Share：
Share this content in WeChat
X
[Chinese][PDF]3504
• Clinical Article •
The value of machine-learning-based radiomics models for predicting disease-free survival and immune levels in endometrial cancer patients
CHEN Shuqing  ZHANG Yu  CHEN Dong  CHANG Xibao  LIU Jingjing  CHEN Lei  QIAN Yinfeng 

Cite this article as: CHEN S Q, ZHANG Y, CHEN D, et al. The value of machine-learning-based radiomics models for predicting disease-free survival and immune levels in endometrial cancer patients[J]. Chin J Magn Reson Imaging, 2024, 15(9): 107-113. DOI:10.12015/issn.1674-8034.2024.09.018.


[Abstract] Objective To investigate the predictive value of machine learning-based radiomics model for disease-free survival (DFS) in endometrial cancer patients. Matirials andMethods Data from 212 endometrial cancer patients who had undergone radical surgery in a dual-center were retrospectively analyzed. Radiomics features of tumor and peri-tumor 5 mm region in T2WI sequences were extracted for all patients. Five machine learning methods (gradient boosting machines, the least absolute shrinkage and selection operator, random survival forest, support vector machine, and extreme gradient boosting) were used to construct the radiomics model and calculate the best radiomics score (Radscore). The incremental value of Radscore to existing clinical predictors was analysed and a combined model was constructed. Finally, bioinformatics analysis was used to reveal the biological mechanisms of the radiomics models.Results The combined radiomics model based on gradient boosting machines showed the best predictive efficacy, with AUC of 0.977, 0.986, 0.995 and 0.745, 0.764, 0.802 for predicting 1-, 3-, and 5-year DFS in the training and validation sets, respectively. Multifactorial Cox regression analyses showed that clinical stage, carbohydrate antigen 125 (CA125), and Radscore were the independent predictors of DFS. The area under the curve (AUC) of the combined model in the training and validation sets were 0.926, 0.894, 0.864 and 0.828, 0.839, 0.873 for predicting 1-, 3-, and 5-year DFS. Meanwhile, bioinformatics analysis suggested that Radscore was significantly correlated with the immune level of endometrial cancer patients.Conclusions Machine learning-based radiomics model is helpful for the prediction of DFS and immune levels in endometrial cancer patients. The combination of radiomics and clinical indicators can further improve the accuracy of prediction and provide a reference basis for prognostic prediction and individualized treatment of endometrial cancer patients.
[Keywords] endometrial cancer；magnetic resonance imaging；radiomics；machine learning；disease-free survival

CHEN Shuqing1   ZHANG Yu2   CHEN Dong2   CHANG Xibao1   LIU Jingjing1   CHEN Lei1   QIAN Yinfeng3*  

1 Department of Imaging, Funan Hospital Affiliated Fuyang Normal University Medical College, Fuyang 236300, China

2 Department of Imaging, the First Affiliated Hospital, University of Science and Technology of China, Hefei 230001, China

3 Department of magnetic resonance, the First Affiliated Hospital of Anhui Medical University, Hefei 230022, China

Corresponding author: QIAN Y F, E-mail: 894206876@qq.com

Conflicts of interest   None.

Received  2024-04-28
Accepted  2024-08-26
DOI: 10.12015/issn.1674-8034.2024.09.018
Cite this article as: CHEN S Q, ZHANG Y, CHEN D, et al. The value of machine-learning-based radiomics models for predicting disease-free survival and immune levels in endometrial cancer patients[J]. Chin J Magn Reson Imaging, 2024, 15(9): 107-113. DOI:10.12015/issn.1674-8034.2024.09.018.

