Better Prediction Of Survival Research Articles

Establishment of a prognostic nomogram and discussion on optimal treatment for cervical adenocarcinoma:a retrospective study based on SEER database and Chinese single-center data

Objective: To establish and validate a predicting nomogram for cervical adenocarcinoma based on surveillance, epidemiology and end results (SEER) database and Chinese single-center data, and to explore the optimal treatment for cervical adenocarcinoma. Methods: This study selected 2 478 cervical adenocarcinoma patients from the SEER database as the training cohort, and 195 cervical adenocarcinoma patients from Cancer Hospital of Dalian University of Technology, Liaouing Cancer Hospital and Institute as an external validation cohort. Clinicopathological information and follow-up data of the two cohorts were collected. The radiotherapy group was defined as receiving comprehensive treatment based on concurrent chemoradiotherapy after initial diagnosis, while the surgery group was defined as receiving comprehensive treatment based on radical surgery. Log-rank test and cox regression were used to evaluate factors affecting the prognosis of cervical adenocarcinoma patients. A nomogram was drawn to predict the 3-year and 5-year overall survival rates of cervical adenocarcinoma patients, and then internal validation of the training cohort from SEER database and external validation of the hospital cohort were conducted. Results: (1) In the SEER database training cohort, there were 385 patients (15.54%, 385/2 478) in the radiotherapy group and 2 093 patients (84.46%, 2 093/2 478) in the surgery group. Overall survival time of the radiotherapy group was (55.8±51.3) months, while that of the surgery roup was (94.4±61.7) months, the difference between the two groups was statistically significant (χ2=256.44, P<0.001). Log-rank test showed that age, marital status, maximum of tumor diameters, pathological grade, International Federation of Gynecology and Obstetrics (FIGO) stage, and treatments were all significant factors affecting the overall survival time of cervical adenocarcinoma patients (all P<0.001). Multivariate Cox regression analysis showed that elder (>50 years old), single status, huge tumors (>4 cm), high pathological grades (G2, G3), and advanced FIGO stages (≥Ⅱa2 stage) were independent risk factors for the overall survival time of cervical adenocarcinoma patients (all P<0.05); compared with radiotherapy, surgery was a protective factor for the prognosis of cervical adenocarcinoma patients (HR=0.619, 95%CI: 0.494-0.777; P<0.001). Further analysis of locally advanced stage and Ⅲc stage of patients showed that surgery was a protective factor for the prognosis of cervical adenocarcinoma patients with a maximum tumor diameter >4 to <6 cm (HR=0.414, 95%CI: 0.182-0.942; P=0.036) in locally advanced stage and Ⅲc T1 to T2 stage (HR=0.473, 95%CI: 0.307-0.728; P=0.001). (2) The external validation cohort consisted of 39 patients (20.00%, 39/195) in the radiotherapy group and 156 patients (80.00%, 156/195) in the surgery group. The overall survival time of patients in the radiotherapy group was (51.7±34.3) months, while that of the surgery group was (63.1±26.6) months (χ2=28.41, P<0.001). Further analysis was conducted on locally advanced stage and Ⅲc stage patients, and multivariate Cox regression analysis was performed after propensity score matching, which showed that surgery was a protective factor for the prognosis of cervical adenocarcinoma patients with a maximum tumor diameter >4 to <6 cm in locally advanced stage (HR=0.141, 95%CI: 0.023-0.843; P=0.032) and Ⅲc T1 to T2 stage (HR=0.184, 95%CI: 0.036-0.947; P=0.043). (3) Establishment and internal and external validation of nomogram: based on the six factors screened out by the multivariate Cox regression model, the nomogram was developed to predict the prognosis of cervical adenocarcinoma patients. The consistency index of the internal and external validation were 0.801 and 0.766, respectively, and the calibration curves matched well with the ideal fitting line. Conclusions: The key to the treatment of cervical adenocarcinoma is to prioritize radical surgery for patients with conditions for radical tumor resection. Compared with concurrent chemoradiotherapy, patients with locally advanced stages (Ⅰb3, Ⅱa2), and Ⅲc (T1, T2) stages cervical adenocarcinoma could benefit from comprehensive treatment based on radical surgery. The nomogram of this study has been validated internally and externally, and show good survival prediction efficacy for cervical adenocarcinoma patients.

Construction of a ferroptosis and hypoxia-related gene signature in cervical cancer to assess tumour immune microenvironment and predict prognosis

Background This study aimed to investigate the potential role of ferroptosis/hypoxia-related genes in cervical cancer to improve early management and treatment of cervical cancer. Methods All data were downloaded from public databases. Ferroptosis/hypoxia-related genes associated with cervical cancer prognosis were selected to construct a risk score model. The relationship between risk score and clinical features, immune microenvironment and prognosis were analysed. Results Risk score model was constructed based on eight signature genes. Drug prediction analysis showed that bevacizumab and cisplatin were related to vascular endothelial growth factor A. Risk score, as an independent prognostic factor of cervical cancer, had a good survival prediction effect. The two groups differed significantly in degree of immune cell infiltration, gene expression, tumour mutation burden and somatic variation. Conclusions We developed a novel prognostic gene signature combining ferroptosis/hypoxia-related genes, which provides new ideas for individual treatment of cervical cancer.

