GBM Model Research Articles

CTIM-21. FIRST-IN-HUMAN PHASE I CLINICAL TRIAL USING 8R-70CAR T CELLS FOR NEWLY DIAGNOSED ADULT GLIOBLASTOMA

Abstract BACKGROUND To address the challenges of chimeric antigen receptor (CAR) T cell therapy in GBM, we utilized IL-8 release from tumor cells as a ‘GPS homing signal’ to enhance T-cell trafficking with a novel CD70-specific CAR design (8R-70CAR) that targets GBM. The 8R-70CAR offers several advantages: 1) Improved CAR T-cell trafficking, enhancing antitumor efficacy; 2) Reduced CD70, reshaping the tumor microenvironment (TME) and boosting endogenous immunity; 3) Utilized CD27 as the CAR, a co-stimulatory molecule that enhances T-cell functions; 4) Enabled CAR T cells to act as a sink to neutralize or remove IL-8, a pro-cancer chemokine, from the TME; and 5) Required only 120 ml of whole blood (not leukapheresis) for CAR T cell production. Preclinically, 8R-70CAR significantly enhanced CAR-T cell tumor migration and persistence, leading to complete tumor regression and sustained immune memory in a chemo/radiation/PD-1-blockade-resistant GBM model. These findings led to a phase-I trial (IMPACT trial: FDA-IND#23881, NCT05353530) using 8R-70CAR T cells for CD70 -positive newly diagnosed adult GBM. METHODS We aim to recruit 18 adult patients (3 + 3 dose-escalation design) with CD70-positive (>20%) tumors. The CAR T cells will be given intravenously after completing standard-of-care (SOC) therapy. Primary endpoints: Evaluate safety, tolerability, and dose-limiting toxicities (DLTs) to determine the maximum tolerated dose. Secondary endpoints: Assess preliminary efficacy. Immune monitoring and correlative studies will also be conducted. RESULTS To date, four patients have been enrolled in the IMPACT trial, with 2 eligible patients participating in treatment at dose level 1 (1x10^6/kg). One patient has completed treatment with 8R-70CAR T cells, experienced no DLTs, and is currently in follow-up. The remaining patient is under SOC therapy before receiving CAR T cell treatment. CONCLUSIONS This study aims to address current therapeutic gaps and offers options for implementing a novel CAR design for the development of a more effective therapeutic modality for GBM.

IMMU-30. IGV-001: A BIOLOGIC-DEVICE COMBINATION PRODUCT ELICITS POTENT ANTI-GBM RESPONSES AS A MONOTHERAPY AND IN COMBINATION WITH CHECK POINT INHIBITION

Abstract INTRODUCTION Imvax is developing GoldspireTM, a personalized immunotherapy that combines whole tumor-derived cells with an antisense oligonucleotide against insulin-like growth factor 1 receptor (IGF-1R; IMV-001) in biodiffusion chambers (BDCs; 0.1 µm pore). BDCs are irradiated and implanted at abdominal sites for ~48h to deliver an antigenic payload and immunostimulatory factors that together train the immune system to attack tumor cells. The lead product, IGV-001, was evaluated in newly diagnosed glioblastoma (ndGBM) patients in a phase 1b study. Median overall survival (OS) of highest exposure IGV-001-treated Stupp-eligible patients (n=10) was 38.2 mos compared with 16.2 mos in standard-of-care-treated patients (p=0.044; ClinicalTrials.gov NCT02507583). A Phase 2b study in ndGBM is currently underway (ClinicalTrials.gov NCT04485949), where a primary endpoint of progression-free survival and key secondary endpoints of overall survival and safety will be evaluated. METHODS The orthotopically implanted GBM (luciferase-expressing GL261-luc2) model was utilized to evaluate the efficacy of mIGV-001 (the murine version of IGV-001) alone or in combination with anti-PD-1 immunotherapy. The following four groups were compared: mIGV-001 or BDC placebo + isotype or anti-PD-1 antibody (4 doses total every 3-4 days). GL261-luc2 cells were treated with IGF-1R antisense IMV-001. The treated cell suspension was loaded into clinical-grade BDCs (0.1-µm pore size, polyvinylidene difluoride membrane) at a density of 1 × 106 cells per BDC and then irradiated with 5–6 Gy. mIGV-001 or BDCs filled with saline were administered 28 days prior to tumor challenges and tumor progression and survival were monitored for an additional 100 days. RESULTS Preventative treatment with mIGV-001 limited tumor progression and extended OS in mice orthotopically (i.e., intracranially) challenged with GL261-luc2 cells. The therapeutic benefit of mIGV-001 was enhanced with anti-PD-1 treatment and the combination therapy was superior to either monotherapy.

