Trastuzumab, combined with chemotherapy, is the current standard treatment for both metastatic and early-stage HER2-positive (HER2 +) breast cancer. One of the mechanisms of action of trastuzumab is antibody-dependent cellular cytotoxicity (ADCC), which involves engaging FcγRIIIA (CD16) on natural killer (NK) cells. A competent immune system and properly functioning NK cells are crucial for effective ADCC, as they can influence favorable clinical outcomes. Resistance to trastuzumab often develops after about one year. We previously reported that elevated levels of miR-19a-3p in the serum of patients with metastatic HER2 + breast cancer treated with trastuzumab were associated with a favorable prognosis. Here, we aim to identify the mechanism and the immune cells responsible for elevated serum levels of miR-19a-3p. Peripheral blood mononuclear cells (PBMCs) from healthy individuals were used to isolate naïve CD4 + T cells and NK cells. Naïve CD4 + T cells were polarized into CD4 + Th1 and CD4 + Th2 cells. NK cells were utilized for the ADCC assay. Levels of transcription factors, cytokines, and miR-19a-3p were measured using RT-qPCR. Surface markers and cytokines were analyzed by flow cytometry to characterize immune cell phenotypes. In vitro NK cell-mediated ADCC resulted in increased levels of miR-19a-3p released into the supernatants after killing breast cancer cells. In vitro polarized CD4 + Th1 cells expressed and secreted higher levels of miR-19a-3p than CD4 + Th2 cells. Over a long-term in vitro culture (24 days), anti-CD3/CD28 restimulation sustained higher levels of miR-19a-3p in CD4 + Th1 cells compared to CD4 + Th2 cells and their respective supernatants. CD4 + Th1 cells developed a central memory T (TCM) phenotype (CD45RO + CCR7 + CD62L +) and expressed and secreted higher levels of miR-19a-3p than CD4 + Th2 cells. In patients with HER2 + metastatic breast cancer, those with elevated serum levels of miR-19a-3p and a favorable prognosis had a larger percentage of circulating activated T cells and NK cells in their blood compared to patients with lower serum levels of miR-19a-3p and a poor prognosis. The small cohort (n = 15) limits the statistical power of our retrospective study. Our findings suggest that elevated levels of miR-19a-3p in the serum of patients with HER2 + metastatic breast cancer may result from effective NK cell-mediated ADCC and activation of CD4 + Th1 cells, which could be responsible for the anti-tumor immune response associated with a favorable prognosis. Blood levels of miR-19a-3p might help identify breast cancer patients who have effective trastuzumab-induced anti-tumor immune responses.

e20172 Background: Thymic epithelial tumors (TETs) are rare neoplasms arising in the anterior mediastinum. TETs are stratified using the WHO classification system, which remains fundamental to prognostication and clinical decision-making. However, metastasis and aggressive tendencies are observed in all WHO subtypes, warranting further understanding of the pathobiology of TETs. Although the expression of some lineage-defining transcription factors (LTFs) in TETs has been reported previously, their clinical significance remains largely unexplored. We, herein, evaluate expression of 14 LTFs and the corresponding impact on patient outcomes across two large clinicogenomic cohorts. Methods: Expression of LTFs previously identified in thymic tissues ( AIRE , FOXA1 , FOXA2 , FOXA3 , FOXJ1 , GRHL1 , GRHL2 , GRHL3 , HNF4A , HNF4G , MYOG , POU2F3 , SOX8 and SPIB ) was evaluated in two cohorts (Caris and TCGA). The Caris cohort consisted of whole-transcriptome data from FFPE-preserved TETs using the Caris Life Sciences platform, which was also supplemented with clinicopathological data curated from the Caris data repository as well as insurance claims data. Blinded histology review with WHO subtype classification was performed by a pathologist with expertise in thymic tumors (S.S.B.). TCGA data was obtained from publicly available resources. Statistical methods employed included quantitative Cox regression and Log-Rank methods for analyses with overall survival (OS) as well as the Mann-Whitney U test for single gene expression comparisons. Results: After filtering out LTFs with low expression (median TPM<1) from the Caris dataset (n=107, 71% WHO type B3/C), seven LTFs were retained for further analysis including FOXA1 , GRHL1 , GRHL2 , GRHL3 , HNF4G , POU2F3 , and SPIB . Univariate survival analyses revealed positive associations with FOXA1 , HNF4G , and POU2F3 , while only FOXA1 and POU2F3 remained independent biomarkers in a multivariate model ( FOXA1 : HR = 1.02; p = 0.05; POU2F3 : HR = 1.01; p = 0.03). Multivariate analysis from the less aggressive TCGA cohort (n=117; 18% WHO type B3/C) also validated FOXA1 and POU2F3 associations with poor outcomes ( FOXA1 : HR = 1.14, p = 0.008; POU2F3 : HR = 1.04, p = 0.05). Tumors highly expressing both LTFs were associated with exceptionally poor outcomes as compared to tumors with low expression of both (median OS: 8 mos. vs 58 mos.; p<0.0001). Thymic carcinomas were associated with higher expression of both LTFs when compared to thymomas. Robust correlation between FOXA1 and POU2F3 was observed as well as with markers of tuft cell differentiation, KIT and GFI1B . Conclusions: FOXA1 and POU2F3 expression are markers of poor outcomes and likely related to tuft cell differentiation in thymic epithelial tumors. Testing these biomarkers prospectively may improve diagnostic resolution for previously established classification systems.

