Predictor Of Overall Survival Research Articles

MPN-086 Time-Dependent Transfusion Status as an Independent Predictor of Overall Survival: A Post Hoc Analysis of 3 Phase 3 Trials

MPN-086 Time-Dependent Transfusion Status as an Independent Predictor of Overall Survival: A Post Hoc Analysis of 3 Phase 3 Trials

Time-Dependent Transfusion Status as an Independent Predictor of Overall Survival: A Post Hoc Analysis of 3 Phase 3 Trials

Time-Dependent Transfusion Status as an Independent Predictor of Overall Survival: A Post Hoc Analysis of 3 Phase 3 Trials

Comprehensive Analysis and Experimental Validation of TLL2 as a Potential New Prognostic Biomarker Associated with Immune Infiltration in Lung Adenocarcinoma.

The precise function of Tolloid Like 2 (TLL2) remains uncertain within the context of Lung Adenocarcinoma (LUAD). The primary objective of this investigation was to conduct a thorough analysis. To assess its diagnostic utility, data from The Cancer Genome Atlas (TCGA) database were used to assess TLL2 expression in pan-cancer and LUAD. The study has also investigated the correlation between TLL2 expression levels and LUAD symptoms and prognosis. Furthermore, the study has explored possible regulatory networks involving TLL2, including its association with immune infiltration, tumor stemness index (mRNAsi), and drug sensitivity in LUAD. We have explored TLL2 expression in single-cell sequencing of LUAD and the genomic variation and clinical significance of TLL2 in LUAD. The expression of TLL2 has been validated in GSE87340 and cell lines by quantitative Real-time PCR (qRT-PCR). An abnormal expression of TLL2 has been found in pan-cancer and LUAD. In LUAD patients, elevated levels of TLL2 were significantly related to the T stage (p = 0.046) and the pathological stage (p = 0.016). The expression of TLL2 in patients with LUAD was significantly associated with poorer Overall Survival (OS) (p < 0.001). The expression of TLL2 was determined to be an independent predictor of poorer OS (p = 0.042). TLL2 was associated with ribosome, neuroactive ligand-receptor interaction, allograft rejection, ECM receptor interaction, asthma, porphyrin and chlorophyll metabolism, focal adhesion, pentose and glucuronate inter-conversions, and ascorbate and aldarate metabolism. The expression of TLL2 in LUAD was correlated with immune infiltration and mRNAsi. The expression of TLL2 was significantly and negatively correlated with TAK-715, XMD13-2, STF-62247, OSI-930, and EHT-1864 in LUAD. The TLL2 gene was up- -regulated in multiple individual LUAD cells. LUAD patients with altered TLL2 had a shorter PFS as opposed to those with unaltered TLL2. The expression of TLL2 was significantly increased in LUAD cells. For patients with LUAD, TLL2 may serve as an immunotherapeutic target and a useful prognosis biomarker.

Expression analysis of SOX2 and SOX9 in patients with oral squamous cell carcinoma.

Lately SOX2 and SOX9, transcription factors associated with stemness-like phenotypes of cancer cells, have been linked to tumor growth, metastasis, and resistance to therapy. This study aimed on evaluating the expression of SOX2 and SOX9 in a large cohort of patients with OSCC including primary and recurrent tumors and corresponding lymph node metastases. Semiautomatic digital pathology scoring was used to determine protein expression and survival analysis was performed to evaluate its prognostic significance. We found a significant downregulation of SOX9 from primary disease to lymph node metastases (p < 0.001). SOX9 expression and the subgroup SOX2lowSOX9high were significantly correlated with worse overall survival (p < 0.05). Additionally, SOX2lowSOX9high expression pattern was confirmed as independent prognosticator for overall survival. These results indicate the relevant role of SOX2 and SOX9 in patients with OSCC and show the clinical relevance for further investigation on the molecular mechanisms underlying SOX-related gene expression.

Sexually dimorphic computational histopathological signatures prognostic of overall survival in high-grade gliomas via deep learning.

