Survival Of Patients Research Articles

IMMU-12. TRANSLATING COMBINED PD1 AND LAG3 INHIBITION FROM PRECLINICAL MODELS TO PATIENTS WITH REFRACTORY, DNA REPLICATION REPAIR DEFICIENT (RRD) GLIOBLASTOMA: AN IRRDC STUDY

Abstract BACKGROUND AND AIMS RRD-glioblastoma harbour high tumor mutation burden (TMB) and respond to anti-PD1 immune-checkpoint inhibition (ICI). However, not all respond, and the majority progress, highlighting the need for combinatorial therapies for sustained immune-surveillance. METHODS We performed transcriptomic analyses of human RRD-glioblastoma specimens for immune checkpoint expression, and accordingly, tested combined ICI in immunocompetent murine models. Based on these preclinical data, we treated refractory patients using a combination of anti-PD1+anti-LAG3 through single-patient trial/compassionate access. Complimentary immuno-genomic biomarker analyses including circulating tumor DNA (ctDNA) were performed to investigate mechanisms and track responses. RESULTS Human RRD-glioblastoma (n=80) demonstrated high LAG3 expression, providing a strong rationale for therapeutic targeting. We tested combined anti-PD1+anti-LAG3 inhibition in three immunocompetent RRD-glioblastoma murine models. In the anti-PD1-responsive (Nestin-CreMSH2LoxP/LoxP-POLES459F/+) model, anti-PD1+anti-LAG3 resulted in universal tumor response and superior survival to ICI-monotherapy. In the anti-PD1 resistant models (Mlh1-/-/NestinCre+/Trp53LoxP/LoxP and therapy-induced hypermutant ENU/Trp53-/- gliomas), anti-PD1+antiLAG3 improved survival, overcoming the lack of response to ICI-monotherapy. Biologically, high LAG3 expression and immune-exhaustion observed in CD8 T-cells after treatment with anti-PD1 was ablated following the addition of anti-LAG3. Serially transplanted, post-anti-PD1 treated tumors showed response, confirming, in-vivo, that resistance to anti-PD1 could be abrogated by anti-PD1+anti-LAG3. Seven children with refractory RRD-glioblastoma who had progressed after anti-PD1 treatment were treated using anti-PD1+anti-LAG3, resulting in objective radiological responses and prolonged ongoing survival. Tolerability was better than a previous study of combined CTLA4 and PD1 inhibition for similar patients. Paired immuno-genomic tumor analyses, serial blood flow-cytometry, T-cell receptor clonotype, and CSF ctDNA analyses provided novel insights into the mechanisms of immunological invigoration and first-in-human, radiological responses. CONCLUSIONS LAG3 is an effective target in refractory RRD-glioblastoma. Combined inhibition with anti-PD1 inhibition demonstrated radiological response, prolonged survival and manageable toxicities in patients, and unearthered mechanisms of immune-responses. The combination will now be tested in biomarker-driven clinical trials in RRD-glioblastoma and other immune-inflamed solid tumors.

Intestinal microbiome changes and mechanisms of maintenance hemodialysis patients with constipation

BackgroundConstipation is a common symptom in maintenance hemodialysis patients and greatly affects the quality of survival of hemodialysis patients. Fecal microbiota transplantation and probiotics are feasible treatments for functional constipation, but there is still a gap in the research on the characteristics of gut flora in patients with maintenance hemodialysis combined with constipation. The aim of this study is to clarify the characteristics of the intestinal flora and its changes in maintenance hemodialysis patients with constipation.MethodsFecal samples were collected from 45 participants, containing 15 in the maintenance hemodialysis constipation group,15 in the maintenance hemodialysis non-constipation group and 15 in the healthy control group. These samples were analyzed using 16S rRNA gene sequencing. The feature of the intestinal microbiome of maintenance hemodialysis constipation group and the microbiome differences among the three groups were elucidated by species annotation analysis, α-diversity analysis, β-diversity analysis, species difference analysis, and predictive functional analysis.ResultsThe alpha diversity analysis indicated that maintenance hemodialysis constipation group was less diverse and homogeneous than maintenance hemodialysis non-constipation group and healthy control group. At the genus level, the top ten dominant genera in maintenance hemodialysis constipation group patients were Enterococcus, Escherichia-Shigella, Bacteroides, Streptococcus, Bifidobacterium, Ruminococcus_gnavus_group, Lachnospiraceae_unclassified, Faecalibacterium, Akkermansia and UCG-002. Compared with non-constipation group, the Enterococcus, Rhizobiales_unclassified, Filomicrobium, Eggerthella, Allobaculum, Prevotella_7, Gordonibacter, Mitochondria_unclassified, Lachnoanaerobaculum were significantly higher in constipation group (p<0.05). Compared with non-constipation group, the Kineothrix, Rhodopirellula, Weissella were significantly lower in constipation group (p<0.05). The predictive functional analysis revealed that compared with non-constipation group, constipation group was significantly enriched in pathways associated with pyruate metabolism, flavonoid biosynthesis.ConclusionsThis study describes for the first time the intestinal microbiome characteristics of maintenance hemodialysis patients with constipation. The results of this study suggest that there is a difference in the intestinal flora between maintenance hemodialysis patients with constipation and maintenance hemodialysis patients without constipation.