NIMG-32. ORAL ADMINISTRATION OF DEUTERATED CHOLINE AS A NEW APPROACH TO PROVIDE HIGH TUMOR-TO-BRAIN IMAGE CONTRAST IN DEUTERIUM METABOLIC IMAGING (DMI)

Abstract BACKGROUND Metabolic imaging using FDG-PET is a key component of disease management in many cancers outside of the brain, but in brain tumors it often shows low tumor-to-brain contrast because of high background signal from normal brain. We explore generating cancer-specific metabolic image contrast using Deuterium Metabolic Imaging (DMI), which is a novel 2H magnetic resonance spectroscopic imaging-based technique. DMI was applied after oral administration of deuterated choline, an essential nutrient needed for membrane synthesis whose uptake is often increased in tumors. METHODS Total choline (tCho) DMI was performed in an orthotopic model of glioblastoma, F344 rats (n=11) implanted with RG2 cells, on an 11.7 tesla MRI scanner, with isotropic spatial resolution of 2.5 mm. tCho-DMI data were collected for 36 min during intravenous infusion (IV, n=6) or after 3 days of oral administration (PO, n=5) of [2H9]-choline chloride dissolved in sterile water. At a dose of 285 mg/kg body weight, bolus-continuous IV infusion via the tail vein required 1.57 ml for a 250-gram rat. PO administration via gavage on 3 consecutive days used the max daily dose recommended for human adults, 50 mg/kg. Results - In the PO group, average tumor tCho concentration was 0.78±0.14 mM and 0.20±0.05 mM in contralateral normal brain. In the IV group, the tCho concentration was 0.88±0.18 mM in the tumor, and 0.18±0.03 mM in normal brain (PO vs. IV: p=0.52). Tumor-to-brain contrast was 5.9±1.2 in PO-administered animals, and 7.1±3.3 for animals receiving IV 2H-choline, (p=0.54). CONCLUSIONS - In a preclinical GBM model oral administration of [2H9]-choline resulted in comparably high tumor-to-brain image contrast on tCho-DMI-based maps as during IV infusion. Work is ongoing to identify the different choline metabolites that contribute to the tCho signal. Because oral administration of choline is very safe, this approach could facilitate the translation of tCho-DMI to human subjects.

DDDR-39. PI3 KINASE AND CHECKPOINT INHIBITION: A POTENTIALLY NOVEL THERAPEUTIC STRATEGY FOR GLIOBLASTOMA (GBM)

Abstract Glioblastoma is the most common and aggressive malignant brain tumor with a median survival of less than 2 years, highlighting the need for more effective therapeutics. Dual pathway synergism between PI3K inhibitors and check-point inhibitors is a promising therapeutic avenue that has not been explored. Based on this possibility, we hypothesized that the combination of PI3K and checkpoint inhibitors will improve survival in our native glioblastoma mouse model. Treatment with GCT-007, a small molecule PI3K inhibitor, induces growth arrest of GBM cells that is accompanied by increases in the cell surface expression of molecules targeted by the immune system and immunotherapy drugs, including MHC Class II and PD-L1. Treatment of mouse GL261 and human U251 GBM cell lines with GCT-007 caused growth arrest, increased the cell surface expression of PD-L1 and MHC class II, and synergized with radiation to cause growth arrest and cell death. We compared GCT-007 to Paxalisib in our in vitro studies and observed that GCT-007 outperformed Paxalisib at increasing MHC class II and PD-L1 expression on the tumor cells at equivalent doses. We then performed in vivo animal studies using our native mouse GBM model. We treated mice orally with GCT-007 and observed a significantly prolonged survival compared to placebo. Mice treated with anti-PD1 resulted in a sexually dimorphic response consisting of prolonged remission of all female mice, now surviving beyond 1 year, but no improvement in the survival of male mice compared to placebo. Adding GCT-007 to anti-PD1 did not alter the prolonged remission outcome in female mice, it did result in prolonged remission of 25% of the male mice, now beyond 1 year. In conclusion, these results indicate that GCT-007 treatment alone improves survival in our native GBM mouse model and it may also improve the response to immune checkpoint inhibitor therapy in at least a fraction of GBM tumors. These data warrant further investigation as a potentially novel therapeutic approach to GBM.

Machine learning prediction for prognosis and long-term effectiveness for transcatheter ventricular septal defect closure: a 5-year single center experience

Abstract Background Transcatheter therapy, with minimal invasiveness and rapid recovery advantages, has emerged as the primary choice for treating perimembranous ventricular septal defect. However, patients receiving transcatheter occlusion may confront conduction abnormalities due to the occluder's proximity to the cardiac conduction bundle during implantation. Few have assessed the risks of postoperative conduction block (CB) via robust machine learning models. Purpose The purpose of this study was to evaluate our long-term follow up results of pmVSD occlusion and establish a prediction model to identify risk predictors of conduction abnormalities via robust machine learning methods. Methods Five methods including logistic regression with a stepwise selection of variables, lasso regularization; random forest; gradient descent boosting (GBM); and support vector machine, were used to train models for predicting risks of late-onset and persistent conduction block through 5 years of follow-up and were validated using 5-fold cross-validation. Receiver-operating characteristic curves and Brier scores were utilized to evaluate model discrimination and calibration, respectively. The top prediction variables were identified by using the best performing models, using the relative influence score of each variable in 5-fold cross-validation. Results The GBM model performed the best with an AUC of 0.82 (95% confidence interval [CI]: 0.75 to 0.89) for predicting late-onset CB (Brier score: 0.0204), and 0.70 (95% CI: 0.62 to 0.78) for persistent CB (Brier score: 0.0003) over entire study period. Conduction block on admission, advanced age and aortic regurgitation before discharge were strongly associated with late-onset CB, whereas decreased left ventricular ejection fraction, enlarged left ventricle end diastolic diameter at 1-month follow-up were the most significant predictors of persistent CB. Conclusions These models predict the risks of temporary and persistent postoperative conduction block in patients with pmVSD occlusion and shed light on preventing long-term complications.Study FlowchartCentral Illustration

