Abstract Background: Interstitial pneumonia increases the risk of lung cancer; however, the underlying carcinogenic pathways remain poorly understood. Usual interstitial pneumonia (UIP) frequently co-localizes with lung adenocarcinoma (LADC), providing a model to understand chronic inflammation-driven carcinogenesis. Method: We conducted whole-genome and RNA sequence of 44 UIP (+) LADCs and compared them with 216 UIP (-) LADCs. To address intra-tumoral heterogeneity and characterize the tumor microenvironment, we performed spatial transcriptomics using a whole-transcriptome platform (Visium, 55 µm resolution) and single-cell-resolution platforms (Xenium lung panel, 302 genes; Xenium 5K, 5,018 genes). Results: UIP (+) LADCs exhibited unique molecular features: low tumor mutation burden, absence of EGFR mutations, and a pervasive presence of myofibroblastic cancer-associated fibroblasts (myCAFs), likely derived from fibrotic myofibroblasts in UIP. The most common genetic alteration was in NKX2-1 (41%), encoding the lung-lineage transcription factor TTF-1. A dichotomous immunogenetic landscape was observed: NKX2-1-deficient tumors showed gastrointestinal differentiation, frequent ERBB2 amplification with super-enhancer formation, and immune "cold" characteristics; NKX2-1-proficient tumors exhibited immune "hot" characteristics with abundant intra-tumoral CD8+ T cells and elevated PD-L1 expression. Spatial transcriptomics analysis revealed intratumor heterogeneity associated with NKX2-1 expression status, suggesting illegitimate gastrointestinal differentiation. In addition, the presence of WNT5A- and CTHRC1-expressing myCAFs were identified, which may originate from UIP-associated myofibroblasts. Paired tumor and UIP regions demonstrated that CTHRC1+WNT5A+ myCAF population was markedly enriched in tumor tissues relative to matched UIP regions, suggesting their expansion during carcinogenesis. Conclusion: We propose a dual model of UIP-associated lung carcinogenesis determined by NKX2-1 status, wherein myCAFs derived from UIP-associated myofibroblasts drive both carcinogenic pathways. These findings offer targeted therapeutic strategies based on the dichotomous immunogenetic landscape and highlight targeting the myofibroblast-to-myCAF transition as a potential approach for cancer prevention in chronic pulmonary fibrosis. Citation Format: Akifumi Mochizuki, Ayako Suzuki, Kouya Shiraishi, Masahiro Torasawa, Takayuki Honda, Susumu Kirimura, Yasunari Miyazaki, Kenichi Okubo, Syuzo Kaneko, Yukihiro Yoshida, Shun-ichi Watanabe, Masahiro Tsuboi, Genichiro Ishii, Issei Imoto, Ryuji Hamamoto, Yasushi Yatabe, Yutaka Suzuki, Takashi Kohno. Dichotomous immunogenetic landscape of lung adenocarcinoma associated with usual interstitial pneumonia [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 2272.

Background Adjuvant alectinib for resected ALK-positive non-small cell lung cancer (NSCLC) offers a window for fertility preservation (FP) for adolescent and young adult (AYA) patients not previously available for patients with advanced disease. Case presentation A 39-year-old nulliparous woman was incidentally diagnosed with Stage IIB (pT2aN1M0) ALK-positive lung adenocarcinoma and underwent lobectomy. Due to her desire for future childbearing, she was referred to a reproductive specialist one month postoperatively. She successfully underwent controlled ovarian stimulation, resulting in the cryopreservation of three oocytes. The process took approximately two weeks, allowing adjuvant alectinib to be initiated two months postoperatively without delay. Conclusion This case demonstrates that multidisciplinary collaboration between oncology and reproductive specialists ensures that FP is feasible for AYA patients with early stage ALK-positive NSCLC. Comprehensive survivorship care must include oncofertility in the era of effective targeted adjuvant therapies.

