Predictor Of Overall Survival In Patients Research Articles

Establishment and validation of a nomogram for predicting overall survival in patients with vulvar carcinoma based on the SEER database

Background: Vulvar carcinoma (VC) is a rare female gynecological malignancy, and optimizing prognostic factors for VC requires large-scale research containing various clinical indicators of patients. Our study attempted to develop and validate a detailed survival nomogram for predicting the overall survival (OS) probability in patients diagnosed with VC. Methods: Patients diagnosed with VC between 2004 and 2015 were extracted from the Surveillance, Epidemiology, and End Results (SEER) database. Univariate and multivariate Cox regression analyses were performed followed by the construction of the nomogram for OS. The performance of this model was evaluated using the concordance index (C-index), area under the time-dependent receiver operating characteristics curve (AUC), net reclassification improvement (NRI), integrated discrimination improvement (IDI), calibration plots and decision curve analysis (DCA). In addition, the C-index, AUC and DCA of the model and the 2009 International Federation of Gynecology and Obstetrics (FIGO) staging system were compared. Results: A total of 6275 patients were randomly assigned to the training cohort (n=4392) and the validation cohort (n=1883). Multivariate analysis identified independent prognostic factors (p<0.05) for OS, including histological type, age, surgery, T stage, N stage, M stage, grade, summary stage, chemotherapy, race, marital status and size. Finally, a nomogram was constructed to predict the 3-, 5-, and 8-year OS probabilities for patients with VC, and the C-index, NRI, IDI and calibration plotting all showed that the model has good discrimination. Additionally, the nomogram also showed better clinical validity of the DCA and AUC compared to that of the FIGO system. Conclusions: We developed and validated a nomogram for individual OS prediction in patients with VC. While further validation is required, this nomogram may be a useful comprehensive prognostic tool to give patients a better idea of prognosis during counseling.

Fully Automatic Calibration of Tumor-Growth Models Using a Single mpMRI Scan.

Our objective is the calibration of mathematical tumor growth models from a single multiparametric scan. The target problem is the analysis of preoperative Glioblastoma (GBM) scans. To this end, we present a fully automatic tumor-growth calibration methodology that integrates a single-species reaction-diffusion partial differential equation (PDE) model for tumor progression with multiparametric Magnetic Resonance Imaging (mpMRI) scans to robustly extract patient specific biomarkers i.e., estimates for (i) the tumor cell proliferation rate, (ii) the tumor cell migration rate, and (iii) the original, localized site(s) of tumor initiation. Our method is based on a sparse reconstruction algorithm for the tumor initial location (TIL). This problem is particularly challenging due to nonlinearity, ill-posedeness, and ill conditioning. We propose a coarse-to-fine multi-resolution continuation scheme with parameter decomposition to stabilize the inversion. We demonstrate robustness and practicality of our method by applying the proposed method to clinical data of 206 GBM patients. We analyze the extracted biomarkers and relate tumor origin with patient overall survival by mapping the former into a common atlas space. We present preliminary results that suggest improved accuracy for prediction of patient overall survival when a set of imaging features is augmented with estimated biophysical parameters. All extracted features, tumor initial positions, and biophysical growth parameters are made publicly available for further analysis. To our knowledge, this is the first fully automatic scheme that can handle multifocal tumors and can localize the TIL to a few millimeters.

Preoperative platelet distribution width-to-platelet count ratio as a prognostic factor in patients with glioblastoma multiforme

Backgrounds The prognostic significance of the platelet volume indices (PVIs), including the platelet distribution width (PDW), mean platelet volume (MPV), and platelet distribution width-to-platelet count ratio (PDW/P) has been demonstrated in a variety of malignancies. This study aimed to evaluate the prognostic value of PVIs in patients with a newly diagnosed glioblastoma multiforme (GBM). Methods We retrospectively evaluated the clinical data of 143 patients with GBM who managed at our center between May 2010 and May 2019. Receiver operating characteristic curves (ROC) for cutoff value determination, Kaplan–Meier survival analysis, and univariate and multivariate Cox regression analyses were performed. Results The corresponding cutoff values for MPV, PDW, and PDW/P were 9.05, 14.7, and 0.51, respectively. The Kaplan-Meier survival analyses showed that patients with an MPV < 9.05 and those with PDW <14.7 and cases with PDW/p < 0.51 had a longer overall survival (OS) (p < 0.05). Based on univariate analysis, age, Karnofsky Performance Status scores (KPS), tumor focality, MPV, PDW, and PDW/P were predictors of OS (p < 0.05). Final multivariate Cox regression analyses showed age (HR 1.040, 95% CI 1.009–1.071, P,0.011), KPS (HR 2.208, 95% CI 1.107–4.405, P,0.025), tumor focality (HR 4.596, 95% CI 1.988–10.626, p < 0.001), and PDW/P (HR 1.786, 95% CI 1.103–3.072, P,0.037) as the independent predictors of OS in patients with newly diagnosed glioblastoma. Conclusions Our results suggest an elevated preoperative PDW/P, along with previously established variables, as a simple and inexpensive prognostic factor for patients with GBM.