BIOM-46. REAL-WORLD EXPERIENCE WITH CIRCULATING TUMOR DNA IN CEREBROSPINAL FLUID IN PATIENTS WITH CENTRAL NERVOUS SYSTEM TUMORS

Abstract Next generation sequencing (NGS) of circulating tumor DNA (ctDNA) from cerebrospinal fluid (CSF) offers a minimally invasive approach to identify somatic alterations for individuals with central nervous system (CNS) cancers. Here, we review our ‘real world’ experience using a clinically validated, FDA-authorized platform, MSK-IMPACTTM, to comprehensively profile ctDNA from CSF samples with and without clinically documented CNS involvement by cancer (1007 samples, 711 patients). The patient cohort included patients with over 90 distinct tumor types of primary CNS and metastatic origin, with lung cancer (n = 188), breast cancer (n = 150), and glioma (n = 148) representing the most common broad categories. We identified genetic alterations (ctDNA-positive) in 489/922 (53.0%) CSF samples from patients with clinically documented CNS cancer. Over half of these ctDNA-positive samples (248, 50.7%) possessed a biomarker that was predictive of response to an FDA-approved drug (OncoKB level 1). Furthermore, the distribution of clinically actionable somatic alterations was consistent with tumor-type specific alterations across the AACR GENIE cohort. Of patients without clinical documentation of CNS cancer, none of these 85 CSF samples were ctDNA-positive (specificity = 100%). Successive CSF sequencing over time identified clonal evolution and the emergence of resistance mechanisms to therapies. ctDNA positivity corresponded with shortened overall survival after CSF collection (HR: 3.23, 95% CI: 2.58-4.05, P < 0.001). NGS of CSF cfDNA can detect clinically actionable somatic alterations in a large subset of CNS cancer patients and allows tracking of tumor evolution. Detection of somatic alterations in CSF may be a useful prognostic biomarker for survival in cancer patients.

Predicting first-line VEGFR-TKI resistance and survival in metastatic clear cell renal cell carcinoma using a clinical-radiomic nomogram.

This study aims to construct predicting models using radiomic and clinical features in predicting first-line vascular endothelial growth factor receptor-tyrosine kinase inhibitor (VEGFR-TKI) early resistance in metastatic clear cell renal cell carcinoma (mccRCC) patients. We also aim to explore the correlation of predicting models with short and long-term survival of mccRCC patients. In this retrospective study, 110 mccRCC patients from 2009 to 2019 were included and assigned into training and test sets. Radiomic features were extracted from tumor 3D-ROI of baseline enhanced CT images. Radiomic features were selected by Lasso method to construct a radiomic score. A combined nomogram was established using the combination of radiomic score and clinical factors. The discriminative abilities of the radiomic, clinical and combined nomogram were quantified using ROC curve. Cox regression analysis was used to test the correlation of nomogram score with progression-free survival (PFS) and overall survival (OS). PFS and OS were compared between different risk groups by log-rank test. The radiomic, clinical and combined nomogram demonstrated AUCs of 0.81, 0.75, and 0.83 in training set; 0.79, 0.77, and 0.88 in test set. Nomogram score ≥ 1.18 was an independent prognostic factor of PFS (HR 0.22 (0.10, 0.47), p < 0.001) and OS (HR 0.38 (0.20, 0.71), p = 0.002), in training set. PFS in low-risk group were significantly longer than high-risk group in training (p < 0.001) and test (p < 0.001) set, respectively. OS in low-risk group were significantly longer than high-risk group in training (p = 0.003) and test (p = 0.009) set, respectively. A nomogram combining baseline radiomic signature and clinical factors helped detecting first-line VEGFR-TKI early resistance and predicting short and long-term prognosis in mccRCC patients.

NIMG-17. PREDICTIVE VALUE OF THE BRAIN TUMOR REPORTING AND DATA SYSTEM SCORE (BT-RADS) ON FUNCTIONAL OUTCOMES AND PROGRESSION-FREE SURVIVAL IN GLIOBLASTOMA PATIENTS

Abstract The Brain Tumor Reporting and Data System Score (BT-RADS) is a novel eight-point classification system that standardizes magnetic resonance imaging (MRI) reporting in glioblastoma (GB) patients. While previous studies have demonstrated BT-RADS predictive value in the prognosis and overall survival of high-grade glioma patients, its association with functional outcomes and progression-free survival (PFS) in GB patients is unclear. We aim to evaluate the BT-RADS predictive value on functional outcomes assessed through the Karnofsky Performance Scale (KPS) at one year post-diagnosis and on PFS. BT-RADS from MRI ~3 months post-diagnosis after chemo-radiation therapy (BT-RADS_1), and ~6 months post-diagnosis after the 6-month maintenance temozolomide (TMZ) cycle (BT-RADS_2) were analyzed. BT-RADS was classified as low (BT-RADS score 0-3a) and high (BT-RADS score 3b-4). In a univariate logistic regression analysis, BT-RADS_1 was a significant predictor (p=0.04) in discriminating an unfavorable KPS from a favorable one. In a multivariate logistic regression, in which the best model fit was the model containing BT-RADS_1 and MGMT tumor status (model p value=0.03), the ROC analyses indicate that BT-RADS_1 is not better than chance in differentiating an unfavorable KPS from a favorable one (model AUC=0.65; 95% CI=0.55, 0.76). The discriminatory ability of BT-RADS scores alone in predicting KPS outcomes was modest, as indicated by the AUC. This underscores the need for larger sample sizes. In a final model of BTRADS_1 and BT-RADS_2, accounting for MGMT tumor status, the risk of tumor progression was 50% lower (p value=0.003) with a low BT-RADS (0-3a) in the BT-RADS _1 group and 38% lower (p value=0.03) with a low BT-RADS (0-3a) in the BT-RADS_2 group. Considering the above-discussed findings, BT-RADS may represent a new tool in the clinical management of glioblastoma patients by informing on disease progression, prognosis, and quality of life.

