments on molecular and clinical features in FGFR altered NSCLC are needed to identify sensitivity to FGFR inhibition. Clinical trials with specific FGFR inhibitors are ongoing.

Treatment outcome and functional characterization of uncommon EGFR mutations in the German National Network Genomic Medicine Lung Cancer (nNGM)

9036

Background: The nNGM centralizes molecular diagnostics, treatment recommendations and follow-up reporting in NSCLC in Germany. Uncommon EGFR mutations pose a clinical challenge because they comprise a heterogenous group and analyses of treatment outcome are still scarce. Here, we analyzed follow-up data of patients with rare EGFR mutations and performed functional characterization of recurrent mutations with unknown function. Methods: This multicenter, retrospective analysis of uncommon EGFR mutations (excluding L858R-, T790M mutations and exon 19 deletions) includes stage IV patients with NSCLC from 12 nNGM centers. We categorized EGFR-mutations into 3 groups: uncommon EGFR mutations with known driver function, for instance E709X, G719X, S768I and L861Q (group 1), exon 20 insertions (group 2) and all other very rare mutations (group 3). Functional characterization of unknown mutations was performed by insertion mutagenesis in Ba/F3 cells and monitoring of growth factor-independent proliferation. Results: In total, 834 cases with uncommon EGFR mutations were reported. Follow-up data after EGFR-TKI (Erlotinib, Gefitinib, Afatinib and Osimertinib), chemotherapy and/or mono-PD(L)1 blockade was available for 252 patients. Mean progression free survival (mPFS) on EGFR-TKIs vs. chemotherapy was 6.6 months vs. 5.0 months (HR 0.54, 95%CI 0.35 to 0.81, P =.003) in group 1 (n = 84), and 6.7 months vs. 3.4 months (HR 0.66, 95%CI 0.47 to 0.92, P =.015) in group 3 (n = 104). Mono-anti-PD(L1) blockade was not superior to chemotherapy (group 1, mPFS 3.0 months, HR 1.32, 95% 0.55 – 3.15, P =.535 and group3, mPFS 4.3 months, HR 1.02, 95% CI 0.64 – 1.62, P = 0.951). Exon 20 insertions (group 2, n = 63) did not benefit from EGFR-TKIs or anti-PD(L1) blockade vs. chemotherapy. Overall survival (OS) analysis (n = 218) following chemotherapy (56%) or EGFR-TKI treatment (44%) showed median OS (mOS) of 18.0 months vs. 13.9 months in patients treated with EGFR-TKI and chemotherapy, respectively in group 1 (HR 0.97, 95%CI 0.54 to 1.75, P =.929). In group 3 patients treated with EGFR-TKI and chemotherapy had a mOS of 35.4 months vs. 12.0 months, respectively (HR 0.59, 95%CI 0.35 to 1.01, P =.056). In the Ba/F3 system we could identify 8 recurrent driver and 12 non-driver mutations with a clinically applicable assay turnaround time of 4 weeks to inform clinical decision-making in the future. Conclusions: This real-world dataset confirms that patients with group 1(uncommon) EGFR mutations benefit from EGFR-TKIs and indicates that mono-anti PD(L)1 blockade is not superior to chemotherapy. Furthermore, patients with very rare EGFR mutations (group 3) also experienced a PFS benefit from EGFR-TKI compared to chemotherapy while immune therapy was not beneficial.

Genetic Heterogeneity of MET-Aberrant NSCLC and Its Impact on the Outcome of Immunotherapy

INTRODUCTION

Robust data on the outcome of MET-aberrant NSCLC with nontargeted therapies are limited, especially in consideration of the heterogeneity of MET-amplified tumors (METamp).

METHODS

A total of 337 tumor specimens of patients with MET-altered Union for International Cancer Control stage IIIB/IV NSCLC were analyzed using next-generation sequencing, fluorescence in situ hybridization, and immunohistochemistry. The evaluation focused on the type of MET aberration, co-occurring mutations, programmed death-ligand 1 expression, and overall survival (OS).

RESULTS

METamp tumors (n = 278) had a high frequency of co-occurring mutations (>80% for all amplification levels), whereas 57.6% of the 59 patients with MET gene and exon 14 (METex14) tumors had no additional mutations. In the METamp tumors, with increasing gene copy number (GCN), the frequency of inactivating TP53 mutations increased (GCN < 4: 58.2%; GCN ≥ 10: 76.5%), whereas the frequency of KRAS mutations decreased (GCN < 4: 43.2%; GCN ≥ 10: 11.8%). A total of 10.1% of all the METamp tumors with a GCN ≥ 10 had a significant worse OS (4.0 mo; 95% CI: 1.9-6.0) compared with the tumors with GCN < 10 (12.0 mo; 95% confidence interval [CI]: 9.4-14.6). In the METamp NSCLC, OS with immune checkpoint inhibitor (ICI) therapy was significantly better compared with chemotherapy with 19.0 months (95% CI: 15.8-22.2) versus 8.0 months (95% CI: 5.8-10.2, p < 0.0001). No significant difference in median OS was found between ICI therapy and chemotherapy in the patients with METex14 (p = 0.147).

CONCLUSIONS

METex14, METamp GCN ≥ 10, and METamp GCN < 10 represent the subgroups of MET-dysregulated NSCLC with distinct molecular and clinical features. The patients with METex14 do not seem to benefit from immunotherapy in contrast to the patients with METamp, which is of particular relevance for the prognostically poor METamp GCN ≥ 10 subgroup.