High-grade glioma (HGG) is an aggressive brain tumor. Sex is an important factor that differentially affects survival outcomes in HGG. We used an end-to-end deep learning approach on hematoxylin and eosin (H&E) scans to (i) identify sex-specific histopathological attributes of the tumor microenvironment (TME), and (ii) create sex-specific risk profiles to prognosticate overall survival. Surgically resected H&E-stained tissue slides were analyzed in a two-stage approach using ResNet18 deep learning models, first, to segment the viable tumor regions and second, to build sex-specific prognostic models for prediction of overall survival. Our mResNet-Cox model yielded C-index (0.696, 0.736, 0.731, and 0.729) for the female cohort and C-index (0.729, 0.738, 0.724, and 0.696) for the male cohort across training and three independent validation cohorts, respectively. End-to-end deep learning approaches using routine H&E-stained slides, trained separately on male and female patients with HGG, may allow for identifying sex-specific histopathological attributes of the TME associated with survival and, ultimately, build patient-centric prognostic risk assessment models.

Neutrophil-to-Lymphocyte Ratio and Prognostic Nutritional Index Are Predictors for Overall Survival after Primary Pancreatic Resection of Pancreatic Ductal Adenocarcinoma: A Single Centre Evaluation.

Prognostic inflammation-based parameters have been reported as useful tools in various oncologic diseases. Pancreatic ductal adenocarcinoma (PDAC) is characterized by a high mortality rate, making reliable prognostic markers highly desirable. However, there is still inconsistency in the literature regarding the efficacy of the different available scores. A total of 207 patients, who underwent primary resection of PDAC from January 2000 to December 2018 at the University Hospital of Erlangen, were included in this retrospective single-center study. Different biomarkers, including the preoperative neutrophil-lymphocyte ratio (NLR), the platelet-lymphocyte ratio (PLR), the c-reactive protein (CRP)-albumin ratio (CAR), the lymphocyte-CRP ratio (LCR), the prognostic nutritional index (PNI) and the modified Glasgow prognostic score (mGPS) were analyzed for their ability to predict overall survival (OS). In our cohort, the median overall survival was 20.7 months. Among the investigated biomarkers, NLR and PNI were identified as independent prognostic markers (Hazard Ratio (HR) 1.6 (1.0-2.5), p = 0.048 and HR 0.6 (0.4-0.9), p = 0.018), whereas PLR, CAR, LCR and mGPS did not reach significance in the multivariate analysis. Subgroup analysis revealed that the prognostic value of NLR and PNI is particularly evident in locally advanced tumor stages (pT3/4 and pN+). The NLR and PNI could serve as valuable tools for estimating prognosis in patients with PDAC undergoing pancreatic resection in curative intention, especially in locally advanced tumor stages. However, conflicting results in the current literature highlight the need for further prospective studies to validate these findings.

Impact of Surgical and Anesthetic Procedures after Colorectal Cancer Surgery: A Propensity Score-Matched Cohort Study (The PROCOL Study).

Background: Surgical inflammatory pain decreases the innate and adaptive immune antitumor response and favors residual circulating tumor cells. Objectives: This study investigated whether minimally invasive surgeries (laparoscopic and robotic procedures), which are less painful and inflammatory, improved oncological outcomes after colorectal resection compared to laparotomy. Methods: This research was a single-center propensity score-matched study involving patients who underwent colectomy and rectum resection from July 2017 to December 2019. Results: Seventy-four laparotomies and 211 minimally invasive procedures were included. Minimally invasive procedures were associated with less blood loss (0 mL vs. 75 mL, p < 0.001), shorter length of stay (8 days vs. 12 days, p < 0.001), and fewer complications at 3 months (11.8% vs. 29.4%, p = 0.02) compared to laparotomies. No difference in overall survival (OS) and recurrence-free survival (RFS) at 3 years between groups was observed. Univariate Cox regression analyses demonstrated that age and ASA > 3 can negatively impact OS, while adjuvant chemotherapy can positively influence OS. pT3-T4 stage and postoperative pain could negatively influence RFS. Multivariate Cox regression analyses concluded that age (HR 1.08, p < 0.01) and epidural analgesia (HR 0.12, p = 0.03) were predictors for OS. Lidocaine infusion (HR 0.39, p = 0.04) was a positive predictor for RFS. Conclusions: Minimally invasive procedures reduce postoperative complications and shorten the length of hospital stay compared to major surgeries without improving prognosis. However, the administration of local anesthetics through neuraxial anesthesia or intravenous infusion could improve survival and decrease the occurrence of relapses.