Abstract Background: Cervical adenocarcinoma exhibits poor responsiveness to radiotherapy and inferior survival compared with squamous cell carcinoma. The molecular mechanisms underlying its intrinsic radiation resistance remain largely unknown. Methods: Integrative analysis of TCGA and our RNA-seq cohort identified CYP4A22-AS1 as one of the most upregulated lncRNAs in cervical adenocarcinoma. qRT-PCR confirmed higher CYP4A22-AS1 expression in tumors with short disease-free survival (DFS). RNA pulldown, mass spectrometry, and RIP assays defined YBX1 as a direct CYP4A22-AS1-binding partner. Immunofluorescence and ChIP-qPCR assays examined YBX1 nuclear localization and promoter occupancy of PGK1, respectively. Functional effects of CYP4A22-AS1, YBX1, and PGK1 silencing were assessed by CCK-8, colony, EdU, and TUNEL assays in HeLa and C33A cells, and validated in xenograft models. Results: CYP4A22-AS1 was markedly overexpressed in cervical adenocarcinoma relative to normal cervix. Clinically, high CYP4A22-AS1 and PGK1 levels correlated with shorter DFS. Mechanistically, CYP4A22-AS1 binds YBX1 and enhances its nuclear translocation, thereby promoting YBX1 recruitment to the PGK1 promoter and activating glycolytic metabolism. Knockdown of CYP4A22-AS1 or PGK1 suppressed proliferation and markedly increased radiosensitivity both in vitro and in vivo. Conclusions: This study identifies CYP4A22-AS1 as a novel oncogenic lncRNA that drives radiation resistance through YBX1-mediated PGK1 transactivation. Clinically, CYP4A22-AS1 overexpression predicts poor outcome, while its inhibition restores radiation sensitivity. Targeting the CYP4A22-AS1-YBX1-PGK1 signaling axis offers a promising therapeutic strategy to overcome radioresistance in cervical adenocarcinoma. Citation Format: Mingyi Zhou, Chunlai Li, Cristina Ivan, Simone Anfossi, Linda Fabris, Melanie Winkle, Recep Bayraktar, Meng Chen, Lan Pang, Masayoshi Shimizu, Francois Claret, George Calin, . CYP4A22-AS1-YBX1 axis drives radiation resistance in cervical adenocarcinoma via PGK1 transactivation [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 1366.

MiR-210 is widely recognized as the quintessential hypoxia-responsive miRNA and is thought to fine-tune various facets of cellular homeostasis. We hereby present an integrative appraisal of the phenotypic and molecular repercussions of disrupting the corresponding locus in human and mouse cells using multiple genetic strategies. In brief, MIR210 deletion led to decreased cellular fitness and suboptimal responses to several stress types. Transcriptomic comparisons via different profiling platforms, performed independently by members of this collaboration, revealed consistent deregulation of neighboring genes, in locus-disrupted cells. Interestingly, the anticipated enrichment of miR-210 targets failed to materialize in unbiased analyses. Our results point to the biological significance of unrecognized regulatory elements that overlap miRNA genes and should serve as a note of caution for studies based on the genetic disruption of such loci.

Despite the clinical use of anti-vascular endothelial growth factor (VEGF) antibodies (AVAs) in cancer therapy, resistance frequently develops, leading to disease progression. To address this, we identify a previously unknown role for breast cancer type 1 susceptibility protein (BRCA1)-associated RING domain 1 (BARD1) in modulating AVA sensitivity. Epigenetic modulation-via global and targeted DNA methylation-reveals BARD1 as a key regulator of angiogenesis. Sequential treatment with azacytidine overcomes AVA resistance in vivo. To enable precise epigenetic reactivation, we develop a liposomal CRISPR-deactivated Cas9 (dCas9)-TET1 system guided by BARD1-targeting single-guide RNAs (sgRNAs). This platform achieves CpG-specific demethylation of the BARD1 promoter, restores expression, and enhances AVA response. Additionally, BARD1 restoration, through either dCas9-TET1 or small interfering RNA (siRNA), significantly reduces tumor growth in combination with AVA in ovarian cancer models. These findings uncover a previously unrecognized function of BARD1 in tumor angiogenesis and demonstrate the potential of gene-specific epigenetic targeting to overcome AVA resistance.

Background: Proteins secreted by cancer-associated fibroblasts (CAFs) can influence inflammation and recruit external immune cells to the tumor microenvironment. However, the cross[1]talk between CAFs and immune cells in the tumor microenvironment is not well understood. Therefore, in this study, we determined the association of proteins secreted by the CAF secretome with immune cell infiltration in patients diagnosed with breast cancer (BC) and ovarian cancer (OC). Methods: The Human Protein Atlas was explored to identify potential gene targets associated with the CAFs secretome, including inhibin subunit beta A (INHBA), Dickkopf-related protein 3 (DKK3), and follistatin (FST). BC and OC patient data sets (n=14 samples) in the Gene Expression Omnibus were analyzed as discovery cohorts to compare CAFs and normal fibroblasts. Kaplan-Meier Plotter, Gene Expression Profiling Interactive Analysis, and Tumor Immune Estimation Resource bioinformatics databases were explored to computationally understand relationships between gene expression, tumor prognosis, and immune infiltration. Single cell analysis datasets and opal multiple staining were used to validate the contribution of CAFs to overexpression of the gene targets in BC and OC cancer patients. Proteogenomic BC and OC data sets (n=46) from Clinical Proteomic Tumor Analysis Consortium were used for validation of the correlation between marker expression and immune cell infiltration. Results: Increased expression of INHBA, DKK3, and FST correlated with shortened survival of patients with BC and OC. All three genes were negatively associated with multiple tumor-infiltrating immune cell types, both in RNA and protein levels. INHBA expression was positively correlated with the expression of T cell exhaustion markers. Pathway analysis demonstrates that INHBA’s role in immune infiltration may be a result of its association with the Smad2 signaling pathway, which promotes the transition of fibroblasts to activated CAFs via an inflammatory response. In BC and OC, the three genes were generally positively correlated with immune cells characterized as “protumor”, such as macrophages, CAFs, and neutrophils and negatively correlated with those considered as “antitumor,” such as B cells. Conclusion: INHBA, DKK3, and FST are candidate therapeutic targets for the treatment of breast, ovarian, and other cancers. Their therapeutic effect should be evaluated in the context of regulating infiltration and inflammation of the tumor microenvironment. Citation Format: Anjali Agrawal, Elaine Stur, Emine Bayraktar, Sara Corvigno, Kirill Pevzner, Gali Arad, Yibo Dai, Sisy Chen, Robiya Joseph, Nitzan Simchi, Eran Seger, Cristina Ivan, Anil Sood. Potential Impact of the Cancer-Associated Fibroblast Secretome in the Tumor Microenvironment [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2024; 2024 Dec 10-13; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2025;31(12 Suppl):Abstract nr P2-06-04.