MET amplification correlates with poor prognosis and immunotherapy response as a subtype of melanoma: a multicenter retrospective study.

Mesenchymal epithelial transition factor (MET) variant is an independent prognostic factor for worse prognosis in patients with lung cancer or gastroesophageal adenocarcinoma. MET gene variants can be regarded as a subtype of melanoma but there is a lack of studies regarding the frequency of MET genetic alterations and the efficacy of immunotherapy in melanoma patients. The purpose of this study is to explore potential therapeutic strategies for melanoma subtypes with MET alterations. A total of 1751 malignant melanomas were analyzed to illustrate the landscape of MET mutations. We collected 55 melanoma cases from multicenter for a retrospective cohort from 2010 to 2023. We analyzed the impact of MET amplification on the efficacy of immunotherapy in the retrospective cohort after propensity score matching (PSM) and a pancancer cohort. CIBERSORT was used to evaluate the immune infiltration. There were no instances of MET 14 exon skipping, and only instances of MET amplification were found in the 1751 melanomas and our retrospective cohort. Cox proportional hazards model analysis showed that MET amplification (P = 0.006) was significantly associated with poorer overall survival (OS) in patients who received immunotherapy as the first-line treatment. Compared with patients with MET amplification, patients in the negative control (NC) group had a significantly better OS (P = 0.022) after PSM. Analysis of 1661 pancancer cases with the MSK-IMPACT assay showed that patients receiving immunotherapy in the MET amplification group had a trend toward worse OS than those without MET amplification (P = 0.025). This database analysis showed that the main type of MET mutation is amplification in malignant melanoma. MET-amplified solid tumors might be considered for targeted therapy, as MET amplification can be regarded as a risk factor affecting the prognosis of patients with tumors treated with immunotherapy.

RADT-47. CLINICAL OUTCOMES AFTER SURGICAL RESECTION OF BRAIN METASTASES IN SMALL CELL LUNG CANCER PATIENTS

Abstract Small cell lung cancer (SCLC) has a high propensity to metastasize to the brain. Radiotherapy is the primary treatment for SCLC brain metastases. Whole brain radiotherapy (WBRT) has been the standard of care, but recent studies show similar overall survival (OS) in patients treated with upfront stereotactic radiosurgery (SRS). Surgery does not typically have a role in SCLC although it can provide relief of neurologic symptoms and pathologic diagnosis. We sought to determine the clinical outcomes of patients who underwent surgical resection of SCLC brain metastases. A retrospective chart review of SCLC patients treated at UPMC from 2010 – 2022 was performed. OS after resection was estimated using the Kaplan-Meier method with log-rank test. 4,213 patients were diagnosed with SCLC, of which 715 patients (17%) had synchronous brain metastases at initial diagnosis. Forty patients underwent resection of SCLC brain metastases, all of whom were symptomatic. The median diameter of the resected brain metastases was 3.8 cm (range 1.7 – 8 cm). Seventy-five percent (n=30) had resection of synchronous brain metastases and 25% (n=10) had resection of metachronous brain metastases. Median OS for the surgical cohort was 13.6 months (interquartile range (IQR), 3.4 – 21.9 months). Patients with metachronous and synchronous brain metastases had median OS of 6.8 months (IQR, 2.8 – 24.6 months) and 14.3 months (IQR, 3.6 – 20.4), respectively. Of patients with synchronous brain metastases (n=30), 53% (n=16) received post-operative WBRT, 40% (n=11) received post-operative SRS and 10% (n=3) died within 7 weeks of surgery. Median OS of patients with synchronous brain metastases treated with WBRT and SRS were 11.9 months (IQR, 3.8 – 21.7) and 19.0 months (IQR, 14.6 – 30.3), respectively, and the difference was not statistically significant (p=0.53). Surgery for SCLC brain metastases is uncommon. Patients who received SRS versus WBRT post-operatively for synchronous brain metastases had no difference in OS.

