Patrick, C.

Abstract Pulmonary large cell neuroendocrine carcinoma (LCNEC) is a rare, aggressive lung tumor marked by significant molecular heterogeneity. In a study of 590 patients across two independent cohorts, we observed comparable overall survival across treatment regimens (chemotherapy, chemoimmunotherapy, immunotherapy) without unexpected adverse events. Genomic analysis identified distinct NSCLC-like (KEAP1, KRAS, STK11 mutations) and SCLC-like (RB1, TP53 mutations) LCNEC subtypes, with 80% aligning with SCLC transcriptional profiles. Serial sampling revealed stable mutational but shifting transcriptomic landscapes over time. NSCLC-like LCNECs showed elevated FGL-1 (a LAG-3 ligand) and SPINK1 expression, while SCLC-like subtypes expressed higher levels of DLL3. Immunofluorescence confirmed FGL-1 in NSCLC-like LCNECs, and H&E slide analyses indicated fewer tumor-infiltrating lymphocytes in LCNECs versus other lung cancers. These findings highlight LCNEC’s distinct immunogenomic profile, supporting future investigations into LAG-3, SPINK1, and DLL3-targeted therapies.

PURPOSE KRAS is the most commonly mutated driver oncogene in non–small cell lung cancer (NSCLC). Sotorasib and adagrasib, KRAS G12C inhibitors, have been granted accelerated US approval; however, hepatotoxicity is a common side effect with higher rates in patients treated with sotorasib proximal to checkpoint inhibitor (CPI) therapy. The aim of this study was to assess the feasibility and safety of adagrasib after discontinuation of sotorasib because of treatment-related grade 3 hepatotoxicity through real-world and clinical cases. METHODS Medical records from five patients treated in real-world settings were retrospectively reviewed. Patients had locally advanced or metastatic KRAS G12C -mutated NSCLC and received adagrasib after sotorasib in the absence of extracranial disease progression. Additional data were collected for 12 patients with KRAS G12C -mutated NSCLC enrolled in a phase Ib cohort of the KRYSTAL-1 study and previously treated with sotorasib. The end points associated with both drugs included timing and severity of hepatotoxicity, best overall response, and duration of therapy. RESULTS All patients were treated with CPIs followed by sotorasib (initiated 0-64 days after CPI). All five real-world patients experienced hepatotoxicity with sotorasib that led to treatment discontinuation, whereas none experienced treatment-related hepatotoxicity with subsequent adagrasib treatment. Three patients from KRYSTAL-1 transitioned from sotorasib to adagrasib because of hepatotoxicity; one experienced grade 3 ALT elevation on adagrasib that resolved with therapy interruption and dose reduction. CONCLUSION Adagrasib may have a distinct hepatotoxicity profile from sotorasib and is more easily combined with CPIs either sequentially or concurrently. These differences may be used to inform clinical decisions regarding an initial KRAS G12C inhibitor for patients who recently discontinued a CPI or experience hepatotoxicity on sotorasib.

9069 Background: With the emergence of effective targeted-therapies for KRAS G12C, and the development of promising agents for KRAS G12V and G12D among others, the identity of unique KRAS mutations in non-small cell lung cancer (NSCLC) has become increasingly relevant. Acquired KRAS mutations are a known mechanism of resistance observed across driver mutation-positive (DM+) NSCLC. The incidence and diversity of these acquired alterations, and whether they differ from those observed in de novo KRAS-mutated NSCLC, is unknown. Methods: NSCLC molecular profiles were obtained using next-generation sequencing (Caris Life Sciences) with paired whole-transcriptomic sequencing (Illumina NovaSeq) and immunohistochemistry (Caris Life Sciences). Demographic data were abstracted from medical records. KRAS mutation diversity was defined for de novo KRAS-mutated ( KRASmt) NSCLC ( KRAS as only identified driver – KRASmt_dn) and DM+ NSCLC with acquired KRAS mutations (concurrent KRASmt with other known driver – KRASmt_acq). Fisher’s exact test was used to compare the distribution of unique KRAS mutations between the groups. Due to the unique biology of NSCLC with class II/III BRAF mutations, this subset was removed from the KRASmt_acq subgroup for the final analysis. Results: A total of 5932 KRASmt NSCLC samples were identified, among which 5879 were KRASmt_dn and 53 were KRASmt_acq (see table). The distribution of unique KRAS mutations was not significantly different between groups (p = 0.14). Within the KRASmt_dn group, KRAS G12C was most common (40%), followed by G12V (19%), G12D (14%), G12A (7%) and Q61H/G13C (4% each). In the KRASmt_acq group, KRAS G12C was most common (30%), followed by G12D (19%), G12V (17%), Q61H (11%), G12A (7%), and G13C (6%). The most common observed driver mutations in the KRASmt_acq group were EGFR (38%) and MET (30%). In the EGFR group, KRAS G12D (25%), Q61H/G12C (20% each), G12V/G13C (10% each), and G12F/A/R (5% each) were observed. In the MET group, KRAS G12C (25%), G12D/V/A (19% each), Q61H (12%) and G12S (6%) were seen. Among 7 patients with a documented history of smoking in the KRASmt_acq group, KRAS G12D and Q61H were most common (29% each), followed by G12V/F and G13C (14% each). Conclusions: While the distribution of unique KRAS mutations did not differ significantly between KRASmt_dn and KRASmt_acq groups, acquired KRAS mutations were seen across DM+ NSCLC subsets, among which the frequency of observed mutations appeared to vary. The functional and immunological significance of these mutations, and their impact on clinical outcomes, warrants further investigation. [Table: see text]