Background:Altered expression of microRNAs (miRNAs) is implicated in lung carcinogenesis, but little research has investigated the association of miRNA alterations with lung cancer stage or smoking status. To identify such alterations in lung adenocarcinoma, we conducted miRNA profiling. Methods:Lung adenocarcinoma specimens and paired nonneoplastic lung tissues from 58 patients who had received no preoperative therapy and underwent tumor resection from 1991 to 2006 were collected from tumor tissue banks. Thirty (52%) of the tumors were stage I and 28 (48%) stage II or higher. Twenty-five (43%) patients were nonsmokers and 33 (57%) were smokers. To identify miRNAs of which its expression was associated with disease stage and/or smoking status, we performed subgroup analyses based on disease stage and smoking status, alongside overall profiling of all samples. miRNA expression was profiled using a microarray kit in tumors and paired nonneoplastic lung tissues. We assessed the fold changes (FCs) and local false discovery rates (FDRs) for the intercept in addition to P values. Results:Of the 706 miRNAs examined, 64 had overall altered expression. Of these, 36 (56%) were associated with all stages of adenocarcinoma in both smokers and nonsmokers. In stage-based analysis, 20 altered miRNAs were associated with stage I disease and 38 with stage II or higher disease. Of the 20 miRNAs associated with stage I adenocarcinoma, 4 were miR-200 family members, and miR-200b was the most highly upregulated. Analysis based on smoking status identified 38 altered miRNAs in nonsmokers and 6 altered miRNAs in smokers. Three miR-200 family members were significantly altered in nonsmokers. Among the miRNA alterations associated with stage I adenocarcinoma in nonsmokers, we found alterations in all 5 miR-200 family members (miR-200b, miR-200a, miR-141, miR-429, and miR-200c). Conclusions:Our identification of miRNA alterations associated with early-stage lung adenocarcinoma in nonsmokers, especially alterations in the miR-200 family, suggests that these miRNAs may play a unique role in the early stages of lung carcinogenesis and progression in nonsmokers and that they may be useful as markers for the early detection of lung cancer.

Pancreatic cancer is known to be the third leading cause of death in the United States resulting from a high frequency of KRAS and TP53 mutations. Several meta-analyses have examined the presence of fusion genes in pancreatic tumors; however, larger cohort evaluation on a single platform is lacking. The primary objective of this study was to characterize the oncogenic kinase gene fusions and co-occurring pathogenic variants including mutations, amplifications and deletions. We performed molecular profiling with WES and WTS of 12418 pancreatic primary and metastatic tumor specimens submitted for clinical molecular profiling at Caris Life Sciences via Caris MI Tumor Seek (Phoenix, AZ). Targeted NGS evaluating single nucleotide alterations, copy number alterations, and fusions was performed on these specimens. Our research found that fewer than 2% of pancreatic tumor samples contained recurrent oncogenic kinase fusions. We identified fusions in several well-known genes, including BRAF, FGFR2, RAF1, PRKACA, RET, ALK, and MET, which have been previously documented. In addition, we observed recurring fusions in RPS6KB1, ROCK1, and STK3. The most frequently involved fusion partners in pancreatic tumors were ATP1B1, CLTC, DNAJB, EML4, and SND1, all of which are highly expressed in these tumors. As a result, chimeric kinase fusion proteins exhibited higher gene expression in fusion-positive cases compared to tumors lacking these fusions. Most kinase fusions are 3’ partners, although ROCK1 fusions, like the more common FGFR2 fusions, are 5’ partners, which are controlled by ROCK1 and FGFR2 promoters for their gene expression. Notably, fusions involving RPS6KB1, ROCK1, and STK3 were more frequently associated with co-occurring KRAS mutations compared to other kinase fusions. Mutational analysis also revealed co-occurring pathogenic mutations in TP53, CDKN2A, SMAD4, ARID1A, GNAS, BCOR, RNF4, ATM, and BRCA2. Furthermore, the frequency of kinase gene fusions was higher in metastatic sites, particularly the liver, compared to primary tumors, suggesting metastatic progression. Histological analysis showed that over 90% of kinase gene fusions were found in ductal adenocarcinoma. We report kinase fusions from largest pancreatic cancer cohort on the same platform, aiming to identify recurrent, oncogenic kinase fusions. In addition to confirming known fusions, we found new partners for several oncogenic kinases. The identified oncogenic kinase fusions are typically 3' partners, with their expression regulated by the corresponding 5' partners. It was noted that most kinase fusions showed higher expression in fusion-positive cases compared to fusion-negative cases for respective genes. Additionally, we have observed increase in kinase gene fusions in metastatic sites. Shivani Jagannathan Murali, Bharath Kumar Karre, Cristina Ivan, Chao Sima, Asfetaw Abera, David Spetzler, Milan Radovich, Heather O'neill, Ravi Chakra Turaga. Kinase fusion landscape of pancreatic cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 3781.

