Access the full text.
Sign up today, get DeepDyve free for 14 days.
Quan Quan, Qianqian Liao, Wanchun Yin, Shuwei Zhou, Sainan Gong, Xiaoling Mu (2021)
Serum HE4 and CA125 combined to predict and monitor recurrence of type II endometrial carcinomaScientific Reports, 11
F. Teng, yanfang zhang, Yingmei Wang, Jing Yu, Xu Lang, W. Tian, C. Jiang, F. Xue (2015)
Contrast‐enhanced MRI in preoperative assessment of myometrial and cervical invasion, and lymph node metastasis: diagnostic value and error analysis in endometrial carcinomaActa Obstetricia et Gynecologica Scandinavica, 94
Xueyan Jiang, Haodong Jia, Zhongyuan Zhang, Chao Wei, Chuanbin Wang, Jiangning Dong (2022)
The Feasibility of Combining ADC Value With Texture Analysis of T2WI, DWI and CE-T1WI to Preoperatively Predict the Expression Levels of Ki-67 and p53 of Endometrial CarcinomaFrontiers in Oncology, 11
B. Geppert, C. Lönnerfors, M. Bollino, Anastasija Arechvo, J. Persson (2017)
A study on uterine lymphatic anatomy for standardization of pelvic sentinel lymph node detection in endometrial cancer.Gynecologic oncology, 145 2
Wenhui Zhong, Chunyu Zhou, Lufei Chen, Zhenna Wang, Hongxing Lin, Kunhai Wu, Sujiao Zhang (2021)
The Coefficient of Variation of Red Blood Cell Distribution Width Combined with Cancer Antigen 125 Predicts Postoperative Overall Survival in Endometrial CancerInternational Journal of General Medicine, 14
Evis Sala, E. Mema, E. Mema, Yuki Himoto, H. Veeraraghavan, James Brenton, A. Snyder, B. Weigelt, H. Vargas (2017)
Unravelling tumour heterogeneity using next-generation imaging: radiomics, radiogenomics, and habitat imaging.Clinical radiology, 72 1
Gillian Lai, A. Rockall (2010)
Lymph node imaging in gynecologic malignancy.Seminars in ultrasound, CT, and MR, 31 5
(2021)
CA: A
Jiangdian Song, Jingyun Shi, D. Dong, M. Fang, W. Zhong, Kun Wang, N. Wu, Yanqi Huang, Zhenyu Liu, Yue Cheng, Y. Gan, Yongzhao Zhou, P. Zhou, Bojiang Chen, C. Liang, Zaiyi Liu, Wei-min Li, Jie Tian (2018)
A New Approach to Predict Progression-free Survival in Stage IV EGFR-mutant NSCLC Patients with EGFR-TKI TherapyClinical Cancer Research, 24
Huifang Lei, Shuxia Xu, X. Mao, Xiaoying Chen, Yaojia Chen, Xiaoqi Sun, P. Sun (2021)
Systemic Immune-Inflammatory Index as a Predictor of Lymph Node Metastasis in Endometrial CancerJournal of Inflammation Research, 14
M. Plante, Jonathan Stanleigh, M. Renaud, A. Sebastianelli, K. Grondin, J. Grégoire (2017)
Isolated tumor cells identified by sentinel lymph node mapping in endometrial cancer: Does adjuvant treatment matter?Gynecologic oncology, 146 2
Antonio Bandala-Jacques, D. Cantú-de-León, D. Pérez‐Montiel, R. Salcedo-Hernández, D. Prada, A. González-Enciso, A. Gonzalez-Valdés, S. Barquet-Muñoz (2020)
Diagnostic performance of intraoperative assessment in grade 2 endometrioid endometrial carcinomaWorld Journal of Surgical Oncology, 18
G. Nakai, M. Matsuki, Y. Inada, F. Tatsugami, M. Tanikake, I. Narabayashi, Takashi Yamada (2008)
Detection and Evaluation of Pelvic Lymph Nodes in Patients With Gynecologic Malignancies Using Body Diffusion-Weighted Magnetic Resonance ImagingJournal of Computer Assisted Tomography, 32
Erica Cappelletti, Jonas Humann, R. Torrejón, P. Gambadauro (2021)
Chances of pregnancy and live birth among women undergoing conservative management of early-stage endometrial cancer: a systematic review and meta-analysisHuman Reproduction Update, 28
Yu Zhang, Kaiyue Zhang, Hao Jia, Xin Fang, Ting-Ting Lin, Chao Wei, L. Qian, Jiangning Dong (2021)
Feasibility of Predicting Pelvic Lymph Node Metastasis Based on IVIM-DWI and Texture Parameters of the Primary Lesion and Lymph Nodes in Patients with Cervical Cancer.Academic radiology
W. Marsden (2012)
I and J
Ying Yuan, Jiliang Ren, X. Tao (2021)
Machine learning–based MRI texture analysis to predict occult lymph node metastasis in early-stage oral tongue squamous cell carcinomaEuropean Radiology, 31
(2014)
WHO Classification of Tumors of Female Reproductive Organs
Yujie Li, Pei-shan Cong, Pan Wang, Chong Peng, Mingjun Liu, Guirong Sun (2019)
Risk factors for pelvic lymph node metastasis in endometrial cancerArchives of Gynecology and Obstetrics, 300