Predictive Models for Long Term Survival of AML Patients Treated with Venetoclax and Azacitidine or 7+3 Based on Post Treatment Events and Responses: Retrospective Cohort Study.

The treatment of acute myeloid leukemia (AML) in older or unfit patients typically involves a regimen of venetoclax plus azacitidine (ven/aza). Toxicity and treatment responses are highly variable following treatment initiation and clinical decision-making continually evolves in response to these as treatment progresses. To improve clinical decision support (CDS) following treatment initiation, predictive models based on evolving and dynamic toxicities, disease responses, and other features should be developed. This study aims to generate machine learning (ML)-based predictive models that incorporate individual predictors of overall survival (OS) for patients with AML, based on clinical events occurring after the initiation of ven/aza or 7+3 regimen. Data from 221 patients with AML, who received either the ven/aza (n=101 patients) or 7+3 regimen (n=120 patients) as their initial induction therapy, were retrospectively analyzed. We performed stratified univariate and multivariate analyses to quantify the association between toxicities, hospital events, and short-term disease responses and OS for the 7+3 and ven/aza subgroups separately. We compared the estimates of confounders to assess potential effect modifications by treatment. 17 ML-based predictive models were developed. The optimal predictive models were selected based on their predictability and discriminability using cross-validation. Uncertainty in the estimation was assessed through bootstrapping. The cumulative incidence of posttreatment toxicities varies between the ven/aza and 7+3 regimen. A variety of laboratory features and clinical events during the first 30 days were differentially associated with OS for the two treatments. An initial transfer to intensive care unit (ICU) worsened OS for 7+3 patients (aHR 1.18, 95% CI 1.10-1.28), while ICU readmission adversely affected OS for those on ven/aza (aHR 1.24, 95% CI 1.12-1.37). At the initial follow-up, achieving a morphologic leukemia free state (MLFS) did not affect OS for ven/aza (aHR 0.99, 95% CI 0.94-1.05), but worsened OS following 7+3 (aHR 1.16, 95% CI 1.01-1.31) compared to that of complete remission (CR). Having blasts over 5% at the initial follow-up negatively impacted OS for both 7+3 (P<.001) and ven/aza (P<.001) treated patients. A best response of CR and CR with incomplete recovery (CRi) was superior to MLFS and refractory disease after ven/aza (P<.001), whereas for 7+3, CR was superior to CRi, MLFS, and refractory disease (P<.001), indicating unequal outcomes. Treatment-specific predictive models, trained on 120 7+3 and 101 ven/aza patients using over 114 features, achieved survival AUCs over 0.70. Our findings indicate that toxicities, clinical events, and responses evolve differently in patients receiving ven/aza compared with that of 7+3 regimen. ML-based predictive models were shown to be a feasible strategy for CDS in both forms of AML treatment. If validated with larger and more diverse data sets, these findings could offer valuable insights for developing AML-CDS tools that leverage posttreatment clinical data.

Real-world efficacy, safety data and predictive clinical parameters for treatment outcomes in advanced soft tissue sarcoma treated with combined immunotherapy and antiangiogenic therapy

