|Table of Contents|

Development of machine learning models and nomogram to predict epithelial ovarian cancer recurrence

Journal Of Modern Oncology[ISSN:1672-4992/CN:61-1415/R]

Issue:
2023 07
Page:
1276-1280
Research Field:
Publishing date:

Info

Title:
Development of machine learning models and nomogram to predict epithelial ovarian cancer recurrence
Author(s):
YANG Lirong; TAN Jincheng; GAO Xia; DAI Xiao; LI Feng; LIU Zhijin; WANG Yufeng
Department of Geriatric Oncology,Yunan Cancer Hospital,Yunnan Kunming 650100,China.
Keywords:
epithelial ovarian cancer; progression free survival; prognostic analysis; machine learning; nomogram
PACS:
R737.31
DOI:
10.3969/j.issn.1672-4992.2023.07.018
Abstract:
Objective:To develop machine learning models and a nomogram to predict epithelial ovarian cancer recurrence based on machine learning and Cox regression.Methods:The medical records of 739 patients diagnosed with stage III-IV EOC in Yunnan Cancer Hospital from January 2010 to December 2020 were retrospectively analyzed.Basic information,surgery,chemotherapy details and survival results were collected.Univariate and multivariate logistic regression and Cox regression were used to screen variables.Five machine learning algorithms were used to construct models based on the results of multivariate logistic regression.The model performance was evaluated by 10-fold cross-validation.The nomogram was developed based on the Cox regression results.Results:A total of 739 patients entered the study,among them,399(54.0%) eventually recurred and 340(46%) were censored.Stage ⅢC predominates in relapsed populations,accounting for 59.1%.The pathological type was mainly serous carcinoma,accounting for 91.0%.Multivariate logistic regression showed that the perioperative chemotherapy cycle,residual tumorsize,surgical method,and neoadjuvant chemotherapy were four variables independently related to recurrence.Based on these variables and the FIGO stage,five machine learning models were established.The XGBoost-based model performed best with an AUC of 0.775.Cox regression showed that local infusion chemotherapy before surgery,residual tumor,perioperative chemotherapy cycle,and surgical approach were independent prognostic risk factors.A nomogram for predicting recurrence in patients with advanced epithelial ovarian cancer was developed based on the four factors.Conclusion:Machine learning models and nomogram could identify ovarian cancer recurrence early,demonstrating the potential to improve the prognosis of advanced ovarian cancer through early detection.

References:

[1]SUNG H,FERLAY J,SIEGEL RL,et al.Global cancer statistics 2020:GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries ［J］.CA:A Cancer Journal for Clinicians,2021,71(3):209-249.
[2]ARMSTRONG DK,ALVAREZ RD,BAKKUM-GAMEZ JN,et al.Ovarian cancer,version 2.2020,NCCN clinical practice guidelines in oncology ［J］.Journal of the National Comprehensive Cancer Network,2021,19(2):191-226.
[3]LIM MC,WON YJ,KO MJ,et al.Incidence of cervical,endometrial,and ovarian cancer in Korea during 1999-2015 ［J］.J Gynecol Oncol,2019,30(1):e38.
[4]中国抗癌协会妇科肿瘤专业委员会.卵巢恶性肿瘤诊断与治疗指南(2021年版)［J］.中国癌症杂志,2021,31(6):490-500. Committee of Gynecological Oncology,Chinese Anti-Cancern Association.Guidelines for the diagnosis and treatment of ovarian Malignancies(2021 edition)［J］.China Oncology,2021,31(6):490-500.
[5]HARTER P,SEHOULI J,VERGOTE I,et al.Randomized trial of cytoreductive surgery for relapsed ovarian cancer ［J］.The New England Journal of Medicine,2021,385(23):2123-2131.
[6]HWANGBO S,KIM SI,KIM JH,et al.Development of machine learning models to predict platinum sensitivity of high-grade serous ovarian carcinoma ［J］.Cancers(Basel),2021,13(8):1-14.
[7]KIM SI,SONG M,HWANGBO S,et al.Development of web-based nomograms to predict treatment response and prognosis of epithelial ovarian cancer ［J］.Cancer Res Treat,2019,51(3):1144-1155.
[8]WEN KC,SUNG PL,LAI HC.The prognostic nomogram in platinum-resistant ovarian cancer:how to develop and validate［J］.Chin Clin Oncol,2021,10(3):31.
[9]MA J,YANG J,JIN Y,et al.Artificial intelligence based on blood biomarkers including CTCs predicts outcomes in epithelial ovarian cancer:A prospective study ［J］.Onco Targets Ther,2021,14:3267-3280.
[10]GOULD MK,HUANG BZ,TAMMEMAGI MC,et al.Machine learning for early lung cancer identification using routine clinical and laboratory data ［J］.American Journal of Respiratory and Critical Care Medicine,2021,204(4):445-453.
[11]AKAZAWA M,HASHIMOTO K.Artificial intelligence in ovarian cancer diagnosis ［J］.Anticancer Res,2020,40(8):4795-4800.
[12]GRIMLEY PM,LIU Z,DARCY KM,et al.A prognostic system for epithelial ovarian carcinomas using machine learning ［J］.Acta Obstet Gynecol Scand,2021,100(8):1511-1519.
[13]DINCA AL,BIRLA RD,DINCA VG,et al.Prognostic factors in advanced ovarian cancer - a clinical trial ［J］.Chirurgia(Bucur),2020,115(1):50-62.
[14]DE LIMA CA,SILVA RODRIGUES IS,MARTINS-FILHO A,et al.Cytokines in peritoneal fluid of ovarian neoplasms ［J］.J Obstet Gynaecol,2020,40(3):401-405.
[15]JIA D,NAGAOKA Y,KATSUMATA M,et al.Inflammation is a key contributor to ovarian cancer cell seeding ［J］.Sci Rep,2018,8(1):1-7.
[16]SONE K,TOYOHARA Y,TAGUCHI A,et al.Application of artificial intelligence in gynecologic malignancies:A review ［J］.J Obstet Gynaecol Res,2021,47(8):2577-2585.
[17]GONG TT,HE XH,GAO S,et al.Application of machine learning in prediction of chemotherapy resistant of ovarian cancer based on gut microbiota ［J］.J Cancer,2021,12(10):2877-2885.
[18]MCDONALD JF.Back to the future-the integration of big data with machine learning is re-establishing the importance of predictive correlations in ovarian cancer diagnostics and therapeutics ［J］.Gynecol Oncol,2018,149(2):230-231.
[19]VAN DRIEL WJ,KOOLE SN,SIKORSKA K,et al.Hyperthermic intraperitoneal chemotherapy in ovarian cancer ［J］.The New England Journal of Medicine,2018,378(3):230-240.
[20]KEHOE S,HOOK J,NANKIVELL M,et al.Primary chemotherapy versus primary surgery for newly diagnosed advanced ovarian cancer(CHORUS):an open-label,randomised,controlled,non-inferiority trial ［J］.Lancet(London,England),2015,386(9990):249-257.
[21]SINUKUMAR S,DAMODARAN D,RAY M,et al.Pattern of recurrence after interval cytoreductive surgery and HIPEC following neoadjuvant chemotherapy in primary advanced stage IIIC/IVA epithelial ovarian cancer ［J］.Eur J Surg Oncol,2021,47(6):1427-1433.
[22]MUELLER JJ,ZHOU QC,IASONOS A,et al.Neoadjuvant chemotherapy and primary debulking surgery utilization for advanced-stage ovarian cancer at a comprehensive cancer center ［J］.Gynecol Oncol,2016,140(3):436-442.
[23]SNEIGE N,THOMISON JB,MALPICA A,et al.Peritoneal washing cytologic analysis of ovarian serous tumors of low malignant potential to detect peritoneal implants and predict clinical outcome ［J］.Cancer Cytopathology,2012,120(4):238-244.
[24]GULZAR R,SHAHID R,MUMTAZ S,et al.Significance of peritoneal washing cytology in the accurate staging of malignant ovarian tumors ［J］.Pakistan Journal of Medical Sciences,2022,38(1):128-132.
[25]ZHOU J,LI L,WANG L,et al.Establishment of a SVM classifier to predict recurrence of ovarian cancer ［J］.Mol Med Rep,2018,18(4):3589-3598.
[26]HUEMAN M,WANG H,LIU Z,et al.Expanding TNM for lung cancer through machine learning ［J］.Thorac Cancer,2021,12(9):1423-1430.
[27]KIM SI,HA HI,EOH KJ,et al.Trends in the incidence and survival rates of primary ovarian clear cell carcinoma compared to ovarian serous carcinoma in Korea ［J］.Front Oncol,2022,12:874037.
[28]HADA T,MIYAMOTO M,ISHIBASHI H,et al.Comparison of clinical behavior between mucinous ovarian carcinoma with infiltrative and expansile invasion and high-grade serous ovarian carcinoma:a retrospective analysis ［J］.Diagnostic Pathology,2022,17(1):12.
[29]RAY-COQUARD I,PAUTIER P,PIGNATA S,et al.Olaparib plus bevacizumab as first-line maintenance in ovarian cancer ［J］.The New England Journal of Medicine,2019,381(25):2416-2428.
[30]NAKAMURA M,BAX HJ,SCOTTO D,et al.Immune mediator expression signatures are associated with improved outcome in ovarian carcinoma ［J］.Oncoimmunology,2019,8(6):e1593811.
[31]HART SN,POLLEY EC,SHIMELIS H,et al.Prediction of the functional impact of missense variants in BRCA1 and BRCA2 with BRCA-ML ［J］.NPJ Breast Cancer,2020,6(13):1-4.

Memo

Memo:
昆明医科大学研究生创新基金项目（编号：2022S317）
Last Update: 2023-02-28