H. Eggemann, T. Ignatov, K. Kaiser, E. Burger, S. Costa, A. Ignatov (2016)
Survival advantage of lymphadenectomy in endometrial cancerJournal of Cancer Research and Clinical Oncology, 142
N. Dubrawsky (1989)
Cancer statisticsCA: A Cancer Journal for Clinicians, 39
X. Hua, Wei Zhao, A. Pesatori, D. Consonni, N. Caporaso, Tongwu Zhang, B. Zhu, Mingyi Wang, Kristine Jones, B. Hicks, L. Song, J. Sampson, D. Wedge, Jianxin Shi, M. Landi (2019)
Genetic and epigenetic intratumor heterogeneity impacts prognosis of lung adenocarcinomaNature Communications, 11
P. Buamah (2000)
Benign conditions associated with raised serum CA‐125 concentrationJournal of Surgical Oncology, 75
F. Kato, K. Kudo, H. Yamashita, M. Baba, Ai Shimizu, N. Oyama-Manabe, R. Kinoshita, Ruijiang Li, H. Shirato (2019)
Predicting metastasis in clinically negative axillary lymph nodes with minimum apparent diffusion coefficient value in luminal A-like breast cancerBreast Cancer
Yosuke Ito, Y. Terao, Shohei Noma, M. Tagami, E. Yoshida, Y. Hayashizaki, M. Itoh, H. Kawaji (2021)
Nanopore sequencing reveals TACC2 locus complexity and diversity of isoforms transcribed from an intronic promoterScientific Reports, 11
E. Yoshida, Y. Terao, N. Hayashi, K. Mogushi, A. Arakawa, Y. Tanaka, Yosuke Ito, Hiroko Ohmiya, Y. Hayashizaki, S. Takeda, M. Itoh, H. Kawaji (2017)
Promoter-level transcriptome in primary lesions of endometrial cancer identified biomarkers associated with lymph node metastasisScientific Reports, 7
J. Casarin, G. Bogani, E. Piovano, F. Falcone, F. Ferrari, F. Odicino, A. Puppo, F. Bonfiglio, N. Donadello, C. Pinelli, A. Laganà, A. Ditto, M. Malzoni, S. Greggi, F. Raspagliesi, F. Ghezzi (2020)
Survival implication of lymphadenectomy in patients surgically treated for apparent early-stage uterine serous carcinomaJournal of Gynecologic Oncology, 31
C. Roy, G. Bierry, A. Matau, G. Bazille, R. Pasquali (2010)
Value of diffusion-weighted imaging to detect small malignant pelvic lymph nodes at 3 TEuropean Radiology, 20
Kaiyue Zhang, Yu Zhang, Xin Fang, Jiangning Dong, Liting Qian (2021)
MRI-based radiomics and ADC values are related to recurrence of endometrial carcinoma: a preliminary analysisBMC Cancer, 21
J. Reyes-Pérez, Y. Villaseñor-Navarro, Mayra Santos, I. Pacheco-Bravo, Maricela Calle-Loja, I. Sollozo-Dupont (2020)
The apparent diffusion coefficient (ADC) on 3-T MRI differentiates myometrial invasion depth and histological grade in patients with endometrial cancerActa Radiologica, 61
F. Amant, M. Mirza, M. Koskas, C. Creutzberg (2015)
Cancer of the corpus uteriInternational Journal of Gynecology & Obstetrics, 131
R. Siegel, K. Miller, Hannah Fuchs, A. Jemal (2021)
Cancer Statistics, 2021CA: A Cancer Journal for Clinicians, 71
Xiaomin Zheng, Weiqian Guo, Jiangning Dong, Liting Qian (2020)
Prediction of early response to concurrent chemoradiotherapy in cervical cancer: Value of multi-parameter MRI combined with clinical prognostic factors.Magnetic resonance imaging
S. Kakimoto, M. Miyamoto, T. Einama, Y. Takihata, H. Matsuura, H. Iwahashi, H. Ishibashi, T. Sakamoto, T. Hada, J. Suminokura, Tsubasa Ito, R. Suzuki, A. Suzuki, M. Takano (2021)
Significance of mesothelin and CA125 expression in endometrial carcinoma: a retrospective analysisDiagnostic Pathology, 16
Hao Jia, Xueyan Jiang, Kaiyue Zhang, Jin Shang, Yu Zhang, Xin Fang, Fei Gao, Nai-Yu Li, Jiangning Dong (2022)
A Nomogram of Combining IVIM‐DWI and MRI Radiomics From the Primary Lesion of Rectal Adenocarcinoma to Assess Nonenlarged Lymph Node Metastasis PreoperativelyJournal of Magnetic Resonance Imaging, 56
Kornél Lakatos, Germán González, Jawad Hoballah, Jeff Brooker, Sinyoung Jeong, C. Evans, P. Krauledat, W. Hansen, K. Elias, M. Patankar, Vilmos Fülöp, P. Konstantinopoulos, D. Cramer (2022)
Application of a novel microscopic technique for quantifying CA125 binding to circulating mononuclear cells in longitudinal specimens during treatment for ovarian cancerJournal of Ovarian Research, 15
P. Widschwendter, E. Bauer, N. Gregorio, I. Bekes, W. Janni, C. Scholz, T. Friedl (2018)