[1]
CROSBIE E J, KITSON S J, MCALPINE J N, et al. Endometrial cancer[J]. Lancet, 2022, 399(10333): 1412-1428. DOI: 10.1016/S0140-6736(22)00323-3.
[2]
MAHESHWARI E, NOUGARET S, STEIN E B, et al. Update on MRI in evaluation and treatment of endometrial cancer[J]. Radiographics, 2022, 42(7): 2112-2130. DOI: 10.1148/rg.220070.
[3]
SHI R, ZHANG W, ZHANG J, et al. CircESRP1 enhances metastasis and epithelial-mesenchymal transition in endometrial cancer via the miR-874-3p/CPEB4 axis[J/OL]. J Transl Med, 2022, 20(1): 139 [2024-08-08]. https://pubmed.ncbi.nlm.nih.gov/35317822/. DOI: 10.1186/s12967-022-03334-6.
[4]
SHIN D W, JUNG K W, HA J, et al. Conditional relative survival of patients with endometrial cancer: a Korean National Cancer Registry study[J/OL]. J Gynecol Oncol, 2022, 33(2): e23 [2024-08-08]. https://pubmed.ncbi.nlm.nih.gov/35245001/. DOI: 10.3802/jgo.2022.33.e23.
[5]
WANG X M, GLUBB D M, O'MARA T A. Dietary factors and endometrial cancer risk: a Mendelian randomization study[J/OL]. Nutrients, 2023, 15(3): 603 [2024-08-08]. https://pubmed.ncbi.nlm.nih.gov/36771310/. DOI: 10.3390/nu15030603.
[6]
CONTRERAS N A, SABADELL J, VERDAGUER P, et al. Fertility-sparing approaches in atypical endometrial hyperplasia and endometrial cancer patients: current evidence and future directions[J/OL]. Int J Mol Sci, 2022, 23(5): 2531 [2024-08-08]. https://pubmed.ncbi.nlm.nih.gov/35269674/. DOI: 10.3390/ijms23052531.
[7]
ESKANDER R N, SILL M W, BEFFA L, et al. Pembrolizumab plus Chemotherapy in Advanced Endometrial Cancer[J]. N Engl J Med, 2023, 388(23): 2159-2170. DOI: 10.1056/NEJMoa2302312.
[8]
MIRZA M R, CHASE D M, SLOMOVITZ B M, et al. Dostarlimab for primary advanced or recurrent endometrial cancer[J]. N Engl J Med, 2023, 388(23): 2145-2158. DOI: 10.1056/NEJMoa2216334.
[9]
TRONCONI F, NERO C, GIUDICE E, et al. Advanced and recurrent endometrial cancer: state of the art and future perspectives[J/OL]. Crit Rev Oncol Hematol, 2022, 180: 103851 [2024-08-08]. https://pubmed.ncbi.nlm.nih.gov/36257537/. DOI: 10.1016/j.critrevonc.2022.103851.
[10]
OROPEZA-DE LARA S A, GARZA-VELOZ I, BERTHAUD-GONZÁLEZ B, et al. Circulating and endometrial tissue microRNA markers associated with endometrial cancer diagnosis, prognosis, and response to treatment[J/OL]. Cancers, 2023, 15(10): 2686 [2024-08-08]. https://pubmed.ncbi.nlm.nih.gov/36257537/. DOI: 10.3390/cancers15102686.
[11]
ZHANG M, JING M Y, CAO Y T, et al. Clinical value of apparent diffusion coefficient in preoperative prediction of myometrial infiltration, Ki-67 and P53 expression levels in endometrial cancer[J]. Chin J Magn Reson Imaging, 2023, 14(7): 78-85. DOI: 10.12015/issn.1674-8034.2023.07.014.
[12]
ZHANG Y, LIU L, ZHANG K Y, et al. Nomograms combining clinical and imaging parameters to predict recurrence and disease-free survival after concurrent chemoradiotherapy in patients with locally advanced cervical cancer[J]. Acad Radiol, 2023, 30(3): 499-508. DOI: 10.1016/j.acra.2022.08.002.
[13]
LEFEBVRE T L, UENO Y, DOHAN A, et al. Development and validation of multiparametric MRI-based radiomics models for preoperative risk stratification of endometrial cancer[J]. Radiology, 2022, 305(2): 375-386. DOI: 10.1148/radiol.212873.
[14]
KIDO A, NISHIO M. MRI-based radiomics models for pretreatment risk stratification of endometrial cancer[J]. Radiology, 2022, 305(2): 387-389. DOI: 10.1148/radiol.221398.
[15]
MATUREN K E, MARTIN M F, CHAPMAN C H, et al. Pelvic recovery after endometrial cancer treatment: patient-reported outcomes and MRI findings[J]. Acad Radiol, 2023, 30(Suppl 2): S202-S210. DOI: 10.1016/j.acra.2023.03.031.
[16]
LIU C L, CAI Q, SHEN Y Y, et al. Establishment of predictive model for surgical resection of ground glass nodules combined with peritumoral radiomics[J]. J Clin Radiol, 2024, 43(3): 376-382. DOI: 10.3969/j.issn.1006-5741.2021.04.004.
[17]
COADA C A, SANTORO M, ZYBIN V, et al. A radiomic-based machine learning model predicts endometrial cancer recurrence using preoperative CT radiomic features: a pilot study[J/OL]. Cancers, 2023, 15(18): 4534 [2024-08-08]. https://pubmed.ncbi.nlm.nih.gov/37760503/. DOI: 10.3390/cancers15184534.
[18]
LI X F, MARCUS D, RUSSELL J, et al. An integrated clinical-MR radiomics model to estimate survival time in patients with endometrial cancer[J]. J Magn Reson Imaging, 2023, 57(6): 1922-1933. DOI: 10.1002/jmri.28544.
