321 RANDOM SURVIVAL FOREST FOR PERSONALISED PROGNOSTICATION AFTER ESOPHAGECTOMY

2021 ◽  
Vol 34 (Supplement_1) ◽  
Author(s):  
Saqib Rahman ◽  
Robert Walker ◽  
Tom Crosby ◽  
Nicholas Maynard ◽  
Nigel Trudgill ◽  
...  
Keyword(s):  
Overall Survival ◽  
Random Forest ◽  
Prediction Models ◽  
Cox Regression ◽  
Cox Model ◽  
Neoadjuvant Treatment ◽  
Net Benefit ◽  
Term Survival ◽  
Post Surgery ◽  
Curative Esophagectomy

Abstract   For patients with esophageal cancer, producing accurate prediction models for long-term survival after esophagectomy has proved challenging. We investigated whether Random Survival Forests (RSF), a machine learning method, could produce an accurate prognostic model for overall survival after esophagectomy. Methods The study used data from the 'National Oesophago-Gastric Cancer Audit' (NOGCA) and included patients diagnosed with esophageal adenocarcinoma or squamous cell carcinoma between 2012 and 2018 in England and Wales and who underwent a curative esophagectomy with adequate lymphadenectomy (15 lymph nodes) and survived to discharge (n = 6838). Missing data was handled using multiple imputation. 15 variables were selected for inclusion using Random Forest variable importance and used to train the final model. The same variables with non-linearity transformations were used to develop a traditional Cox regression model for comparison. Results Median survival was 50 months. The final RSF model had good discrimination on internal validation with a C-index of 0.7627 (0.7625–0.7629), exceeding the cox model 0.7539 (0.7541–0.7537). At 3 years post-surgery, overall survival was 56.2%. The RSF yielded a mean predicted survival of 59.3% (IQR 33.3–87.1%) with good calibration (Figure 1) compared to 57.4% (38.4%–79.8%) for the cox model. The most influential variables were lymph node involvement and pT/ypT stage, however other variables including neoadjuvant treatment completion and surgical complications were also important. Decision curve analysis was undertaken which also showed an increased net benefit with the RSF model. Conclusion A Random Forest survival model provided better performance in predicting survival after curative esophagectomy. This will allow more personalised predictions to be delivered clinicians and patients. An online web app is provided at https://uoscancer.shinyapps.io/NOGCA/

scholarly journals O14: RANDOM FOREST MODELS FOR PREDICTING SURVIVAL AFTER OESOPHAGECTOMY

2021 ◽  
Vol 108 (Supplement_1) ◽  
Author(s):  
SA Rahman ◽  
RC Walker ◽  
T Crosby ◽  
N Maynard ◽  
DA Cromwell ◽  
...  
Keyword(s):  
Overall Survival ◽  
Random Forest ◽  
Prediction Models ◽  
Cox Regression ◽  
Cox Model ◽  
Neoadjuvant Treatment ◽  
Lymph Node Involvement ◽  
Oesophageal Adenocarcinoma ◽  
Survival Models ◽  
Post Surgery

Abstract Introduction For patients with oesophageal cancer, producing accurate prediction models for survival after oesophagectomy has proved challenging. We investigated whether Random Survival Forests (RSF), a novel machine learning method, could produce an accurate prognostic model for overall survival after oesophagectomy. Method The study used data from the National Oesophago-Gastric Cancer Audit and included patients diagnosed with oesophageal adenocarcinoma or squamous cell carcinoma between 2012 and 2018 in England and Wales and who underwent a curative oesophagectomy with adequate lymphadenectomy (15 LN) and survived to discharge (n=6198). Missing data was handled using multiple imputation and the data was split into training and validation cohorts. 13 variables were selected for inclusion using Random Forest variable importance and used to train the final model. The same variables were used to develop a traditional Cox regression model. Result Median survival was 53 months in both cohorts. The final RSF model had good discrimination in the validation cohort with a C-index of 0.757(0.755-0.759), exceeding the Cox model; 0.748(0.746-0.750). At 3 years post-surgery, overall survival was 56.2%. The RSF yielded a mean predicted survival of 55.8%(IQR 29.5%-81.7%) compared to 55.4%(40.0%-77.7%) for the Cox model. The most important variables were lymph node involvement and pT/ypT stage, however other variables including neoadjuvant treatment completion and surgical complications were also found to be important. Conculsion A Random Forest survival model provided better performance in predicting survival after curative oesophagectomy. This will allow more personalised predictions to be delivered clinicians and patients. Take-home message Random Forest survival models can accurately predict post-operative prognosis after oesophagectomy.