DIAPH3 predicts survival of patients with MGMT-methylated glioblastoma.

Glioblastoma is one of the most aggressive primary brain tumors, with a poor outcome despite multimodal treatment. Methylation of the MGMT promoter, which predicts the response to temozolomide, is a well-established prognostic marker for glioblastoma. However, a difference in survival can still be detected within the MGMT methylated group, with some patients exhibiting a shorter survival than others, emphasizing the need for additional predictive factors. We analyzed DIAPH3 expression in glioblastoma samples from the cancer genome atlas (TCGA). We also retrospectively analyzed one hundred seventeen histological glioblastomas from patients operated on at Saint-Luc University Hospital between May 2013 and August 2019. We analyzed the DIAPH3 expression, explored the relationship between mRNA levels and Patient's survival after the surgical resection. Finally, we assessed the methylation pattern of the DIAPH3 promoter using a targeted deep bisulfite sequencing approach. We found that 36% and 1% of the TCGA glioblastoma samples exhibit copy number alterations and mutations in DIAPH3, respectively. We scrutinized the expression of DIAPH3 at single cell level and detected an overlap with MKI67 expression in glioblastoma proliferating cells, including neural progenitor-like, oligodendrocyte progenitor-like and astrocyte-like states. We quantitatively analyzed DIAPH3 expression in our cohort and uncovered a positive correlation between DIAPH3 mRNA level and patient's survival. The effect of DIAPH3 was prominent in MGMT-methylated glioblastoma. Finally, we report that the expression of DIAPH3 is at least partially regulated by the methylation of three CpG sites in the promoter region. We propose that combining the DIAPH3 expression with MGMT methylation could offer a better prediction of survival and more adapted postsurgical treatment for patients with MGMT-methylated glioblastoma.

A Prognostic Nomogram Survival Model for Newly Diagnosed Patients with AIDS-Related Diffuse Large B-Cell Lymphoma: A Multicenter Cohort Study in China

BACKGROUND The prognosis of patients with AIDS-related diffuse large B-cell lymphoma(AR-DLBCL) becomes very poor when the lymphoma relapses or is refractory to first line immunochemotherapy. We aimed to develop a novel ARDPI nomogram prognostic model for risk stratification so as to guide individualized treatment to achieve sustained remission and improve the overall prognosis for newly-diagnosed AR-DLBCL patients. METHODS We interrogated data from 306 patients with newly-diagnosed AR-DLBCL. We filtered variables using LASSO regression and Cox regression to identify prognostic co-variates and develop a survival model, we termed AR-DLBCL Prognostic Index (ARDPI). We evaluated model discrimination, calibration and clinical benefit by Area Under the Receiver-Operator Characteristic (AUROC), calibration plots and decision curve analysis (DCA). Next, we compared the ARDPI model discrimination, calibration and clinical benefit with the IPI and NCCN-IPI models using the same methods. Lastly, we stratified patients into three survival risk cohorts based on ARDPI model by X-tile selecting cutoff point. RESULTS 7 co-variates were independently correlated with survival and were used to develop the ARDPI model including age, lymphocyte monocyte ratio (LMR), CD5 expression on lymphoma cells, blood EBV-DNA copy number, CD4/CD8 ratio, central nervous system (CNS) involvement and anti-HIV therapy (ART). AUROCs of ARDPI model for 1-, 3-, and 5-year were 0.80 (95% Confidence Interval [CI], 0.72, 0.88), 0.78 (0.69, 0.87) and 0.77 (0.63, 0.91). Predicted and calibrated values were concordant. The DCA curve had higher net benefit using the ARDPI model. Prediction accuracy of the ARDPI model was better compared with the IPI and NCCN-IPI models. For example, 3-year survival AUROC in the ARDPI model was 0.78 (0.69, 0.87) compared with the IPI (0.53 [0.43, 0.63] P&amp;lt; 0.001) and the NCCN-IPI (0.52 [0.42, 0.62] P&amp;lt; 0.001). Using the ARDPI model, we identified 3 survival risk cohorts with 3-year survivals of 0.80 , 0.38 and 0.09 ( P&amp;lt;0.001). CONCLUSION The ARDPI has good survival prediction accuracy in newly-diagnosed persons with AIDS-related DLBCL and using it has clinical benefit.Accuracy is better than the IPI and NCCN-IPI models. Validation of our conclusions is needed.

Predicting Survival Signature of Bladder Cancer Related to Cancer-Associated Fibroblast (CAF) Constructed by Intersecting Genes in TCGA and GEO.