Retrospective survival analysis of stage II-III rectal cancer: tumour regression grade, grading and lymphovascular invasion are the only predictors.

Tumour regression grade is gaining interest as a prognostic factor of patients undergoing neoadjuvant chemoradiotherapy and surgery for locally advanced rectal cancer. A series of 68 consecutive patients with locally advanced rectal cancer treated by neoadjuvant chemoradiotherapy and surgery between 2010 and 2016 was retrospectively studied. The impact on disease-free survival (DFS) and overall survival (OS) of several criteria was analysed. Univariate analysis was performed through Kaplan-Meier statistics. Multivariate analysis was performed through Cox regression model. Using criteria found to be related to long-term outcomes, a predictive model of patient's OS was calculated. Poor tumour regression grade - TRG3 (P = 0.010), poor grading - G3 (P = 0.001) and lymphovascular invasion (LVI; P = 0.030) were associated with short OS at univariate analysis. OS was associated with TRG3 and G3 at multivariate analysis (P = 0.016 and P = 0.027, respectively). DFS was associated with LVI (P = 0.001), G3 tumours (P = 0.046) and TRG3 (P = 0.045) at univariate analysis. At multivariate analysis, only LVI was associated with DFS (P = 0.041). A score, pondering the impact of three parameters (2 points for TRG3, 2 for G3 and 1 for LVI), was created and resulted to predict patient OS (P = 0.008), ranging from 94.5 months (score = 0-1) to 32 months (score = 3-5). TRG3 and G3 were associated with poor OS, and LVI was the most significant predictor of DFS. An easy-to-use score may allow for a more accurate prediction of OS.

PATH-13. PLEOMORPHIC XANTHOASTROCYTOMA INTEGRATED GENOMIC CHARACTERIZATION - WHAT HAVE WE LEARNED?

Pleomorphic xanthoastrocytoma (PXA) is a rare astrocytoma occurring predominantly in children and young adults. It is characterized histologically by large pleomorphic, spindled and lipidized cells with frequent eosinophilic granular bodies and pericellular reticulin deposition. BRAF p.V600E mutation and CKDN2A/B deletion are the most common genetic alterations. We report the integrated genomic characterization of a cohort of 67 patients (37 F, 30 M; median age 20.3 years (interquartile 13.4–32.9) with histologically defined PXA (52, 78%) or anaplastic PXA (A-PXA) (15, 22%), using genome-wide cytogenetic (ThermoFisher Oncoscan, n=67), methylation profiling (Illumina EPIC array, n=43), and targeted next generation sequencing (n=32). BRAF p.V600E mutation (n=51, 76.1%) and CDKN2A/B deletion (n=63; 94%) were the most frequent alterations. Of 16 BRAF p.V600E negative cases, 7 showed an alternative BRAF activating mutation (n=2), NF1 (n=3) mutation or ATG7-RAF1 fusion (n=2). Targeted TERT analysis found promoter mutations in 3 (of 58) cases, but TERT amplification was absent. Supervised and unsupervised methylation profiling against a comprehensive reference cohort demonstrated consensus grouping with the PXA class in 36 of 43 cases; while the minority grouped with a ganglioglioma class (n=3), with reactive brain or had no resolvable subgroup (n=4). Follow-up was available in 61 patients (91.0%) (median 63 months). Overall survival was significantly different between PXA and A-PXA (5-year:80.4% vs. 55.1%; p=0.001), but not progression-free survival (5-year:61.7% vs. 39.8%; p=0.128). Our data confirm the high frequency of MAP-K abnormalities and CDKN2A/B deletion in PXA. WHO grade remains a strong predictor of patient overall survival.