O45 THE ASSOCIATION OF PERIOPERATIVE QUALITY-OF-CARE PARAMETERS (TEXTBOOK OUTCOME) WITH LONG TERM OUTCOME AFTER ESOPHAGECTOMY FOR ESOPHAGEAL CANCER

2019 ◽  
Vol 32 (Supplement_2) ◽  
Author(s):  
M C Kalff ◽  
I Vesseur ◽  
W Eshuis ◽  
D Heineman ◽  
F Daams ◽  
...  
Keyword(s):  
Overall Survival ◽  
Esophageal Cancer ◽  
Cox Regression ◽  
Neoadjuvant Treatment ◽  
Recurrence Rates ◽  
Term Survival ◽  
Long Term Outcome ◽  
Asa Score

Abstract Aim The objectives of this study were to confirm the association of textbook outcome (TO) and overall long-term survival after esophagectomy for esophageal cancer, to investigate the relationship of TO and recurrence rates and to identify clinicopathological predictors for not achieving TO. Background & Methods Despite current improvements in the multimodal treatment of esophageal cancer, surgery remains the key component. Therefore, it is essential to optimize the surgical procedure and to pursue the highest surgical quality. TO is a composite measure of ten perioperative parameters reflecting the quality of surgical care concerning esophagectomy. All patients with esophageal cancer who underwent a transthoracic or transhiatal esophagectomy with curative intent in two tertiary referral centers in The Netherlands between 2007-2016 were included. Patients with a carcinoma in situ, patients undergoing salvage or emergency procedure and patients that applied for opt-out were excluded. Clinicopathological predictors for not achieving TO were identified using univariate and multivariate logistic regression. Survival was compared using Kaplan-Meier life-table estimates and cox regression. Results In total, 1057 patients were included. Over time, the percentage of patients who achieved TO increased from 28.9% in 2007 to 37.5% in 2016. BMI under 18.5, ASA score above one and age above 65 years were associated with a worse TO rate (OR 2.72 [1.02-7.24], ASA 2 OR 1.57 [1.13-2.17] and ASA 3+4 OR 2.33 [1.56-3.48], OR 1.387 [1.06-1.81], respectively), whereas neoadjuvant treatment predicted a better TO rate (OR 0.58 [0.41-0.81]). The median overall survival was 53 months (95% CI 42 – 63) for patients with TO and 35 months (95% CI 29 – 41) for patients without TO; resulting in an overall survival benefit of 18 months (HR 0.759, 95% CI 0.636 – 0.906, P = 0.002). The recurrence rates between TO and no-TO differed, but was not statistically significant (47.1% vs 42.8%, P = 0.177). Conclusion BMI less than 18.5, ASA-score higher than one and age older than 65 were characteristics associated with not achieving TO. Neoadjuvant therapy was associated with a better TO rate. Achieved TO resulted in a better overall five-year survival indicating the importance of pursuing TO.

Nomograms Predicting Overall Survival and Cancer Special Survival for Cervical Cancer in stage IIIC1

2021 ◽  
Author(s):  
Yifan Feng ◽  
Ye Wang ◽  
Yangqin Xie ◽  
Shuwei Wu ◽  
Yuyang Li ◽  
...  
Keyword(s):  
Cervical Cancer ◽  
Overall Survival ◽  
Tumor Size ◽  
Random Sampling ◽  
Prediction Models ◽  
Cox Regression ◽  
Characteristic Curve ◽  
Training Group ◽  
Data Sets ◽  
External Data