BACKGROUND:Dual inhibition of epidermal growth factor receptor (EGFR) and vascular endothelial growth factor (VEGF) pathways may improve outcomes in metastatic EGFR-mutant NSCLC, but VEGF inhibitors are not universally effective. We evaluated the impact of VEGFA and VEGFR2 expressions on EGFR-TKI outcomes. METHODS:EGFR-mutant NSCLC patients from the National Cancer Center Hospital, were retrospectively analysed. The early-stage cohort comprised stage I-IIIA patients (1997-2019). The advanced-stage cohort included metastatic patients (2018-2022). VEGFA/VEGFR2 expressions were dichotomised by median transcripts per million. RESULTS:Among 447 early-stage patients (median age 66), high VEGFA was associated with smoking, TP53 co-mutation, higher Brinkman Index, and higher tumour mutation burden (all p < 0.01). High VEGFA predicted shorter relapse-free survival (HR 2.10, 95% CI 1.63-2.71, p < 0.01) and overall survival (HR 2.07, 95% CI 1.46-2.95, p < 0.01). VEGFR2 expression showed no prognostic impact. In 146 relapsed patients receiving first-line EGFR-TKIs, high VEGFA was linked to shorter progression-free survival (PFS) overall (HR 1.70,95%CI 1.13-2.56, p = 0.009), particularly for first/second-generation EGFR-TKIs (HR 1.66,95%CI 1.08-2.54 p = 0.023), but not for osimertinib (p = 0.491). In 60 advanced-stage patients on osimertinib, PFS was unaffected by VEGFA (p = 0.102). CONCLUSIONS:High VEGFA is associated with aggressive biology and inferior outcomes, correlating with shorter PFS for first/second-generation EGFR-TKIs but not for osimertinib.

INTRODUCTION:Tarlatamab, a delta-like ligand 3-targeting bispecific T-cell engager, has demonstrated promising efficacy in relapsed SCLC. Data on immune effector cell-associated neurotoxicity syndrome (ICANS) in patients with central nervous system disease remain limited. In particular, tumor inflammation-associated neurotoxicity (TIAN), proposed in chimeric antigen receptor T-cell therapy as localized neurotoxicity linked to peritumoral inflammation, has not been characterized in this setting. CASE PRESENTATION:Here, we describe a single case of severe neurotoxicity in a young woman with SCLC and multiple brain metastases who had a history of symptomatic epilepsy and received tarlatamab as fourth-line therapy. She developed sudden loss of consciousness (immune effector cell-associated encephalopathy score: 0) and seizures 27 hours after the initial 1 mg dose of tarlatamab. The patient's presentation met the criteria for grade 4 neurotoxicity. Brain imaging revealed enlarged brain metastases with worsened peritumoral edema, and electroencephalography demonstrated focal epileptiform discharges at the tumor site. The symptoms resolved rapidly (within 36 h) after intensified corticosteroid therapy. Tarlatamab was cautiously continued without recurrence of severe neurotoxicity, and subsequent imaging revealed marked tumor regression. The event was interpreted as TIAN driven by local tumor inflammation rather than ICANS. CONCLUSIONS:Severe acute neurotoxicity may indicate TIAN in patients with SCLC and brain metastases treated with tarlatamab. TIAN presents as severe but reversible symptoms that mimic high-grade ICANS. Differentiating TIAN from ICANS is crucial because TIAN may be manageable, and its recognition may inform individualized decisions regarding therapy continuation under careful monitoring.

INTRODUCTION:Immune checkpoint inhibitors (ICIs) have enabled durable responses in a subset of patients with advanced NSCLC; however, the clinical trajectories and patterns of late disease progression among long-term survivors remain poorly characterized. METHODS:This multicenter retrospective study enrolled patients with advanced NSCLC who received ICI-containing therapy between January 2015 and April 2020. Long-term beneficiaries (LTBs) were defined as patients who achieved both an overall survival (OS) of more than or equal to 4 years and a time to next cytotoxic chemotherapy of more than or equal to 4 years. We performed 4-year landmark analyses, characterized late progression (first radiographic progression ≥4 y from ICI initiation), and evaluated cause-specific mortality using competing risk analysis. RESULTS:Of 3144 patients, 537 (17%) survived more than or equal to 4 years, and 295 (9.4%) were LTBs (median follow-up: 68.6 mo); only 8.1% were lost to follow-up at the data cutoff. ICI was initiated as first-line therapy in 63% of the patients. Among the 235 LTBs without progression at 4 years, 66% discontinued ICI. The lung cancer-specific OS rates were 97.8% and 88.0% at 6 and 8 years, respectively. Late progression occurred in 26 patients (11.1%); all had less than or equal to five lesions, and 84.6% demonstrated indolent progression with growth speeds less than or equal to 10 mm/mo. Local therapy was selected in 42% of the patients after progression. Of 22 deaths more than or equal to 4 years post-ICI, only seven (32%) were lung cancer related, whereas 15 (68%) were from other causes, including secondary malignancies. Competing risk analysis revealed that other-cause mortality consistently exceeded lung cancer-related mortality. CONCLUSIONS:LTBs without progression at 4 years achieved durable control, with lung cancer-specific OS approaching 90% at 8 years of follow-up. Late progression is uncommon, limited in extent, and often treated locally. The predominance of non-lung cancer-related mortality underscores the importance of comprehensive survivorship care.

