Predictive Model For Overall Survival Research Articles

Identification of a novel matrix metalloproteinases-related prognostic signature in hepatocellular carcinoma.

Hepatocellular carcinoma (HCC) is the most common primary liver cancer worldwide. Cancer cells' local infiltration, proliferation, and spread are mainly influenced by the protein hydrolyzing function of different matrix metalloproteinases (MMPs). However, no study has determined the relationship between MMPs and prognostic prediction in HCC. Expression profiles of mRNA and MMPs-related genes were obtained from publicly available databases. Cox regression and LASSO Cox regression analysis were used to identify and predict MMPs-related prognostic signature and construct predictive models for overall survival (OS). A nomogram was used to validate the accuracy of the prediction model. Drug prediction was performed using the Genomics of Drug Sensitivity in Cancer (GDSC) dataset, and single-cell clustering analysis was performed to further understand the significance of the MMPs-related signature. A MMPs-related prognostic signature (including RNPEPL1, ADAM15, ADAM18, ADAMTS5, CAD, YME1L1, AMZ2, PSMD14, and COPS6) was identified. Using the median value, HCC patients in the high-risk group showed worse OS than those in the low-risk group. Immune microenvironment analysis showed that patients in the high-risk group had higher levels of M0 and M2 macrophages. Drug sensitivity analysis revealed that the IC50 values of sorafenib, cisplatin, and cytarabine were higher in the high-risk group. Finally, the single-cell cluster analysis results showed that YME1L1 and COPS6 were the major genes expressed in the monocyte cluster. A novel MMPs-related signature can be used to predict the prognosis of HCC. The findings of this research could potentially impact the predictability of the prognosis and treatment of HCC.

Identification and validation of neutrophils-related subtypes and prognosis model in triple negative breast cancer

BackgroundNeutrophils are considered to be crucial players in the initiation and progression of cancer. However, the complex relationship between neutrophils and cancer prognosis remains elusive, mainly due to the significant plasticity and diversity exhibited by these immune cells.MethodsAs part of our thorough investigation, we examined 38 Neutrophils-Related Genes (NRGs) and the associated copy number variations (CNV), somatic mutations, and gene expression patterns in relation to triple negative breast cancer (TNBC). The interactions between these genes, their biological roles, and their possible prognostic significance were then examined. With the NRGs as our basis, we applied Lasso and Cox regression analyses to create a predictive model for overall survival (OS). Furthermore, TNBC tissue and a public database were used to assess changes in MYO1D expression (MYO1D is characterized as a member of the myosin-I family, a group of motor proteins based on actin), its connection to neutrophil infiltration, and the clinical importance of MYO1D in TNBC.ResultsFour neutrophil-related genes were included in the development of a prognostic model based on neutrophils. The model was further shown to be an independent predicted factor for overall survival by multivariate Cox regression analysis. According to this study, neutrophil subtype B as well as gene subtype B, were associated with activated cancer immunity and poor prognosis of TNBC patients. Furthermore, considering that poor OS was linked to increased MYO1D expression, MYO1D was increased in TNBC tissues and associated with neutrophil infiltration. In vitro experiments also confirmed that MYO1D facilitates breast cancer invasion and metastasis.ConclusionBased on the degree of gene expression linked to neutrophils, a unique prognostic model was created. MYO1D could be a potential prognostic biomarker in TNBC patients and also a prospective target for therapy.

Colorectal cancer with low SLC35A3 is associated with immune infiltrates and poor prognosis