Bladder cancer was one of the most common carcinomas around the world. However, the mechanism of the disease still remained to be investigated. We expected to establish a prognostic survival model with 9 prognostic genes to predict overall survival (OS) in patients of bladder cancer. The gene expression data of bladder cancer were obtained from TCGA and GEO datasets. TCGA and GEO datasets were used for screening prognostic genes along with developing and validating a 9-gene prognostic survival model by method of weighted gene co-expression network analysis (WGCNA) and LASSO with Cox regression. The relative analysis of evaluate tumor burden mutation (TBM), GO, KEGG, chemotherapy drug and functional pathway were also performed based on CAF-related mRNAs. 151 Overlapping CAF-related genes were distinguished after intersecting differentially expressed genes from 945 genes in TCGA and 491 genes in GEO dataset. 9 Prognostic genes (MSRB2, AGMAT, KLF6, DDAH2, GADD45B, SERPINE2, STMN3, TEAD2, and COMP) were used for construction of prognostic model after LASSO with Cox regression. Receiver operating characteristic (ROC) curves showed a good survival prediction by this model. Functional analysis indicated chemokine, cytokine, ECM interaction, oxidative stress and apoptosis were highly appeared. Potential drugs targeted different CAF-related genes like vemurafenib, irofulven, ghiotepa, and idarubicin were found as well. We constructed a novel 9 CAF-related mRNAs prognostic model and explored the gene expression and potential functional information of related genes, which might be worthy of clinical application. In addition, potential chemotherapy drugs could provide useful insights into the potential clinical treatment of bladder cancer.

A novel and robust pyroptosis-related prognostic signature predicts prognosis and response to immunotherapy in esophageal squamous cell carcinoma.

Esophageal squamous cell carcinoma (ESCC) is a highly malignant gastrointestinal tumor, has a poor prognosis and high mortality rate. Pyroptosis could regulate tumor cell proliferation, invasion, and metastasis, thereby affecting the prognosis of cancer patients. However, the role of pyroptosis-related genes (PRGs) in ESCC remains unclear. This study selected 33 PRGs, and finally identified 29 PRGs that were differentially expressed between ESCC and normal esophageal tissues. The genetic mutation variation landscape of PRG in ESCC was also summarised. Based on consensus clustering for the 33 PRGs, all ESCC patients could be divided into two subtypes. Functional enrichment analysis revealed that these 33 PRGs were mainly involved in cytokine production, interleukin-1 production, and the NOD-like receptor signalling pathway. We created a prognostic PRG signature based on least absolute shrinkage and selection operator regression and Cox regression analysis with good survival prediction ability in both GEO and TCGA cohorts. Combined with the clinical characteristics, signature-based risk score was found to be an independent factor for predicting the OS of ESCC patients. A nomogram with enhanced precision for forecasting ESCC was established based on various independent prognostic elements. Significant correlation was observed between prognostic PRGs and immune-cell infiltration, tumor mutation burden, microsatellite instability, immune checkpoint, and drug sensitivity. Finally, we validated the expression of four PRGs in ESCC cell lines and tissues samples. In conclusion, the PRGs exerted significant effects on tumor immunity and prognosis of ESCC.

An 8-gene predicting survival model of hepatocellular carcinoma (HCC) related to pyroptosis and cuproptosis

BackgroundThe study aimed to establish a prognostic survival model with 8 pyroptosis-and-cuproptosis-related genes to examine the prognostic effect in patients of hepatocellular carcinoma (HCC).MethodsWe downloaded gene expression data and clinical information of HCC patients from The Cancer Genome Atlas (TCGA), International Cancer Genome Consortium (ICGC) and Gene Expression Omnibus (GEO). The clustering analysis and cox regression with LASSO were used for constructing an 8 PCmRNAs survival model. Using TCGA, ICGC and GEO cohort, the overall survival (OS) between high- and low- risk group was determined. We also evaluated independent prognostic indicators using univariate and multivariate analyses. The relatively bioinformatics analysis, including immune cell infiltration, function enrichment and drug sensitivity analyses, was performed as well. The gene expression of 8 PCmRNAs in vitro were validated in several HCC cell lines by qRT-PCR and Western blot. The relationship between GZMA and Fludarabine were further checked by CCK-8 assay.ResultsThe survival prognostic model was constructed with ATP7A, GLS, CDKN2A, BAK1, CHMP4B, NLRP6, NOD1 and GZMA using data from TCGA cohort. The ICGC and GEO cohort were used for model validation. Receiver operating characteristic (ROC) curves showed a good survival prediction by this model. Risk scores had the highest predictable value for survival among Stage, Age, Gender and Grade. Most Immune cells and immune functions were decreased in high-risk group. Besides, function enrichment analyses showed that steroid metabolic process, hormone metabolic process, collagen − containing extracellular matrix, oxidoreductase activity and pyruvate metabolism were enriched. Potential drugs targeted different PCDEGs like Nelarabine, Dexamethasone and Fludarabine were found as well. ATP7A, GLS, CDKN2A, BAK1, CHMP4B, NOD1 were upregulated while NLRP6 and GZMA were downregulated in most HCC cell lines. The potential therapy of Fludarabine was demonstrated when GZMA was low expressed in Huh7 cell line.ConclusionWe constructed a novel 8-gene (ATP7A, GLS, CDKN2A, BAK1, CHMP4B, NLRP6, NOD1 and GZMA) prognostic model and explored potential functional information and microenvironment of HCC, which might be worthy of clinical application. In addition, several potential chemotherapy drugs were screened and Fludarabine might be effective for HCC patients whose GZMA was low expressed.