Association between number of dissected lymph nodes and survival in stage IA non-small cell lung cancer: a propensity score matching analysis

BackgroundFor patients with stage IA non-small cell lung cancer (NSCLC) with tumor size ≤ 2 cm, the prognostic significance of the number of removed lymph nodes (NLNs) through different surgical methods remains unclear. To determine the association of NLNs with cancer-specific survival (CSS) and overall survival (OS) in patients with stage IA NSCLC with tumor size ≤ 2 cm who underwent different lung surgeries.MethodsWe retrospectively enrolled 7293 patients from the Surveillance, Epidemiology and End Results database. Median NLNs was used to classify the patients into two groups: group A with NLNs ≤ 5 and group B with NLNs > 5. Propensity score matching (PSM) was performed to decrease selection bias. Kaplan–Meier analysis and Cox regression analysis were performed to identify the association between NLNs and survival outcomes.ResultsGroup B had better survival than group A in the unmatched cohort and matched cohort (all P < 0.05). Multivariable analyses revealed that the NLNs significantly affected CSS and OS of eligible cases in the unmatched cohort and matched cohort. Additionally, we found that the NLNs was a protective prognostic predictor of OS for patients who underwent wedge resection, segmental resection, or lobectomy.ConclusionThe NLNs was a protective prognostic factor in NSCLC patients with tumor size ≤ 2 cm. We demonstrated that patients with > 5 NLNs in the cohort of wedge resection, segmental resection, or lobectomy exhibited a significantly better OS.

NIMG-09. PREDICTING OVERALL SURVIVAL OF GLIOBLASTOMA PATIENTS ON MULTI-INSTITUTIONAL HISTOPATHOLOGY STAINED SLIDES USING DEEP LEARNING AND POPULATION-BASED NORMALIZATION

Abstract MOTIVATION Glioblastoma is the most common and aggressive adult brain tumor. Clinical histopathologic evaluation is essential for tumor classification, which according to the World Health Organization is associated with prognostic information. Accurate prediction of patient overall survival (OS) from clinical routine baseline histopathology whole slide images (WSI) using advanced computational methods, while considering variations in the staining process, could contribute to clinical decision-making and patient management optimization. METHODS We utilize The Cancer Genome Atlas glioblastoma (TCGA-GBM) collection, comprising multi-institutional hematoxylin and eosin (H&amp;E) stained frozen top-section WSI, genomic, and clinical data from 121 subjects. Data are randomly split into training (80%), validation (10%), and testing (10%) sets, while proportionally keeping the ratio of censored patients. We propose a novel deep learning algorithm to identify survival-discriminative histopathological patterns in a WSI, through feature maps, and quantitatively integrate them with gene expression and clinical data to predict patient OS. The concordance index (C-index) is used to quantify the predictive OS performance. Variations in slide staining are assessed through a novel population-based stain normalization approach, informed of glioblastoma distinct histologic sub-regions and their appearance from 509 H&amp;E stained slides with corresponding anatomical annotations from the Ivy Glioblastoma Atlas Project (IvyGAP). RESULTS C-index was equal to 0.797, 0.713, and 0.703 for the training, validation, and testing data, respectively, prior to stain normalization. Following normalization, staining variations in H&amp;E and ‘E’ gained significant improvements in IvyGAP (pWilcoxon&amp;lt; 0.01) and TCGA-GBM (pWilcoxon&amp;lt; 0.0001) data, respectively. These improvements contributed to further optimizing the C-index to 0.871, 0.777, and 0.780 for the training, validation, and testing data, respectively. CONCLUSIONS Appropriate normalization and integrative deep learning yield accurate OS prediction of glioblastoma patients through H&amp;E slides, generalizable in multi-institutional data, potentially contributing to patient stratification in clinical trials. Our computationally-identified survival-discriminative histopathological patterns can contribute in further understanding glioblastoma.

Utility of Radiomics for Predicting Patient Survival in Hepatocellular Carcinoma With Portal Vein Tumor Thrombosis Treated With Stereotactic Body Radiotherapy.

Introduction: This study aimed to develop and validate the combination of radiomic features and clinical characteristics that can predict patient survival in hepatocellular carcinoma (HCC) with portal vein tumor thrombosis (PVTT) treated with stereotactic body radiotherapy (SBRT).Materials and Methods: The prediction model was developed in a primary cohort of 70 patients with HCC and PVTT treated with SBRT, using data acquired between December 2015 and June 2017. The radiomic features were extracted from computed tomography (CT) scans. A least absolute shrinkage and selection operator regression model was used to build the model. Multivariate Cox-regression hazard models were created for analyzing survival outcomes and the radiomic features and clinical characteristics were presented with a nomogram. The area under the receiver operating characteristic curve (AUROC) was used to evaluate the model. Participants were divided into a high-risk group and a low-risk group based on the radiomic features.Results: A total of four radiomic features and six clinical characteristics were extracted for survival analysis. A combination of the radiomic features and clinical characteristics resulted in better performance for the estimation of overall survival (OS) [area under the curve (AUC) = 0.859 (CI: 0.770–0.948)] than that with clinical characteristics alone [AUC = 0.761 (CI: 0.641–0.881)]. These patients were divided into high-risk and low-risk groups according to the radiomic features.Conclusion: This study demonstrated that a nomogram of combined radiomic features and clinical characteristics can be conveniently used to assess individualized preoperative prediction of OS in patients with HCC with PVTT before SBRT.