The expression level of SLC35A3 is associated with the prognosis of many cancers, but its role in colorectal cancer (CRC) is unclear. The purpose of our study was to elucidate the role of SLC35A3 in CRC. The expression levels of SLC35A3 in CRC were evaluated through tumor immune resource assessment (TIMER), The Cancer Genome Atlas (TCGA), Gene Expression Omnibus (GEO), International Cancer Genome Consortium (ICGC), Human Protein Atlas (HPA), qRT-PCR, and immunohistochemical evaluation. TCGA, GEO, and ICGC databases were used to analyze the diagnostic and prognostic value of SLC35A3 in CRC. A overall survival (OS) model was constructed and validated based on the expression level of SLC35A3 and multivariable analysis results. The cBioPortal tool was used to analyze SLC35A3 mutation in CRC. The UALCAN tool was used to analyze the promoter methylation level of SLC35A3 in colorectal cancer. In addition, the role of SLC35A3 in CRC was determined through GO analysis, KEGG analysis, gene set enrichment analysis (GSEA), immune infiltration analysis, and immune checkpoint correlation analysis. In vitro experiments validated the function of SLC35A3 in colorectal cancer cells. Compared with adjacent normal tissues and colonic epithelial cells, the expression of SLC35A3 was decreased in CRC tissues and CRC cell lines. Low expression of SLC35A3 was associated with N stage, pathological stage, and lymphatic infiltration, and it was unfavorable for OS, disease-specific survival (DSS), recurrence-free survival (RFS), and post-progression survival (PPS). According to the Receiver Operating Characteristic (ROC) analysis, SLC35A3 is a potential important diagnostic biomarker for CRC patients. The nomograph based on the expression level of SLC35A3 showed a better predictive model for OS than single prognostic factors and TNM staging. SLC35A3 has multiple types of mutations in CRC, and its promoter methylation level is significantly decreased. GO and KEGG analysis indicated that SLC35A3 may be involved in transmembrane transport protein activity, cell communication, and interaction with neurotransmitter receptors. GSEA revealed that SLC35A3 may be involved in energy metabolism, DNA repair, and cancer pathways. In addition, SLC35A3 was closely related to immune cell infiltration and immune checkpoint expression. Immunohistochemistry confirmed the positive correlation between SLC35A3 and helper T cell infiltration. In vitro experiments showed that overexpression of SLC35A3 inhibited the proliferation and invasion capability of colorectal cancer cells and promoted apoptosis. The results of this study indicate that decreased expression of SLC35A3 is closely associated with poor prognosis and immune cell infiltration in colorectal cancer, and it can serve as a promising independent prognostic biomarker and potential therapeutic target.

Imaging-Based Biomarkers Predict Programmed Death-Ligand 1 and Survival Outcomes in Advanced NSCLC Treated With Nivolumab and Pembrolizumab: A Multi-Institutional Study

BackgroundAlthough the immune checkpoint inhibitors, nivolumab and pembrolizumab, were found to be promising in patients with advanced NSCLC, some of them either do not respond or have recurrence after an initial response. It is still unclear who will benefit from these therapies, and, hence, there is an unmet clinical need to build robust biomarkers. MethodsPatients with advanced NSCLC (N = 323) who were treated with pembrolizumab or nivolumab were retrospectively identified from two institutions. Radiomics features extracted from baseline pretreatment computed tomography scans along with the clinical variables were used to build the predictive models for overall survival (OS), progression-free survival (PFS), and programmed death-ligand 1 (PD-L1). To develop the imaging and integrative clinical-imaging predictive models, we used the XGBoost learning algorithm with ReliefF feature selection method and validated them in an independent cohort. The concordance index for OS, PFS, and area under the curve for PD-L1 was used to evaluate model performance. ResultsWe developed radiomics and the ensemble radiomics-clinical predictive models for OS, PFS, and PD-L1 expression. The concordance indices of the radiomics model were 0.60 and 0.61 for predicting OS and PFS and area under the curve was 0.61 for predicting PD-L1 in the validation cohort, respectively. The combined radiomics-clinical model resulted in higher performance with 0.65, 0.63, and 0.68 to predict OS, PFS, and PD-L1 in the validation cohort, respectively. ConclusionsWe found that pretreatment computed tomography imaging along with clinical data can aid as predictive biomarkers for PD-L1 and survival end points. These imaging-driven approaches may prove useful to expand the therapeutic options for nonresponders and improve the selection of patients who would benefit from immune checkpoint inhibitors.

A Predictive Model Combining Clinical Characteristics and Nutritional Risk Factors for Overall Survival after Umbilical Cord Blood Transplantation

Background: Umbilical cord blood transplantation (UCBT) is a curable therapy for hematological disease; however, the impact of nutritional status on UCBT outcomes remains controversial. To evaluate the joint effect of clinical characteristics and nutritional status on the prognosis of patients who underwent UCBT, we screened various factors to establish a predictive model of overall survival (OS) after UCBT. Methods: We performed an integrated clinical characteristic and nutritional risk factor analysis and established a predictive model that could be used to identify UCBT recipients with poor OS. Results : Four factors, including disease status, conditioning regimen, calf skinfold thickness and albumin level, were identified and used to develop a risk score for OS, which showed a positive predictive value of 84.0%. A high-risk score (≥2.225) was associated with inferior 3-year OS post-UCBT [67.5% (95% CI, 51.1%-79.4%), P=0.001]. Then, we built a nomogram based on the four factors that showed good discrimination with a C-index of 0.833 (95% CI, 0.743-0.922). Internal validation was performed by using the bootstrap method (corrected C-index:0.804). Multivariate analysis suggested that a high calf skinfold thickness (≥20.5mm) and a low albumin level (<33.6g/L) conferred poor disease-free survival (DFS). Conclusion: The predictive model combining clinical and nutritional factors could be used to predict OS in UCBT recipients, which may promote preemptive treatment.