Only 15% of young-onset breast cancers have identifiable hereditary germline pathogenic variants (PVs) in an established breast cancer susceptibility gene. However, it is believed that a significant proportion of these breast cancers have additional monogenic or rare risk variants that require identification. To uncover novel cancer susceptibility genes, we performed germline whole-exome/genome sequencing of samples from 564 patients with young-onset breast cancer (aged <40 years), as well as samples from 4032 female controls. The identified candidate variants were further genotyped in 6,967 independent breast cancer cases across all age groups. We identified two PVs that were significantly associated with the risk of hormone receptor-negative young-onset breast cancer: POLH p.K589T (OR = 3.65, 95% confidence interval [CI] = 1.28-10.4, P = 0.0095) and RAD51 p.M1fs (OR = 2.15, 95% CI = 1.15-4.02, P = 0.014). When BRCA1/2 PV carriers were excluded from the analysis, only RAD51 p.M1fs retained a significant association. Whole-genome sequencing of tumor samples carrying these germline risk variants revealed that they harbored mutational signatures indicative of a deficiency of homologous recombination. These findings suggest that hereditary POLH p.K589T and RAD51 p.M1fs are candidate variants associated with an increased risk of hormone receptor-negative breast cancer.

INTRODUCTION:Up to 54% of all lung adenocarcinoma (LADC) cases in Asian populations occur in never-smoking women, suggesting that the impact of smoking and other environmental factors on the risk of early-onset LADC is minimal. Genetic factors may play a crucial role in disease development. METHODS:The prevalence of germline pathogenic variants (GPVs) of 454 hereditary cancer and DNA repair genes was evaluated by whole-exome and whole-genome sequencing of 348 early-onset LADC (aged ≤ 40 y) and 1425 later-onset LADC (aged ≥ 41 y) cases. A case-control study comprising 10,302 LADC cases and 7898 healthy controls was performed to identify moderate-risk genetic factors for the disease. Analysis of somatic mutations in 1278 patients with LADC, including 31 patients with GPVs, was also performed. RESULTS:The frequency of GPVs of TP53 and BRCA2 was significantly higher in those with early-onset LADC than in those with later-onset LADC. The detection rates for TP53 and BRCA2 GPVs were 2.9% and 1.7%, respectively, in patients with early-onset LADC, and 0.14% and 0.21%, respectively, in patients with later-onset LADC. Patients with BRCA1 GPVs exhibited a high incidence of concurrent TP53 somatic mutations. Patients with BRCA2 GPVs exhibited deficient homologous recombination in tumors by means of loss of the wild-type allele. A germline ALKBH2 variant, p.Glu35Alafs∗54, was associated with the risk of early-onset LADC, and patients with a deleterious variant exhibited a correlation between SBS4-related somatic mutations and the Brinkman index. CONCLUSION:TP53 and BRCA2 GPVs and the ALKBH2 novel variant are associated with early-onset LADC in Asians.