Influence of Prognostic Factors on Lymph Node Involvement in Endometrial Cancer: A Single-Center ExperienceInternational Journal of Gynecologic Cancer, 28
(2022)
(e feasibility of combining ADC value with texture analysis of T2WI
and P
Hindawi Journal of Oncology Volume 2022, Article ID 3335048, 10 pages https://doi.org/10.1155/2022/3335048 Research Article Preoperative Prediction Value of Pelvic Lymph Node Metastasis of Endometrial Cancer: Combining of ADC Value and Radiomics Features of the Primary Lesion and Clinical Parameters 1 1 2 3 3 4 Juan Bo , Haodong Jia , Yu Zhang , Baoyue Fu , Xueyan Jiang , Yulan Chen , 4 4 1,4 Bin Shi , Xin Fang , and Jiangning Dong Department of Radiology, Anhui Provincial Hospital Aliated to Anhui Medical University, Hefei, Anhui 230001, China Department of Radiation Oncology, Anhui Provincial Hospital Aliated to Anhui Medical University, Hefei, Anhui 230001, China Bengbu Medical College, Bengbu, Anhui 233030, China Department of Radiology, e First Aliated Hospital of University of Science and Technology of China, Hefei, Anhui 230001, China Correspondence should be addressed to Jiangning Dong; dongjn@ustc.edu.cn Received 17 April 2022; Accepted 8 June 2022; Published 30 June 2022 Academic Editor: Emmanuel Navarro Flores Copyright © 2022 Juan Bo et al. �is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Objective. To investigate the value of apparent di„usion coe…cient (ADC) value of endometrial cancer (EC) primary lesion and magnetic resonance imaging (MRI) three-dimensional (3D) radiomics features combined with clinical parameters for preop- erative prediction of pelvic lymph node metastasis (PLNM). Methods. A total of 136 patients with EC con“rmed by postoperative pathology were retrospectively reviewed and analyzed. Patients were randomly divided into training set (n 95) and test set (n 41) at a ratio of 7 : 3. Radiomics features based on T WI, DWI, and contrast-enhanced T WI (CE-T WI) sequence were 2 1 1 extracted and screened, and then radiomics score (Rads-score) was calculated. Clinical parameters and ADC value of EC primary lesion were measured and collected, and their correlation with PLNM was analyzed. Receiver operating characteristic (ROC) curve was plotted to assess the diagnostic e…cacy of the model. A nomogram for PLNM was created based on the multivariate logistic regression model. Results. �e ADC value of the EC primary lesion showed inverse correlation with PLNM, while CA125 and Rads-score were positively associated with PLNM. A predictive model was proposed based on ADC value, Rads-score, CA125, and MR-reported pelvic lymph node status (PLNS) for PLNM in EC. �e area under the curve (AUC) of the model is 0.940; the sensitivity and speci“city (87.1% and 90.6%) of the model were signi“cantly higher than that of the MRI morphological signs. Conclusion. A combination of ADC value, MRI 3D radiomics features of the EC primary lesion, and clinical parameters generated a prediction model for PLNM in EC and had a good diagnostic performance; it was a useful supplement to MR-reported PLNS based on MRI morphological signs. controversial. On the one hand, surgical complications pose 1. Introduction signi“cant challenges to EC patients. On the other hand, Endometrial cancer (EC) is one of the most common gy- some studies have con“rmed that low-risk patients do not necological malignancies worldwide [1], and its morbidity bene“t from lymphadenectomy. Emerging evidence con- and mortality are increasing over years. Lymph node me- “rmed the survival bene“t of systematic lymphadenectomy tastasis (LNM) is one of the major metastatic routes of EC in patients with EC with intermediate or high risk for LNM [4], suggesting that preoperative precise LN risk strati“ca- and the main adverse prognostic factor [2]. �e overall “ve- year survival rate was signi“cantly lower for those with LNM tion is conducive to balancing the bene“ts of treatment and in EC compared to those without LNM [3]. Internationally, surgical complications. �erefore, preoperative prediction of the use and the extent of lymphadenectomy in EC have been LNM is crucial for EC treatment plan selection. 