Predictive Ability of Different Scoring Models in Severe Alcoholic Hepatitis

ABSTRACT Introduction: Alcoholic hepatitis is a major health problem which carries a high mortality despite treatment. Predicting severity of disease is of utmost importance to avoid overzealous and unnecessary treatment with steroids in each and every case. Maddrey’s discriminant function (DF) may not always be reliable in predicting such prognosis. Aim of the Study: To observe the predictive ability of Glasgow alcoholic hepatitis score (GAHS), age, bilirubin, international normalized ratio (INR) and creatinine (ABIC) score, model for end stage liver disease (MELD) score, and Lille’s score in determining the prognosis in severe alcoholic hepatitis. Materials and Methods: Patients of severe alcoholic hepatitis were enrolled in this study as per inclusion and exclusion criteria and a DF score of &gt;32. Once the decision to initiate corticosteroids was made, DF scores, MELD scores, GAHS scores, and ABIC scores were calculated on the day on initiation of therapy. On day 7, patients were re-assessed and Lille score was calculated in addition. Follow-up was obtained on repeat out-patient visit or telephonically after every 2 weeks till 90 days or till mortality, whichever occurred first. Statistical Analysis: The scores were compared among the survival and mortality groups to see association between mortality and severity of scores using Chi-square test of significance. Cut-offs providing a composite of best sensitivity, specificity, negative predictive value, and positive predictive value were derived on the basis of data of our study. P value of ≤ 0.05 was taken as a cut-off for significance. Results: Our study comprised 30 patients with severe alcoholic hepatitis. At the end of 90-day follow-up, 23 patients survived with 7 (23%) mortalities because of liver failure and sepsis. On day 7, barring the ABIC score (0.713), the areas under curve of all scores were above 0.8 and were hence of comparable efficacy. Lille score performed the best at assessing treatment response but had the limitation of not being applicable at day 1. Conclusion: Severe alcoholic hepatitis carries a bad prognosis. Cut-offs of DF and MELD need to be raised, whereas they need to be lowered in the case of GAHS, ABIC, and Lille score to improve the predictive value of score. Steroid-treated patients should be evaluated preferably with Lille score for a better prediction of survival.

Comprehensive analysis of the prognosis and immune infiltrates for the BET protein family reveals the significance of BRD4 in glioblastoma multiforme.

Background: Glioblastoma multiforme (GBM) is the most common and invasive primary central nervous system tumor. The prognosis after surgery, radiation and chemotherapy is very poor. Bromodomain (BRD) proteins have been identified in oncogenic rearrangements, and play a key role in the development of multiple cancers. However, the relationship between BET proteins and prognosis of GBM are still worth exploring, and the distinct functions of BET proteins and tumor immunology in GBM have not been fully elucidated. Therefore, it is particularly important to develop effective biomarkers to predict the prognosis of GBM patients. Methods: Metascape, David, Kaplan-Meier Plotter, Oncomine, GEPIA, TCGA, TIMER, and LinkedOmics databases were used to assess the expression and prognosis for BET proteins in GBM. ROC analysis of risk model was established to identify the correlation between BET genes and overall survival in GBM patients. TIMER and GEPIA databases were used to comprehensively investigate the correlation between BET genes and tumor immune infiltration cells. Moreover, the image of immunohistochemistry staining of BET proteins in normal tissue and tumor tissue were retrived from the HPA database. In addition, differential analysis and pathway enrichment analysis of BRD4 gene expression profile were also carried out. Finally, immune-fluorescence and Western blot were used to clarify the expression of BRD4 in GBM cells. Results: Bioinformatics analysis showed that the expression levels of BET genes in GBM may play an important role in oncogenesis. Specifically, bioinformatic and immunohistochemistry analysis showed that BRD4 protein was more highly expressed in tumor tissues than that in normal tissues. The high expression of BRD4 was associated with poor prognosis in GBM. The expression of BET genes were closely related to the immune checkpoint in GBM. The correlation effect of BRD4 was significantly higher than other BET genes, which represented negative correlation with immune checkpoint. The expression of BRD4 was positively associated with tumor purity, and negatively associated with immune infiltration abundance of macrophage, neutrophil and CD8+ T-cell, respectively. Cox analysis showed that the model had good survival prediction and prognosis discrimination ability. In addition, the expression levels of BRD4 protein was significantly higher in U-251MGcells than that in normal cells, which was consistent with the results of bioinformatics data. Conclusion: This study implied that BRD4 could be hopeful prognostic biomarker in GBM. The increased expression of BRD4 may act as a molecular marker to identify GBM patients with high-risk subgroups. BRD4 may be a valuable prognostic biomarker, and a potential target of precision therapy against GBM.