Low Skeletal Muscle Mass Is a Risk Factor for Aspiration Pneumonia During Chemoradiotherapy.

This study aimed to investigate whether pretreatment skeletal muscle mass index (SMI) is a predictor for the risk of aspiration pneumonia and to explore the relationship between low SMI and overall survival (OS) in patients with head and neck squamous cell carcinoma (HNSCC) receiving chemoradiotherapy (CRT). We retrospectively reviewed the data of patients with HNSCC who received CRT during 2010-2019. Patients received a combination of radiotherapy and cisplatin-based chemotherapy (3 cycles of 80 mg/m2 cisplatin on days 1, 22, and 43). Aspiration pneumonia were defined as the presence of both subjective and objective symptoms. Kaplan-Meier curves were generated to analyze survival. Among the 159 patients, 36 (22.6%) developed aspiration pneumonia during treatment. Median SMI in patients with and without pneumonia was 12.4 cm2 /m2 (9.0-20.7) and 13.6 cm2 /m2 (8.1-19.7), respectively (P < .01). Multivariate logistic regression revealed that SMI was the only independent predictor of aspiration pneumonia (P = .0026). Mean OS was significantly shorter for patients with low SMI than for patients with normal SMI (66.9 months vs. 92.7 months, P = .001). Pretreatment low SMI predicts development of aspiration pneumonia and is a strong negative prognostic predictor for OS in patients with HNSCC undergoing CRT. Supportive treatment can be provided to patients at high risk of a low SMI. This study is the first to report SMI as a prognostic predictor in HNSCC. Laryngoscope, 131:E1524-E1529, 2021.

Integrating baseline MR imaging biomarkers into BCLC and CLIP improves overall survival prediction of patients with hepatocellular carcinoma (HCC).

We aimed to evaluate the independent predictive role of baseline imaging biomarkers for overall survival (OS) and transplant-free survival (TFS) in patients with HCC and assess the incremental value of these biomarkers to current staging systems. In this retrospective IRB approved study, the clinical, laboratory, and imaging parameters of 304 HCC patients were collected. Cox regression model was utilized to identify the potential predictors of survival. Recursive partitioning test was utilized to identify the optimal ADC cutoff for stratifying patients' OS. Patients were stratified based on Barcelona Clinic Liver Cancer (BCLC) and Cancer of the Liver Italian Program (CLIP). Binary ADC value (above vs. below the cutoff) and tumor margin (well- vs. ill-defined) were integrated into BCLC and CLIP. OS and TFS was compared for patients based on standard criteria with and without imaging biomarkers. At baseline, patients with low tumor ADC and well-defined tumor margin (favorable imaging biomarkers) had longer survival, as compared to those with high ADC and ill-defined tumor margin (unfavorable imaging biomarkers) (median OS of 43months vs. 7months, respectively) (p < 0.001). Tumor ADC and tumor margin remained strong independent predictors of survival after adjustment for demographics, BCLC and CLIP staging, and tumor burden. Incorporating ADC and tumor margin improved performance of OS prediction by 9% in BCLC group and 6% in CLIP group. Incorporating ADC and tumor margin to current staging systems for HCC significantly improve prediction of OS and TFS of these criteria. • ADC and tumor margin are predictors of overall survival in HCC patients, independent of clinical, laboratory, and other imaging variables. • Adding ADC and tumor margin improved the prognostic value of BCLC and CLIP criteria by 9% and 6%, respectively. • High ADC and ill-defined tumor margin at baseline predicted poor survival, regardless of patient's liver function and general health status.

Endothelial activation and stress index (EASIX) is a reliable predictor for overall survival in patients with multiple myeloma