BACKGROUND:The risk factors associated with treatment resistance to consolidation durvalumab following chemoradiotherapy (CRT) for locally advanced non-small cell lung cancer (NSCLC) have not been well established. METHODS:Extracellular vesicles (EVs) were isolated from the pretreatment serum of 73 patients treated with consolidation durvalumab. Isolation was performed using CD9/CD63 antibodies, and EV proteins were identified using liquid chromatography-tandem mass spectrometry (LC-MS). The relationship between the expression of these proteins and progression-free survival (PFS) after durvalumab initiation was analyzed. RESULTS:The median PFS was not reached(NR) (95 % confidence interval [CI], 17.2 months-NR), andthe 24-month PFS rate was 54.7 % (95 % CI, 44.4-67.4). Proteomic analysis of circulating EVs identified RPS27A, SAA1, and S100A7 as the primary candidate biomarkers. Notably, patients with high RPS27A expression exhibited a significantly shorter PFS compared with those with low expression; hazard ratio (HR) 2.93 (95 % CI: 1.48-5.79),P = 0.002. Similarly, high SAA1 expression was associated with a shorter PFS; HR 2.94 (95 % CI: 1.37-6.30), P = 0.006. High S100A7 expression also correlated with poorer outcomes; HR 2.94 (95 % CI: 1.43-6.04), P = 0.003. In multivariate analysis, a high expression level of RPS27A was identified as an independent predictor of poor PFS; HR 2.29 (95 % CI: 1.01-5.17), P = 0.047. Multivariate receiver operating characteristic (ROC) analysis incorporating these three proteins yielded an area under the curve (AUC) of 0.71 (95 % CI: 0.59-0.83). CONCLUSION:This study demonstratedfavorable PFS outcomes in patients receiving durvalumab consolidation therapy. Circulating EV proteomic analysis identified RPS27A, SAA1, and S100A7, particularly RPS27A, as potential biomarkers for predicting resistance to durvalumab.

8523 Background: Driver-negative lung adenocarcinoma (DN-LUAD) without actionable genomic alterations (AGAs) has a poor prognosis. A deeper understanding of its molecular background, particularly across the entire genome, is crucial for improving risk assessment and therapeutic development. Methods: We performed deep whole-genome sequencing (WGS; tumor depth: 100–120x, normal depth: 30x, paired-end reads: 150 bp) and RNA sequencing using fresh-frozen tissues from four institutions. DN-LUAD was defined as LUAD without known actionable mutations or gene fusions. We analyzed somatic mutations, copy number alterations (CNAs), and structural variations (SVs). Tumors were classified as whole-genome doubling (WGD) if more than half of the autosomal tumor genome showed at least two copies in the major copy numbers of somatic cells. HRDetect score ≥0.7 was used to determine the presence of homologous recombination DNA repair deficiency (HRD) (Davies H, et al. Nat Med. 2017). Gene Set Enrichment Analysis (GSEA) was used for pathway analysis. Results: Among the 745 patients (pts), 517 were classified as having DN-LUAD. WGS identified AGAs undetected by whole-exome sequencing (WES) in 33 (4.4%) pts ( EGFR 13, KRAS 7, BRAF 5, ERBB2 4, MET 4, and HRAS 1). The frequently observed genomic alterations in DN-LUAD are shown in the Table. WGD was observed in 62.1% (n = 321) of DN-LUADs and was significantly associated with higher TMB, CNA, and SV burden. Tumor suppressor gene (TSG) mutations in TP53 , STK11 , and KEAP1 , as well as CDKN2A copy number loss and SVs, were significantly more frequent in WGD pts. HRD was identified in 19 DN-LUAD pts (3.7%), with a significantly higher frequency in the WGD group (WGD vs. without WGD; 4.98% vs. 1.5%, p = 0.016). GSEA showed significant (q < 0.0001) upregulation of cell cycle pathways (E2F targets, G2M checkpoint, and MYC targets) and downregulation of immune pathways (allograft rejection and interferon-gamma response) in DN-LUADs with WGD. Conclusions: This largest-ever WGS study identified AGAs undetectable by WES and uncovered a subgroup of DN-LUAD characterized by increased genomic instability driven by multiple TSG alterations associated with WGD, which was also more likely to exhibit HRD. WGD was associated with the upregulation of cell cycle pathways and the downregulation of immune pathways. These findings highlight the critical role of WGS in elucidating the pathogenesis of DN-LUAD. AllN = 517 WGDn = 321 Without WGDn = 196 p Median TMB (Mutations/Mb) 6.6 10.2 3.0 <0.0001 Median No. of CNAs 101 117 77 <0.0001 Median No. of SVs 185 244 103 <0.0001 TP53 381 (74) 277 (86) 104 (53) <0.0001 SMARCA4 59 (11) 43 (13) 16 (8) 0.094 STK11 56 (11) 42 (13) 14 (7) 0.049 CDKN2A 44 (9) 33 (10) 11 (6) 0.092 KEAP1 40 (8) 34 (11) 6 (3) 0.003 CDKN2A loss 106 (21) 78 (24) 28 (14) 0.009 MET amplification 44 (9) 34 (11) 10 (5) 0.045 CDKN2A 135 (26) 95 (30) 40 (20) 0.028 FHIT 125 (24) 99 (31) 26 (13) <0.0001