A predictive model combining clinical characteristics and nutritional risk factors for overall survival after umbilical cord blood transplantation

BackgroundUmbilical cord blood transplantation (UCBT) is a curable therapy for hematological disease; however, the impact of nutritional status on UCBT outcomes remains controversial. To evaluate the joint effect of clinical characteristics and nutritional status on the prognosis of patients who underwent UCBT, we screened various factors to establish a predictive model of overall survival (OS) after UCBT.MethodsWe performed an integrated clinical characteristic and nutritional risk factor analysis and established a predictive model that could be used to identify UCBT recipients with poor OS. Internal validation was performed by using the bootstrap method with 500 repetitions.ResultsFour factors, including disease status, conditioning regimen, calf skinfold thickness and albumin level, were identified and used to develop a risk score for OS, which showed a positive predictive value of 84.0%. A high-risk score (≥ 2.225) was associated with inferior 3-year OS post-UCBT [67.5% (95% CI 51.1–79.4%), P = 0.001]. Then, we built a nomogram based on the four factors that showed good discrimination with a C-index of 0.833 (95% CI 0.743–0.922). The optimism-corrected C-index value of the bootstrapping was 0.804. Multivariate analysis suggested that a high calf skinfold thickness (≥ 20.5 mm) and a low albumin level (< 33.6 g/L) conferred poor disease-free survival (DFS).ConclusionThe predictive model combining clinical and nutritional factors could be used to predict OS in UCBT recipients, thereby promoting preemptive treatment.

Integration and deconvolution methodology deciphering prognosis-related signatures in lung adenocarcinoma.

This study aims to establish a risk prediction model based on prognosis-related genes (PRGs) and clinicopathological factors, and investigate the biological activities of PRGs in lung adenocarcinoma (LUAD). Risk score signatures were developed by employing multiple algorithms and their amalgamations. A predictive model for overall survival was established through the integration of risk score signatures and several clinicopathological parameters. A comprehensive single-cell atlas, gene set enrichment analysis (GSEA) and gene set variation analysis (GSVA) were used to investigate the biological activities of prognosis-related genes in LUAD. A risk prediction model was established based on 16 PRGs, exhibiting robust performance in predicting overall survival. The single-cell analysis revealed that epithelial cells were primarily associated with worse survival of LUAD, and PRGs were predominantly enriched in malignant epithelial cells and influenced epithelial cell growth and progression. Furthermore, GSEA and GSVA analysis showed that PRGs were involved in tumor pathways such as epithelial-mesenchymal transition, hypoxia and KRAS_UP, and high GSVA scores are correlated with worse outcome in LUAD patients. The constructed risk prediction model in this study offers clinicians a valuable tool for tailoring treatment strategies of LUAD and provides a comprehensive interpretation on the biological activities of PRGs in LUAD.

Internal Validation of a Predictive Model for Overall Survival in Patients with FIGO stages I–IV Cervical Cancer

Internal Validation of a Predictive Model for Overall Survival in Patients with FIGO stages I–IV Cervical Cancer

A model-based head-to-head comparison of single-agent lurbinectedin in the pivotal ATLANTIS Study.