This study aims to carry out a pan cancer analyses of SMARCA4 genomic variants, including fusions, protein truncating variants (PTVs) and non-truncating variants(non-PTVs). Histological correlation of genomic variants for SMARCA4 expression may help identify tumors deficient in SMARCA4, making them potential candidates for treatment with synthetically lethal SMARCA2 inhibitors. We analyzed 350, 413 samples from Caris Life Sciences for clinical molecular profiling at Caris Life Sciences via Caris MI Tumor Seek (Phoenix, AZ). Gene fusion detection was performed on mRNA from FFPE tumor samples using the Illumina NovaSeq platform. Fusions with ≥ 3 junction reads were reviewed for domain prediction, breakpoints, frame retention, and co-occurring alterations. WES data was also examined for pathogenic mutations to assess fusion oncogenicity. IHC for SMARCA4 (N-terminal & C-terminal) and SMARCA2 was conducted on FFPE sections from 35 patients with SMARCA4 variants, with staining scored by a board-certified pathologist. Pan-Cancer analysis of SMARCA4 alterations revealed that NSCLC (30.7%) cases harbored the largest percentage of mutations followed by the CRC (15%) and UTNE (10%). Further, analysis through IHC revealed that patients with PTVs such as nonsense and frameshift mutations, were typically characterized by shorter/ truncated SMARCA4 protein and exhibited complete loss of SMARCA4 expression through IHC, often accompanied by an increase in SMARCA2 expression by IHC. In contrast, patients with missense mutations (non-PTVs) retained SMARCA4 expression, with minimal changes in SMARCA2 staining. In cases with SMARCA4 fusions detected, IHC analysis revealed an ambiguous result with some fusions retaining expression of SMARCA4 protein and others that did not. Patient cases harboring SMARCA4 fusions that displayed loss of SMARCA4 and concordant increase in SMARCA2 staining, would make good candidates for SMARCA2 inhibitor treatment. WTS analyses also revealed a decrease in SMARAC4 expression in PTVs, but not in non-PTVs. Further, we noted that other prevalent co-occurring mutations included TP53(55.7%), KRAS(19.7%) and ARID1A(16.7%) to be the most prominent alterations in patients with SMARCA4 alterations. Patients with SMARCA4 PTVs had poor prognosis compared to those without SMARCA4 variants in the NSCLC and UTNE indication. SMARCA4 mutations, including frameshift and nonsense mutations, are common in NSCLC, UTNE, and OSEC. These alterations, particularly PTVs, make SMARCA4-deficient tumors potential targets for SMARCA2 inhibitor therapy. Some cases with SMARCA4 fusions or missense mutations may also be eligible for treatment, but these should be assessed individually, considering co-occurring mutations, copy number amplifications and fusion partners. This will aid in refining patient selection and improving treatment outcomes using companion diagnostics. Shivani Jagannathan Murali, Bharath Kumar karre, Cristina Ivan, Kelly Craven, Stephanie Williams, Shikha Mahajan, Asfetaw Abera, Chao Sima, David Spetzler, Milan Radovich, Heather O'neill, Ravi Chakra Turaga. Functional characterization of SMARCA4 genomic variants [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 674.

Protein kinase C (PKC) is a phospholipid-dependent, calcium-independent Ser/Thr protein kinase family with eight members, each involved in processes including apoptosis, migration, and proliferation. PRKCI promotes tumor growth in many types of cancer, including pancreatic, lung, ovarian, and renal cell carcinoma. Structurally, PKCι has an N-terminal regulatory domain and a C-terminal catalytic domain. While PRKCI fusions have been reported, they remain uncharacterized. This study aims to investigate these fusions in gastric cancer and esophageal junction carcinoma, exploring potential mechanisms and co-occurring mutations to inform targeted therapies that could improve patient outcomes. We performed a comprehensive analysis of PRKCI fusions identified from whole transcriptome sequencing (WTS) and associated co-occurring pathogenic mutations from whole exome sequencing (WES). We screened 6437 EJC and 3615 gastric adenocarcinoma cancer cases retrospectively and calculated the frequency of occurrence of PRKCI fusions. Additionally, we characterized and mapped the domains of PRKCI fusions and its partners. We further analyzed patient demographics, co-occurring mutations and copy number amplifications. PRKCI fusions were identified in 0.19% and 0.3% of EEJC and Gastric adenocarcinoma cancer cases. All the PRKCI fusions are formed with other genes on chromosome 3 from intra-chromosomal rearrangement (PHC3, TNIK, SKIL and YES1) proximal to the PRKCI gene. The most common fusions of PRKCI are formed with partner PHC3 and SKIL which account for 11 patient cases. This is followed by 4 cases of TNIK and unique cases of fusions with LRRC31, PCYT1A, ERBB2, VPS13D and YES. Further, the genomic landscape of PRKCI indicated co-occurring mutations in TP53 (46.7%), ARID1A (13.3%), 6.7% inactivating mutations of ATM, FBXW7, CDH1. Expression of PRKCI in PRKCI fusion positive cases was significantly higher as opposed to patient cases that did not harbor gene fusions. PRKCI fusion cases were commonly observed to have PRKCI amplification (46.7%). PRKCI fusions lack the regulatory domain, leading to constitutive kinase activation, which may drive tumor progression by disrupting normal regulation and altering protein localization. Our studies are the first to identify PRKCI fusions in gastric cancer and EEJC patients, along with associated mutations and amplifications suggesting a novel mechanism. This highlights the importance of personalized molecular diagnostics to guide tailored treatments. PKCι kinase inhibitors may improve survival and outcomes for patients with PRKCI fusions in gastric cancer and EEJC. Shivani Jagannathan Murali, Bharath Kumar Karre, Cristina Ivan, Chao Sima, Asfetaw Abera, David Spetzler, Milan Radovich, Heather O'neill, Ravi Chakra Turaga. Recurrent PRKCI gene fusions represent a drug target in gastric cancer and esophageal junction cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 1474.

The mutational landscape of phylogenetically ultraconserved elements (UCEs), especially those in noncoding DNAs (ncUCEs), and their functional relevance in cancers remain poorly characterized. Here, we perform a systematic analysis of whole-genome and in-house targeted UCE sequencing datasets from more than 3000 patients with cancer of 13,736 UCEs and demonstrate that ncUCE somatic alterations are common. Using a multiplexed CRISPR knockout screen in colorectal cancer cells, we show that the loss of several altered ncUCEs significantly affects cell proliferation. In-depth functional studies in vitro and in vivo further reveal that specific ncUCEs can be enhancers of tumor suppressors (such as ARID1B) and silencers of oncogenic proteins (such as RPS13). Moreover, several miRNAs located in ncUCEs are recurrently mutated. Mutations in miR-142 locus can affect the Drosha-mediated processing of precursor miRNAs, resulting in the down-regulation of the mature transcript. These results provide systematic evidence that specific ncUCEs play diverse regulatory roles in cancer.