BackgroundThe combination of immunotherapy and antiangiogenic therapy has shown potential in the treatment of numerous malignant tumors, but limited evidence was available for soft tissue sarcomas (STS). Therefore, the aim of the present study is to assess the efficacy and safety of immunotherapy in conjunction with antiangiogenic therapy in patients diagnosed with advanced STS (aSTS).MethodsThe study enrolled patients with aSTS from January 2014 to October 2022. Eligible participants had previously received anthracycline-based chemotherapy, presented with an anthracycline-resistant sarcoma subtype, or were ineligible for anthracycline treatment due to medical conditions. Following enrollment, these patients received a combination of immunotherapy and antiangiogenic therapy. The primary endpoints were the objective response rate (ORR) and progression-free survival (PFS), while the secondary endpoints included the disease control rate (DCR), overall survival (OS), and the incidence of adverse events.ResultsFifty-one patients were included in this cohort study. The median duration of follow-up was 15.8 months. The ORR and DCR were 17.6%, and 76.5%, respectively. The median PFS (mPFS) was 5.8 months (95% CI: 4.8–6.8) for all patients, and the median OS had not been reached as of the date cutoff. Multivariate analysis indicated that Eastern Cooperative Oncology Group performance status of 0–1 and ≤ second-line treatment were positive predictors for both PFS and OS. Patients with alveolar soft part sarcoma or clear cell sarcoma had longer mPFS (16.2 months, 95% CI: 7.8–25.6) when compared to those with other subtypes of STS (4.4 months, 95% CI: 1.4–7.5, P < 0.001). Among the observed adverse events, hypertension (23.5%), diarrhea (17.6%), and proteinuria (17.6%) were the most common, with no treatment-related deaths reported.ConclusionThe combination of immunotherapy and antiangiogenic agents showed promising efficacy and acceptable toxicity in patients with aSTS, especially those with alveolar soft part sarcoma or clear cell sarcoma.

Impact of pre-sarcopenia on outcomes of transarterial chemoembolization in unresectable hepatocellular carcinoma

Sarcopenia’s impact on hepatocellular carcinoma (HCC) outcomes is well-documented, but the effects of pre-sarcopenia remain unclear. This study investigates the impact of pre-sarcopenia on tumor response and survival in patients with unresectable HCC undergoing transarterial chemoembolization (TACE). We retrospectively evaluated muscle volume using the SliceOmatic software in patients with unresectable HCC treated with TACE. Pre-sarcopenia was defined by Japan Society of Hepatology standards (men: 42 cm2/m2; women: 38 cm2/m2). Pre-sarcopenia and non-pre-sarcopenia groups were compared, and Cox proportional hazards model was used to identify survival-influencing variables. Subgroup analysis was conducted stratified by the tumor burden, using serum alpha-fetoprotein (AFP) levels at a diagnostic cutoff value of 200 ng/mL. Of the 100 patients, 39 had pre-sarcopenia. The presence of pre-sarcopenia was not associated with tumor complete response achievement. The median overall survival (OS) was significantly lower in the pre-sarcopenia group (18 months) than in the non-pre-sarcopenia group (30 months; log-rank P = 0.039). Subgroup analysis among 77 patients with AFP < 200 ng/mL revealed that OS was particularly poor in the pre-sarcopenia group (16 vs. 34 months; log-rank P < 0.001). Multivariate analysis identified increased AFP (adjusted hazard ratio [HR] per 10-unit increase 1.142; P < 0.001), higher Model for End-Stage Liver Disease score (adjusted HR per 1-unit increase 1.176; P < 0.001), and pre-sarcopenia (adjusted HR 2.965; P < 0.001) as predictors of shorter OS. Pre-sarcopenia is a significant predictor of increased mortality in patients with unresectable HCC undergoing TACE, especially in those with AFP < 200 ng/mL, suggesting its potential as a target for early intervention.

Prognostic and Predictive Significance of Primary Tumor Localization and HER2 Expression in the Treatment of Patients with KRAS Wild-Type Metastatic Colorectal Cancer: Single-Centre Experience from Serbia.

The treatment of patients with metastatic colorectal cancer (mCRC) is complex and is impacted by the location of the primary tumor (LPT). Our study aims to emphasize the importance of LPT as a prognostic and predictive marker as well as to examine the significance of HER2 overexpression in patients with mCRC, particularly in relation to the response to Epidermal Growth Factor Receptor Antibody treatment (anti-EGFR therapy). In this study, 181 patients with Kirsten RAS (KRAS) wild-type mCRC who received anti-EGFR therapy were included. Among them, 101 had left colon cancer (LCC) and 80 had right colon cancer (RCC). Results demonstrated that patients with KRAS wild-type LCC had better median overall survival (OS) (43 vs. 33 months, p = 0.005) and progression-free survival (PFS) (6 vs. 3 months, p < 0.001) compared to those with RCC. Multivariate analysis identified mucinous adenocarcinoma (p < 0.001), RCC location (p = 0.022), perineural invasion (p = 0.034), and tumors at the resection margin (p = 0.001) as independent predictors of OS, while mucinous adenocarcinoma (p = 0.001) and RCC location (p = 0.004) independently correlated with significantly shorter PFS. In addition, human epidermal growth factor receptor 2 (HER2) positive expression was significantly associated with worse PFS compared to HER2 negative results (p < 0.001). In conclusion, LPT is an important marker for predicting outcomes in the treatment of wild-type mCRC using anti-EGFR therapy, since patients with RCC have a statistically significantly shorter PFS and OS. Further investigation is needed to understand the role of HER2 overexpression in wild-type mCRC, as these patients also exhibit shorter survival.