INTRODUCTION:Brain metastasis (BM) is a devastating complication of NSCLC, particularly in NSCLC tumors harboring EGFR mutations. However, the genomic alterations driving BM remain poorly understood. METHODS:We analyzed three independent cohorts of resected NSCLCs from Asian patients, including 1081 primary tumor (PT) samples analyzed by whole-genome sequencing (WGS, discovery cohort, n = 180) or whole-exome sequencing (WES, validation cohort, n = 901), eight BM samples, and 17 matched primary-BM pairs analyzed by WGS (BM cohort). Furthermore, RNA sequencing of the available 172 samples in the validation cohort and drug sensitivity profiling using NSCLC cell lines were performed. RESULTS:In the discovery cohort, deletions of the 15q15 chromosomal segment were more enriched in PTs from patients with BM than in those without this deletion (73% versus 39%, p = 0.004). Cumulative BM incidence was significantly higher in PTs with the 15q15 deletion (subdistribution hazard ratio = 3.9, p = 0.008 [Fine-Gray competing risk analysis]), whereas metastases at other organ sites did not differ significantly. These findings were obtained using the validation cohort. The 15q15 deletions significantly co-occurred with EGFR mutations (p = 2.8 × 10-7). In matched PT-BM pairs, the 15q15 deletion was detected exclusively in BMs in 46.7% of patients. The deleted region includes the MGA gene, encoding a suppressor of MYC signaling. Transcriptomic analysis revealed MYC signaling and oxidative phosphorylation (OXPHOS) activation in PTs from patients with BM. NSCLC cell lines harboring the 15q15 deletion were selectively sensitive to elesclomol, an OXPHOS inhibitor. CONCLUSIONS:The 15q15 deletion promotes BM development through aberrant MYC signaling and the subsequent reprogramming of carbohydrate metabolism.

BACKGROUND:In EGFR-mutated lung adenocarcinoma (EGFRm LUAD), EGFR mutations do not necessarily result in increased EGFR expression (EGFR-exp), which differs among patients. However, the factors influencing EGFR-exp and the impact of EGFR-exp on tumor characteristics in patients with EGFRm LUAD remain unclear. PATIENTS AND METHODS:Whole-exome and RNA sequencing were performed for patients with early- and advanced-stage EGFRm LUAD. The patients were classified into low or high EGFR-exp groups based on the median transcripts per million. We retrospectively examined the association between EGFR-exp, genomic characteristics, downstream EGFR signaling activity, tumor microenvironment (TME) status, and clinical outcomes. RESULTS:This study included 450 and 45 patients in the early- and advanced-stage cohorts, respectively. In both cohorts, the EGFR-exp low group exhibited a lower incidence of TP53 co-mutations and EGFR amplification and a higher incidence of EGFR subclonal mutations than the EGFR-exp high group. Furthermore, downstream EGFR signaling pathways, such as the MAPK signaling, were less activated in the EGFR-exp low group. However, this group showed significantly enriched adaptive immune response pathways (Q < 0.0001) and an immune-inflamed TME. Additionally, a low EGFR-exp was a significantly favorable factor for postoperative relapse (odds ratio [OR], 0.6; P = 0.04). However, in the advanced-stage cohort, a low EGFR-exp was a significant risk factor for non-responders to osimertinib (OR, 17.5; P = 0.03). CONCLUSIONS:In EGFRm LUAD, significant associations were observed between EGFR-exp levels and both EGFR signaling pathways and adaptive immune status, which in turn influence clinical outcomes. This large-scale multi-omics analysis highlights the heterogeneity among patients with EGFRm LUAD and emphasizes the need to assess EGFR-exp levels alongside mutation status for optimal treatment strategies in EGFRm LUAD.