Tissue transglutaminase (TG2) is a multifunctional protein with roles in multiple diseases including cancer. It was found overexpressed in many solid tumors in the past two decades. In ovarian cancer, it emerged as a potential therapeutic target due to its involvement in processes like metastasis and chemo- and radiotherapy resistance. Generally, TG2 expression in cancer cells has been extensively studied and linked to increased tumor progression. However, its role in the host was less studied. Our research explored the anti-tumor immune response using a TG2KO syngeneic ovarian cancer mouse model, as compared to wild-type animals. We observed decreased tumor burden and increased survival upon i.p. injection of ID8 mouse ovarian cancer cells in TG2KO mice, as compared to wild-type. In the absence of TG2 in the host, an increased infiltration of CD8+ T cells in ascites was evidenced by FACS, while myeloid cells were less present. The TG2KO CD8+ T cells showed increased activation and increased effector function. Moreover, these cells showed an increased cancer cell killing capacity. At the molecular level, this phenotype was supported by attenuated STAT3 phosphorylation. Cancer cells in ascites collected from TG2KO mice showed a gene signature corresponding to IFN-γ response. Overall, these data show decreased tumor progression and increased effector phenotype in CD8+ T cells when TG2 is absent in the host (Sima LE et al. (2021) Journal for ImmunoTherapy of Cancer 9, e002682). When looking to confirm these results in human OC tumor microarrays (TMAs), we found an inverse correlation between human stromal (but not tumor) TG2 expression and CD8+ T cells infiltration by IHC staining followed by StrataQuest image cytometry. Using Opal-based multiplex IHC staining, we identified TG2 in a subset of cancer associated fibroblasts (CAFs). TG2 was previously reported as a non-toxic druggable target involved in “outside-in” signaling downstream of integrins. Further, we used inhibitors developed towards TG2-fibronectin interaction (Sima LE et al. (2019) Mol Cancer Ther 18, 1057-1068) to evaluate how they affect the cancer cells-fibroblasts cross-talk. This treatment prevented the self-assembly of heterospheroids. This is relevant for the situation in the tumor microenvironment, where we found TG2-FN interaction present in both stroma and tumor areas. These results support the potential use of TG2 directed agents in ovarian cancer therapy, as we propose TG2 as a new immunomodulatory target. Our current efforts also include using FLASH radiation to sensitize cancer cells to immunotherapy. This research was supported by funding from the US Department of Veterans Affairs, Robert H Lurie Comprehensive Cancer Center, UEFISCDI (PN-III- P1-1.1-TE- 2019-0670; PN-III-P2-2.1-PED-2019-1543) and IFA (ELI-RO_10/2024).

Background and aim:Tacrolimus, a widely used immunosuppressive drug in kidney transplant recipients, exhibits a narrow therapeutic window necessitating careful monitoring of its concentration to balance efficacy and minimize dose-related toxic effects. Although essential, this approach is not optimal, and tacrolinemia, even in the therapeutic interval, might be associated with toxicity and rejection within range. This study aimed to identify specific urinary metabolites associated with tacrolimus levels in kidney transplant patients using a combination of serum high-precision liquid chromatography-mass spectrometry (HPLC-MS) and machine learning algorithms. Methods:A cohort of 42 kidney transplant patients, comprising 19 individuals with high tacrolimus levels (>8 ng/mL) and 23 individuals with low tacrolimus levels (<5 ng/mL), were included in the analysis. Urinary samples were subjected to HPLC-MS analysis, enabling comprehensive metabolite profiling across the study cohort. Additionally, tacrolimus concentrations were quantified using established clinical assays. Results:Through an extensive analysis of the HPLC-MS data, a panel of five metabolites were identified that exhibited a significant correlation with tacrolimus levels (Valeryl carnitine, Glycyl-tyrosine, Adrenosterone, LPC 18:3 and 6-methylprednisolone). Machine learning algorithms were then employed to develop a predictive model utilizing the identified metabolites as features. The logistic regression model achieved an area under the curve of 0.810, indicating good discriminatory power and classification accuracy of 0.690. Conclusions:This study demonstrates the potential of integrating HPLC-MS metabolomics with machine learning algorithms to identify urinary metabolites associated with tacrolimus levels. The identified metabolites are promising biomarkers for monitoring tacrolimus therapy, aiding in dose optimization and personalized treatment approaches.

ERBB4, a member of the ERBB family of receptor tyrosine kinases, has received less attention than its counterparts, ERBB1 (EGFR), ERBB2 (HER2) and ERBB3(HER3), despite its established involvement in oncogenesis and its potential as a therapeutic target. While the role of ERBB4 and its isoforms in neuronal development is well understood, mutations and fusions of ERBB4 have been linked to several types of cancer, however they have not been fully explored. Further research into ERBB4 fusions and their oncogenic mechanisms is essential for advancing targeted therapies. We performed a comprehensive analysis of ERBB4 fusions using next-generation sequencing (NGS) data from a total of 216, 176 formalin-fixed paraffin-embedded (FFPE) tumor samples across various solid tumors. NGS was conducted on DNA isolated from these samples and RNA fusions were detected using whole transcriptome sequencing (WTS). We analyzed patient demographics, fusion partners, ERBB4 isoforms, and co-occurring mutations. Additionally, tumor mutational burden (TMB) was assessed, and data from public bioinformatic resources were used for comparison. ERBB4 fusions were detected in 0.25% of the 24 solid tumor types analyzed, with the highest prevalence observed in OSEC, BRCA, BLCA, UTNE, and NSCLC. Notably, 63.1% of ERBB4 fusions involved the 3' end of the gene and included various fusion partners. The most common fusion partner was IKZF2:ERBB4, accounting for 58.2% of cases. Exon 2 was identified as the most frequent breakpoint in ERBB4 fusions. As a result, 5' ERBB4 fusions retain only the signal peptide, while 3' fusions lose the signal peptide, which may impair translocation to the cell membrane. The most common isoform observed in ERBB4 fusions was the JMa-Cyt2 isoform, seen in over 83% of cases. This isoform skips exon 26, which contains a PI3K phosphorylation site critical for ubiquitination and degradation. Further analyses revealed a significant upregulation of ERBB4 expression in ERBB4 fusion positive cases in BLCA(P>0.001), OSEC(P>0.001), and NSCLC(P>0.001). Pathway analyses revealed upregulation of Myc targets indicating ERBB4-ICD-STAT5A complex in promoting cell proliferation. Co-occurring mutations are enriched in TP53, PIK3CA, and TERT in ERBB4 fusions. Our study highlights the frequent occurrence of ERBB4 fusions across various cancer types, with IKZF2 being the most common fusion partner. The prevalence of the ubiquitination-resistant JMa-Cyt2 isoform and the retention of cytoplasmic (kinase) domain in many fusions suggest that ERBB4 fusions may contribute to oncogenesis through a STAT5A-dependent mechanism, independent of PI3K. These findings underscore the potential of ERBB4 fusions as therapeutic targets and suggest that gamma-secretase inhibitors could be a promising treatment option for patients with these fusions. Shivani Jagannathan Murali, Bharath Kumar Karre, Cristina Ivan, Chao Sima, Asfetaw Abera, David Spetzler, Milan Radovich, Heather O'neill, Ravi Chakra Turaga. The ERBB4 exon skipping isoform JMA-CYT2 is the dominant isoform of ERBB4 gene fusions [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 4761.