Abstract BackgroundThe purpose of this study is to explore the factors that affect the prognosis of overall survival (OS) and cancer special survival (CSS) in cervical cancer with stage IIIC1 and establish nomogram models to predict this prognosis.MethodsData from The Surveil-lance, Epidemiology, and End Results (SEER) Program meeting the inclusion criterions were classified into training group, and data of validation were obtained from the First Affiliated Hospital of Anhui Medical University from 2010 to 2019. The incidence, Kaplan‐Meier curves, OS and CSS of stage IIIC1 were evaluated according to the training group. Nomograms were established according to the results of univariate and multivariate Cox regression models. Harrell’s C-index and receiver operating characteristic curve (ROC) were calculated to measure the accuracy of the prediction models. Calibration plots show the relationship between the predicted probability and the actual outcome. Decision-curve analysis (DCA) was applied to evaluate the clinical applicability of the constructed nomogram.ResultsThe incidence of pelvic lymph node metastasis, a high-risk factor for prognosis in cervical cancer, decreased slightly over time. There are eight independent prognostic variables for OS, including age, race, histology, differentiation, extension range, tumor size, radiation recode and surgery, but seven for CSS with age excluded. Nomograms of OS and CSS were established based on the results. The C-index for the nomograms of OS and CSS were 0.692, 0.689 respectively when random sampling of SEER data sets, and 0.706, 0.737 respectively when random sampling of external data sets. AUCs for the nomogram of OS were 0.648, 0.644 respectively, and 0.683, 0.675 for the nomogram of CSS. Calibration plots for the nomograms were almost identical to the actual observations. The DCA also proved the value of the two models.ConclusionAge, race, histology, differentiation, extension range, tumor size, radiation recode and surgery were all independent prognosis factors for OS. Only age excepts in CSS. OS and CSS nomograms were established in our study based on the result of multivariate Cox proportional hazard regression, and both own good predictive and clinical application value after validation.

O30 NON-RADICAL ESOPHAGECTOMY FOR ESOPHAGEAL CANCER; A NATIONWIDE STUDY ON INCIDENCE, RISK FACTORS AND OUTCOMES IN SWEDEN

2019 ◽  
Vol 32 (Supplement_2) ◽  
Author(s):  
David Edholm ◽  
Petter Hollertz ◽  
Per Sandström ◽  
Bergthor Björnsson ◽  
Dennis Björk ◽  
...  
Keyword(s):  
Risk Factors ◽  
Esophageal Cancer ◽  
Cox Regression ◽  
Regional Lymph Node ◽  
Neoadjuvant Treatment ◽  
Resected Specimen ◽  
Radical Esophagectomy ◽  
Term Survival ◽  
R1 Resection ◽  
Independent Risk Factors

Abstract Aim To identify potential risk factors for a microscopically non-radical esophageal cancer resection (R1) and investigate how such a resection affects long-term survival. Background & Methods Esophageal cancer resections that are considered R1 have been associated with worse survival. The Swedish National Register for Esophageal and Gastric Cancer includes information on all esophageal cancer resections in Sweden. All patients having undergone esophageal resection with curative intent 2006-2017 were included. Risk factors for R1 resection were assessed through logistic regression. Factors predicting five-year survival were assessed through Cox-regression, adjusted for T-stage, N-stage, age and R-status. Results The study included 1,504 patients. The margins were microscopically involved in 146 patients (10%). Of these the circumferential margin was involved in 115 (8%). The proximal margin was involved in 55 patients (4%) and the distal in 30 (2%). In 54 (4%) specimens two margins were involved. Independent risk factors for R1-resection were absence of neoadjuvant treatment and clinical T3 stage or higher. The 5-year survival for the entire cohort was 41%, but only 19% for those with an R1 resection. Independent risk factors for death within 5-year from resection were regional lymph node metastasis (Hazard Ratio (HR) 2.6 (95% CI 2.2-3.1), histopathological stage T3 or higher (HR 1.2 95% CI 1.1-1.5), age above 60 years and R1-resection (HR 1.6 95% CI 1.4-2.0) Conclusion Involved margin in the resected specimen is an independent risk factor predicting worse 5-year survival. Besides striving for adequate surgical margins, the rate of R1-resections could be decreased through neoadjuvant treatment in fit patients.

scholarly journals GATA binding protein 6 (GATA6) is co-amplified with PIK3CA in patients with esophageal adenocarcinoma and is linked to neoadjuvant therapy

2020 ◽  
Author(s):  
Patrick Sven Plum ◽  
Heike Löser ◽  
Thomas Zander ◽  
Ahlem Essakly ◽  
Christiane J. Bruns ◽  
...  
Keyword(s):  
Overall Survival ◽  
Esophageal Adenocarcinoma ◽  
Cox Regression ◽  
Neoadjuvant Treatment ◽  
Driver Mutations ◽  
Large Tumor ◽  
Cox Regression Analysis ◽  
Entire Cohort ◽  
Independent Prognostic Marker ◽  
Survival Difference