Development of a novel transcription factor signature for accurate cervical cancer prognosis

Cervical cancer (CC) is a leading cause of cancer-related deaths in women. During tumor development, transcriptional factors regulate the transcription of proto-oncogenes and tumor suppressor genes. We examined the possibility of using transcription factors as prognostic biomarkers for patients with cervical cancer. Single-cell RNA-sequencing data were downloaded from the Gene Expression Omnibus database to identify specific activated transcription factors in different types of cells from CC. Publicly available bulk RNA-sequencing and clinical data of CC were obtained to identify associated prognostic transcription factors using survival analysis and the random survival forest methods. Accuracy and effectiveness of the established transcription factor-related predictive random survival forest model were verified using training and test datasets. We identified specific activated transcription factors in tissue cells of cervical cancer. A 3-transcription factors (PBX4 (PBX Homeobox 4), EBF2 (EBF Transcription Factor 2) and ZNF696 (Zinc Finger Protein 696)) prognostic signature for patients with cervical cancer was constructed showing good survival prediction. Gene function enrichment analysis indicated a correlation between the prognostic characteristics and different signaling pathways associated with cancer. Using the random survival forest model based on the 3-transcription factor signature, patients with cervical cancer were stratified into low- and high-risk groups with significant variations in overall survival (p &lt; 0.001). The area under the curve of the time-dependent receiver operator characteristic revealed a strong predictive accuracy for training and test datasets of the corresponding signature. CC has cellular heterogeneity of transcriptional activation. Our analyses provide a novel transcription factor-associated prognostic model for CC. These transcription factors could be used as effective prognostic biomarkers and potential therapeutic targets for patients with cervical cancer.

A survival prediction model and nomogram based on immune-related gene expression in chronic lymphocytic leukemia cells.

There are many different chronic lymphoblastic leukemia (CLL) survival prediction models and scores. But none provide information on expression of immune-related genes in the CLL cells. We interrogated data from the Gene Expression Omnibus database (GEO, GSE22762; Number = 151; training) and International Cancer Genome Consortium database (ICGC, CLLE-ES; Number = 491; validation) to develop an immune risk score (IRS) using Least absolute shrinkage and selection operator (LASSO) Cox regression analyses based on expression of immune-related genes in CLL cells. The accuracy of the predicted nomogram we developed using the IRS, Binet stage, and del(17p) cytogenetic data was subsequently assessed using calibration curves. A survival model based on expression of 5 immune-related genes was constructed. Areas under the curve (AUC) for 1-year survivals were 0.90 (95% confidence interval, 0.78, 0.99) and 0.75 (0.54, 0.87) in the training and validation datasets, respectively. 5-year survivals of low- and high-risk subjects were 89% (83, 95%) vs. 6% (0, 17%; p < 0.001) and 98% (95, 100%) vs. 92% (88, 96%; p < 0.001) in two datasets. The IRS was an independent survival predictor of both datasets. A calibration curve showed good performance of the nomogram. In vitro, the high expression of CDKN2A and SREBF2 in the bone marrow of patients with CLL was verified by immunohistochemistry analysis (IHC), which were associated with poor prognosis and may play an important role in the complex bone marrow immune environment. The IRS is an accurate independent survival predictor with a high C-statistic. A combined nomogram had good survival prediction accuracy in calibration curves. These data demonstrate the potential impact of immune related genes on survival in CLL.

Identification and validation of a prognostic model based on ferroptosis-associated genes in head and neck squamous cancer.

Ferroptosis is that under the action of ferrous iron or ester oxygenase, unsaturated fatty acids highly expressed on the cell membrane are catalyzed to undergo lipid peroxidation, thereby inducing cell death. In this study, we used ferroptosis marker genes to identify 3 stable molecular subtypes (C1, C2, C3) with distinct prognostic, mutational, and immune signatures by consensus clustering; TP53, CDKN2A, etc. Have higher mutation frequencies in the three subtypes. C3 has a better prognosis, while the C1 subtype has a worse prognosis. WGCNA is used to identify molecular subtype-related gene modules.After filting, we obtained a total of 540 genes related to the module feature vector (correlation>0.7).We performed univariate COX regression analysis on these genes, and identified a total of 97 genes (p < 0.05) that had a greater impact on prognosis, including 8 ''Risk" and 89 ''Protective" genes. After using lasso regression, we identified 8 genes (ZNF566, ZNF541, TMEM150C, PPAN, PGLYRP4, ENDOU, RPL23 and MALSU1) as ferroptosis-related genes affecting prognosis. The ferroptosis prognosis-related risk score (FPRS) was calculated for each sample in TCGA-HNSC dataset. The results showed that FPRS was negatively correlated with prognosis.The activated pathways in the PFRS-high group mainly include immune-related pathways and invasion-related pathways. We assessed the extent of immune cell infiltration in patients in our TCGA-HNSC cohort by using the expression levels of gene markers in immune cells. The FPRS-high group had a higher level of immune cell infiltration. We found that the expression of immune checkpoints was significantly up-regulated in the FPRS-low group and the FPRS-high group had a higher probability of immune escape and a lower probability of benefiting from immunotherapy. In this work, we constructed a scoring Ferroptosis-related prognostic model that can well reflect risk and positive factors for prognosis in patients with head and neck squamous cell carcinoma. It can be used to guide individualized adjuvant therapy and chemotherapy for patients with head and neck cancer. Therefore, it has a good survival prediction ability and provides an important reference for clinical treatment.