Supplementary Figure S1 Expression levels of miR-196b-5p in different tumor stages and tumor grade. (A-B) In cohort 1, miR-196b-5p expression decreased with (A) higher tumor stage, whereas no difference could be observed for (B) tumor grade. (C-D) In cohort 2, (C) stage IV tumors and (D) high grade tumors showed the lowest levels of miR-196b-5p. Supplementary Figure S2 Kaplan-Meier curve for 5-year overall survival in tumor stage II/III patients in (A) cohort 1 (n=51) and (B) cohort 2 (n=122). In both cohorts, low miR-196b-5p expression levels were associated with poor clinical outcome (p <0.05, log-rank test). (C) Kaplan-Meier curve for overall survival in stage IV metastatic disease in cohort 1 (n=59, p<0.05, log-rank test). Supplementary Figure S3 (A-B) Relative quantitative RT-PCR and absolute digital PCR quantification of miR-196b-5p in eight colorectal cancer cell lines. Different colors indicate each cell line, black vertical line separates the groups intoa low and high miR-196b-5p category. A degree of correlation (R=0.881, p=0.004) was found for both methods. (C-D) Relative and absolute quantification of miR-196b-5p in tumor tissue of ten selected colorectal cancer patients. Different colors indicate each patient (numbered with #1-10), black vertical line separates the groups into low and high miR-196b-5p category. Again, a high degree of correlation (R=0.794, p=0.006) was found for both methods. Supplementary Figure S4 (A-B) Comparison of miR-196b-5p mean expression levels in cell lines and patient samples by relative and absolute quantification methods. The cell lines and patient samples were categorized into low and high according to the threshold in previous Supplementary Figure. The cells belonging to the high category showed a mean 4 fold higher expression level of miR-196b-5p than the group of low expressing cell lines, independently which quantification method were used. Interestingly, the tissue levels of miR196b-5p were higher in the group of patients with high levels (ranging from 16 fold to 1000 fold, depending on the method, in comparison to the cells with low expression). In contrast, the group of patients belonging to the low category showed significantly less levels (ranging from 0.9 to 16 fold, depending on the method used) in comparison with cell lines of the with low expression category. (C-D) Quantitative RT-PCR confirmed a statistically significant overexpression (between 80 to 100 fold) and silencing (between 22 to 50%) of miR-196b-5p after 48 hours of transient transfection in all three CRC cell lines. Transfection of inhibitor led to a 55% reduction of miR-196b expression compared to control in HCT116 cells. In RKO and HRT18 transfection of miR-196b-5p inhibitor resulted in 23% reduction of miR-196b-5p expression compared to the control. (E-F) Successful transfection was validated by detecting well-known target genes GATA6 und HOXA9 mRNA expression levels 48 hours after miR-196b-5p mimic exposure in HCT116 and HRT18 cell lines (* = significant, p < 0.05; **, p < 0.01; ***, p < 0.001 Supplementary Figure S5 (A-C) The WST-1 cellular growth assay showed no significant differences in HCT116 and HRT18 cell lines transfected with control, miR-196b-5p mimic or inhibitor. In contrast, RKO cells showed a slight but significant (p=0.0028) growth inhibition after 96 hours in miR196b-5p overexpressing cells. (p-value <0.05 considered as significant). Supplementary Figure S6: (A-C) No significant effects of miR-196b-5p expression levels on drug sensitivity against commonly used colorectal cancer drugs were observed in HCT116 cells after transfection with miR-196b-5p mimic, inhibitor or the respective control as measured by WST-1 assay. Supplementary Figure S7: (A-C) No significant effects of miR-196b-5p expression levels on drug sensitivity against commonly used colorectal cancer drugs were observed in HRT18 cells after transfection with miR-196b-5p mimic, inhibitor or the respective control as measured by WST-1 assay. Supplementary Figure S8: (A-C) No significant effects of miR-196b-5p expression levels on drug sensitivity against commonly used colorectal cancer drugs were observed in RKO cells after transfection with miR-196b-5p mimic, inhibitor or the respective control as measured by WST-1 assay. Supplementary Figure S9 (A-C): Decreased expression of miR-196b-5p led to a higher migration rate in (A) HCT116 cells and (B) HRT18 cells as measured by the xCELLigence Real Time system (p<0.05, n=3). (C).Forced miR196b-5p expression resulted in decreased migration rates in RKO cells Supplementary Figure S10 (A-B) Scratch assays in SW48 (with low endogenous miR196b-5p expression) and DLD1 cells after transient miR-196b-5p overexpression or inhibition. Representative examples of scratch assays in SW48 and DLD1 after transfection of respective control, miR-196b-5p mimic (for SW480) or inhibitor (for DLD1). 0h = scratch at beginning, 16h = scratch after 16 hours, 36h = scratch after 36 hours. Bar charts graphs demonstrating the results of measurement of scratch closure after 36h and 16 hours for the miR-196-5p overexpression in SW48 cells or miR-196-5p inhibition in DLD1 cells. The miR-196b-5p inhibited DLD1 cells closed the scratch significantly earlier than the inhibitor control cells, whereas the miR-196b-5p overexpression in SW480 resulted in decreased scratch closure (* = significant, p < 0.05; **, p < 0.01; ***, p < 0.001; studentÂ's t-test). Supplementary Figure S11 (A) Decreased miR-196b-5p expression resulted in increased invasion in HCT116 cells as measured by transwell invasion assay. (B) Bar charts graph showing the results of the transwell invasion assay in HCT116 cells (p value <0.05 considered as significant). (C) Low expression of miR-196b-5p led to a higher rate of invasion in HCT116 colorectal cancer cells as measured by the xCELLigence Real Time Invasion Assay (n=3). (* = significant, p < 0.05; **, p < 0.01; ***, p < 0.001). Supplementary Figure S12 (A) quantitative RT-PCR confirmed a statistically significant overexpression (OE) of miR-196-5p in the lentiviral transfected CRC cells (p value <0.05 considered as significant). (B) Silencing (Inhibitor) of miR-196b-5p silencing was measured by qRT PCR. Supplementary Figure S13: (A) Bar chart of photon intensities in mice injected with control HCT116 cells and miR-196b-5p silenced cells (p=0.09). (B) Macroscopic assessment of all livers from seven controls and eight miR196b-5p silenced cells. White arrow heads indicate visible metastases (metastases were found in 7 out of 8 mice of the miR-196b-5p silenced cells versus 2 out of 7 control cells, p<0.05). (C) Hematoxylin-eosin staining of representative slides of livers from mice xenografts that had intra-spleen injection of stable HCT116 miR-196b-5p inhibitor cells (index mice I1-I8) or control cells (index mice C1-C7). Stars (*) indicate metastases in some areas of mice (metastases were found in 7 out of 8 mice of the miR-196b-5p silenced cells versus 2 out of 7 control cells, p<0.05). Note that for Inhibitor mouse 8 (I8) the whole liver was filled up with metastases as indicated by the stars, which corresponded to the macroscopic picture (D) Number of detectable metastases on hematoxylin -eosin stained slides for three different liver sections explored. Supplementary Figure S14: Relative mRNA quantification of markers for epithelial-mesenchymal transition was performed after stable miR196b-5p overexpression (A) or silencing (B) of three different CRC cells lines. E-Cadherin and N-Cadherin could not be detected in RKO cells. HCT116 cells did not show an N-Cadherin expression. There was no clear trend or opposite effects with regard to differential expressed EMT markers (* = significant, p < 0.05; **, p < 0.01; ***, p < 0.001). Supplementary Figure S15: (A) Confirmation of Array data in HCT116 cells. Downregulation of genes related to cell migration after miR-196b-5p overexpression (OE) in HCT116 were selected and validated by qRT-PCR. A downregulation of all seven selected genes could be observed in stably transduced HCT116 cells. (B) Validation of the candidate genes in HRT18 cell line. In this cell line we could only identify a downregulation of HOXB7 and GALNT5. CPA4 and DKK1 genes could not be detected in HRT18 cells. (C-E) qRT-PCR and Western blot confirmed a statistically significant upregulation of HOXB7 and GALNT5 mRNA expression after transient miR196b-5p inhibition in HCT116 and HRT18 cells. (F-G) In stable HCT116 miR-196b-5p silenced cells a statistically significant upregulation of HOXB7 and GALNT5 expression was detected by qRT PCR and western blot analysis. (* = significant, p < 0.05; **, p < 0.01; ***, p < 0.001; studentÂ's t-test) Supplementary Figure S16: (A) Human specific HOXB7 mRNA expression in liver tissue of mice xenografts injected intra-splenic with stable HCT116 miR-196b-5p inhibitor cells or control cells measured by qRT PCR. (B) Human specific GALNT5 mRNA expression in liver tissue of mice xenografts injected intra-splenic with stable HCT116 miR-196b-5p inhibitor cells or control cells measured by qRT PCR. (C-D) Luciferase activity after co-transfection of the HOXB7 and GALNT5 wild-type (WT) or mutated (MT) constructs and miR-196b-5p inhibitor/control in HCT116 stable miR196b-5p overexpression cells. Three independent biological experiments were performed and the means and standard deviations are shown (* = significant, p < 0.05; **, p < 0.01; ***, p < 0.001). Supplementary Figure S17 Spearman correlation analysis showed a negative and significant correlation between miR-196b-5p and its target genes HOXB7 and GALNT5. Supplementary Figure S18 Kaplan Meier curve analysis of 439 colorectal cancer patients showed a significant poor prognosis for patients with high HOXB7 expression levels (p<0.05, log-rank test). Data was generated by using the online tool http://www.oncolnc.org/ Supplementary Figure S19 Kaplan Meier curve analysis of 62 colorectal cancer patients showed a significant poor prognosis for patients with high HOXB7 expression levels (p<0.05, log-rank test). Data derived from GSE12945 (Staub et al. J Mol Med (Berl) 2009 Jun;87(6):633-44). Supplementary Figure S20 Kaplan Meier curve analysis of 290 colorectal cancer patients showed a significant poor prognosis for patients with high GALNT5 expression levels (p<0.05, log-rank test). R2: Genomics analysis and visualization platform (https://hgserver1.amc.nl/cgi-bin/r2/main.cgi?&species=hs). Supplementary Figure S21 (A,B) GALNT5 and HOXB7 mRNA expression after 48 hours of siRNA transfection measured by qRT-PCR in HCT116 und HRT18 cells. A statistically significant knockdown of HOXB7 and GALNT5 compared to control was confirmed in both cell lines. (C,D) Expression levels of well-known HOXB7 effector downstream genes CCND1 and CDKN1B were measured after 72h hours of transient miR-196b-5p mimic transfection in (C) HCT116 and (D) HRT18 cells (*= significant, *p < 0.05; **, p < 0.01; ***, p < 0.001).