INTRODUCTION:Programmed cell death ligand-1 (PD-L1) expression is a predictive biomarker for the efficacy of anti-programmed cell death receptor-1/PD-L1 antibodies in advanced non-small cell lung cancer (NSCLC). Although several assays have been approved for evaluating PD-L1 expression status, inter-assay discordance has been observed between some assays. The clinical significance of these discrepancies is still unclear. METHODS:We retrospectively reviewed treatment-naïve NSCLC patients whose PD-L1 expression was evaluated using both 22C3 and SP142 assays. Among those, efficacy analysis was performed for patients with PD-L1 tumor proportion score (TPS) ≥ 50 % (22C3), who had received first-line pembrolizumab monotherapy. Additionally, transcriptome analysis was conducted in the available tumors with TPS ≥ 50 % to investigate the distinct immune profiles that accompany inter-assay discordance. RESULTS:In total, 611 patients were eligible. Among 198 patients with TPS ≥ 50 %, 91 (46 %) had tumor cell score ≤ 1 (SP142, i.e., inter-assay discrepancy). In the 52 patients who received first-line pembrolizumab monotherapy, treatment efficacy was significantly lower in patients with the discrepancy than that in those without (objective response rate: 18 % vs. 83 %, p < 0.001; median progression-free survival [months]: 3.2 vs. 8.3, p < 0.001). Transcriptome analysis revealed significantly more CD274 splice variants with aberrant 3'-terminal sequences in tumors with the inter-assay discrepancy than in those without. CONCLUSION:The inter-assay discrepancy in the PD-L1 status of tumor cells between the 22C3 and SP142 assays, reflecting an imbalance in the CD274 splice variants, could be a biomarker for primary resistance against pembrolizumab monotherapy in high PD-L1-expressing NSCLCs.

Abstract Introduction: The expression of programmed cell death ligand-1 (PD-L1) is a predictive biomarker for the efficacy of pembrolizumab in advanced non-small cell lung cancer (NSCLC). While several assays have been approved for evaluating PD-L1 expression status, the inter-assay discordances are observed. The clinical significance of these discrepancies is still unclear. Methods: We retrospectively reviewed treatment-naïve NSCLC patients with available PD-L1 expression status by both 22C3 and SP142 assays. Among these patients, efficacy analysis was performed in those with PD-L1 tumor proportion score (TPS) ≥ 50% (22C3) who received the first-line pembrolizumab monotherapy. Additionally, we conducted RNA-sequence analysis in the patients with TPS ≥ 50% to investigate the distinct immune profiles with the inter-assay discordance. Results: A total of 611 patients were evaluated for PD-L1 expression status by both assays. Of 198 patients with TPS ≥ 50%, 91 (46%) had TC score ≤ 1 (SP142, i.e., inter-assay discrepancy). In 52 patients who received the first-line pembrolizumab monotherapy, the efficacy in patients with the inter-assay discrepancy was significantly lower than those without the discrepancy (shown in Table). While there was no difference in tumor microenvironments such as the tumor-infiltrating lymphocytes, the transcriptome analysis revealed significantly more CD274 splicing variants with aberrant 3’-terminal sequences in tumors with the inter-assay discrepancy than those without discrepancy, leading to the suppressed PD-1/PD-L1 pathway in those with the discrepancy. Conclusion: The inter-assay discrepancy in the PD-L1 expression status on tumor cells between the 22C3 and SP142 assays, reflecting an imbalance of the CD274 splicing variants, could be a primary-resistance mechanism to the pembrolizumab monotherapy in highly PD-L1-expressing NSCLCs. PD-L1(22C3) Discrepancy Pts. ORR(%) p value Median PFS(months) 95%CI p value TPS ≥ 50% yes 17 18 < 0.001 3.2 0.8–5.5 < 0.001 no 35 83 8.3 5.7–24.2 Citation Format: Jun Miyakoshi, Jumpei Kashima, Masayuki Shirasawa, Masahiro Torasawa, Yuji Matsumoto, Ken Masuda, Yuki Shinno, Yusuke Okuma, Tatsuya Yoshida, Yasushi Goto, Hidehito Horinouchi, Kouya Shiraishi, Takashi Kohno, Noboru Yamamoto, Yasushi Yatabe, Yuichiro Ohe. Clinical significance of inter-assay discrepancy in PD-L1 expression evaluation for the efficacy of pembrolizumab in advanced NSCLC patients with high PD-L1 expression [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 5146.

