Risk Stratification for Diffuse Large B-Cell Lymphoma by Integrating Interim Evaluation and International Prognostic Index: A Multicenter Retrospective Study

The baseline International Prognostic Index (IPI) is not sufficient for the initial risk stratification of patients with diffuse large B-cell lymphoma (DLBCL) treated with R‐CHOP (rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone). The aims of this study were to evaluate the prognostic relevance of early risk stratification in DLBCL and develop a new stratification system that combines an interim evaluation and IPI. This multicenter retrospective study enrolled 314 newly diagnosed DLBCL patients with baseline and interim evaluations. All patients were treated with R-CHOP or R-CHOP-like regimens as the first-line therapy. Survival differences were evaluated for different risk stratification systems including the IPI, interim evaluation, and the combined system. When stratified by IPI, the high-intermediate and high-risk groups presented overlapping survival curves with no significant differences, and the high-risk group still had >50% of 3-year overall survival (OS). The interim evaluation can also stratify patients into three groups, as 3-year OS and progression-free survival (PFS) rates in patients with stable disease (SD) and progressive disease (PD) were not significantly different. The SD and PD patients had significantly lower 3-year OS and PFS rates than complete remission and partial response patients, but the percentage of these patients was only ~10%. The IPI and interim evaluation combined risk stratification system separated the patients into low-, intermediate-, high-, and very high-risk groups. The 3-year OS rates were 96.4%, 86.7%, 46.4%, and 40%, while the 3-year PFS rates were 87.1%, 71.5%, 42.5%, and 7.2%. The OS comparison between the high-risk group and very high-risk group was marginally significant, and OS and PFS comparisons between any other two groups were significantly different. This combined risk stratification system could be a useful tool for the prognostic prediction of DLBCL patients.

Development of a new risk stratification system for patients with newly diagnosed multiple myeloma using R-ISS and 18F-FDG PET/CT

In multiple myeloma (MM), a high number of focal lesions (FL) detected using positron emission tomography/computed tomography (PET/CT) was found to be associated with adverse prognosis. To design a new risk stratification system that combines the Revised International Staging System (R-ISS) with FL, we analyzed the data of 380 patients with newly diagnosed MM (NDMM) who underwent 18F-fluorodeoxyglucose (18F-FDG) PET/CT upon diagnosis. The K-adaptive partitioning algorithm was adopted to define subgroups with homogeneous survival. The combined R-ISS with PET/CT classified NDMM patients into four groups: R-ISS/PET stage I (n = 31; R-ISS I with FL ≤ 3), stage II (n = 156; R-ISS I with FL > 3 and R-ISS II with FL ≤ 3), stage III (n = 162; R-ISS II with FL > 3 and R-ISS III with FL ≤ 3), and stage IV (n = 31; R-ISS III with FL > 3). The 2-year overall survival rates for stages I, II, III, and IV were 96.7%, 89.8%, 74.7%, and 50.3%. The 2-year progression-free survival rates were 84.1%, 64.7%, 40.8%, and 17.1%, respectively. The new R-ISS/PET was successfully validated in an external cohort. This new system had a remarkable prognostic power for estimating the survival outcomes of patients with NDMM. This system helps discriminate patients with a good prognosis from those with a poor prognosis more precisely.

Development and validation of a prognostic nomogram for predicting overall survival in patients with primary bladder sarcoma: a SEER-based retrospective study

Background Primary bladder sarcoma (PBS) is a rare malignant tumor of the bladder with a poor prognosis, and its disease course is inadequately understood. Therefore, our study aimed to establish a prognostic model to determine individualized prognosis of patients with PBS. Patients and Methods Data of 866 patients with PBS, registered from 1973 to 2015, were extracted from the surveillance, epidemiology, and end result (SEER) database. The patients included were randomly split into a training (n = 608) and a validation set (n = 258). Univariate and multivariate Cox regression analyses were employed to identify the important independent prognostic factors. A nomogram was then established to predict overall survival (OS). Using calibration curves, receiver operating characteristic curves, concordance index (C-index), decision curve analysis (DCA), net reclassification improvement (NRI) and integrated discrimination improvement (IDI), the performance of the nomogram was internally validated. We compared the nomogram with the TNM staging system. The application of the risk stratification system was tested using Kaplan–Meier survival analysis. Results Age at diagnosis, T-stage, N-stage, M-stage, and tumor size were identified as independent predictors of OS. C-index of the training cohort were 0.675, 0.670, 0.671 for 1-, 3- and 5-year OS, respectively. And that in the validation cohort were 0.701, 0.684, 0.679, respectively. Calibration curves also showed great prediction accuracy. In comparison with TNM staging system, improved net benefits in DCA, evaluated NRI and IDI were obtained. The risk stratification system can significantly distinguish the patients with different survival risk. Conclusion A prognostic nomogram was developed and validated in the present study to predict the prognosis of the PBS patients. It may assist clinicians in evaluating the risk factors of patients and formulating an optimal individualized treatment strategy.