Abstract Purpose Driver mutations are typically absent in esophageal adenocarcinoma (EAC). Mostly, oncogenes are amplified as driving molecular events (including GATA6-amplification in 14% of cases). However, only little is known about its biological function and clinical relevance. Methods We examined a large number of EAC (n = 496) for their GATA6 amplification by fluorescence in situ hybridization (FISH) analyzing both primary resected (n = 219) and neoadjuvant treated EAC (n = 277). Results were correlated to clinicopathological data and known mutations/amplifications in our EAC-cohort. Results GATA6 amplification was detectable in 49 (9.9%) EACs of our cohort. We observed an enrichment of GATA6-positive tumors among patients after neoadjuvant treatment (12,3% amplified tumors versus 6,8% in the primary resected group; p = 0.044). Additionally, there was a simultaneous amplification of PIK3CA and GATA6 (p < 0.001) not detectable when analyzing other genes such as EGFR, ERBB2, KRAS or MDM2. Although we did not identify a survival difference depending on GATA6 in the entire cohort (p = 0.212), GATA6 amplification was associated with prolonged overall survival among patients with primary surgery (median overall-survival 121.1 vs. 41.4 months, p = 0.032). Multivariate cox-regression analysis did not confirm GATA6 as an independent prognostic marker, neither in the entire cohort (p = 0.210), nor in the subgroup with (p = 0.655) or without pretreatment (p = 0.961). Conclusions Our study investigates the relevance of GATA6 amplification on a large tumor collective, which includes primary resected tumors and the clinically relevant group of neoadjuvant treated EACs. Especially in the pretreated group, we found an accumulation of GATA6-amplified tumors (12.3%) and a frequent co-amplification of PIK3CA. Our data suggest an increased resistance to radio-chemotherapy in GATA6-amplified tumors.

MR-guided laser-induced thermotherapy (LITT) in liver metastases of colorectal cancer: Long-term survival data.

2012 ◽  
Vol 30 (15_suppl) ◽  
pp. e14131-e14131
Author(s):  
Thomas J. Vogl ◽  
Alena Dommermuth ◽  
Katrin Eichler ◽  
Stephan Zangos
Keyword(s):  
Overall Survival ◽  
Survival Analysis ◽  
Survival Rate ◽  
Lymph Nodes ◽  
Liver Metastases ◽  
Median Survival ◽  
Cox Regression ◽  
Term Survival ◽  
Median Survival Rate

e14131 Background: To evaluate retrospectively long-term survival of 594 patients with colorectal liver metastases treated with MR-guided laser-induced thermotherapy (LITT) depending on different factors. Methods: 594 patients with liver metastases from colorectal carcinoma treated with MR-guided LITT between 01/99 and 12/10 were included. For survival analysis tumor localization, TNM classification, number of metastases, diameter and volume of metastases and necrosis, lobular spread, number of treatment sessions, performance of adjuvant chemotherapy and transarterial chemoembolisation were considered. The Kaplan-Meier method was used to conduct this survival analysis. Results: Log-rank test showed statistically significant differences between survival curves, multivariate Cox-regression-analysis (p<0.05) showed prognostic factors regarding overall survival like number of metastases pre intervention, adjuvant chemotherapy, diameter of metastases, ratio of volumes of necrosis and metastases, and affected lymph nodes. Median overall survival rate at the time of first LITT was 25 months, 1-year survival: 78%, 2-year survival: 50.1%, 3-year survival: 28%; 4-year survival: 16.4%; 5-year survival: 7.8%. Numbers of metastases pre intervention: 1-2 metastases with a median survival rate of 60 months; 3-4 metastases: 45 months; ≥5 metastases: 42 months. Median survival rate for metastases <20mm in diameter 36 months; 20-30mm 27 months, 30-40mm 24 months and >40mm 21 months. Affected lymph nodes: median survival rate for patients with N0-classification 30 months, N1-classification 24 months; N2/N3/N4-classification 22 months. Conclusions: Multivariate Cox regression model provided the minimal number of significant variables with the maximal prognostic value concerning overall survival for MR-guided LITT, i.e., diameter and number of metastases and primary classification of lymph nodes.

Overall survival with preoperative biliary drainage in patients with resectable pancreatic cancer.