The development and progression of chronic lymphocytic leukemia (CLL) depend on genetic abnormalities and on the immunosuppressive microenvironment. We have explored the possibility that genetic drivers might be responsible for the immune cell dysregulation that shapes the protumor microenvironment. We performed a transcriptome analysis of coding and non-coding RNAs (ncRNAs) during leukemia progression in the Rag2(-/-)gamma(-/-)(c) MEC1-based xenotransplantation model. The DLEU2/miR-16 locus was found downmodulated in monocytes/macrophages of leukemic mice. To validate the role of this cluster in the tumor immune microenvironment, we generated a mouse model that simultaneously mimics the overexpression of hTCL1 and the germline deletion of the minimal deleted region (MDR) encoding the DLEU2/miR-15a/miR-16-1 cluster. This model provides an innovative and faster CLL system where monocyte differentiation and macrophage polarization are exacerbated, and T-cells are dysfunctional. MDR deletion inversely correlates with the levels of predicted target proteins including BCL2 and PD1/PD-L1 on murine CLL cells and immune cells. The inverse correlation of miR-15a/miR-16-1 with target proteins has been confirmed on patient-derived immune cells. Forced expression of miR-16-1 interferes with monocyte differentiation into tumor-associated macrophages, indicating that selected ncRNAs drive the protumor phenotype of non-malignant immune cells.