[19]
JACOB H, DYBVIK J A, YTRE-HAUGE S, et al. An MRI-based radiomic prognostic index predicts poor outcome and specific genetic alterations in endometrial cancer[J/OL]. J Clin Med, 2021, 10(3): 538 [2024-08-08]. https://pubmed.ncbi.nlm.nih.gov/33540589/. DOI: 10.3390/jcm10030538.
[20]
ABU-RUSTUM N, YASHAR C, AREND R, et al. Uterine neoplasms, version 1.2023, NCCN clinical practice guidelines in oncology[J]. J Natl Compr Canc Netw, 2023, 21(2): 181-209. DOI: 10.6004/jnccn.2023.0006.
[21]
YANG J H, SHI W J, YANG Z, et al. Establishing a predictive model for tumor mutation burden status based on CT radiomics and clinical features of non-small cell lung cancer patients[J]. Transl Lung Cancer Res, 2023, 12(4): 808-823. DOI: 10.21037/tlcr-23-171.
[22]
CHEN Q Y, ZHANG L, LIU S Y, et al. Radiomics in precision medicine for gastric cancer: opportunities and challenges[J]. Eur Radiol, 2022, 32(9): 5852-5868. DOI: 10.1007/s00330-022-08704-8.
[23]
FASMER K E, HODNELAND E, DYBVIK J A, et al. Whole-volume tumor MRI radiomics for prognostic modeling in endometrial cancer[J]. J Magn Reson Imaging, 2021, 53(3): 928-937. DOI: 10.1002/jmri.27444.
[24]
ZHANG K Y, ZHANG Y, FANG X, et al. MRI-based radiomics and ADC values are related to recurrence of endometrial carcinoma: a preliminary analysis[J/OL]. BMC Cancer, 2021, 21(1): 1266 [2024-08-08]. https://pubmed.ncbi.nlm.nih.gov/34819042/. DOI: 10.1186/s12885-021-08988-x.
[25]
GILL R R. Virtual image-based biopsy of lung metastases: the promise of radiomics[J]. Acad Radiol, 2023, 30(1): 47-48. DOI: 10.1016/j.acra.2022.10.030.
[26]
ZHENG Y L, ZHOU D, LIU H, et al. CT-based radiomics analysis of different machine learning models for differentiating benign and malignant parotid tumors[J]. Eur Radiol, 2022, 32(10): 6953-6964. DOI: 10.1007/s00330-022-08830-3.
[27]
GITTO S, CUOCOLO R, VAN LANGEVELDE K, et al. MRI radiomics-based machine learning classification of atypical cartilaginous tumour and grade Ⅱ chondrosarcoma of long bones[J/OL]. EBioMedicine, 2022, 75: 103757 [2024-08-08]. https://pubmed.ncbi.nlm.nih.gov/34933178/. DOI: 10.1016/j.ebiom.2021.103757.
[28]
YU Y F, HE Z F, OUYANG J, et al. Magnetic resonance imaging radiomics predicts preoperative axillary lymph node metastasis to support surgical decisions and is associated with tumor microenvironment in invasive breast cancer: A machine learning, multicenter study[J/OL]. EBioMedicine, 2021, 69: 103460 [2024-08-08]. https://pubmed.ncbi.nlm.nih.gov/34233259/. DOI: 10.1016/j.ebiom.2021.103460.
[29]
LI W, LI Y, LIU X L, et al. Machine learning-based radiomics for predicting BRAF-V600E mutations in ameloblastoma[J/OL]. Front Immunol, 2023, 14: 1180908 [2024-08-08]. https://pubmed.ncbi.nlm.nih.gov/37646022/. DOI: 10.3389/fimmu.2023.1180908.
[30]
BEREK J S, MATIAS-GUIU X, CREUTZBERG C, et al. FIGO staging of endometrial cancer: 2023[J]. Int J Gynaecol Obstet, 2023, 162(2): 383-394. DOI: 10.1002/ijgo.14923.
[31]
ZHOU Z Q, SUN M, DAI A J. Relationship between preoperative serum carbohydrate antigen 125, vascular endothelial growth factor and pyruvate kinase M2 expression and recurrence of endometrial cancer[J]. Chin J Clin, 2020, 48(8): 968-971. DOI: 10.3969/j.issn.2095-8552.2020.08.029.
[32]
WU H H, CHOU H T, TSENG J Y, et al. The relationship between serum CA-125 level and recurrence in surgical stage I endometrial cancer patients[J]. J Chin Med Assoc, 2023, 86(11): 1001-1007. DOI: 10.1097/JCMA.0000000000000985.
[33]
SALMINEN L, NADEEM N, ROLFSEN A L, et al. Exploratory analysis of CA125-MGL and-STn glycoforms in the differential diagnostics of pelvic masses[J]. J Appl Lab Med, 2020, 5(2): 263-272. DOI: 10.1093/jalm/jfz012.
[34]
TONG H P, SUN J J, FANG J Q, et al. A machine learning model based on PET/CT radiomics and clinical characteristics predicts tumor immune profiles in non-small cell lung cancer: a retrospective multicohort study[J/OL]. Front Immunol, 2022, 13: 859323 [2024-08-08]. https://pubmed.ncbi.nlm.nih.gov/35572597/. DOI: 10.3389/fimmu.2022.859323.

PREV Study of ADC minimum in combination with clinical and imaging features for prediction of HIFU efficacy in treatment of uterine fibroids
NEXT MRI features and clinical correlation analysis of thigh muscle in children with spinal muscular atrophy
  


LINK


Tel & Fax: +8610-67113815    E-mail: editor@cjmri.cn