Machine Learning versus Cox Models for Predicting Overall Survival in Patients with Osteosarcoma: A Retrospective Analysis of the EURAMOS-1 Clinical Trial Data.

Since the mid-1980s, there has been little progress in improving survival of patients diagnosed with osteosarcoma. Survival prediction models play a key role in clinical decision-making, guiding healthcare professionals in tailoring treatment strategies based on individual patient risks. The increasing interest of the medical community in using machine learning (ML) for predicting survival has sparked an ongoing debate on the value of ML techniques versus more traditional statistical modelling (SM) approaches. This study investigates the use of SM versus ML methods in predicting overall survival (OS) using osteosarcoma data from the EURAMOS-1 clinical trial (NCT00134030). The well-established Cox proportional hazard model is compared to the extended Cox model that includes time-varying effects, and to the ML methods random survival forests and survival neural networks. The impact of eight variables on OS predictions is explored. Results are compared on different model performance metrics, variable importance, and patient-specific predictions. The article provides comprehensive insights to aid healthcare researchers in evaluating diverse survival prediction models for low-dimensional clinical data.

Association between dichotomized VASARI feature and overall survival in glioblastoma patients: a single-institution propensity score matching analysis

ObjectivesThis study aimed to investigate the intra- and inter-observer consistency of the Visually Accessible Rembrandt Images (VASARI) feature set before and after dichotomization, and the association between dichotomous VASARI features and the overall survival (OS) in glioblastoma (GBM) patients.MethodsThis retrospective study included 351 patients with pathologically confirmed IDH1 wild-type GBM between January 2016 and June 2022. Firstly, VASARI features were assessed by four radiologists with varying levels of experience before and after dichotomization. Cohen’s kappa coefficient (κ) was calculated to measure the intra- and inter-observer consistency. Then, after adjustment for confounders using propensity score matching, Kaplan-Meier curves were used to compare OS differences for each dichotomous VASARI feature. Next, patients were randomly stratified into a training set (n = 211) and a test set (n = 140) in a 3:2 ratio. Based on the training set, Cox proportional hazards regression analysis was adopted to develop combined and clinical models to predict OS, and the performance of the models was evaluated with the test set.ResultsEleven VASARI features with κ value of 0.61–0.8 demonstrated almost perfect agreement after dichotomization, with the range of κ values across all readers being 0.874–1.000. Seven VASARI features were correlated with GBM patient OS. For OS prediction, the combined model outperformed the clinical model in both training set (C-index, 0.762 vs. 0.723) and test set (C-index, 0.812 vs. 0.702).ConclusionThe dichotomous VASARI features exhibited excellent inter- and intra-observer consistency. The combined model outperformed the clinical model for OS prediction.

Biologically interpretable multi-task deep learning pipeline predicts molecular alterations, grade, and prognosis in glioma patients