IMMU-35. TARGETING THE AHR IMMUNOMETABOLIC HUB DURING VIROIMMUNOTHERAPY OF GLIOMAS

Abstract Glioblastoma (GBM) poses a significant therapeutic challenge in oncology. Recent clinical trials have shown the feasibility of virotherapy to treat patients with GBM. Our laboratory developed an oncolytic adenovirus termed Delta-24-RGD that has been tested in clinical trials for recurrent GBM patients with encouraging results (NCT00805376, NCT02798406). These clinical studies, together with preclinical data, showed that the efficacy of Delta-24-RGD was due to direct tumor cell oncolysis and indirect activation of anti-tumor immune response. To further leverage this immune component, our group generated Delta-24-RGDOX, an armed version of Delta-24-RGD expressing the T-cell activator OX40-L, which exhibits a superior anti-cancer effect. In this project, we aim to further boost the effectiveness of Delta-24-RGDOX by targeting factors that maintain the immunosuppressive characteristic of gliomas. Specifically, we identified the environmental sensor and transcription factor Aryl Hydrocarbon Receptor (AhR) as one of the immunosuppressive factors. We show that AhR is overexpressed in GBM surgical specimens compared to healthy brain tissue, and high AhR expression correlates with poorer survival in glioma patients. Immunofluorescence staining of glioma cells treated with Delta-24-RGDOX demonstrated that AhR translocates to the nucleus, and qPCR showed that translocation resulted in the enhanced transcription of AhR-responsive elements. Additionally, bulk RNA-seq reveals that in vivo glioma models treated with Delta-24-RGDOX exhibit activation of the AhR pathway network. This unexpected and paradoxical activation may be responsible for immune suppression mechanisms, which might hinder the effect of Delta-24-RGDOX. Based on these data, we combined two innovative strategies, the oncolytic virus Delta-24-RGDOX and AhR inhibitors. Survival analysis in immunocompetent models of gliomas treated with this combined therapy showed the greatest therapeutic benefit compared to monotherapy controls. These findings highlight the critical role of AhR activation triggered by Delta-24-RGDOX and position AhR as a potential therapeutic target for enhancing the efficacy of oncolytic adenovirus therapy in GBM.

EPID-23. HISTORY OF TYPE III IMMUNE HYPERSENSITIVITY IMPROVES SURVIVAL IN GLIOBLASTOMA

Abstract Preexisting atopy, particularly allergic rhinitis, is consistently associated with decreased risk of glioma, and there is evidence that preexisting autoimmune conditions may also decrease risk. Glioblastoma is both locally and systemically immunosuppressive and, unlike many other forms of cancer, its initiation is not strongly linked with chronic inflammation. Therefore, history of immune hypersensitivity could be associated with improved anti-glioma immunosurveillance. We leveraged Surveillance, Epidemiology and End Results (SEER) data combined with Medicare claims to investigate the effect of preexisting immune hypersensitivities on overall survival of glioblastoma patients receiving resection, radiation, and temozolomide (standard of care, or SOC) from 2008-2019. Patients &amp;lt;66 years of age, enrolled in Medicare Advantage, not continuously enrolled in Medicare Part A/B, missing follow-up, reported by autopsy/death certificate only, or who did not receive surgery were excluded. Prior history of immune hypersensitivity (≥6 months prior to primary cancer diagnosis) was identified by ICD-9/ICD-10 codes. Treatment pattern was determined using ICD-9/ICD-10 and HCPCS/CPT codes. Conditions were classified into broad immune hypersensitivity types: type I (IgE-mediated, including atopy), type II (IgG- or IgM-mediated), type III (immune complex-mediated, including lupus and rheumatoid arthritis), and type IV (cell-mediated). Survival comparisons were made using Cox proportional hazard models adjusted for demographics, extent of resection, and Charlson comorbidity score (excluding immune conditions) to estimate hazard ratios, 95% confidence intervals, and p-values. Overall, 44% of 3,426 glioblastoma patients who received SOC had an immune hypersensitivity (type I: 33%, type II: 4%, type III: 8%, and type IV: 15%). These conditions were significantly more common in females (47% vs 41%, p&amp;lt;0.001). After adjustment, only type III immune sensitivity was associated with improved OS (HR=0.81, 95% CI:0.71-0.92, p=0.002). These results suggest that underlying immune-complex mediated hypersensitivity may improve treatment response and prognosis in glioblastoma, and that these pathways could potentially be exploited for therapy.

DISP-14. THE IMPACT OF DISTANCE IN TREATING GLIOBLASTOMA: A RETROSPECTIVE SINGLE CENTER STUDY

Abstract INTRODUCTION Glioblastoma (GBM) is an aggressive primary brain tumor associated with poor survival. Research shows that patients living in rural areas have poorer outcomes than those in metropolitan areas; however, little is known about how distance between patients and their subspecialist impacts outcomes. We explored the effect of distance from a National Cancer Institute-designated Comprehensive Cancer Center (CCC) has on overall survival (OS) and clinical trial enrollment for patients with GBM. METHODS Retrospective data was collected from patients with newly diagnosed GBM per the World Health Organization 2021 who received treatment at Huntsman Cancer Institute (HCI) from 2018 to 2022. Distance was evaluated as a categorical (near: 0-25 miles, intermediate: 25-40 miles, far: &amp;gt;40 miles) and continuous variable for analyses. In addition to univariable models, multivariable models adjusting for known prognostic factors were also performed. RESULTS A total of 167 patients were identified. Baseline characteristics including age, sex, race, and KPS scores were similar across groups. The univariable COX regression model found no significant difference in OS when comparing the near group to the intermediate (p=0.125) and far (p=0.599). However, the multivariable model found poorer OS in the intermediate group that was statistically significant when compared to the near group (HR (95% CI) = 1.55 (1.01, 2.39), p = 0.047). When comparing distances as a continuous variable, patients in the far group had statistically significant lower clinical trial enrollment rates (near group 43%, intermediate 35%, far 18%; p=0.0001). DISCUSSION Patients with GBM seeing a neuro-oncologist at our CCC did have a difference in OS as well as clinical trial enrollment when considering distance traveled. Clinical trials are standard of care for patients with GBM; our data provides evidence that distance from a CCC significantly affects trial enrollment, underscoring the importance of ensuring availability of resources for clinical trial participants.