A Stratification System of Ferroptosis and Iron-Metabolism Related Lncrnas Guides the Prediction of The Survival of Patients with Esophageal Squamous Cell Carcinoma

Background: As the major type of esophageal cancer (ESCA), esophageal squamous cell carcinoma (ESCC) is also related to the highest malignant level and low survival rates across the world. Increasing people recognize long non-coding RNAs (lncRNAs) as significant mediators in regulating ferroptosis and iron-metabolism. Determining the prognostic value of ferroptosis and iron-metabolism related lncRNAs (FIRLs) in ESCC is thus critical. Methods: Pearson’s correlation analysis was carried out between ferroptosis and iron-metabolism-related genes (FIRGs) and all lncRNAs to derive the FIRLs. Based on weighted gene co-expression network exploration (WCGNA), least absolute shrinkage and selection operator (LASSO) regression and Cox regression analysis, a risk stratification system was established. According to Kaplan-Meier analysis, receiver operating characteristic (ROC) curve analysis, and univariate and multivariate Cox regression analyses, the predictive ability and clinical relevance of the risk stratification system were evaluated. The validity of the established prognostic signature was further examined in TCGA (training set) and GEO (validation set) cohorts. A nomogram with enhanced precision for forecasting OS was set up on basis of the independent prognostic elements. Gene Ontology (GO) term enrichment analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis were applied for the identification of pathways in which FIRLs significantly enriched. we used cell culture, transfection, CCK-8, and qRT-PCR as in vitro assays. Results: An 3-FIRLs risk stratification system was developed by multivariate Cox regression analysis to divide patients into two risk groups. Patients in the high-risk group had worse prognosis than patients in the low-risk group. Multivariate Cox regression analysis showed the risk stratification system was an independent prognostic indicator. Receiver operating characteristic curve (ROC) analysis proved the predictive accuracy of the signature. The area under time-dependent ROC curve (AUC) reached 0.853 at 1 year, 0.802 at 2 years, 0.740 at 5 years in the training cohort and 0.712 at 1 year, 0.822 at 2 years, 0.883 at 5 years in the validation cohort. Functional enrichment analysis predicted potential associations of 49 possible upstream regulated FIRGs with ferroptosis and iron-metabolism processes and oncological signatures. Analysis of the immune cell infiltration landscape showed that ESCC in the high-risk group tended be immunologically “cold”. In vitro experiments suggested that LINC01068 promoted ESCC cell proliferation. Conclusion: The risk stratification system based on FIRLs could serve as a reliable tool for forecasting the survival of patients with ESCC.

Does Ethnicity Matter in Multiple Myeloma Risk Prediction in the Era of Genomics and Novel Agents? Evidence From Real-World Data

IntroductionCurrent risk predictors of multiple myeloma do not integrate ethnicity-specific information. However, the impact of ethnicity on disease biology cannot be overlooked. In this study, we have investigated the impact of ethnicity in multiple myeloma risk prediction. In addition, an efficient and robust artificial intelligence (AI)-enabled risk-stratification system is developed for newly diagnosed multiple myeloma (NDMM) patients that utilizes ethnicity-specific cutoffs of key prognostic parameters.MethodsK-adaptive partitioning is used to propose new cutoffs of parameters for two different datasets—the MMIn (MM Indian dataset) dataset and the MMRF (Multiple Myeloma Research Foundation) dataset belonging to two different ethnicities. The Consensus-based Risk-Stratification System (CRSS) is designed using the Gaussian mixture model (GMM) and agglomerative clustering. CRSS is validated via Cox hazard proportional methods, Kaplan–Meier analysis, and log-rank tests on progression-free survival (PFS) and overall survival (OS). SHAP (SHapley Additive exPlanations) is utilized to establish the biological relevance of the risk prediction by CRSS.ResultsThere is a significant variation in the key prognostic parameters of the two datasets belonging to two different ethnicities. CRSS demonstrates superior performance as compared with the R-ISS in terms of C-index and hazard ratios on both the MMIn and MMRF datasets. An online calculator has been built that can predict the risk stage of a multiple myeloma (MM) patient based on the values of parameters and ethnicity.ConclusionOur methodology discovers changes in the cutoffs with ethnicities from the established cutoffs of prognostic features. The best predictor model for both cohorts was obtained with the new ethnicity-specific cutoffs of clinical parameters. Our study also revealed the efficacy of AI in building a deployable risk prediction system for MM. In the future, it is suggested to use the CRSS risk calculator on a large dataset as the cohort size of the present study is 25% of the cohort used in the R-ISS reported in 2015.