CHARACTERISATION OF THE RESPONSE OF PRE-CLINICAL MODELS OF GLIOMA TO METABOLIC THERAPIES USING IN VIVO MAGNETIC RESONANCE SPECTROSCOPY

Abstract AIMS Magnetic resonance imaging (MRI) is routinely used to diagnose and monitor brain tumours. Magnetic resonance spectroscopy (MRS) allows us to record the metabolic profile of tumours in situ. This provides a non-invasive method to monitor the response of tumours to metabolic therapies. We aimed to characterise the response of pre-clinical models of glioma to metabolic therapies such as arginine deprivation and the ketogenic diet (KD) by MRS, to better understand what metabolic changes are occurring in the tumour. METHOD We performed orthotopic injection of three different cell lines, including the syngeneic models GL261 and CT2A, as well as a primary GBM cell line, into the striatum of C57BL/6 and athymic nude mice respectively. Mice were randomised to either arginine deprivation treatment using pegylated arginine deiminase (ADI), or KD, plus controls. In each animal, we collected single voxel MRS data using STEAM on both the tumour itself, and healthy brain. MRS data was exported in jMRUI format and analysed in R using the spant package, followed by PCA and OPLS-DA. RESULTS PCA showed that there is clear distinction between healthy and tumour spectra across all models. OPLS-DA indicated that the lipid and choline spectral regions were significantly associated with tumour, whereas the NAA and glutamine/glutamate regions were significantly associated with healthy brain. Integration of specific peaks associated with these metabolites confirmed these results. Subset analysis indicated distinct metabolic signatures between the syngeneic models and the primary GBM model. Furthermore, there is evidence to suggest that tumours in animals treated with arginine deprivation have a unique metabolic signature. CONCLUSION This data indicates that the tumour specific response to metabolic therapies can be measured by routine MRS screening. This would allow a non-invasive method by which clinicians could monitor the effectiveness of metabolic therapies in patients with gliomas.

Development and In vitro and In vivo efficacy investigation of multifunctional, targeted, theranostic liposomes for imaging and photodynamic therapy of glioblastoma

The challenge of effectively diagnosing and treating glioblastoma, which has a high incidence and low survival rate, remains unresolved. Consequently, research efforts have increased to develop specific and effective diagnostic and therapeutic agents. In this study, N-acetyl glucosamine-modified, IR780-encapsulated, PEG-coated, nanosized, 68Ga-labeled, neutral, and positively charged theranostic liposomes were developed for GLUT1 targeting for PET/CT and NIR imaging and PDT/PTT of GBM. Characterization, in vitro release profile, stability, radiolabeling efficiency, and labeling stability were evaluated. In vitro cell uptake studies were conducted in RG2 and U87 GBM cell lines, NIR images were obtained in the RG2 and U87 cell lines, and fluorescence microscope images were obtained in the RG2 cell line. PDT/PTT activities were assessed after exposure to NIR light (808 nm, 0.8–1 W/cm2). Cytotoxicity was studied in RG2, U87, and L929 cell lines. In vivo studies were performed in GBM model nude mice. The liposomes had an optimum mean particle size (181–193 nm), high encapsulation efficiency (79–84 %), and zeta potential (−5 to −7 mV). The maximum 68Ga labeling efficiency was achieved with an incubation at 80 °C, pH: 3.5, and 5 min. Liposomes remained stable for 30 days at 4 °C. While the intracellular uptake of targeted formulations was higher than non-targeted ones, it was similar for positively charged and neutral formulations. In U87 and RG2 cell lines, cell viability remained above 50 % for liposomal formulations at a concentration of 1.5 μg/mL exposed to NIR light (808 nm, 0.8–1 W/cm2). The formulations showed significant PDT/PTT activity under NIR light and lower cytotoxicity in the L929 than in RG2 and U87 cell lines. In vivo bioluminescence images and biodistribution studies in U87 tumor-bearing nude mice revealed higher tumor uptake of targeted, IR780-encapsulated liposomal formulations in tumor tissue compared to non-targeted ones. Liposomal formulations showed higher tumor uptake than the free IR780 solution. The data suggest that targeted liposomes have potential as theranostic agents for PDT/PTT and multifunctional glioblastoma imaging.