CNSC-16. ASSOCIATION BETWEEN RESTING-STATE-NETWORK FUNCTIONAL CONNECTIVITY AND SURVIVAL IN RECURRENT GLIOMAS DIAGNOSED USING FET PET

Abstract BACKGROUND Functional connectivity between glioma-infiltrated brain tissue and remote cortical regions has been linked to overall survival (OS) in patients with newly diagnosed glioblastoma. We studied the association of functional connectivity between tumor and resting-state networks and OS in patients with recurrent gliomas. PATIENTS AND METHODS The study included 82 patients (26-81 years; median ECOG performance score, 0) with recurrent gliomas (grade 4 glioblastoma, n=57; grade 3 and 4 astrocytoma, n=12; grade 2 and 3 oligodendroglioma, n=13) who had undergone first-line therapy. Diagnosis of recurrence was confirmed by amino acid PET using the tracer O-(2-[18F]-fluoroethyl)-L-tyrosine (FET). FET PET and resting-state functional MRI were simultaneously acquired on a 3T hybrid PET/MR scanner. We performed a seed-based dual regression analysis to determine the functional connectivity between metabolically active tumors and seven canonical resting state networks. RESULTS Average functional connectivity was highest in oligodendrogliomas (z-score, 6.9±1.9) and was inversely related to the tumor grade (p=0.031). Univariate Cox regression analysis revealed a significant (p&amp;lt;0.05) positive association between OS and tumor connectivity with the visual, somatomotor, and dorsal attention networks. Subsequent stepwise multivariate Cox regression analysis refined this association to the dorsal attention network alone (HR, 0.88; 95% CI, 0.80-0.97; p=0.007). Patients with the highest functional connectivity (z-score&amp;gt;8.3, n=21) had a median OS of 31.4 months, while those with the lowest functional connectivity (z-score&amp;lt;4.3, n=21) had 10.6 months. Functional connectivity between tumors and the dorsal attention network showed an independent association with OS (HR, 0.90; 95% CI, 0.81-0.99; p=0.033) in a multivariate model, including various prognostic factors. CONCLUSION Recurrent gliomas exhibit significant functional connectivity to most of the major known resting-state networks, highest in oligodendrogliomas. The data suggest that high functional connectivity between the tumor and the dorsal attention network is an independent prognostic factor for improved OS.

NIMG-83. THE PROGNOSTIC VALUE OF PARAMETRIZED KI-67 MAPS DERIVED FROM CONVENTIONAL MRI AND MACHINE LEARNING IN A LARGE RETROSPECTIVE COHORT OF ADULT-TYPE DIFFUSE GLIOMAS

Abstract BACKGROUND Ki-67 is a well-established proliferation marker relied upon for glioma grade assessment. Advances in machine learning (ML) now allow researchers to non-invasively predict the voxelwise Ki-67 index with reasonable accuracy based on anatomical MRI.The purpose of this study is to evaluate the prognostic ability of these predictions in adult-type diffuse gliomas. METHODS We included adult patients with untreated gliomas of various histopathologic grades who underwent their first resection at a comprehensive cancer center from 1993 to 2018. Anatomic MR images were pre-processed, segmented, normalized, then run through a pre-trained ML algorithm to generate Ki-67 parametric maps. The maximum estimated Ki-67 (maxKi-67) for each patient was documented, and the optimal threshold stratifying the cohort into low- and high-risk patients was computed using 10-fold cross-validation. Kaplan-Meier survival curves were then generated for both risk groups and compared using the log-rank and Breslow tests. MaxKi-67 was correlated with OS using Cox proportional hazard regression analysis adjusted for age and preoperative KPS. RESULTS 1,179 patients (age 52±15; 701 males) were included in this study. 727 tumors were histologic grade IV, while 246 and 206 were histologic grade III and II. The mean maxKi-67 was 15.9± 5.2. A maxKi-67 threshold of 13 was computed from the 10-fold cross-validation to dichotomize the cohort by OS. The low-risk group (maxKi-67&amp;lt;13) had a median OS = 2,435 days, significantly different from the high-risk group (maxKi-67&amp;gt;13) which had median OS = 551 days (p&amp;lt;.001). A maxKi-67 index above this threshold significantly increased the hazard of death: unadjusted HR = 2.53 [95% CI 2.17-2.96] and adjusted HR = 1.55 [1.31-1.83]. Age &amp;gt; 55 and preoperative KPS &amp;lt; 90 were also negative prognosticators in the Cox model: adjusted HR for age = 2.79 [2.36-3.31] and for KPS = 2.08 [1.77-2.46]. CONCLUSION Image-based derivation of maxKi-67 index using ML accurately predicts survival in glioma patients with a cut-off of 13 segregating these patients into 2 risk groups.