BackgroundRecently, the endothelial Activation and Stress Index (EASIX) score has been reported to predict overall survival (OS) after allogeneic stem cell transplantation. This study evaluated the prognostic role of EASIX score in patients with newly diagnosed multiple myeloma (MM).MethodsThis retrospective study analyzed the records of 1177 patients with newly diagnosed MM between February 2003 and December 2017 from three institutions in the Republic of Korea. Serum lactate dehydrogenase (LDH), creatinine, and platelet count at diagnosis were measured in all included patients. EASIX scores were calculated using the formula-LDH (U/L) × Creatinine (mg/dL) / platelet count (109/L) and were evaluated based on log2 transformed values.ResultsThe median age of patients was 63 years (range, 22–92), and 495 patients (42.1%) underwent autologous stem cell transplantation (ASCT). The median log2 EASIX score at diagnosis was 1.1 (IQR 0.3–2.3). Using maximally selected log-rank statistics, the optimal EASIX cutoff value for OS was 1.87 on the log2 scale (95% CI 0.562–0.619, p < 0.001). After median follow-up for 50.0 months (range, 0.3–184.1), the median OS was 58.2 months (95% CI 53.644–62.674). Overall, 372 patients (31.6%) showed high EASIX scores at diagnosis, and had significantly inferior OS compared to those with low EASIX (log2 EASIX ≤1.87) (39.1 months vs. 67.2 months, p < 0.001). In multivariate Cox analysis, high EASIX was significantly associated with poor OS (HR 1.444, 95% CI 1.170–1.780, p = 0.001). In the subgroup analysis of patients who underwent ASCT, patients with high EASIX showed significantly inferior OS compared to those with low EASIX (52.8 months vs. 87.0 months, p < 0.001). In addition, in each group of ISS I, II, and III, high EASIX was associated with significantly inferior OS (ISS 1, 45.2 months vs. 76.0 months, p = 0.001; ISS 2, 42.3 months vs. 66.5 months, p = 0.002; ISS 3, 36.8 months vs. 55.1 months, p = 0.001).ConclusionEASIX score at diagnosis is a simple and strong predictor for OS in patients with newly diagnosed MM.

High Mean Corpuscular Volume as a Predictor of Poor Overall Survival in Patients with Esophageal Cancer Receiving Concurrent Chemoradiotherapy.

BackgroundIncreasing numbers of recent studies have demonstrated that high mean corpuscular volume (MCV) is a predictor of poor overall survival (OS) and therapeutic response in patients with solid tumors. The aim of the present study was to explore the association between high MCV and OS in patients with advanced esophageal cancer (EC) undergoing concurrent chemoradiotherapy.Patients and MethodsEnrolled in this study were 249 patients with advanced EC who underwent concurrent chemoradiotherapy. Pre-treatment MCV values were collected in all patients and their correlations with OS and pathophysiological characteristics were analyzed. The chi-square test was used to explore the correlation between MCV and various clinical pathophysiological characteristics, and the prognostic significance of high MCV using Kaplan–Meier curves and the Cox proportional hazards model. All P-values were two-tailed and a P-value <0.05 was considered statistically significant.ResultsAccording to ROC curve analysis, the optimal cut-off value of MCV was 93.6 fL. The mean OS was 14.7 months in all 249 EC patients, 10.9 months in patients with MCV >93.6 fL, and 18.8 months in patients with MCV <93.6 fL; the difference is statistically significant (P<0.05). Chi-square test showed that the MCV value was correlated with the N stage of the tumor and the therapeutic effect, indicating that the higher the MCV was, the higher the T stage of the tumor and the worse the therapeutic effect would be (p=0.012 and p <0.01). Multivariate analysis showed that MCV (OR = 1.864, 95% CI: 1.439–2.415) was an independent prognostic factor for OS in EC patients.ConclusionHigh MCV is a poor predictor of OS in patients with advanced EC receiving concurrent chemoradiotherapy.

High expression of EMP1 predicts a poor prognosis and correlates with immune infiltrates in bladder urothelial carcinoma.

Epithelial membrane protein 1 (EMP1) is a key gene that regulates cell proliferation and metastatic capability in various types of cancer, and serves an important role in tumor-immune interactions. However, the association between EMP1 and clinical prognosis, as well as the presence of tumor-infiltrating lymphocytes in bladder urothelial carcinoma (BLCA) remains unclear. The present study aimed to explore the relationship between EMP1 expression and tumor immune cell infiltration in BLCA. In the present study, EMP1 expression in BLCA was analyzed using the Oncomine database, The Cancer Genome Atlas (TCGA) and the Tumor Immune Estimation Resource (TIMER). The effects of EMP1 on clinical prognosis were evaluated using the Kaplan-Meier plotter and Gene Expression Profiling Interactive Analysis. The correlations between EMP1, cancer immune infiltrates and lymphocyte abundance were determined using the TIMER and Tumor immune system interaction database. In addition, correlations between EMP1 expression and gene markers in immune infiltrates were analyzed using cBioportal. The results demonstrated that, compared with adjacent normal tissues, EMP1 was downregulated in BLCA tissues. High expression of EMP1 was significantly associated with poor overall survival (OS) in BLCA cases obtained from TCGA. Multivariate Cox analysis revealed that EMP1 was an independent predictor of OS in patients with BLCA. Gene set enrichment analysis revealed that EMP1 was associated with cancer-related pathways and was positively correlated with the levels of infiltrating CD8+ T cells, macrophages, neutrophils and dendritic cells in BLCA. Further analysis demonstrated that EMP1 was significantly associated with the enrichment of multiple types of lymphocyte. EMP1 expression exhibited a strong correlation with a range of immune markers in BLCA. In conclusion, the results of the present study demonstrated that EMP1 was associated with a poor prognosis in patients with BLCA, and that the levels of immune infiltration and multiple immunomarker groups were associated with EMP1 expression. These results suggested that EMP1 may be used as a predictive biomarker to determine the prognosis and immune infiltration in BLCA.