Deep learning models have been developed for various predictions in glioma; yet, they were constrained by manual segmentation, task-specific design, or a lack of biological interpretation. Herein, we aimed to develop an end-to-end multi-task deep learning (MDL) pipeline that can simultaneously predict molecular alterations and histological grade (auxiliary tasks), as well as prognosis (primary task) in gliomas. Further, we aimed to provide the biological mechanisms underlying the model’s predictions. We collected multiscale data including baseline MRI images from 2776 glioma patients across two private (FAHZU and HPPH, n = 1931) and three public datasets (TCGA, n = 213; UCSF, n = 410; and EGD, n = 222). We trained and internally validated the MDL model using our private datasets, and externally validated it using the three public datasets. We used the model-predicted deep prognosis score (DPS) to stratify patients into low-DPS and high-DPS subtypes. Additionally, a radio-multiomics analysis was conducted to elucidate the biological basis of the DPS. In the external validation cohorts, the MDL model achieved average areas under the curve of 0.892–0.903, 0.710–0.894, and 0.850–0.879 for predicting IDH mutation status, 1p/19q co-deletion status, and tumor grade, respectively. Moreover, the MDL model yielded a C-index of 0.723 in the TCGA and 0.671 in the UCSF for the prediction of overall survival. The DPS exhibits significant correlations with activated oncogenic pathways, immune infiltration patterns, specific protein expression, DNA methylation, tumor mutation burden, and tumor-stroma ratio. Accordingly, our work presents an accurate and biologically meaningful tool for predicting molecular subtypes, tumor grade, and survival outcomes in gliomas, which provides personalized clinical decision-making in a global and non-invasive manner.

Lipiodol accumulation patterns and their impact on survival outcomes in transarterial chemoembolization for hepatocellular carcinoma: a single institution retrospective analysis

Conventional Transarterial chemoembolization (TACE) using Lipiodol is a pivotal therapeutic modality for hepatocellular carcinoma (HCC). The link between Lipiodol accumulation patterns and patient survival outcomes remains underexplored. This study assesses the impact of these patterns on the prognosis of HCC patients undergoing TACE. We evaluated HCC patients treated with selective TACE between July 2015 and March 2020, classifying post-procedure Lipiodol accumulation observed on CT scans into four distinct patterns: homogeneous, heterogeneous, defective, and deficient. We analyzed cumulative local tumor recurrence (LTR), progression-free survival (PFS), and overall survival (OS) rates across these groups. Univariate and multivariate logistic regression analyses were performed to identify potential prognostic factors influencing PFS and OS. Among 124 HCC nodules, the distribution of Lipiodol patterns was: 65 homogeneous, 24 heterogeneous, 10 defective, and 25 deficient. Median PFS was 33.2, 9.1, 1.1, and 1.0 months, respectively, while median OS spanned 54.8, 44.5, 25.0, and 29.1 months for these groups. A significant difference in survival was found only between the homogeneous and defective patterns (hazard ratio, 2.33; confidence interval 1.25–4.36). Multivariate analyses revealed nonhomogeneous patterns as significant predictors of shorter PFS (HR 6.45, p < 0.001) and OS (HR 1.73, p = 0.033). Nonhomogeneous Lipiodol patterns in HCC following TACE significantly correlate with higher recurrence and decreased survival rates, especially with defective patterns. Early detection of these patterns may guide timely intervention strategies, potentially enhancing survival outcomes for patients with HCC.

Liver Oligometastasis in Biliary Tract Cancer and Impact on Survival Outcomes.

Outcomes of unresectable biliary tract cancer (BTC) with varying extents of liver involvement remain unclear. We evaluated characteristics and outcomes of BTC patients with liver metastases who underwent chemotherapy. We retrospectively reviewed consecutive BTC patients with synchronous or metachronous intrahepatic metastases who started first-line chemotherapy at our institution between January 2016 and December 2021. Ninety-six patients were included, of which 57 only had liver metastases and 39 had multiorgan involvement. The liver only group had longer median overall survival (OS) (11.8 vs. 7.4months, P = 0.006) and median progression-free survival (PFS) (4.1 vs. 2.7months, P = 0.035) than the multiorgan group. Patients with oligometastases (defined as no more than three liver metastases) achieved longer OS than those with polymetastases (four or more liver metastases) in the entire cohort. Within the liver only group, there were no significant differences in OS or PFS between the oligometastasis and polymetastasis groups. Patients who underwent subsequent surgery had significantly longer median OS than those who did not (44.4 vs. 7.7months, P < 0.001). Age ≥ 75years, liver-only metastasis, modified Glasgow prognostic score ≥ 1 carcinoembryonic antigen ≥ 5μg/L, and subsequent surgery were independent predictors of OS. Liver oligometastasis was only a significant predictor of longer OS in univariate Cox analysis. Outcomes in BTC patients with metastases limited to the liver, particularly those with oligometastasis, were more favorable than those with multiorgan metastases. Selected cases, generally with liver oligometastases, may achieve prolonged OS through subsequent surgery.