Developing Model-Agnostic Meta-Learning Enabled Lightbgm Model Asthma Level Prediction in Smart Healthcare Modeling

Millions of people across the world suffer from the chronic respiratory condition known as asthma. Predicting the severity of asthma based on a variety of personal and environmental characteristics might yield useful information for preventative measures. LightGBM model is a gradient-boosted model with the potential for great accuracy, but it requires careful hyperparameter adjustment to reach its full potential. Common tuning techniques have a hard time generalizing to new data distributions. The dataset was used, and its many subsets were used to represent various demographics and geographic areas. LightGBM was configured with hyperparameters, trained on a sample dataset, and then verified for each job. To quickly adjust to new tasks, the MAML method sought to find the optimal values for its hyperparameters. After the meta-training step was complete, the generalizability of the hyperparameters was tested on new data. After including MAML for hyperparameter adjustment, the Light- GBM model showed a gain of 7% in accuracy, coming in at 98.5%. Predictions of severe asthma had a crucially high 97.8 percent degree of accuracy. The model’s recall rate for severe asthma levels was 97.4%, demonstrating its capacity to reliably detect and anticipate important instances. An F1-score of 97.1%, a metric that averages the accuracy and recall of a model, is indicative of good overall performance.. For gradient-boosted model applications like asthma level prediction, MAML provides a viable path for hyperparameter adjustment. Although there are obstacles to be overcome, this method has the potential to greatly improve the flexibility and precision of predictive healthcare models. More effective implementations and a wider range of applications can be explored in future studies.

Construction of a clinical prediction model for complicated appendicitis based on machine learning techniques

Acute appendicitis is a typical surgical emergency worldwide and one of the common causes of surgical acute abdomen in the elderly. Accurately diagnosing and differentiating acute appendicitis can assist clinicians in formulating a scientific and reasonable treatment plan and providing high-quality medical services for the elderly. In this study, we validated and analyzed the different performances of various machine learning models based on the analysis of clinical data, so as to construct a simple, fast, and accurate estimation method for the diagnosis of early acute appendicitis. The dataset of this paper was obtained from the medical data of elderly patients with acute appendicitis attending the First Affiliated Hospital of Anhui University of Chinese Medicine from January 2012 to January 2022, including 196 males (60.87%) and 126 females (39.13%), including 103 (31.99%) patients with complicated appendicitis and 219 (68.01%) patients with uncomplicated appendicitis. By comparing and analyzing the prediction results of the models implemented by nine different machine learning techniques (LR, CART, RF, SVM, Bayes, KNN, NN, FDA, and GBM), we found that the GBM algorithm gave the optimal results and that sensitivity, specificity, PPV, NPV, precision, recall, F1 and brier are 0.9167, 0.9739, 0.9429, 0.9613, 0.9429, 0.9167, 0.9296, and 0.05649, respectively. The GBM model prediction results are interpreted using the SHAP technology framework. Calibration and Decision curve analysis also show that the machine learning model proposed in this paper has some clinical and economic benefits. Finally, we developed the Shiny application for complicated appendicitis diagnosis to assist clinicians in quickly and effectively recognizing patients with complicated appendicitis (CA) and uncomplicated appendicitis (UA), and to formulate a more reasonable and scientific clinical plan for acute appendicitis patient population promptly.

A multimodal machine learning model for predicting dementia conversion in Alzheimer’s disease

Alzheimer’s disease (AD) accounts for 60–70% of the population with dementia. Mild cognitive impairment (MCI) is a diagnostic entity defined as an intermediate stage between subjective cognitive decline and dementia, and about 10–15% of people annually convert to AD. We aimed to investigate the most robust model and modality combination by combining multi-modality image features based on demographic characteristics in six machine learning models. A total of 196 subjects were enrolled from four hospitals and the Alzheimer’s Disease Neuroimaging Initiative dataset. During the four-year follow-up period, 47 (24%) patients progressed from MCI to AD. Volumes of the regions of interest, white matter hyperintensity, and regional Standardized Uptake Value Ratio (SUVR) were analyzed using T1, T2-weighted-Fluid-Attenuated Inversion Recovery (T2-FLAIR) MRIs, and amyloid PET (αPET), along with automatically provided hippocampal occupancy scores (HOC) and Fazekas scales. As a result of testing the robustness of the model, the GBM model was the most stable, and in modality combination, model performance was further improved in the absence of T2-FLAIR image features. Our study predicts the probability of AD conversion in MCI patients, which is expected to be useful information for clinician’s early diagnosis and treatment plan design.