Development and Validation of an Individualized Nomogram for Predicting Overall Survival in Patients With Typical Lung Carcinoid Tumors.

We aim to develop and validate an effective nomogram prognostic model for patients with typical lung carcinoid tumors using a large patient cohort from the Surveillance, Epidemiology, and End Results (SEER) database. Data from patients with typical lung carcinoid tumors between 2010 and 2015 were selected from the SEER database for retrospective analysis. Univariate and multivariate Cox analysis was performed to clarify independent prognostic factors. Next, a nomogram was formulated to predict the probability of 3- and 5-year overall survival (OS). Concordance indexes (c-index), receiver operating characteristic analysis and calibration curves were used to evaluate the model. The selected patients were randomly divided into a training and a validation cohort. A nomogram was established based on the training cohort. Cox analysis results indicated that age, sex, T stage, N stage, surgery, and bone metastasis were independent variables for OS. All these factors, except surgery, were included in the nomogram model for predicting 3- and 5-year OS. The internally and externally validated c-indexes were 0.787 and 0.817, respectively. For the 3-year survival prediction, receiver operating characteristic analysis showed that the areas under the curve in the training and validation cohorts were 0.824 and 0.795, respectively. For the 5-year survival prediction, the area under the curve in the training and validation cohorts were 0.812 and 0.787, respectively. The calibration plots for probability of survival were in good agreement. The nomogram brings us closer to personalized medicine and the maximization of predictive accuracy in the prediction of OS in patients with typical lung carcinoid tumors.

Risk of disease recurrence and survival in patients with multiple myeloma: A German Study Group analysis using a conditional survival approach with long-term follow-up of 815 patients.

Unlike the traditional method of overall survival prediction in patients with cancer, conditional survival predicts the survival of patients dynamically throughout the course of disease, identifying how a prognosis evolves over time. The authors assessed 815 consecutive patients with multiple myeloma through the German Study Group on Multiple Myeloma (Deutsche Studiengruppe Multiples Myelom; DSMM) incentive. Over 10 variables, including patient-specific and multiple myeloma-specific parameters, were analyzed at the time of initial diagnosis and repeatedly during follow-up. The probability of survival for another 5years was calculated according to disease-related and host-related risks. Multivariate Cox models were used to determine baseline and updated prognostic factors for survival. The median follow-up and overall survival were 10.3years and 5.1years, respectively. When comparing 5-year conditional survival probabilities from the data derived at the time of initial diagnosis with those updated over time, substantially differing prognoses were observed when follow-up data were used. Multivariate Cox regression models for cohorts surviving 0 to 5years demonstrated hazard ratios (HRs) for patients aged <60years, 60 to 69years, and >70years of 1, 1.68, and 3.17, respectively. These HRs for age were found to decline for patients surviving 5years, as well as for those with advanced stages of disease (II/III) and unfavorable cytogenetics, whereas progressive disease remained an important factor in patients surviving 1year, 3years, and 5years, with HRs of 1.85, 2.11, and 2.14, respectively. To the authors' knowledge, the current study is the first analysis of conditional survival in patients with multiple myeloma using both baseline and follow-up risk parameters, demonstrating that regular risk assessment throughout the course of disease and complete follow-up provide a more reliable conditional survival estimation than baseline assessment alone.

A risk scoring system to predict overall survival for patients with liver cancer.