BIOM-27. VOXEL-WISE SURVIVAL ANALYSIS IN GLIOBLASTOMA: TUMOR PROBABILITY MAPS REVEAL CRITICAL REGIONS NEAR WHITE MATTER TRACTS

Abstract BACKGROUND Glioblastoma is the most aggressive primary brain tumor, characterized by rapid growth and poor prognosis. Understanding spatial determinants of survival in GBM is crucial for improving prognostic accuracy and treatment strategies. Our study employs voxel-wise survival analysis (VSA) of tumor probability maps (TPMs) derived from MRI to identify high-risk brain regions. TPMs are generated through radio-pathomic mapping and can predict non-enhancing tumor regions not captured by traditional imaging. We aim to determine if high tumor probability regions are near previously identified critical white matter (WM) tracts and assess their impact on survival outcomes. METHODS We analyzed data from 374 GBM patients from the UCSF-PDGM-v2 dataset. Using a pretrained radio-pathomic model, TPMs were predicted based on T1, T1C, ADC, and FLAIR MRI sequences. All TPMs were rigidly registered to MNI space using ANTs. A Cox proportional hazards model was applied voxel-wise to assess the association between tumor probability and overall survival (OS). A p-value map was generated, thresholded, and cluster corrected (p &amp;lt; 0.05) for further analysis. RESULTS The VSA revealed that high tumor probability regions were significantly associated with poorer OS in GBM. Gray matter (GM) regions near WM tracts with high tumor probability showed markedly reduced survival. GM regions adjacent to the left IFOF (p &amp;lt; 0.05, HR = 1.19, 845 vs. 961 days for high/low TPM), Corpus Callosum Forceps Minor (p = 0.0001, HR = 1, 740 vs. 902 days), Body of the Corpus Callosum (p = 0.005, HR = 0.97, 748 vs. 798 days for high/low TPM) and other WM tracts, such as the left thalamic radiation superior and left reticulospinal tracts had similar trends. CONCLUSIONS Our study demonstrates that regions with high tumor probability, particularly those in the GM near critical WM tracts, are strongly associated with worse overall survival in GBM patients.

IMMU-15. CD74/MIF SIGNALING AXIS DISRUPTION IN BONE MARROW-DERIVED MYELOID CELLS DELAYS MALIGNANT PROGRESSION IN A PRECLINICAL MODEL OF GLIOMA

Abstract INTRODUCTION CD74, a chaperone associated with the major histocompatibility complex class II (MHC-II), plays a crucial role in assembling and transporting MHC-II molecules on the cellular surface. Beyond its physiological function in antigen presentation, CD74 is a negative regulator of inflammation, and its dysregulation is associated with inflammatory disorders and cancer. Notably, tumor stem cells release macrophage migration inhibitory factor (MIF) to trigger CD74 immunosuppressive signaling in tumor-infiltrating immune cells, thereby abrogating CD8+ T cell activation and promoting angiogenesis. While CD74 expression is known in dendritic cells, macrophages and B cells, our recent findings reveal its upregulation in immunosuppressive macrophages within the low-grade glioma (LGG) tumor microenvironment, and its negative correlation with the survival of adolescent and young adult patients with LGG. HYPOTHESIS. We hypothesize that inhibiting CD74/MIF in immunosuppressive myeloid cells may affect tumor progression and prevent malignant transformation (MT) to high-grade glioma (HGG). METHODS To test our hypothesis, we isolated bone marrow cells from immunocompetent RCAS n/tv-a LGG bearing animals, differentiated them into bone marrow-derived myeloid cells (BMDMs), silenced CD74 expression, and co-cultured them with primary CD8+T cells. Moreover, we inhibited CD74/MIF signaling in vivo. RESULTS. Deconvolution analysis of tumor-infiltrating immune cells revealed upregulation of CD74 and its network in myeloid cells during LGG compared to HGG. Silencing CD74 in LGG BMDMs in vitro resulted in the downregulation of genes associated with migration, proliferation, and immunosuppression. The co-culture of LGG CD74KD-BMDMs with CD8+ T cells increased the expression of T cell activation genes. In vivo treatment of LGG mice with MIF inhibitors significantly decreased myeloid cell infiltration in the tumor microenvironment (TME) and prolonged survival. CONCLUSIONS These preclinical studies highlight a role for the CD74/MIF signaling axis in glioma and present a novel immunotherapeutic target to modulate the activity of immunosuppressive myeloid cells in order to delay MT.