Abstract Background: While most EGFR-mutant NSCLC tumors are resistant to checkpoint inhibitors, a small subset has durable responses. Certain molecular subtypes of EGFR may be more responsive. There is a lack of clarity on the genomic and immune profiles of EGFR subtypes, and further elucidation of this may assist in optimally identifying those likely to respond to immune-based therapies. Methods: Profiles of 5,510 lung adenocarcinoma specimens were obtained utilizing next-generation sequencing of 592 genes (Caris Life Sciences) and categorized by EGFR subtype. PD-L1 expression (Ab: 22c3), tumor mutational burden (TMB), immune cell type fractions, and co-mutations in TP53 were assessed by EGFR subtype and compared to wild-type (WT) tumors. Results: Of the total cohort, 791 (14.4%) were EGFR-mutant (Table). Among EGFR-mutant tumors, exon 19 deletions were most common (49.7%), followed by L858R, (32.9%), exon 20 insertions (13.0%), L861Q (2.9%), and G719X (1.5%). PD-L1 positivity of each EGFR subtype as compared to WT did not vary, except for L858R tumors which had a significantly lower percentage of PD-L1 positive cases. Exon 19 deletion, L858R, and exon 20 insertion tumors were significantly less likely to have high PD-L1 or high TMB compared to WT. Among the EGFR subtypes, L861Q and G719X tumors had the greatest percentage with high TMB. TP53 co-mutations were frequent in EGFR tumors, especially among L861Q and G719X tumors. Exon 19 deletion and L858R tumors had significantly less CD8+ and greater CD4+ T cell fractions compared to WT. Neutrophil and M2 macrophage cell fractions were significantly higher in exon 19 deletion, L858R, and exon 20 insertion tumors compared to WT. Conclusions: Most subtypes of EGFR have profiles consistent with decreased immunogenicity; however L861Q and C719X tumors have a greater percentage with high TMB or TP53 co-mutations. Such features in uncommon EGFR subtypes may correlate with responsiveness to immune-based therapies and warrant further study. Tumor Characteristic EGFR WT(n=4719) Exon 19 del(n=393) L858R(n=260) Exon 20 ins(n=103) L861Q(n=23) G719X(n=12) PD-L1 positive (≥1%) 57.7% 51.0% 49.0% a 49.5% 47.8% 50.0% PD-L1 high (≥50%) 30.6% 19.7% a 18.8% a 14.4% a 21.7% 16.7% TMB high (≥10 mut/Mb) 35.3% 1.0% a 3.7% a 1.0% a 9.5% 18.2% TP53 co-mutation 53.9% 61.7% 63.6% a 59.2% a 69.6% a 83.3% a Median % cell fraction CD8+ 0.7 0.3 a 0.4 a 0.3 0.6 0.2 CD4+ 0.0 0.6 a 0.8 a 1.3 a 1.7 1.1 Neutrophils 5.5 8.0 a 7.3 a 7.8 a 7.4 6.4 M2 macrophages 5.5 6.4a 6.9a 6.7a 5.8 4.3 aq-value < 0.05 when compared to EGFR WT (corrected for multiple comparisons) Citation Format: J. Nicholas Bodor, Joanne Xiu, Vinicius Ernani, Supreet Kaur, Hirva Mamdani, Pavel Brodskiy, Sai-Hong I. Ou, Patrick C. Ma, Margie L. Clapper, W. Michael Korn, Joseph Treat. Genomic and immune characteristics of EGFR subtypes in non-small cell lung cancer (NSCLC) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 4119.