e16635 Background: Surgical resection is a common curative treatment for patients with primary hepatocellular carcinoma (HCC), and conventional strategies of assessing clinical and pathologic risk factors have been adopted to predict clinic outcomes in patients after curative surgery. We hypothesized that an ensemble learning approach which incorporates multidimensional features would enable effective prediction of patient survival. Methods: We analyzed data from 222 stage II-III HCC patients who underwent surgical resection at Eastern Hepatobiliary Surgery Hospital (Shanghai, China). Baseline information for each patient includes clinical and pathologic risk factors, laboratory tests and in situ immunological profiles. Using machine learning, we developed CoxPH, GBS, CGBS, FSVM and NSVM models for patient overall survival (OS). Models were trained on 155 cases with 48 features. Thirteen-fold cross-validation (CV) was used to measure performance with area under the ROC curve (AUC) and C-Index (CI). The ensemble model was used to predict patient OS and validated on the subsequent 67 cases. Results: For all models tested, immune features, including the fraction of CD68+ and CD8+ cells in tumor, and CD8+ cells in stroma, play a crucial role in predicting patient OS. Using the ensemble CoxPH model (with superior value of AUC and CI), a risk scoring system for patient OS was developed. The scoring system could stratify patients into high-risk or low-risk groups, revealing different prognosis (validation cohort: HR, 6.5, 95% CI, 2.4-18, P = 3.4e-05). The CoxPH model was also predictive of the time to-recurrence (p &lt; 0.0001). In validation set, the scoring system predicted half-year mortality of patients with AUC of 0.9, and 1-year mortality of patients with AUC of 0.897. The scoring system could also predict half-year recurrence and 1-year recurrence with AUC more than 0.83. Conclusions: The machine learning-based risk scoring system offers a novel strategy for incorporating multidimensional risk factors to predict clinic outcome and may help medical practitioners to optimize clinical follow-up or therapeutic interventions. [Table: see text]

Prognostic Role of Beta-2 Microglobulin in Patients with Light Chain Amyloidosis Treated with Autologous Stem Cell Transplantation.

The prognostic impact of increased beta-2 microglobulin (B2M) in patients with light chain (AL) amyloidosis undergoing autologous stem cell transplantation (ASCT) is unknown. The Mayo 2012 stage and increased bone marrow plasma cell (BMPC) percentage are known predictors for survival. Increased B2M is predictive of survival in patients with multiple myeloma. We evaluated the prognostic role of B2M in patients with newly diagnosed AL undergoing ASCT. We retrospectively reviewed patients with a diagnosis of AL amyloidosis who were treated with ASCT between July 1996 and September 2017. Patients with a creatinine level >1.2 mg/dL were excluded, because that affects B2M levels. The receiver operator characteristic curve was used to determine the best cutoff for B2M before ASCT in predicting survival, which was 2.5 µg/mL, which was also the upper limit of normal in our laboratory. Baseline characteristics were compared between patients with B2M >2.5 µg/mL and ≤2.5 µg/mL. Progression-free survival (PFS) was defined as the time from ASCT to relapse or death, whichever occurred first. Overall survival (OS) was calculated from the time of ASCT to death of any cause. Univariate and multivariate analyses were done for OS. Five hundred and ten patients were identified, 222 of whom (44%) had a B2M >2.5 µg/mL. These patients were more likely to be older (median age, 61 versus 57 years;P=.0002), to have Mayo 2012 stage III/IV disease (33% versus 8%; P < .0001), to have more than 2 organs involved (25% versus 14%; P=.001), and to have ≥10% BMPCs (56% versus 40%; P=.0002) compared with patients with B2M ≤2.5 µg/mL. The median PFS and OS were shorter in patients with B2M >2.5 µg/mL (median PFS, 64 months versus 80 months [P=.03]; median OS, 104.9 months versus 175.5 months [P < .0001]). On univariate analysis, predictors for OS included age >60 years (hazard ratio [HR], 1.61; P=.001), Mayo 2012 stage III/IV (HR, 3.36; P < .0001), more than 2 organs involved (HR, 1.36; P=.07), ≥10% BMPCs (HR, 1.5; P=.005), melphalan conditioning with 200 mg/m2 (HR, .29; P < .0001), B2M >2.5 µg/mL (HR, 1.82; P < .0001), and transplantation during or after 2010 (HR, .4; P=.0006). On multivariate analysis, only Mayo 2012 stage III/IV (HR, 1.89; P=.005), melphalan conditioning with 200 mg/m2 (HR, .39; P < .0001), B2M >2.5 µg/mL (HR, 1.84; P=.003), and transplantation performed during or after 2010 (HR, .58; P=.03) remained independent predictors of OS. Our findings identify B2M >2.5 µg/dL before ASCT as an independent predictor for OS in patients with AL amyloidosis and normal kidney function and should be routinely measured.

A 14-Methylation-Driven Differentially Expressed RNA as a Signature for Overall Survival Prediction in Patients with Uterine Corpus Endometrial Carcinoma.