SURG-29. EXTENDED RESECTION IMPROVES PATIENT SURVIVAL OUTCOMES AFTER GROSS-TOTAL RESECTION OF GLIOBLASTOMA

Abstract In recent years, the effectiveness of extended resection (ER) beyond the contrast-enhanced (CE) lesion has been reported as a surgical strategy for glioblastoma. This study was to determine whether ER of FLAIR-hyperintense area beyond the CE tumor affected the overall survival (OS) and progression-free survival (PFS) of patients with glioblastoma after gross-total resection (GTR). From October 2019 to March 2022, a total of 62 patients who underwent resection of glioblastoma (GBM) at Kobe University Hospital were retrospectively examined. Volumetric measurements of the CE area and surrounding FLAIR-hyperintense area were performed, and clinical variables related with OS and PFS were analyzed. In total, 33 patients with GBM who underwent GTR of the CE tumor met the inclusion criteria. The mean volume of CE tumor and FLAIR hyperintensity area before resection was 36.5 cm3 and 77.2 cm3 respectively. The mean excision rate of FLAIR-hyperintense area was 18.2%. Removal rates of 20% or more in the FLAIR-hyperintense region significantly increased OS and PFS (p=0.040 and 0.026, respectively). In multivariate analysis, age ≥ 65 years (HR 2.73; 95% CI 1.09–7.22; p=0.032) were associated with shorter OS, but ER ≥ 20% (HR 0.41; 95% CI 0.17–0.99; p=0.049) and MGMT promotor methylation (HR 0.36; 95% CI 0.15–0.85; p=0.020) were associated with longer OS. ER of FLAIR-hyperintense lesions of 20% or more in glioblastoma contributed to prolonging patient survival. It is necessary to validate the effect of ER in prospective clinical trials.

Survival, treatment duration and costs of patients with prostate cancer treated with triptorelin in Italy: a study of administrative databases.

Several data support the efficacy/effectiveness, safety and favorable impact on quality of life of triptorelin treatment in patients with prostate cancer. However, little evidence is available concerning triptorelin use in the long term. We analyzed data on triptorelin treatment in patients with prostate cancer in an integrated Italian administrative database, covering around 6 million health-assisted subjects throughout the country. Patients with at least one prescription for triptorelin in the period 2010-2020 and with no evidence of metastasis were included and followed up until 2021. Overall survival (OS) and duration of treatment were analyzed using Kaplan-Meier curves, starting from the date of first prescription. The cohort included a total of 3,411 patients (mean age: 76.8 ± 8.7 years), of whom 1,326 (38.9%) were treated with triptorelin only and 2,085 (61.1%) with triptorelin combined with an anti-androgen. Overall, 847 (24.8%) patients with prostate cancer died and 1,037 (30.4%) had a treatment switch during the follow-up period, and both the median OS and median duration of treatment were not reached in both groups. The mean annual total cost per patient was estimated as 5,574 €, with almost half of the costs related to medication expenses (2,737 €). We found a long survival and duration of triptorelin treatment in this population of Italian patients with prostate cancer. This study with a long follow-up period further highlights the usefulness of healthcare utilization databases to integrate results obtained from clinical studies with those from everyday clinical practice.

EXTH-14. IRREVERSIBLE TARGETING OF THE C-MYC/TOPOISOMERASE I AXIS IN GLIOBLASTOMA: DUAL INHIBITION USING A NOVEL INDENOISOQUINOLINE INHIBITOR

Abstract Glioblastoma–the most common primary CNS malignancy–is notorious for its cellular and molecular complexity and uniform fatality. Overexpression of the c-MYC and TOPOISOMERASE 1 (TOP1) oncogenes is implicated in more than 50% of human cancers. They are involved in the pathogenesis of glioblastomas with overexpression correlating with decreased overall median survival in glioblastoma patients. LMP744 is a non-camptothecin indenoisoquinoline small molecule inhibitor designed to target the C-MYC/TOP1 axis. The compound is inherently stable, has high brain penetrance, and irreversibly binds to its target providing several advantages over the camptothecin class of drugs. Human glioblastoma cells treated with LMP744 show a dose-dependent decrease in viability with LD50 ranging between 50 and 100 nM compared to 100-200µM following treatment with temozolomide in the same cell lines. Treatment results in a pro-apoptotic effect in exposed cells as well as S-phase arrest. Protein arrays and confirmatory western blots demonstrate increased CHK2 and decreased p53 and GSK-a/ß in treated cells. Orthotopic xenotransplantation via stereotactic injection of human glioblastoma cells (7×105 cells/4 µL) in nude mice showed exquisite sensitivity to LMP744 without adverse events. Administration of 20 mg/kg of LMP744 via tail vein injection resulted in total eradication of tumor signal on bioluminescent imaging. These preliminary findings suggest that LMP744–and perhaps the class of non-camptothecin indenoisoquinolones–is a promising novel candidate for glioblastoma pharmacotherapy.

SURG-08. A PHASE III RANDOMIZED TRIAL OF GROSS TOTAL RESECTION VERSUS POSSIBLE RESECTION OF FLAIR HYPERINTENSITY LESION ON MR IMAGE FOR NEWLY DIAGNOSED SUPRATENTORIAL GLIOBLASTOMA (JCOG2209)