NCOG-05. A BRAIN METASTASES SURVIVAL MODEL USING AN ENSEMBLE TREE APPROACH

Abstract PURPOSE/OBJECTIVE(S) The primary purpose of this study is to determine whether a machine learning approach can estimate survival in patients with brain metastases undergoing stereotactic radiosurgery or fractionated stereotactic radiotherapy (SRS/SRT). The secondary purpose is to identify covariates of importance. MATERIALS/METHODS Data were collected for 377 SRS/SRT treatments in 291 patients done between the years 2008-2021. If a patient was treated with more than one course of SRS/SRT within 30 days, they were counted only once. Twenty-five clinically-relevant variables were identified as covariates and the primary outcome of time from brain metastasis diagnoses to death was used to build a random survival forest model. Brain metastasis location was categorized as infratentorial, supratentorial, or both. An 80/20 split was used for training (n = 302) and test (n = 75) sets. Missing data points were imputed using a just-in-time adaptive tree approach. Minimal depth and variable importance (VIMP) approaches were used to identify prognostic factors. Model performance was assessed using time-dependent area under the receiver operating characteristics curve (tAUC). RESULTS Median survival time was 16 months. The most important variables according to minimal depth analysis (depth threshold 5.23) were Karnofsky Performance Status (KPS), extracranial status, age, insurance status, metastases volume, histology, number of metastases, and location. Error rate on the test set was 0.38. tAUC was found to increase continuously over time and at 6, 12, 24, and 36 months was 0.56, 0.63, 0.74, and 0.84 respectively. CONCLUSION An ensemble tree approach can provide good survival prediction for patients with brain metastases undergoing SRS/SRT. Model performance, as measured by tAUC, increases over time suggesting better predictive capability at longer time intervals. Future directions include collecting more data to increase model performance, comparing to other models, and validating with external data.

A Brain Metastases Survival Model Using an Ensemble Tree Approach

<h3>Purpose/Objective(s)</h3> The primary purpose of this study is to determine whether a machine learning approach can estimate survival in patients with brain metastases undergoing stereotactic radiosurgery or fractionated stereotactic radiotherapy (SRS/SRT). The secondary purpose is to identify variables of importance. <h3>Materials/Methods</h3> Data were collected for 285 SRS/SRT treatments in 228 patients. If a patient was treated with more than one course of SRS/SRT within 30 days, they were counted only once. Twenty-five clinically-relevant variables were identified as covariates and the primary outcome of time from brain metastasis diagnoses to death was used to build a random survival forest model. Brain metastasis location was categorized as infratentorial, supratentorial, or both. An 80/20 split was used for training (n = 228) and test (n = 57) sets. Missing data points were imputed using a just-in-time adaptive tree approach. Minimal depth and variable importance (VIMP) approaches were used to identify prognostic factors. Model performance was assessed using time-dependent area under the receiver operating characteristics curve (tAUC). <h3>Results</h3> Median survival time was 17 months. The most important variables according to minimal depth analysis (depth threshold 5.49) were age, extracranial status, Karnofsky Performance Status (KPS), intracranial metastases volume, primary cancer histology, insurance status, number of intracranial metastases, SRS/SRT dose, molecular marker status, and brain metastasis location (see table). Error rate on the test set was 0.27. tAUC was found to increase continuously over time and at 3, 6, 12, and 24 months was 0.62, 0.72, 0.77, and 0.86 respectively. <h3>Conclusion</h3> An ensemble tree approach can provide good survival prediction for patients with brain metastases undergoing SRS/SRT. Model performance, as measured by tAUC, increases over time suggesting better predictive capability at longer time intervals. Future directions include collecting more data to increase model performance, comparing to other models, and validating with external data.

Identification and validation of a prognostic model of necroptosis-related lncRNAs in hepatocellular carcinoma.

Background: The study focused on establishing a prognostic survival model with six necroptosis-related lncRNAs to predict overall survival (OS) in patients with hepatocellular carcinoma (HCC). Methods: The data of gene expression and clinical information of HCC patients were obtained from The Cancer Genome Atlas (TCGA). Cox regression with LASSO was used for constructing a necroptosis-related lncRNA survival model, which we further validated with qRT-PCR in vitro. The relative bioinformatics analysis and consensus cluster analysis were performed based on six differentially expressed lncRNAs. Results: The survival prognostic model was constructed by using data from TCGA. Receiver operating characteristic (ROC) curves showed a good survival prediction by this model. GSEA showed that several signaling pathways were related to HCC progression. Immune-related functional analysis showed that aDCs, macrophages, Th2 cells, and Tregs have stronger correlation with the high-risk group. The consensus cluster analysis further validated the 6-lncRNA prognostic model. Conclusion: A novel 6-lncRNA (AL606489.1, NRAV, LINC02870, DUXAP8, “ZFPM2-AS1,” and AL031985.3) prognostic model had an accurately predictive power in HCC prognosis, which might be worthy of clinical application.