Lurbinectedin is a selective inhibitor of oncogenic transcription U.S. Food and Drug Administration (FDA)-approved for patients with relapsed small cell lung cancer (SCLC) as monotherapy at 3.2 mg/m2 every 3 weeks (q3wk). ATLANTIS was a phase 3 study in SCLC with lurbinectedin 2.0 mg/m2 plus doxorubicin 40 mg/m2 q3wk vs physician's choice, with overall survival (OS) as the primary endpoint and objective response rate (ORR) as the secondary endpoint. This work aimed to dissect the contribution of lurbinectedin and doxorubicin to antitumor effects in SCLC, and to predict the efficacy of single-agent lurbinectedin at 3.2 mg/m2 in ATLANTIS to allow for a head-to-head comparison with the control arm. The dataset included exposure and efficacy data from 387 patients with relapsed SCLC (ATLANTIS, n=288; study B-005, n=99). Patients in the ATLANTIS control arm (n=289) were used for comparison. Unbound plasma lurbinectedin area under the concentration-time curve (AUCu) and total plasma doxorubicin area under the concentration-time curve (AUCDOX) were used as exposure metrics. Univariate and multivariate analyses were conducted to determine the best predictors and predictive model for OS and ORR. OS baseline hazard was best described by a log-logistic distribution, with chemotherapy-free interval (CTFI), lactate dehydrogenase, albumin, brain metastases, neutrophils/lymphocytes ratio, AUCu, and the interaction between AUCu and AUCDOX as predictors. Effect of AUCu on ORR best fitted to a sigmoid-maximal response (Emax) logistic model, where Emax was dependent on CTFI. Head-to-head comparisons with predicted 3.2 mg/m2 lurbinectedin resulted in a positive outcome in ATLANTIS, with hazard ratio (95% prediction intervals [95% PI]) for OS of 0.54 (0.41, 0.72), and odds ratio (95% PI) for ORR of 0.35 (0.25, 0.5). These results support the superiority of lurbinectedin monotherapy for relapsed SCLC over other approved therapies.

Development and verification of prognostic nomogram for ampullary carcinoma based on the SEER database.

Ampullary carcinoma (AC) is a rare cancer of the digestive system that occurs in the ampulla at the junction of the bile duct and pancreatic duct. However, there is a lack of predictive models for overall survival (OS) and disease -specific survival (DSS) in AC. This study aimed to develop a prognostic nomogram for patients with AC using data from the Surveillance, Epidemiology, and End Results Program (SEER) database. Data from 891 patients between 2004 and 2019 were downloaded and extracted from the SEER database. They were randomly divided into the development group (70%) and the verification group (30%), and then univariate and multivariate Cox proportional hazards regression, respectively, was used to explore the possible risk factors of AC. The factors significantly related to OS and DSS were used to establish the nomogram, which was assessed via the concordance index (C-index), and calibration curve. An internal validation was conducted to test the accuracy and effectiveness of the nomogram. Kaplan-Meier calculation was used to predict the further OS and DSS status of these patients. On multivariate Cox proportional hazards regression, the independent prognostic risk factors associated with OS were age, surgery, chemotherapy, regional node positive (RNP),extension range and distant metastasis with a moderate C-index of 0.731 (95% confidence interval (CI): 0.719-0.744) and 0.766 (95% CI: 0.747-0.785) in the development and verification groups, respectively. While, marital status, surgery, chemotherapy, regional node positive (RNP),extension range and distant metastasis were significantly linked to AC patients' DSS, which have a better C-index of 0.756 (95% confidence interval (CI): 0.741-0.770) and 0.781 (95% CI: 0.757-0.805) in the development and verification groups. Both the survival calibration curves of 3- and 5-year OS and DSS brought out a high consistency. Our study yielded a satisfactory nomogram showing the survival of AC patients, which may help clinicians to assess the situation of AC patients and implement further treatment.

The AUGIS Survival Predictor: Prediction of Long-Term and Conditional Survival After Esophagectomy Using Random Survival Forests.

The aim of this study was to develop a predictive model for overall survival after esophagectomy using pre/postoperative clinical data and machine learning. For patients with esophageal cancer, accurately predicting long-term survival after esophagectomy is challenging. This study investigated survival prediction after esophagectomy using a RandomSurvival Forest (RSF) model derived from routine data from a large, well-curated, national dataset. Patients diagnosed with esophageal adenocarcinoma or squamous cell carcinoma between 2012 and 2018 in England and Wales who underwent an esophagectomy were included. Prediction models for overall survival were developed using the RSF method and Cox regression from 41 patient and disease characteristics. Calibration and discrimination (time-dependent area under the curve) were validated internally using bootstrap resampling. The study analyzed 6399 patients, with 2625 deaths during follow-up. Median follow-up was 41 months. Overall survival was 47.1% at 5 years. The final RSF model included 14 variables and had excellent discrimination with a 5-year time-dependent area under the receiver operator curve of 83.9% [95% confidence interval (CI) 82.6%-84.9%], compared to 82.3% (95% CI 81.1%-83.3%) for the Cox model. The most important variables were lymph node involvement, pT stage, circumferential resection margin involvement (tumor at < 1 mm from cut edge) and age. There was a wide range of survival estimates even within TNM staging groups, with quintiles of prediction within Stage 3b ranging from 12.2% to 44.7% survival at 5 years. An RSF model for long-term survival after esophagectomy exhibited excellent discrimination and well-calibrated predictions. At a patient level, it provides more accuracy than TNM staging alone and could help in the delivery of tailored treatment and follow-up.