Development of a New Risk Stratification System for Patients with Newly Diagnosed Multiple Myeloma Using R-ISS and 18F-FDG PET/CT

Background A high number of focal lesions (FL) detected using PET/CT at diagnosis were found to be associated with adverse prognosis along with Revised International Staging System (R-ISS). In present study, we combined R-ISS with FL using PET/CT to design a reliable and easily applicable risk stratification system in patients with newly diagnosed MM (NDMM). Methods In training cohort, the data of 380 patients with NDMM who underwent 18F-fluorodeoxyglucose (18F-FDG) PET/CT upon diagnosis from 10 hospitals of the Korean Multiple Myeloma Working Party were retrospectively analyzed. All patients were classified by R-ISS and were treated by frontline therapy with proteasome inhibitors (PI) and/or immunomodulatory drugs (IMiD). The K-adaptive partitioning algorithm was adopted to develop the new risk groups with homogeneous survival. Sixty-seven patients in external validation cohort were additionally collected to confirm reproducibility of the new risk groups. Results In the training cohort, 199 patients (52.4%) showed FL > 3 using PET/CT at diagnosis. R-ISS stages I, II, and III were 78 patients (20.5%), 230 (60.5%), and 72 (18.9%), respectively. The combined R-ISS with PET/CT newly allocated NDMM patients into four groups: R-ISS/PET stage I (n=30; R-ISS I with FL≤3), stage II (n=149; R-ISS I with FL>3 and R-ISS II with FL≤3), stage III (n=166; R-ISS II with FL>3 and R-ISS III with FL≤3), and stage IV (n=35; R-ISS III with FL>3). The new R-ISS/PET showed significantly pronounced survival differences according to stages. Two-year overall survival (OS) rates were 96.6%, 89.5%, 75.0%, and 57.9% (p < 0.001), and 2-year progression-free survival (PFS) rates were 86.9%, 65.1%, 41.9%, and 15.2% (p < 0.001) in stages I, II, III, and IV, respectively. The prognostic role of the R-ISS/PET for survival outcomes was also confirmed in different subgroups classified by transplant eligibility and by types of treatments. In the external validation cohort, the new R-ISS/PET was successfully implemented. Two-year OS rates for were 100%, 89.9%, 82.6%, and 42.0% for R-ISS/PET I, II, III, and IV, respectively (p = 0.001). PFS rates at 2 years for each R-ISS/PET were 100%, 74.5%, 57.9%, and 25.6%, respectively (p = 0.004). In the multivariate Cox analysis for survival outcome, R-ISS/PET was a significant factor and could predict long-term outcomes with regard to OS: stage II vs. I (HR 2.50, p = 0.215), (ii) stage III vs. I (HR 5.11, p = 0.025), and (iii) stage IV vs. I (HR 10.3, p = 0.003) and PFS: (i) stage II vs. I (HR 2.21, p = 0.005), (ii) stage III vs. I (HR 4.57, p < 0.001), and (iii) stage IV vs. I (HR 9.48, p < 0.001). Conclusion The new R-ISS/PET had a remarkable prognostic power for estimating the survival outcomes of patients with NDMM. This system helps discriminate patients with a good prognosis from those with a poor prognosis more precisely. Thus, R-ISS/PET is applicable for identifying heterogeneous manifestation of clinical MM. Figure 1 Figure 1. Disclosures No relevant conflicts of interest to declare.