A dual-radiomics model for overall survival prediction in early-stage NSCLC patient using pre-treatment CT images.

Radiation therapy (RT) is one of the primary treatment options for early-stage non-small cell lung cancer (ES-NSCLC). Therefore, accurately predicting the overall survival (OS) rate following radiotherapy is crucial for implementing personalized treatment strategies. This work aims to develop a dual-radiomics (DR) model to (1) predict 3-year OS in ES-NSCLC patients receiving RT using pre-treatment CT images, and (2) provide explanations between feature importanceand model prediction performance. The publicly available TCIA Lung1 dataset with 132 ES-NSCLC patients received RT were studied: 89/43 patients in the under/over 3-year OS group. For each patient, two types of radiomic features were examined: 56 handcrafted radiomic features (HRFs) extracted within gross tumor volume, and 512 image deep features (IDFs) extracted using a pre-trained U-Net encoder. They were combined as inputs to an explainable boosting machine (EBM) model for OS prediction. The EBM's mean absolute scores for HRFs and IDFs were used as feature importance explanations. To evaluate identified feature importance, the DR model was compared with EBM using either (1) key or (2) non-key feature type only. Comparison studies with other models, including supporting vector machine (SVM) and random forest (RF), were also included. The performance was evaluated by the area under the receiver operating characteristic curve (AUCROC), accuracy, sensitivity, and specificity with a 100-fold Monte Carlo cross-validation. The DR model showed highestperformance in predicting 3-year OS (AUCROC=0.81 ± 0.04), and EBM scores suggested that IDFs showed significantly greater importance (normalized mean score=0.0019) than HRFs (score=0.0008). The comparison studies showed that EBM with key feature type (IDFs-only demonstrated comparable AUCROC results (0.81 ± 0.04), while EBM with non-key feature type (HRFs-only) showed limited AUCROC (0.64 ± 0.10). The results suggested that feature importance score identified by EBM is highly correlated with OS prediction performance. Both SVM and RF models were unable to explain key feature type while showing limited overall AUCROC=0.66 ± 0.07 and 0.77 ± 0.06, respectively. Accuracy, sensitivity, and specificity showed a similar trend. In conclusion, a DR model was successfully developed to predict ES-NSCLC OS based on pre-treatment CT images. The results suggested that the feature importance from DR model is highly correlated to the model prediction power.

M2BPGi Correlated with Immunological Biomarkers and Further Stratified Recurrence Risk in Patients with Hepatocellular Carcinoma

Introduction Novel biomarkers reflecting liver fibrosis and the immune microenvironment may correlate with the risk of hepatocellular carcinoma (HCC) recurrence. This study aimed to evaluate the prognostic value of serum biomarkers in predicting HCC recurrence. Methods Serum biomarkers, including M2BPGi, IL-6, IL-10, CCL5, VEGF-A, soluble PD-1, PD-L1, TIM-3, and LAG-3, were measured in 247 patients with HCC undergoing surgical resection. Factors associated with recurrence-free survival (RFS) and overall survival (OS) were evaluated. The ERASL-post model and IMbrave050 criteria were used to define HCC recurrence risk groups. Results Serum M2BPGi levels significantly correlated with FIB-4 score, APRI, ALBI score, AFP, ALT, AST, IL-10, CCL5, VEGF-A, soluble PD-1, PD-L1, TIM-3, and LAG-3 levels. M2BPGi, VEGF-A, soluble PD-1, and TIM-3 levels significantly correlated with RFS. In multivariate analysis, M2BPGi &amp;gt;1.5 COI (hazard ratio (HR) = 2.100, p &amp;lt; 0.001), tumor size &amp;gt;5 cm (HR = 1.859, p = 0.002), multiple tumors (HR = 2.562, p &amp;lt; 0.001), AFP &amp;gt;20 ng/mL (HR = 2.141, p &amp;lt; 0.001), and microvascular invasion (HR = 1.954, p = 0.004) were independent predictors of RFS. M2BPGi levels significantly stratified the recurrence risk in ERASL-post and IMbrave050 risk groups. An M2BPGi-based model could significantly discriminate RFS in the overall cohort as well as in the IMbrave050 low- and high-risk groups. M2BPGi &amp;gt;1.5 COI was also an independent predictor of OS after resection (HR = 2.707, p &amp;lt; 0.001). Conclusion Serum M2BPGi levels significantly correlated with surrogate markers of liver fibrosis, liver function, and immunology. M2BPGi is a significant predictor of HCC recurrence and survival after resection and could be incorporated into recurrence-prediction models.