2014 ◽  
Vol 32 (3_suppl) ◽  
pp. 314-314
Author(s):  
Tobin Joel Crill Strom ◽  
Sarah E. Hoffe ◽  
Shivakumar Vignesh ◽  
Jason Klapman ◽  
Cynthia L. Harris ◽  
...  
Keyword(s):  
Pancreatic Cancer ◽  
Overall Survival ◽  
Cancer Patients ◽  
Tumor Size ◽  
Biliary Drainage ◽  
Cox Regression ◽  
Neoadjuvant Treatment ◽  
Preoperative Biliary Drainage ◽  
Resectable Pancreatic Cancer ◽  
Pathologic Features

314 Background: Resectable pancreatic cancer patients often present with obstructive jaundice necessitating the placement of biliary stents or percutaneouse drainage catheters. We sought to evaluate whether preoperative biliary drainage affects recurrence and survival. Methods: An IRB-approved study was conducted on our institutional tumor registry to identify pancreatic cancer patients who were treated with upfront surgery between 2000 and 2012. Patients were then stratified by preoperative use of endoscopically placed stents (ERCP), percutaneous catheters (PTC), or no biliary drainage (NBD). The primary endpoint was overall survival (OS). Survival curves were calculated using the Kaplan-Meier method and the log-rank test. Multivariate analysis (MVA) was performed with a Cox regression model. Results: We identified 202 patients for the study (21 PTC; 89 ERCP; 92 NBD). Key differences between the 3 groups were mean pathologic tumor size (p=0.005), pathologic T3/4 (p =0.01), and pathologic N1 (p=0.007) status, with more aggressive pathologic features in PTC patients. PTC patients had a non-significant increase in rate of hepatic recurrences compared with ERCP and NBD patients (47.4% vs. 26.6% vs. 28.7%, respectively; p=0.20). PTC patients also had worse median and 3 year survival (21 months and 16%) compared to ERCP (23.3 months and 39%) and NBD patients (29 months and 45%, p=0.02). MVA revealed that PTC was an independent predictor of worse overall survival (HR 2.3[95% CI 1.3-4.0], p=0.005), along with pathologic tumor size (HR 1.1[1.0-1.3], p=0.008), nodes positive (HR 1.1[1.1-1.2], p=0.001), and post-operative CA19-9 >90 (HR 2.6[1.5-4.4], p=0.001). Conclusions: Patients with resectable pancreatic cancer who require a pre-operative PTC drain had a non-significant increase in hepatic recurrence rate and worse overall survival than patients who either had an ERCP stent placed or no biliary decompression prior to surgery. Given their worse prognosis, patients who require PTC placement might also benefit from neoadjuvant treatment with restaging prior to surgery.

Prognosis of gastric cancer (GC) patients with positive peritoneal cytology.

2015 ◽  
Vol 33 (3_suppl) ◽  
pp. 41-41
Author(s):  
Hironori Shiozaki ◽  
Elena Elimova ◽  
Rebecca Slack ◽  
Hsiang-Chun Chen ◽  
Gregg A Staerkel ◽  
...  
Keyword(s):  
Overall Survival ◽  
Diagnostic Laparoscopy ◽  
Cox Model ◽  
Peritoneal Metastases ◽  
Patient Treatment ◽  
Cancer Center ◽  
Peritoneal Cytology ◽  
Term Survival ◽  
Risk Of Death ◽  
One Year

41 Background: Laparoscopic staging of patients with GC can disclose peritoneal metastases. Although this finding is associated with a poor prognosis, some patients achieve a long-term survival. In an attempt to provide explanation we compared the overall survival (OS) of patients with GC peritoneal metastases from two settings: cytology positive only (Cy+) and grossly positive (Gross+). Methods: 146 GC patients with peritoneal metastases were identified between 2000 and 2014. Cox-model regression was used for overall survival (OS) analyses. Results: Patient/treatment characteristics were as follows: males (66%), good ECOG scores (0-1; 89%), metastases confirmed by a diagnostic laparoscopy (84%), poorly differentiated histology(92%), received chemotherapy (89%), received chemoradiation (22%), and received surgery (10%). The median follow-up time for all patients was 12.9 months and median OS was 15 months. Patients with Gross+ were at higher risk of death compared to Cy+ patients (50% vs. 83%1-year OS, respectively). Only diagnostic laparoscopy and metastasis type (Gross+ vs. Cy+) were significant in both univariate and multivariate OS models. With both factors in the same model, patients with Gross+ were more than twice as likely to die when compared to those with Cy+ (HR=2.23; p=0.001) while patients having a diagnostic laparoscopy were half as likely to die (HR=0.52; p=0.01). Conclusions: The one-year OS of patients with Cy+ peritoneal metastases is significantly longer than those with Gross+ findings. As such, novel strategies for Cy+ patients may further prolong their survival. From U. T. M. D. Anderson Cancer Center (UTMDACC), Houston, Texas, USA. (Supported in part by UTMDACC, and CA 138671 and CA172741 from the NCI).