The AGAMENON-SEOM model for prediction of survival in patients with advanced HER2-positive oesophagogastric adenocarcinoma receiving first-line trastuzumab-based therapy.

Trastuzumab and chemotherapy is the standard first-line treatment in human epidermal growth factor receptor 2 (HER2)-positive advanced gastro-oesophageal cancer. The objective was to develop a predictive model for overall survival (OS) and progression-free survival (PFS) in patients treated with trastuzumab. Patients with HER2-positive advanced gastro-oesophageal adenocarcinoma (AGA) from the Spanish Society of Medical Oncology (SEOM)-AGAMENON registry and treated first line with trastuzumab and chemotherapy between 2008 and 2021 were included. The model was externally validated in an independent series (The Christie NHS Foundation Trust, Manchester, UK). In all, 737 patients were recruited (AGAMENON-SEOM, n = 654; Manchester, n = 83). Median PFS and OS in the training cohort were 7.76 [95% confidence interval (CI), 7.13-8.25] and 14.0 months (95% CI, 13.0-14.9), respectively. Six covariates were significantly associated with OS: neutrophil-to-lymphocyte ratio, Eastern Cooperative Oncology Group performance status, Lauren subtype, HER2 expression, histological grade and tumour burden. The AGAMENON-HER2 model demonstrated adequate calibration and fair discriminatory ability with a c-index for corrected PFS/OS of 0.606 (95% CI, 0.578-0.636) and 0.623 (95% CI, 0.594-0.655), respectively. In the validation cohort, the model is well calibrated, with a c-index of 0.650 and 0.683 for PFS and OS, respectively. The AGAMENON-HER2 prognostic tool stratifies HER2-positive AGA patients receiving trastuzumab and chemotherapy according to their estimated survival endpoints.

Predictive models for overall survival in breast cancer patients with a second primary malignancy: a real-world study in Shanghai, China

BackgroundThe incidents of second primary malignancy (SPM) is increasing among breast cancer (BC) patients with long-term progression, adversely affecting survival. The purpose of this study was to screen independent overall survival (OS) risk factors and establish nomograms to predict the survival probabilities of BC patients with SPM.MethodA total of 163 BC patients with SPM were recruited during 2002–2015 from a total of 50 hospitals in Shanghai, China. Two nomograms to predict survival from primary BC and SPM diagnosis were constructed based on independent factors screened from multivariable analysis. The calibration and discrimination of nomograms were calculated in the training and validation cohorts.ResultsThe overall survival rates of BC patients with SPM were 88.34%, 64.42% and 54.66% at 5, 10 and 15 years, respectively. Factors of late TNM stage of SPM (HR = 4.68, 95% CI 2.14–10.25), surgery for SPM (HR = 0.60, 95% CI 0.36–1.00), SPM in the colon and rectum (HR = 0.49, 95% CI 0.25–0.98) and thyroid (HR = 0.08, 95% CI 0.01–0.61) independently affected the OS of BC patients with SPM (p < 0.05). In addition, a longer latency (≥ 5 years) was associated with better OS from BC diagnosis (p < 0.001). Older age (≥ 56) was associated with poor OS from SPM diagnosis (p = 0.019). Two nomograms established based on the above factors had better calibration and discrimination.ConclusionThe TNM stage of SPM, surgery for SPM, SPM sites, latency and age at BC diagnosis are independent factors for survival and the two nomograms may provide more personalized management for BC patients with SPM.

A web-based predictive model for overall survival of patients with cutaneous Merkel cell carcinoma: A population-based study.