Abstract Background Breast cancer is the most prevalent cancer among women worldwide. Most breast cancer-related deaths result from metastasis and drug resistance. Novel therapies are imperative for targeting metastatic and drug-resistant breast cancer cells. Accumulating evidence suggests that dysregulated microRNAs (miRNAs) promote breast cancer progression, metastasis, and drug resistance. Compared with healthy breast tissue, miR-660-5p is notably overexpressed in breast cancer tumor tissues. However, the downstream effectors of miR-660-5p in breast cancer cells have not been fully elucidated. Our aim was to investigate the role of miR-660-5p in breast cancer cell proliferation, migration, invasion, and angiogenesis and to identify its potential targets. Results Our findings revealed significant upregulation of miR-660-5p in MDA-MB-231 and MCF-7 cells compared with MCF-10 A cells. Furthermore, inhibiting miR-660-5p led to notable decreases in the proliferation, migration, and invasion of breast cancer cells, as well as angiogenesis, in HUVEC cells. Through bioinformatics analysis, we identified 15 potential targets of miR-660-5p. We validated TMEM41B as a direct target of miR-660-5p via Western blot and dual-luciferase reporter assays. Conclusions Our study highlights the upregulation and involvement of miR-660-5p in breast cancer cell proliferation, migration, invasion, and angiogenesis. Additionally, we identified TMEM41B as a direct target of miR-660-5p in breast cancer cells.

Background and aim:Tacrolimus (TAC) has significantly improved kidney graft survival following transplantation, though it is associated with adverse side effects. The most prevalent complication resulting from excessive TAC exposure is the onset of de novo diabetes mellitus (DM), a condition that can negatively impact both renal graft function and patient outcomes. De novo DM is linked to an increased risk of chronic transplant dysfunction, as well as cardiovascular morbidity and mortality. Although the underlying mechanisms remain unclear, emerging research in the field of omics shows promise. The aim of this study was to investigate the metabolomic profile of kidney transplant patients who developed de novo DM, in comparison to those who did not, following TAC exposure, using untargeted metabolomic analysis through ultra-high-performance liquid chromatography-mass spectrometry (UHPLC-MS) and machine learning algorithms. Methods:A cohort of 34 kidney transplant patients on a Tacrolimus regimen for at least 6 months was enrolled in the study, with serum samples collected from each patient. Comprehensive profiling of serum metabolites was performed, enabling the classification of patients into de novo diabetes mellitus and non diabetes groups. The metabolomic analysis of serum was conducted using UHPLC-MS. Results:Of the 34 patients, 16 were diagnosed with TAC-induced diabetes. A total of 334 metabolites were identified in the serum samples, of which 10 demonstrated a significant correlation with the de novo diabetes mellitus group. Most of these metabolites were linked to alterations in lipid metabolism. Conclusion:The application of metabolomics in kidney transplant patients undergoing a Tacrolimus regimen is both feasible and effective in identifying metabolites associated with de novo diabetes mellitus. This approach may provide valuable insights into the metabolic alterations underlying TAC-induced diabetes.

Historically, CLL prognostication relied on disease burden, reflected in clinical stage. Later, chromosome abnormalities and genomics suggested several CLL subtypes which were aligned with response to therapy. Gene expression profiling data identified pathways associated with CLL progression. We hypothesized that transcriptome and proteome may identify functional omics associated with CLL nosology. As a test cohort, we utilized publicly available treatment-naive CLL transcriptomics data (n = 130) and did consensus clustering that identified BTK-expression-based clusters. The BTK-High and BTK-Low clusters were validated in public and our in-house databases (n = >550 CLL patients). To associate with functional relevance, we took samples from 151 previously treated patient with CLL and analyzed them using RNA sequencing and reverse-phase protein array. Transcript levels were strongly correlated with BTK protein levels. BTK-High subtype showed increased CCL3/CCL4 levels and disease burden such as high WBC. BTK-Low subtype showed down-regulated mRNA/proteins of DNA-repair pathway and increased DNA-damage-response, which may have contributed to enrichment of inflammatory pathway. BTK-Low subtype was rich in proapoptotic gene and protein expression and relied less on BCR pathway. High-BTK subgroup was enriched in replication/repair pathway and transcription machinery. In conclusion, profiling of 5 datasets of similar to 700 patients revealed unique BTK-associated expression clusters in CLL.