Comprehensive analysis of TAMs marker genes in glioma for predicting prognosis and immunotherapy response

Glioma is one of the most frequent types of primary tumors in central nervous system. Previous studies deomostrated that tumor-associated macrophages (TAMs) and their marker genes were significantly associated with immunologic suppression and immune escape of cancer. However, the molecular mechanism between glioma and TAM marker genes is still rarely reported. In this research, we performed a comprehensive analysis of the prognostic prediction value of TAM marker genes in multiple glioma cohorts. Further investigation indicated that the increased expression of TAM marker genes resulted in the immune suppressive microenvironment in glioma through regulating tumor-infiltrating immune cells and Cancer-Immunity Cycle. To better forecast the survival of glioma patients, we then developed gene risk models in four glioma datasets (CGGA, TCGA, Rembrandt and Gravendeel). Univariate and multivariate Cox analysis exhibited the good survival prediction ability and prognostic discrimination ability of our models. The results of immunotherapy prediction indicated that glioma patients with low risk were more likely to benefit from ICB (immune checkpoint blockade) treatment. Altogether, our research provided a comprehensive analysis of TAM marker genes and explored their value for predicting prognosis and immunotherapy response in glioma.

Construction of a novel radiosensitivity- and ferroptosis-associated gene signature for prognosis prediction in gliomas.

Background: Gliomas are the most refractory intracranial disease characterized by high incidence and mortality rates. Therefore, radiotherapy plays a crucial role in the treatment of gliomas. However, recent evidence reveals that ferroptosis is highly associated with radiosensitivity in tumor cells. Therefore, this study aimed to investigate radiosensitivity- and ferroptosis-associated biomarkers. Moreover, the study aimed to provide new strategies for the treatment and evaluation of prognosis in gliomas.Methods: The mRNA sequencing and relevant clinical data were obtained from The Cancer Genome Atlas (TCGA). Secondly, differential analysis was conducted to reveal the radiosensitivity- and ferroptosis-associated differentially expressed genes (DEGs). Further, a predictive model based on the seven genes was constructed, and LASSO regression analysis was carried out. After that, the Chinese Glioma Genome Atlas (CGGA) was used for validation of the results.Results: A total of 36 radiosensitivity- and 19 ferroptosis-associated DEGs with a prognostic value were identified. Moreover, seven intersecting genes (HSPB1, STAT3, CA9, MAP1LC3A, MAPK1, ZEB1, and TNFAIP3) were identified as the risk signature genes. The ROC curves and K-M analysis revealed that the signature genes showed a good survival prediction. Furthermore, the functional analysis revealed that the differentially expressed genes between the high-risk and the low-risk groups were enriched in glioma-related biological processes. In addition, differences were reported in immune function status between the two groups.Conclusion: This study revealed that the seven biomarkers could help predict the prognosis in glioma patients. In addition, this study provides a basis for understanding the molecular mechanisms of radiosensitivity and ferroptosis in the treatment of gliomas.

Development and Validation of a Prediction Model of Overall Survival in High-Risk Neuroblastoma Using Mechanistic Modeling of Metastasis.

Prognosis of high-risk neuroblastoma (HRNB) remains poor despite multimodal therapies. Better prediction of survival could help to refine patient stratification and better tailor treatments. We established a mechanistic model of metastasis in HRNB relying on two processes: growth and dissemination relying on two patient-specific parameters: the dissemination rate μ and the minimal visible lesion size Svis. This model was calibrated using diagnosis values of primary tumor size, lactate dehydrogenase circulating levels, and the meta-iodobenzylguanidine International Society for Paediatric Oncology European (SIOPEN) score from nuclear imaging, using data from 49 metastatic patients. It was able to describe the data of total tumor mass (lactate dehydrogenase, R2 > 0.99) and number of visible metastases (SIOPEN, R2 = 0.96). A prediction model of overall survival (OS) was then developed using Cox regression. Clinical variables alone were not able to generate a model with sufficient OS prognosis ability (P = .507). The parameter μ was found to be independent of the clinical variables and positively associated with OS (P = .0739 in multivariable analysis). Critically, addition of this computational biomarker significantly improved prediction of OS with a concordance index increasing from 0.675 (95% CI, 0.663 to 0.688) to 0.733 (95% CI, 0.722 to 0.744, P < .0001), resulting in significant OS prognosis ability (P = .0422).

Derivation and Comprehensive Analysis of Aging Patterns in Patients with Bladder Cancer.