DCE-CT parameters as new functional imaging biomarkers at baseline and during immune checkpoint inhibitor therapy in patients with lung cancer – a feasibility study

BackgroundWith the development of immune checkpoint inhibitors for the treatment of non-small cell lung cancer, the need for new functional imaging techniques and early response assessments has increased to account for new response patterns and the high cost of treatment. The present study was designed to assess the prognostic impact of dynamic contrast-enhanced computed tomography (DCE-CT) on survival outcomes in non-small cell lung cancer patients treated with immune checkpoint inhibitors.MethodsThirty-three patients with inoperable non-small-cell lung cancer treated with immune checkpoint inhibitors were prospectively enrolled for DCE-CT as part of their follow-up. A single target lesion at baseline and subsequent follow-up examinations were enclosed in the DCE-CT. Blood volume deconvolution (BVdecon), blood flow deconvolution (BFdecon), blood flow maximum slope (BFMax slope) and permeability were assessed using overall survival (OS) and progression-free survival (PFS) as endpoints in Kaplan Meier and Cox regression analyses.ResultsHigh baseline Blood Volume (BVdecon) (> 12.97 ml × 100 g−1) was associated with a favorable OS (26.7 vs 7.9 months; p = 0.050) and PFS (14.6 vs 2.5 months; p = 0.050). At early follow-up on day seven a higher relative increase in BFdecon (> 24.50% for OS and > 12.04% for PFS) was associated with an unfavorable OS (8.7 months vs 23.1 months; p < 0.025) and PFS (2.5 vs 13.7 months; p < 0.018). The relative change in BFdecon (categorical) on day seven was a predictor of OS (HR 0.26, CI95: 0.06 to 0.93 p = 0.039) and PFS (HR 0.27, CI95: 0.09 to 0.85 p = 0.026).ConclusionDCE-CT-identified parameters may serve as potential prognostic biomarkers at baseline and during early treatment in patients with NSCLC treated with immune checkpoint inhibitor therapy.

Proof of concept that melanoma nuclear count compares favourably with the benchmark histological prognostic feature, Breslow thickness.

Breslow thickness (BT) is the most important histological prognostic feature for melanoma prognosis, but it only captures tumour size in one dimension. Adding a further measurement in a different axis has been shown to improve prognostic value. It seems reasonable that further prognostic value could be obtained by estimating the number of invasive melanoma cells using nuclear count. The aim of this study was to show proof of concept that nuclear count has prognostic value independent of BT. Melanoma cell nuclei were labelled with SRY-related HMG-box 10 (SOX10) protein, the sections scanned and StarDist machine-learning algorithm used to count nuclei in 102 cases of primary cutaneous melanoma. Prognostic value was assessed using survival analyses. Nuclear count correlated strongly with T category, BT and calculated tumour area (each P < 0.001), suggesting that it was a valid marker of melanoma burden. Nuclear count was a predictor for overall survival in univariable analysis [hazard ratio (HR) = 2.25, confidence interval (CI) = 1.66-3.06, P < 0.001] and multivariable analysis (HR = 2.60, CI = 1.59-4.24, P < 0.001). BT and ulceration were significant in univariable analyses, but not in multivariable models with nuclear count. Models containing nuclear count showed the best fit. Similar results were seen for melanoma-specific and metastasis-free survival. Nuclear count was able to stratify melanomas within a given T stage. This study demonstrated proof of concept that counting melanoma nuclei may be an improved measure of invasive tumour burden compared to BT. Future studies will need to refine methods of nuclear detection and also to confirm its prognostic value.