Merkel cell carcinoma (MCC) is an aggressive neuroendocrine carcinoma with a high mortality rate, so it is necessary to create models to predict overall survival of MCC. We developed an easy-to-use web-based calculator to predict the OS of MCC patients based on the nomogram. MCC patients between 2004 and 2015 were collected from the Surveillance, Epidemiology, and End Results (SEER) database and randomly assigned to training and validation cohorts. Patients between 2016-2017 serve as the external validation cohort. Relevant risk factors were identified by univariate and multivariate COX hazards regression methods and combined to produce nomograms. The concordance index (C-index), area under the receiver operating characteristic (AUC) curve, and calibration plots have demonstrated the predictive power of the nomograms. Decision curve analysis (DCA) was used to measure nomograms in clinical practice. Patients were divided into three groups according to the scores of the nomogram. A total of 3480 patients were randomly assigned to the training group and validation group in this study. Meaningful prognostic factors were applied to the establishment of nomograms. The C-index for OS was 0.725 (95% CI: 0.706-0.741) in the training cohort and 0.710 (95% CI: 0.683-0.737) in the validation cohort. In the external validation cohort, C-index was 0.763 (95% CI: 0.734-0.792). The C-index of training cohort, validation cohort and external validation cohort for CSS were 0.743 (95% CI:0.725-0.761), 0.739(95%CI:0.712-0.766) and 0.774 (95%CI:0.735-0.813), respectively. The AUC and calibration plots of 1-, 3-, and 5-year OS rates showed that the nomogram had good predictive power. DCA demonstrated that the nomogram constructed in this study could provide a clinical net benefit. Our calculator demonstrated excellent predictive capabilities for better risk grouping of MCC patients. We created novel nomograms of prognostic factors for MCC, which more accurately and comprehensively predicted 1-, 3-, and 5-year OS/CSS in MCC patients. We established a calculator which can easily and quickly calculate the risk grouping of MCC patients by inputting clinically relevant characteristics. This can help clinicians identify high-risk patients as early as possible, carry out personalized treatment, follow-up, and monitoring, and improve the survival rate of MCC patients.

Prediction of survival in patients with colorectal liver metastases- development and validation of a prognostic score model

BackgroundMetastatic spread of colorectal cancer to the liver impacts prognosis. Advances in chemotherapy have resulted in increased resectability rates and thereby improved survival in patients with colorectal liver metastases (CRLM). However, criteria are needed to ensure that patients selected for hepatic resection benefit from the invasive therapy. The study aimed to construct a predictive model for overall survival (OS) in patients with CRLM, based on preoperatively available information. MethodsThe retrospective cohort study reviewed all patients with CRLM discussed at multidisciplinary team conference at Karolinska University Hospital, Stockholm, Sweden, 2013–2018. Independent prognostic factors for OS were identified, based on which a score model was generated. The model was validated on patients treated for CRLM at Hôpital Universitaire Paul Brousse, Villejuif, France, 2007–2018. Calibration and discrimination methods were used for internal and external validation. ResultsThe Swedish development cohort included 1013 patients, the French validation cohort 391 patients. Poor OS was significantly associated with age>60years (hazard ratio (HR) 3.57 (95%CI 2.18–9.94)), number of CRLM (HR 4.59 (2.83–12.20)), diameter of largest CRLM>5 cm (HR 2.59 (1.74–5.03)), right-sided primary tumour (HR 2.98 (2.00–5.80)), extrahepatic disease (HR 4.14 (2.38–15.87)) and non-resectability (HR 0.77 (0.66-0.90)). The C-statistic for prediction of OS was .74, in the development cohort and 0.69 in the validation cohort. ConclusionThe presented predictive score model can adequately predict OS for patients at the initial diagnosis of CRLM. The prognostic model could be of clinical value in the management of all patients with CRLM, by predicting individualized survival and thereby facilitating treatment recommendations.

Development of a prognostic model of overall survival in oropharyngeal cancer from real-world data: PRO.M.E.THE.O.

SUMMARYObjectiveThe PRO.M.E.THE.O. study (PredictiOn Models in Ent cancer for anti-EGFR based THErapy Optimization) aimed to develop a predictive model (PM) of overall survival (OS) for patients with locally advanced oropharyngeal cancer (LAOC) treated with radiotherapy (RT) and cetuximab (Cet) from an Italian dataset.MethodsWe enrolled patients with LAOC from 6 centres treated with RT-Cet. Clinical and treatment variables were collected. Patients were randomly divided into training (TS) (80%) and validation (VS) (20%) sets. A binary logistic regression model was used on the TS with stepwise feature selection and then on VS. Timepoints of 2, 3 and 5 years were considered. The area under the curve (AUC) of receiver operating characteristic of 2, 3 and 5 year and confusion matrix statistics at 5-threshold were used as performance criteria.ResultsOverall, 218 patients were enrolled and 174 (79.8%) were analysed. Age at diagnosis, gender, ECOG performance, clinical stage, dose to high-risk volume, overall treatment time and day of RT interruption were considered in the final PMs. The PMs were developed and represented by nomograms with AUC of 0.75, 0.73 and 0.73 for TS and 0.713, 0.713, 0.775 for VS at 2, 3 and 5 years, respectively.ConclusionsPRO.M.E.THE.O. allows the creation of a PM for OS in patients with LAOC treated with RT-Cet.