BACKGROUND AND OBJECTIVE:Lymphangioleiomyomatosis (LAM) is a rare neoplastic disease associated with the functional tumour suppressor genes TSC1 and TSC2 and causes structural destruction in the lungs, which could potentially increase the risk of lung cancer. However, this relationship remains unclear because of the rarity of the disease. METHODS:We investigated the relative risk of developing lung cancer among patients diagnosed with LAM between 2001 and 2022 at a single high-volume centre in Japan, using data from the Japanese Cancer Registry as the reference population. Next-generation sequencing (NGS) was performed in cases where tumour samples were available. RESULTS:Among 642 patients diagnosed with LAM (sporadic LAM, n = 557; tuberous sclerosis complex-LAM, n = 80; unclassified, n = 5), 13 (2.2%) were diagnosed with lung cancer during a median follow-up period of 5.13 years. All patients were female, 61.5% were never smokers, and the median age at lung cancer diagnosis was 53 years. Eight patients developed lung cancer after LAM diagnosis. The estimated incidence of lung cancer was 301.4 cases per 100,000 person-years, and the standardized incidence ratio was 13.6 (95% confidence interval, 6.2-21.0; p = 0.0008). Actionable genetic alterations were identified in 38.5% of the patients (EGFR: 3, ALK: 1 and ERBB2: 1). No findings suggested loss of TSC gene function in the two patients analysed by NGS. CONCLUSION:Our study revealed that patients diagnosed with LAM had a significantly increased risk of lung cancer. Further research is warranted to clarify the carcinogenesis of lung cancer in patients with LAM.

Objectives: Lorlatinib, a third-generation anaplastic lymphoma kinase (ALK) inhibitor, causes distinct adverse events (AEs), including hyperlipidemia and central nervous system (CNS) disorders. Although dose modifications are recommended to manage these AEs, whether dose modifications can achieve optimal blood lorlatinib concentrations and reduce the incidence of lorlatinib-induced AEs remains unclear. Therefore, we investigated the association between lorlatinib exposure and AEs in each patient. Materials and methods: We retrospectively reviewed patients with advanced ALK-rearranged non-small cell lung cancer treated with lorlatinib between November 2018 and July 2022. Serum lorlatinib concentrations were assessed using high-performance liquid chromatography-tandem mass spectrometry. All AEs were evaluated using the Common Terminology Criteria for Adverse Events version 5.0. Results: The median age of the 55 eligible patients was 59 years (range: 23-79 years). All patients were administered lorlatinib after first line ALK-tyrosine kinase inhibitor failure. Grade >= 3 AEs occurred in 25 patients (25/55, 45 %), including hyperlipidemia in 17 (17/55, 31 %), CNS disorders in 7 (7/55, 13 %), and edema in 6 (6/55, 11 %). Dose modification was required in 23 patients (23/55, 42 %). Among the 36 patients with available data on serum lorlatinib levels at day 28 (+/- 14) +/- 14) and no drug dose modifications, lorlatinib serum concentrations were significantly higher in patients with grade >= 3 AEs than in those without AEs (median: 462 ng/mL vs. 177 ng/mL, p < 0.01). In eight patients with data on serial lorlatinib serum concentrations following dose modifications, lorlatinib serum concentrations were effectively reduced, facilitating the ongoing administration of lorlatinib. Additionally, no significant difference was observed in the landmark analysis of progression- free survival between patients with dose modification within the first 16 weeks and those without (median: 24.8 months vs. 10.1 months, p = 0.46). Conclusion: Dose modification of lorlatinib was associated with successful management of AEs and decreased serum concentration of lorlatinib.