DNA methylation has been implicated as an important mechanism for the development of uterine corpus endometrial carcinoma (UCEC), indicating that methylation-driven genes may be potential biomarkers for survival prediction. In this study, we aimed to identify a new prognostic methylation signature for UCEC based on differentially expressed genes (DEGs) and long noncoding RNAs (lncRNAs) (DELs). Sample-matched RNA-sequencing and methylation-array data were downloaded from The Cancer Genome Atlas database, by analysis of which a total of 269 DEGs and 4 DELs were identified to be methylation driven. Least absolute shrinkage and selection operator analysis screened that 14 methylation-driven genes were significantly associated with overall survival (OS) and thus were used as a signature to establish a prognostic risk model. Based on the median threshold, the patients were divided into the low-risk and the high-risk groups, which showed significantly different survival periods under the Kaplan-Meier curve. The area under receiver operating characteristic curve (AUC) was 0.934, 0.919, and 0.952 for the training, validation, and entire cohort, respectively. Stratification analysis showed that the established risk model may add prognostic values to conventional clinical factors (age, neoplasm histologic grade, and clinical stage). A nomogram was constructed based on the risk model and clinical parameters, with the AUC of 0.978 and c-index of 0.8079. Database for Annotation, Visualization, and Integrated Discovery (DAVID) function enrichment and Human Protein Atlas (HPA) protein expression validation showed 5 of these 14 genes may be especially important for UCEC (hypermethylated lowly expressed: CCBE1, FOXL2, PHLDB2, and DTNA; hypomethylated highly expressed: CCNE1). Comparison with breast cancer in the methylation level indicated ABCA12, CCNE1, and CLRN3 may be specific methylation-driven genes for UCEC. LncRNA HCG11 may function by coexpressing with DTNA. In conclusion, this 14-DNA methylation signature combined with clinical factors may a potentially effective biomarker in predicting OS for UCEC patients.

Sarcopenia Predicts Overall Survival in Patients with Lung, Breast, Prostate, or Myeloma Spine Metastases Undergoing Stereotactic Body Radiation Therapy (SBRT), Independent of Histology.

Predicting survival of patients with spinal metastases would help stratify treatments from aggressive to palliation. To evaluate whether sarcopenia predicts survival in patients with lung, breast, prostate, or multiple myeloma spinal metastases. Psoas muscle measurements in patients with spinal metastasis were taken from computed tomography scans at 2 time points: at first episode of stereotactic body radiation therapy (SBRT) and from the most recent scan available. Overall survival and hazard ratios were calculated with multivariate cox proportional hazards regression analyses. In 417 patients with spinal metastases, 40% had lung cancer, 27% breast, 21% prostate, and 11% myeloma. Overall survival was not associated with age, sex, ethnicity, levels treated, or SBRT volume. Multivariate analysis showed patients in the lowest psoas tertile had shorter survival (222 d, 95% CI=185-323 d) as compared to the largest tertile (579 d, 95% CI=405-815 d), (HR1.54, P=.005). Median psoas size as a cutoff value was also strongly predictive for survival (HR1.48, P=.002). Survival was independent of tumor histology. The psoas/vertebral body ratio was also successful in predicting overall survival independent of tumor histology and gender (HR1.52, P<.01). Kaplan-Meier survival curves visually represent survival (P=.0005). In patients with spine metastases, psoas muscle size as a hallmark of frailty/sarcopenia is an objective, simple, and effective way to identify patients who are at risk for shorter survival, regardless of tumor histology. This information can be used to help with surgical decision making in patients with advanced cancer, as patients with small psoas sizes are at higher risk of death.

Can Avoiding So-Called HLA High Risk Epitope Mismatches (REM) Improve Lung Transplant (LTx) Outcomes?

HLA epitope mismatches can determine risk of graft rejection and de novo Donor Specific Antibodies (dnDSA) formation. Recently (McCaughan 2018) noted some epitopes posed a greater risk of developing dnDSA post-LTx and assumed this correlated with CLAD and survival. Using these reported REM we aim to determine associations between REM and actual LTx outcomes. We retrospectively analysed all 310 primary LTx undertaken Aug 2008-Dec 2015, with both patients and donors' high resolution HLA typed for all HLA loci. Using HLAMatchmaker (v2.1), LTx were assessed for the presence of REM (45GE3 or 45EV/55PP) and DQA1*05. BOS and RAS were determined using the most recent 2019 ISHLT guidelines. Kaplan Meier analysis with log-ranked tests compared groups. Within our cohort, 82 (27%) LTx occurred in the presence of a REM from the donor. Of these, 43 developed CLAD (BOS n=28 RAS n=15). REM was not a predictor of overall patient survival (fig 1a) (p=0.52) or time to CLAD (fig 1b) (p=0.32). As previously published (McCaughan 2018) the development of dnDSA directed to HLA-DQ was frequent (ie 75% of those tested post-LTx) this did not correlate to time to CLAD or overall patient survival in our cohort. Although avoiding REM has been reported to reduce risk of dnDSA following LTx, in our cohort this does not strongly correlate to overall patient survival or time to CLAD. Further studies are still warranted to identify the riskiest of epitopes to avoid in LTx and how the correlate to LTx outcomes.