#2742 Artificial intelligence accurately approximates MGPE dose response relationship in a North American Dialysis Population

Abstract Background and Aims The Anemia Control Model (ACM) is a certified medical device aimed at optimizing the management of CKD-related anemia among patients on dialysis. ACM is based on an artificial neural network (ANN) trained on over 900’000 patients treated in the European Nephrocare Network. Over the years, ACM has offered physicians hundreds of thousands of erythropoietin stimulating agent (ESA) and intravenous iron dosing suggestions and ensured high hemoglobin target achievement rate while minimizing drug use. However, evidence concerning its validity in non-European samples is still limited. Apollo Dial DB is the largest, fully anonymized, collaborative, international database of dialysis patients. The database includes detailed clinical information of more than 500,000 patients over millions of patient-months of observation with an extensive representation over geographies as well as ethnic groups. Given its detailed medical information and size, the database may offer unprecedented opportunities to develop and test equitable and fair decision support systems based on artificial intelligence. In this study we sought to re-train and validate and assess the accuracy of the ACM in an ethnically diverse sample of North American dialysis patients in Apollo Dial DB. Method We constructed the cohort for training and validation of this new ACM version by including a random sample of patients treated in the North American Fresenius Medical Care Dialysis Network. The sample accounted for 20% of the entire North American Section of the Apollo Dial DB dataset. Each patient history has been segmented in a collection of patient-months, which was the unit of analysis. The initial cohort was split in 3 datasets for training (70%), validation (20%) and test (10%). The input variables included laboratory test results, body weight, treatment data, age, sex, and previous ESA and iron prescription. The outcome of interest was the variation in serum hemoglobin concentration over 1 month of follow up. We compared the accuracy of the original ACM based on ANN (baseline model) with a newly trained algorithm based on high-performance light GBM. Results We included 44,220 patients followed by 2,277,276 patient-months overall. The mean age=60.9 ± 13.4, 59.2% were male, 56.4% were Caucasians, 32.2% African American, 3.2% Asians and 8.2% were others. At baseline, mean Hb= 10.9 ± 1.2. The original EMEA version of the Anemia Control Model (ACM) was still accurate in North American patients (mean absolute error, MAE = 0.65 g/dL). The fine-tuned light GBM model achieved a 16% reduction in error rate (MAE = 0.55 g/dL; Fig. 1). No difference in MAE was observed across sex and ethnic groups. Conclusion This study demonstrates the excellent predictive performance of the original ANN-based ACM in the North American dialysis population. Fine tuning by re-training ACM on the dataset further improved the accuracy of the model with substantial equivalence in performance across different demographics, thus ensuring algorithmic fairness. ACM is a promising tool for anemia management in ethnically diverse dialysis population and different healthcare settings.

Machine learning-based radiomics for predicting outcomes in cervical cancer patients undergoing concurrent chemoradiotherapy

PurposesTo investigate the value of machine learning-based radiomics for predicting disease-free survival (DFS) and overall survival (OS) undergoing concurrent chemoradiotherapy (CCRT) for patients with locally advanced cervical cancer (LACC). Materials and methodsIn this multicentre study, 700 patients with IB2-IVA cervical cancer who underwent CCRT with ongoing follow-up were retrospectively analyzed. Three-dimensional radiomics features of primary lesions and its surrounding 5 mm region in T2WI sequences were collected. Six machine learning methods were used to construct the optimal radiomics model for accurate prediction of DFS and OS after CCRT in LACC patients. Eventually, TCGA and GEO databases were used to explore the mechanisms of radiomics in predicting the progression and survival of cervical cancer. This study adhered CLEAR for reporting and its quality was assessed using RQS and METRICS. ResultsIn the prediction of DFS, the RSF model combined tumor and peritumor radiomics demonstrated the best predictive efficacy, with the AUC for predicting 1-year, 3-year, and 5-year DFS in the training, validation, and test sets of 0.986, 0.989, 0.990, and 0.884, 0.838, 0.823, and 0.829, 0.809, 0.841, respectively. In the prediction of OS, the GBM model best performer, with AUC of 0.999, 0.995, 0.978, and 0.981, 0.975, 0.837, and 0.904, 0.860, 0.905. Differential genes in TCGA and GEO suggest that the prediction of radiomics model may be associated with KDELR2 and HK2. ConclusionMachine learning-based radiomics models help to predict DFS and OS after CCRT in LACC patients, and the combination of tumor and peritumor information has higher predictive efficacy, which can provide a reliable basis for therapeutic decision-making in cervical cancer patients.

Application value of the automated machine learning model based on modified CT index combined with serological indices in the early prediction of lung cancer.