Multivariate Model for Predicting Overall Survival in Malignant Pleural Mesothelioma After Adjuvant Radiotherapy

<h3>Purpose/Objective(s)</h3> Malignant pleural mesothelioma (MPM) is a rare but aggressive cancer arising from the cells of the thoracic pleura. Its anatomic complexity, large surface area, and unique location pose challenges for target coverage while sparing adjacent organs at risk (OARs) during treatment planning for adjuvant radiotherapy after pleurectomy and decortication. We aimed to investigate the associations between patient/target characteristics and achievable dosimetric plan quality with overall survival. <h3>Materials/Methods</h3> Sixty patients with MPM patients who were treated on a tomotherapy unit between 2013-2018 at a single institution were included. All patients received helical IMRT with 45 Gy/25 fractions. Dosimetric quantities for the lungs, heart, and liver were systematically examined for 37 right-sided (RSM) and 23 left-sided mesothelioma (LSM) patients. Patient characteristics and/or planning dosimetric metrics were input into machine learning algorithms, e.g., LASSO regression, support vector machine (SVM), and multiple layer perceptron (MLP), in order to construct predictive models for overall survival. The performance of predictive models was evaluated with area under the curve (AUC) using 3-fold cross validation. Normal tissue complication probabilities (NTCP) for selected OARs were estimated using the Lyman-Kutcher-Burman model based on individual dose volume histograms. <h3>Results</h3> The achieved dosimetric endpoints were significantly different (<b>P</b> < 0.05) between LSM vs. RSM patients, respectively: ipsilateral lung V20_Gy of 91.2 ± 6.7 vs. 85.3 ± 6.5%, ipsilateral mean lung doses of 39.1 ± 3.8 vs. 36.9 ± 2.5 Gy, heart mean doses of 25.2 ± 3.6 vs. 18.8 ± 3.7 Gy, and liver mean doses of 10.6 ± 2.7 vs. 23.4 ± 4.2 Gy. There was no significant difference in total lung – planning target volume (PTV) mean dose, 13.4 ± 2.0 vs. 14.4 ± 1.5 Gy, nor expected lung pneumonitis for the LSM vs. RSM. No differences were observed between PTV coverage of 94.2 ± 1.2 vs. 93.4 ± 2.4%, and average survival intervals were 16.8 ± 14.4 months and 18.0 ± 18.0 months after RT between LSM vs. RSM. The MLP model achieved AUCs of 0.65, 0.67 and 0.71 using patient characteristics, dosimetric features, and combined features. The identified predictors most associated with overall survival were age, PTV volume, ipsilateral lung volume/V20_Gy, and ipsilateral lung – PTV V20_Gy <h3>Conclusion</h3> We demonstrated correlations between patient specific characteristics and achieved planning dosimetric parameters for left- and right-sided MPM treatment planning endpoints. Combining patient characteristics and dosimetric endpoints, the multivariate predictive model achieved decent performance in predicting overall survival. The predictive model may guide treatment planning to achieve optimal planning dosimetry, and to improve clinical outcome via personalized treatment for MPM.

Predictive Model for Overall Survival and Cancer-Specific Survival in Patients with Esophageal Adenocarcinoma.

Objective Recent years, there has been a rapid increase in the incidence of esophageal adenocarcinoma (EAC), while the prognosis for patients diagnosed remains poor and has slightly improved. Methods We extracted 6,466 cases with detailed demographical characteristics including age at diagnosis, sex, ethnicity, marital status, and clinical features, involving tumor grade and stage at diagnosis and treatment modalities (radiation therapy, chemotherapy, and surgery) from the Surveillance, Epidemiology, and End Results (SEER) (1975–2017) dataset. They were further randomly divided into the training and validating cohorts. Univariate and multivariate Cox analyses were conducted to determine significant variables for construction of nomogram. The predictive power of the model was then assessed by Harrell concordance index (C-index) and the area under the curve (AUC) of the receiver operating characteristic (ROC) curve. Results Multivariate analysis revealed that age, marital status, insurance, tumor grade, TNM stage, surgery, and chemotherapy all showed a significant association with overall survival (OS) and cancer-specific survival (CSS). These characteristics were employed to build a nomogram. Particularly, the discrimination of nomogram for OS and CSS prediction in the training set were excellent (C-index = 0.762, 95% CI: 0.754–0.770 and C-index = 0.774, 95% CI: 0.766–0.782). The AUC of the nomogram for predicting 2- and 5-year OS was 0.834 and 0.853 and CSS was 0.844 and 0.866. Similar results were observed in the internal validation set. Conclusion We have successfully established a novel nomogram for predicting OS and CSS in EAC patients with good accuracy, which can help clinicians predict the survival of individual patient survival and provide optimal treatment strategies.