Background Aging is an essential risk factor for cancer. However, aging-related genes (ARGs) have not been comprehensively analyzed in bladder cancer (BC). Therefore, the study is aimed at derivating a risk stratification system for BC patients based on ARGs. Methods Public databases were used to acquire ARGs sets, transcriptome files, and clinical data. The “limma” package was then used to screen for differential ARGs while also using univariate Cox regression analysis to explore for prognostic ARGs. The “ConsensusClusterPlus” package was used to perform aging patterns in BC patients based on the above prognostic ARGs. Subsequently, aging patterns were investigated in survival prediction, mutation landscape, immunotherapy, immunological checkpoints, and immune microenvironment. We likewise utilized gene enrichment analysis to explore the biological functions that were behind the findings. To construct a risk signature and nonogram for prognostic prediction, we used LASSO and Cox regression analysis based on differential genes in aging patterns. In addition, we plotted a nomogram and validate the accuracy of the risk signature in GEO and TCGA cohorts. We explored the possible biological mechanism using GSEA analysis and preliminarily identified a hub gene using PPI network. Finally, we validated the expression of hub gene in BC cell lines. Results We screened 84 downregulated ARGs, 74 upregulated ARGs, and 32 prognostic ARGs in the human aging genome resource. The aging patterns based on prognostic genes had excellent survival prediction (p < 0.001) and discriminatory ability in 405 BC patients. In addition, we found no significant differences in aging patterns in mutation analysis, which were all characterized by TP53, TTN, and KMT2D mutations. It is worth noting that cluster B in the aging patterns has a better response to immunotherapy and a more active immune microenvironment (p < 0.05). In addition, gene enrichment analysis showed that aging patterns may be related to biological processes such as Staphylococcus aureus infection, phagosome, and cytokine-cytokine receptor interaction. Subsequently, we constructed a risk signature based on 16 differential genes from different aging patterns and had good survival prediction ability in both GEO and TCGA cohort. Specifically, survival analysis revealed a significantly shorter survival time in the high-risk group than in the low-risk group (TCGA and GEO, p < 0.001). In addition, AUC values in the ROC analysis predicted 1, 3, and 5 years in TCGA cohort that are 0.713, 0.714, and 0.738, respectively. AUC values predicted 1, 3, and 5 years in GEO cohort that are 0.606, 0.663, and 0.718, respectively. There is no doubt that risk score was an independent prognostic factor from results of multivariate Cox regression analysis in BC patients (p < 0.001). There were also significant differences in immune cell infiltration, immune checkpoint, and immune score between the two groups (p < 0.05), but it should not be ignored that the correlation with the HLA expression was weak. Finally, we identified and validated CLIC3 as a hub gene that may be involved in the Wnt signaling pathway, etc. Conclusion We provided robust evidences that aging patterns based on ARGs can guide targeted therapy and survival prediction in BC patients.

Geriatric Nutrition Risk Index: Prognostic factor related to inflammation in elderly patients with cancer cachexia.

BackgroundSystemic inflammation and cachexia are associated with adverse clinical outcomes in elderly patients with cancer. The Geriatric Nutritional Risk Index (GNRI) is a simple and useful tool to assess these conditions, but its predictive ability for elderly patients with cancer cachexia (EPCC) is unknown.MethodsThis multicentre cohort study included 746 EPCC with an average age of 72.00 ± 5.24 years, of whom 489 (65.5%) were male. The patients were divided into two groups (high GNRI group ≥91.959 vs. low GNRI group <91.959) according to the optimal cut‐off value of the ROC curve. The calibration curves were performed to analyse the prognostic, predictive ability of GNRI. Comprehensive survival analyses were utilized to explore the relationship between GNRI and the overall survival (OS) of EPCC. Interaction analysis was used to investigate the comprehensive effects of low GNRI and subgroup parameters on the OS of EPCC.ResultsIn this study, a total of 2560 patients were diagnosed with cancer cachexia, including 746 cases of EPCC. During the 3.6 year median follow‐up, we observed 403 deaths. The overall mortality rate for EPCC at 12 months was 34.3% (95% CI: 62.3% to 69.2%), and resulting in rate of 278 events per 1000 patient‐years. The GNRI score of EPCC was significantly lower than those of young patients with cancer cachexia (P < 0.001). The 1, 3, and 5 year calibration curves showed that the GNRI score had good survival prediction in the OS of EPCC. The GNRI could predict the OS of EPCC, whether as a continuous variable or a categorical variable. Particularly, we also found that low GNRI score (<91.959) of EPCC had a worse prognosis than those with a high GNRI score (≥91.959, P = 0.001, HR = 1.728, 95% CI: 1.244–2.401). Consistent results were observed in the tumour subgroups of gastric cancer and colorectal cancer. Notably, similar results were observed in the sensitivity analysis. In the subgroup analysis, the low GNRI has a combined effect with age (<70 years) on poor OS of EPCC. The results of the prognostic risk model found that the lower the GNRI score, the greater the prognostic risk score, and the greater the risk of death in EPCC.ConclusionsFor the first time, this study found that the GNRI score can serve as an independent prognostic factor for the OS of EPCC.