Accurately predicting the extent of lung tumor infiltration is crucial for improving patient survival and cure rates. This study aims to evaluate the application value of an improved CT index combined with serum biomarkers, obtained through an artificial intelligence recognition system analyzing CT features of pulmonary nodules, in early prediction of lung cancer infiltration using machine learning models. A retrospective analysis was conducted on clinical data of 803 patients hospitalized for lung cancer treatment from January 2020 to December 2023 at two hospitals: Hospital 1 (Affiliated Changshu Hospital of Soochow University) and Hospital 2 (Nantong Eighth People's Hospital). Data from Hospital 1 were used for internal training, while data from Hospital 2 were used for external validation. Five algorithms, including traditional logistic regression (LR) and machine learning techniques (generalized linear models [GLM], random forest [RF], gradient boosting machine [GBM], deep neural network [DL], and naive Bayes [NB]), were employed to construct models predicting early lung cancer infiltration and were analyzed. The models were comprehensively evaluated through receiver operating characteristic curve (AUC) analysis based on LR, calibration curves, decision curve analysis (DCA), as well as global and individual interpretative analyses using variable feature importance and SHapley additive explanations (SHAP) plots. A total of 560 patients were used for model development in the training dataset, while a dataset comprising 243 patients was used for external validation. The GBM model exhibited the best performance among the five algorithms, with AUCs of 0.931 and 0.99 in the validation and test sets, respectively, and accuracies of 0.857 and 0.955 in the validation and test groups, respectively, outperforming other models. Additionally, the study found that nodule diameter and average CT value were the most significant features for predicting lung cancer infiltration using machine learning models. The GBM model established in this study can effectively predict the risk of infiltration in early-stage lung cancer patients, thereby improving the accuracy of lung cancer screening and facilitating timely intervention for infiltrative lung cancer patients by clinicians, leading to early diagnosis and treatment of lung cancer, and ultimately reducing lung cancer-related mortality.

Oncolytic herpes simplex virus expressing IL-2 controls glioblastoma growth and improves survival

BackgroundGlioblastoma (GBM), a highly immunosuppressive and often fatal primary brain tumor, lacks effective treatment options. GBMs contain a subpopulation of GBM stem-like cells (GSCs) that play a central role in...

Abstract 5703: Development of RVL-101, a CDK4/6-selective, small molecule inhibitor with superior CNS penetration and combinability potential for the effective treatment of primary and metastatic brain tumors

Abstract Cyclin-dependent kinases 4 and 6 (CDK4/6) are key regulators of the cell cycle that are frequently dysregulated in cancer, leading to unchecked cell growth. CDK4/6 inhibitors have emerged as a promising class of anticancer agents that selectively target this pathway. The primary mode of action of CDK4/6 inhibitors relies on the disruption of retinoblastoma (Rb) protein phosphorylation, which thereby blocks cell cycle progression from G1 to S phase and reduces proliferation of cancer cells by a cytostatic mechanism. However, despite their efficacy, approved and clinical-stage CDK4/6 inhibitors have modest penetration of unbound drug into the brain, limiting their utility in the treatment of brain metastases and primary brain tumors. Additionally, despite the potential benefits of CDK4/6 inhibitors as combination therapies with cytotoxic agents, adjunctive treatment options remain underexplored due to the poor safety profiles of existing therapies. RVL-101 is a preclinical, investigational drug with a novel pharmacophore that has a low nanomolar activity against CDK4, is CDK2,6,7,9-sparing, and maintains >500-fold selectivity over CDK1. To maximize the potential of RVL-101 as a preferred combination partner, this novel agent has been engineered to have exquisite kinome- and safety-panel selectivity with no measured CYP inhibition or induction liabilities. RVL-101 is differentiated from approved and clinical stage CDK4/6 inhibitors having both high partitioning into the brain (Kp) and high free-brain (Kpu,u = 0.7) levels resulting in equal levels of unbound drug in all relevant compartments at steady state. Like other CDK4/6 agents, RVL-101 is broadly efficacious in vivo in ER+/HER2- xenograft models at well-tolerated doses. However, RVL-101 has also demonstrated robust activity in HR-/HER2+ xenograft models as a single-agent and has uniquely shown improved tumor inhibition effect in combination with approved HER2 cytotoxic therapeutics tucatinib and fam-trastuzumab-deruxtecan. RVL-101 was also tested in a GBM model as a monotherapy and in conjunction with the cytotoxic agent temozolomide (TMZ). Remarkably, QD dosing of the cytostatic agent, RVL-101, after pulsatile pretreatment with TMZ led to dramatic synergistic efficacy compared to either agent alone, leading to 100% survival of all animals on the dual therapy arm for more than 3 weeks post discontinuation of dosing. RVL-101 offers the combination of substantial therapeutic benefit and consistent free-drug levels in plasma and CNS that could lead to safe and effective treatment for a broader patient population. Citation Format: John E. Campbell, Bo-Sheng Pan, Michael Bower, Peter Bertinato, Nisha Perez, Victoria Gras Andreu, Neha Rana, Sophia Tabchouri, Ankit Gupta, Jonah Kallenbach. Development of RVL-101, a CDK4/6-selective, small molecule inhibitor with superior CNS penetration and combinability potential for the effective treatment of primary and metastatic brain tumors [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 5703.