scholarly journals Nomograms Predicting Long-Term Overall Survival and Cancer-Specific Survival in Metastatic Hepatocellular Carcinoma Patients

2020 ◽  
Author(s):  
Chi Cui ◽  
Yaru Duan ◽  
Rui Li ◽  
Hua Ye ◽  
Peng Wang ◽  
...  
Keyword(s):  
Hepatocellular Carcinoma ◽  
Overall Survival ◽  
Cox Regression ◽  
Training Group ◽  
Clinicopathological Characteristics ◽  
Validation Group ◽  
Cancer Specific Survival ◽  
Term Survival ◽  
Metastatic Hepatocellular Carcinoma

Abstract Background This study aims to evaluate the clinicopathological characteristics of metastatic hepatocellular carcinoma (HCC) patients and develop nomograms to predict their long-term overall survival (OS) and cancer-specific survival (CSS). Methods Information on metastatic HCC from 2010 to 2015 was retrieved from the Surveillance, Epidemiology and End Results (SEER) program of the National Cancer Institute. The metastatic HCC patients were divided into a long-term survival (LTS) group and a short-term survival (STS) group with 1 year selected as the cut-off value. Then, we compared the demographic and clinicopathological features between the two groups. Next, all patients were randomly divided into a training group and validation group at a 7:3 ratio. Univariate and multivariate Cox regression analyses were used to identify potential predictors for OS and CSS in the training group, and nomograms of OS and CSS were established. These predictive models were further validated in the validation group. Results A total of 2163 patients were included in the current study according to the inclusion and exclusion criteria. Patients with characteristics including lower T stage and N stage; treatment with surgery, radiation or chemotherapy; no lung metastasis; and AFP negative status showed better survival. The concordance index (C-index) of the OS nomogram was 0.72 based on 9 variables. The C-index of the CSS nomogram was 0.71 based on 8 variables. Conclusions These nomograms may help clinicians make better treatment recommendations for metastatic HCC patients.

scholarly journals Prediction of Overall Survival Rate in Patients With Hepatocellular Carcinoma Using an Integrated Model Based on Autophagy Gene Marker

2021 ◽  
Vol 12 ◽  
Author(s):  
Shuaiqun Wang ◽  
Dalu Yang ◽  
Wei Kong
Keyword(s):  
Hepatocellular Carcinoma ◽  
Overall Survival ◽  
Survival Rate ◽  
Clinical Data ◽  
Cox Regression ◽  
Cox Model ◽  
Pathway Enrichment Analysis ◽  
Gene Marker ◽  
Overall Survival Rate ◽  
Cox Regression Analysis

The autophagy cell, which can inhibit the formation of tumor in the early stage and can promote the development of tumor in the late stage, plays an important role in the development of tumor. Therefore, it has potential significance to explore the influence of autophagy-related genes (AAGs) on the prognosis of hepatocellular carcinoma (HCC). The differentially expressed AAGs are selected from HCC gene expression profile data and clinical data downloaded from the TCGA database, and human autophagy database (HADB). The role of AAGs in HCC is elucidated by GO functional annotation and KEGG pathway enrichment analysis. Combining with clinical data, we selected age, gender, grade, stage, T state, M state, and N state as Cox model indexes to construct the multivariate Cox model and survival curve of Kaplan Meier (KM) was drawn to estimate patients’ survival between high- and low-risk groups. Through an ROC curve drawn by univariate and multivariate Cox regression analysis, we found that seven genes with high expression levels, including HSP90AB1, SQSTM1, RHEB, HDAC1, ATIC, HSPB8, and BIRC5 were associated with poor prognosis of HCC patients. Then the ICGC database is used to verify the reliability and robustness of the model. Therefore, the prognosis model of HCC constructed by autophagy genes might effectively predict the overall survival rate and help to find the best personalized targeted therapy of patients with HCC, which can provide better prognosis for patients.

Sign in / Sign up

Export Citation Format

Share Document