Abstract BACKGROUND Complete resection of contrast-enhanced lesion (gross total resection: GTR) without severe neurological deficits has been generally accepted as the goal of surgery. However, it remains unclear if additional resection of surrounding FLAIR hyper-intense lesion combined with GTR (FLAIRectomy) has survival advantage of primary glioblastoma patients. Purpose: Multicenter, open-label, randomized phase III trial was commenced to confirm the superiority of FLAIRectomy to GTR alone followed by radiotherapy with concomitant and adjuvant temozolomide in terms of overall survival (OS) for primary glioblastoma IDH-wildtype patients. This trial investigates not only survival but also postoperative neurological and neurocognitive deficits in detail. METHODS We assumed a 2-year OS of 50% in the GTR arm and expected a 15% improvement in the FLAIRectomy arm. A total of 130 patients is required with a one-sided alpha of 5%, power of 70%, and will be accrued from 49 Japanese institutions in 4 years and follow-up will last 2.5 years. Patients 18-75 years of age will be registered and randomly assigned to each arm with 1:1 allocation. The primary endpoint is OS, and the secondary endpoints are progression-free survival, frequency of adverse events, proportion of Karnofsky performance status preservation, proportion of National Institutes of Health stroke scale preservation, proportion of mini-mental state examination preservation and proportion of health-related quality of life preservation. The Japan Clinical Oncology Group Protocol Review Committee approved this study protocol in May 2023. Ethics approval was granted by the National Cancer Center Hospital Certified Review Board. Patient enrollment began in July 2023. DISCUSSION If FLAIRectomy is superior to GTR alone, aggressive surgery will become a standard surgical treatment for glioblastoma with resectable contrast-enhanced lesion. Trial registration: Registry number: jRCT1031230245. Date of registration: 19/July/2023. Date of first participant enrollment: 28/July/2023. Funding for this study is provided by National Cancer Center Research and Development Fund (2023-J-03).

Identification of alternative lengthening of telomeres-related genes prognosis model in hepatocellular carcinoma.

Hepatocellular carcinoma (HCC) is a common malignant tumor worldwide, characterized by high mortality. This study aimed to explore the prognostic value and function of alternative lengthening of telomeres (ALT)-related genes in HCC. Differentially expressed genes (DEGs) were identified based on The Cancer Genome Atlas (TCGA) and then intersected with ALT-related genes to obtain ALTDEGs. Risk score model was constructed using the least absolute shrinkage and selection operator (LASSO) algorithm and Cox regression and validated with Gene Expression Omnibus (GEO) datasets. The predictive efficacy of the risk score and ALTs-score was evaluated by Kaplan-Meier curves, time-ROC curves, and the nomogram analyses. The impacts of SMG5 silencing on the HCC cell behaviors were assessed by CCK-8, wound healing, and Transwell assays. A total of 500 ALTDEGs were screened and 13 genes (CDCA8, SMG5, RAD54B, FOXD2, NOL10, RRP12, CCT5, CCT4, HDAC1, DDX1, HRG, HDAC2, and PPP1CB) were identified for constructing a prognostic model. The overall survival (OS) curves, time-ROC curves, and nomograms based on the risk score or ALTs-score were developed to optimally predict the survival of HCC patients. ALTs-score was correlated with immune infiltration and confirmed its value in predicting immunotherapy outcomes. Furthermore, RT-qPCR demonstrated that eight risk signature genes were up-regulated in HCC cells. SMG5 silencing suppressed the proliferation, migration, and invasion of HCC cells. It was also found that SMG5 silencing reduced C-circle level in SNU-387 cells. We identified new ALT-related prognostic biomarkers for HCC. SMG5 knockdown inhibited the HCC progression, which might be a promising target for HCC therapy.

IMMU-46. CROSS-TALK OF INNATE IMMUNE CELLS IN GLIOBLASTOMA PATHOPHYSIOLOGY: UNDERSTANDING NLRP3 AND NLRP12

Abstract Glioblastoma (GBM) are malignant gliomas with an average survival of less than 14 months after diagnosis, despite surgery, chemotherapy, and radiotherapy. Tumor survival is aided by the heterogeneous cell populations in the tumor microenvironment. Two crucial innate immune cells of the brain, namely microglia and macrophages, favor a tumor-promoting microenvironment IN GBM. The nucleotide-binding domain leucine-rich repeat-containing receptor (NLR) family in astrocytes and microglia cells recognizes pathogen- and damage-associated molecular patterns that induce inflammation. NLRP3, upon activation, leads to cleavage and activation of Caspase-1, causing the proteolytic cleavage-mediated activation and conversion of pro-IL-1ß and pro-IL-18 to IL-1ß and IL-18, leading to pyroptosis. NLRP12 downregulates inflammatory responses in microglia and macrophages by regulating the NF-kB pathway. NLRP3 and NLRP12 have tumor-promoting and anti-tumour functions in various cancers. NLRP12 is a prognostic marker for glioblastoma, and its increased expression correlates with poor survival in GBM patients. However, the cellular and molecular mechanisms of NLRP3 and NLRP12 in GBM pathophysiology are largely unknown. This study aims to understand the effect of NLRP3 and NLRP12 expression in GBM, to interpret the pathophysiology, and to help develop future targeted therapeutic interventions. To understand the baseline expression of NLRP3 and NLRP12 in cell lines and patient-derived GBM cells, PCR, western blotting, and Immunocytochemistry were performed comparing WT and nlrp3-/- and nlrp12-/- cells. Differential expression of NLRP3 and NLRP12 in patient-derived cells can be compared to GBM, astrocyte, and microglia cell lines. This may be the underlying cause for intra and inter-tumor heterogeneity and resultant differential therapy outcomes. To understand cell-to-cell communication in GBM-like tumor microenvironments, heterocellular 3D organoids, and patient-derived tumor organoids were developed. Cell viability in 3D patient-derived tumor organoids decreased by up to 10% with Doxorubicin and 40% with Carboplatin treatment decreases compared to untreated controls. This suggests the scope of developing drugs targeting inflammation regulatory genes.