Abstract 2910: Targeting thioredoxin reductase 1 (TrxR1) suppresses TAZ-driven glioblastoma progression

Abstract Glioblastoma (GBM) is the most common and aggressive primary brain malignancy in adults. Poor outcomes for traditional treatments demand targeted therapies based on identified mechanisms that drive tumor development and sustain its malignancy. Molecular pathology studies have classified GBM into subtypes differing in treatment responses and survival rates. Among these subtypes, the mesenchymal (MES) group is associated with the worst prognosis. The Hippo pathway transcriptional co-activator with PDZ-binding motif (TAZ) is one of the three transcriptional regulators that drive the GBM MES gene expression program. Studies from our lab and others have previously shown that TAZ activation is able to promote GBM MES differentiation and progression. Therefore, finding vulnerabilities of TAZ-driven tumors may help to develop targeted therapeutic methods for MES GBM. In an orthotopic GBM mouse model, we found that TAZ-driven tumors show increased expression of thioredoxin reductase 1 (TrxR1), which is responsible for catalyzing the reduction of thioredoxin and maintaining the cellular redox balance. In searching for the cause of increased expression of TrxR1, we found that glucose deprivation can induce TrxR1 expression, suggesting TrxR1 may be an adaptive mechanism for tumor cells to survive in glucose deprivation conditions. Indeed, either depleting TrxR1 genetically or inhibiting TrxR1 pharmacologically can promote tumor cell death under glucose deprivation. Furthermore, knockout TrxR1 in a TAZ-driven mouse GBM model significantly decelerated tumor progression. Collectively, our studies revealed that inhibition of the Trx redox system under glucose deprivation conditions leads to a synergistic cell death induction in GBM cells, therefore making targeting TrxR1 a potential therapeutic strategy for TAZ-driven MES GBM. Citation Format: Miaolu Tang, Kaitlyn Dirks, Soo Yeon Kim, Jessica Thorpe, Wei Li. Targeting thioredoxin reductase 1 (TrxR1) suppresses TAZ-driven glioblastoma progression [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 2910.

Abstract 1441: Establishing a novel clinically relevant disease model of glioblastoma

Abstract Introduction: GBM is the most common primary malignancy of the CNS. Historically pre-clinical models have failed to predict response in humans, despite promising preclinical data. Moreover, current models seldom incorporate surgical resection and/or standard of care chemotherapy treatment. These models also commonly employ young animals exclusively, whose immune contexture differs from older patients. We therefore sought to establish an improved, orthotopic, preclinical GBM model, which better recapitulates patient response to standard-of-care and targeted treatments. Methods: NFpp10a-Luc2 GBM cells were orthotopically implanted into C57BL/6-mice (aged[>18months] and young[6-8weeks]) and weekly bioluminescence imaging performed to monitor tumor growth. Overall survival (OS) and tumor growth were assessed in response to (1) Surgical Resection, (2) Temozolomide (TMZ) (3) anti-PD1 (4) neoadjuvant anti-PD1 and (5) Regorafenib (REGO) therapy. Tissue collected post-mortem underwent bulk RNA sequencing followed by analysis via microenvironment cell population counter (MCP) and gene set enrichment (GSEA) to determine changes in the GBM-TME following treatment. Results: We demonstrated OS advantage in aged mice undergoing surgical resection (Resection:33.5 days vs Non-Resection: 18 days; p=0.0166) and observed age to be a significant prognostic factor (Young:62 days vs Aged: 22 days; p=0.0002). Subsequently, we observed that TMZ and anti-PD1 monotherapies had no impact on NFpp10-Luc2 growth (p=0.9001, p=0.7933) or survival (p=0.3035, p=0.6328). Neoadjuvant anti-PD1 treated mice demonstrated no significant survival advantage compared to IgG control (33 days vs 35 days; p=0.9429). Lastly, REGO treatment demonstrated a trend towards improved OS vs Vehicle (p=0.096). MCP analysis revealed both neoadjuvant anti-PD1 and REGO treatment induced influx of CD8+ T cells, B cells and monocytes into the TME, with neoadjuvant anti-PD1 associated with an upregulation of CXCR3 (p=0.0045). Conclusions: We have, for the first time, established and characterized response of the NFpp10a-C57BL/6 model to surgical resection, TMZ, anti-PD1 and REGO therapy in both young and aged mice. We have shown the model is markedly insensitive to intervention with chemotherapy and immune checkpoint therapy, mirroring what is seen clinically in patients. The model may therefore be employed in future pre-clinical studies to guide clinical trials in the setting of mesenchymal GBM. Citation Format: Kate Connor, Kieron White, James Clerkin, Kieron Sweeney, Liam Shiels, Thomas van Brussel, Ingrid Arijs, Diether Lambrechts, Gautam Shankar, Frederik de Smet, Stephen G. Maher, Laure Marignol, Patrick Dicker, Jochen Prehn, David O'Brien, Annette T. Byrne. Establishing a novel clinically relevant disease model of glioblastoma [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 1441.