Establishment and validation of a nomogram for predicting overall survival of node-negative perihilar cholangiocarcinoma

AimThere lacks a predictive model for overall survival (OS) of node-negative perihilar cholangiocarcinoma (PHC). This study aimed at developing and validating a prognostic nomogram to predict OS of node-negative PHC after resection. MethodsWe established a nomogram via multivariate regression analysis by using the design cohort (n = 410, obtained from Surveillance, Epidemiology, and End Results database), and its external verification was done in the validation cohort (n = 100, the First Affiliated Hospital of Sun Yat-sen University). Predictive accuracy of the nomogram was assessed by concordance-index (C-index), calibration curves, and decision curve analysis (DCA). Performance of the nomogram was compared with the American Joint Committee on Cancer (AJCC) staging system. ResultsMultivariate regression analysis revealed that age, tumor grade, and the count of examined lymph nodes were independent prognostic factors for OS of node-negative PHC. The nomogram had a C-index of 0.603 and 0.626 in design cohort and validation cohort, respectively, which was better than that of AJCC staging system (both p < 0.05). The calibration curves showed good consistency between actual and nomogram-predicted OS probabilities. DCA showed that nomogram had better clinical usefulness. Furthermore, the nomogram-predicted scores could stratify the patients into three risk groups, and patients in higher risk group had worse prognosis than those in lower risk group (all p < 0.05). ConclusionThe proposed nomogram had a better prognostic accuracy than the AJCC staging system in predicting postoperative OS of node-negative PHC. It was helpful to guide the adjuvant therapeutic strategies for node-negative PHC.

Population-based, real-world prognostic factors related to survival among patients with metastatic pancreatic ductal adenocarcinoma (mPDAC).

389 Background: Pancreatic cancer is expected to be the third deadliest cancer in the US in 2020. Many real-world studies of pts with mPDAC are restricted to single centers, limiting the generalizability of the insights they generate. There is a need to understand prognostic factors of survival in a broader setting to aid in tailoring treatment strategies for pts. This study aimed to identify important population-based predictors related to survival among pts diagnosed with mPDAC. Methods: Data were extracted for pts diagnosed with mPDAC between Jan 2017 and Dec 2019 from the Flatiron Health database. Predictive models for overall survival from the start of each treatment were developed using multivariable Cox proportional hazards regression. Treatment specific predictive models were generated for pts treated with first line (1L) gemcitabine + nab-paclitaxel (GNP), 1L FOLFIRINOX, 1L gemcitabine monotherapy (gem-mono), and 2L liposomal irinotecan-based regimens. The holdout method was used for cross-validation, splitting the data into 70% training / 30% validation. Age at diagnosis, sex, body mass index, smoking status, and ECOG performance score (PS) were included in all models due to clinical importance. Demographic, clinical characteristics, hematological labs, liver function tests (LFTs), and serum bilirubin levels were assessed for inclusion into the models. Uno’s concordance statistic (c-statistic) was used to assess the predictive accuracy of the models. Results: Of the 3,572 pts included in the study, 44% (n = 1,557) received 1L GNP, 27% (n = 954) received 1L FOLFIRINOX, 7% (n = 265) received gem-mono, and 22% (n = 796) received other regimens. 38% (n = 1,345) pts received 2L and of those, 17% (n = 222) received liposomal irinotecan-based regimens. Among all 1L pts, the following were included in the final model: prior surgery, white blood cell (WBC) counts, serum albumin, LFTs (ALP and ALT), serum bilirubin, and ascites (c-statistic: 0.65). The model for pts treated with GNP differed from the overall model via the addition of neutrophil counts and removal of serum bilirubin and ascites (c-statistic: 0.67). Stage at initial diagnosis was included in the model only for pts treated with 1L FOLFIRINOX (c-statistic: 0.68). Among pts treated with gem-mono the LFTs were not included in the model (c-statistic: 0.78). ALP, serum albumin, and WBC counts were important predictors of survival among pts treated with 2L liposomal irinotecan-based regimens (c-statistic: 0.70). Across all regimens the strongest predictors of survival were ECOG PS, serum albumin, and ALP. Conclusions: In one of the largest contemporary real-world studies of patients with mPDAC to date, important population predictors of survival in pts receiving systemic treatment were identified. Further validation studies are needed to understand the generalizability of these results.