Abstract 1417: IRF8 driven reprogramming of the immune microenvironment enhances anti-tumor adaptive immunity and reduces immunosuppression in murine glioblastoma

Abstract Glioblastoma (GBM) has an extremely immunosuppressive microenvironment, primarily mediated by the recruitment of immature myeloid cells (IMCs) to the tumor. IMCs suppress T cell function and dampen the anti-tumor response. In this study, we sought to “reprogram” IMCs into antigen-presenting cells (APCs) to enhance anti-tumor T cell responses. To induce IMC differentiation into APCs, we utilized a retroviral replicating vector (RRV) expressing IRF8, a master regulator of type 1 conventional dendritic cell (cDC1) development. We evaluated the impact of IRF8 RRV on in vivo tumor growth kinetics, survival, immunosuppression, and immune microenvironment. Because RRV transduction requires the proliferation of target cells, we assessed the proliferative capacity of myeloid cells in murine and human GBM samples. We utilized the SB28 cell line, a poorly immunogenic murine GBM model. We pre-transduced SB28 cells with IRF8 RRV and implanted them intracranially into syngeneic C57BL/6 mice; controls expressed an empty vector. Intratumoral myeloid cells were proliferative in both murine and human samples. In vivo, tumor and myeloid cells are infected proportionately to their respective levels of proliferation. Mice with IRF8 RRV pre-transduced intracranial tumors had significantly longer survival (median 28 days) compared to controls (median 19 days) (p <0.0001). Additionally, tumor growth was delayed in IRF8 RRV mice. IRF8 RRV mice had significant enrichment of dendritic cells (p = 0.0145), cDC1s (p < 0.0001), T cells (p = 0.0025) and CD8+ T cells (p < 0.0001). Both monocytic and granulocytic IMCs had decreased expression of the immunosuppressive markers Arginase 1 and IDO1. Further, ex vivo co-culture of intra-tumoral myeloid cells with naïve T cells showed a functional decrease in immunosuppression in IRF8 RRV tumors compared to controls (p < 0.0001). To study the mechanism of action, we assessed whether IRF8 transduction of tumor cells alone or both tumor and immune cells, was necessary for the observed survival benefit. To inhibit RRV spread from infected cells to other cells, we treated mice with azidothymidine, a reverse transcription inhibitor. A robust survival benefit and tumor growth inhibition were seen only in mice where the RRV could spread freely (p = 0.0005). This finding shows that both tumor and immune cells must be infected to cause the phenotype and suggests that reprogramming immune cells is required for therapeutic efficacy. Our novel approach used RRV to deliver a myeloid transcriptional regulator in intracerebral GBM. This therapy achieved significant survival benefits and increased intratumoral cytotoxic T cells and APCs, while decreasing myeloid immunosuppression. Further studies will expand upon the mechanism of action and test combination therapies. Citation Format: Megan Montoya, Sara A. Collins, Pavlina Chuntova, Noriyuki Kasahara, Hideho Okada. IRF8 driven reprogramming of the immune microenvironment enhances anti-tumor adaptive immunity and reduces immunosuppression in murine glioblastoma [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 1417.

Hearing Recovery Prediction for Patients with Chronic Otitis Media Who Underwent Canal-Wall-Down Mastoidectomy.

Background: Chronic otitis media affects approximately 2% of the global population, causing significant hearing loss and diminishing the quality of life. However, there is a lack of studies focusing on outcome prediction for otitis media patients undergoing canal-wall-down mastoidectomy. Methods: This study proposes a recovery prediction model for chronic otitis media patients undergoing canal-wall-down mastoidectomy, utilizing data from 298 patients treated at Korea University Ansan Hospital between March 2007 and August 2020. Various machine learning techniques, including logistic regression, decision tree, random forest, support vector machine (SVM), extreme gradient boosting (XGBoost), and light gradient boosting machine (light GBM), were employed. Results: The light GBM model achieved a predictive value (PPV) of 0.6945, the decision tree algorithm showed a sensitivity of 0.7574 and an F1 score of 0.6751, and the light GBM algorithm demonstrated the highest AUC-ROC values of 0.7749 for each model. XGBoost had the most efficient PR-AUC curve, with a value of 0.7196. Conclusions: This study presents the first predictive model for chronic otitis media patients undergoing canal-wall-down mastoidectomy. The findings underscore the potential of machine learning techniques in predicting hearing recovery outcomes in this population, offering valuable insights for personalized treatment strategies and improving patient care.

Comparative Analysis of Machine Learning Models for Prediction of Acute Liver Injury in Sepsis Patients.

Acute liver injury (ALI) is a common complication of sepsis and is associated with adverse clinical outcomes. We aimed to develop a model to predict the risk of ALI in patients with sepsis after hospitalization. Medical records of 3196 septic patients treated at the Lishui Central Hospital in Zhejiang Province from January 2015 to May 2023 were selected. Cohort 1 was divided into ALI and non-ALI groups for model training and internal validation. The initial laboratory test results of the study subjects were used as features for machine learning (ML), and models built using nine different ML algorithms were compared to select the best algorithm and model. The predictive performance of model stacking methods was then explored. The best model was externally validated in Cohort 2. In Cohort 1, LightGBM demonstrated good stability and predictive performance with an area under the curve (AUC) of 0.841. The top five most important variables in the model were diabetes, congestive heart failure, prothrombin time, heart rate, and platelet count. The LightGBM model showed stable and good ALI risk prediction ability in the external validation of Cohort 2 with an AUC of 0.815. Furthermore, an online prediction website was developed to assist healthcare professionals in applying this model more effectively. The Light GBM model can predict the risk of ALI in patients with sepsis after hospitalization.