2 Journal of Oncology (e occurrence of LNM implies the proliferation and were (1) combination of other malignant tumors, (2) spread of tumor, which is a complex process with multiple combined with other pelvic diseases or a history of pelvic factors interacting with each other. Studies have reported surgery, and (3) poor picture quality due to motion artifacts that the intronic promoter p10 of TACC2 in primary lesion caused by respiration, intestinal peristalsis, and so on. Based of EC is more active in those with LNM [5]. (e tran- on exclusion criteria, 136 patients were eventually included scriptional regulatory networks of EC primary lesion are in our study. Patients were divided into training set (n �95) different between LNM and non-LNM states [6], suggesting andtestset(n �41)accordingtorandomdistributionof7:3. that LNM is closely related to the primary EC lesion. Based on the histopathologic examination, patients in (erefore, preoperative evaluation of pelvic lymph node training and test sets were further divided into PLNM (+) metastasis (PLNM) is theoretically feasible from the per- group and PLNM (−) group. spective of EC primary lesion. Conventional MRI mainly providesmorphologicalinformationofLNs,suchasLNsize, 2.2. Imaging Protocol. Pelvic MRI was performed using a enhancement mode, necrosis, signal characteristics, and unit system (GE Signa HDXT 3.0T MRI scanner, GE extranodal expansion [7]. (e morphologic appearance of Healthcare, USA) equipped with an eight-channel phased- metastatic LNs with short diameter is usually similar to that array body coil. All patients fasted for at least 4 hours and of nonmetastatic LNs, and the diagnostic process is sub- were given an intramuscular injection of scopolamine hy- jective. Invasive sentinel node localization is not easy to drochloride half an hour before the MRI examination to implement due to the complexity of EC lymphatic drainage reduce gastrointestinal peristalsis artifacts. Patients main- [8]. In addition, since about half of metastatic sentinels are tained a supine position with an empty urinary bladder. (e small-volume metastases, most of them are not easily de- scan covered the upper edge of the iliac crest to just below tected [9]. (erefore, it is necessary to find an accurate, the pubic symphysis. effective,low-cost,and noninvasivemethod for preoperative Detailed scanning parameters were listed in Table 1. (e assessment of PLNM in EC. plain scan sequence includes axial fast spin-echo (FSE) T - (e apparent diffusion coefficient (ADC) value derived weighted images (T WI), axial FSE T -weighted images 1 2 from diffusion-weighted imaging (DWI) can quantify the (T WI), axial fat suppression (FS) FSE T WI, and sagittal 2 2 degree of restricted diffusion of water molecules. ADC value FSE T WI. (e enhancement sequence includes axial of EC differed among tumor grades and was significantly LAVA-FLEX in arterial phase, venous phase, and delayed lower for patients with G2 and G3 than G1 [10]. A statis- phase, and sagittal LAVA-FLEX in late delayed phase. (e tically significant negative correlation was observed between delay time is 25s, 60s, 150s, and 180s, respectively. (e b ADC value and the Ki-67 index [11]. (erefore, ADC value values of DWI are 0 and 1000s/mm . Contrast agent Gd- could be used as imaging biomarker to reflect the prolif- DTPA (Magnevist, Bayer Schering, Berlin, Germany) was erativecapacityandaggressivenessoftumors.Radiomicsisa injected through the anterior cubital vein with a high- new data mining technique. It evaluates the inhomogeneity pressure syringe at a flow rate of 2.5ml/s at 0.1mmol. of image signals by quantifying the inhomogeneity and regularity of pixel gray values in normal and pathological tissues, through which it can reflect the microscopic het- 2.3. Surgical Procedure and Histopathology. Primary surgical erogeneity at the histopathological level. Radiomics is po- treatment consisted of hysterectomy, bilateral adnexectomy, tentially valuable for assessing tumor efficacy and predicting and pelvic lymphadenectomy. All surgical specimens were recurrence and metastasis [12–15]. Our study attempted to examined and reported by gynecologic pathologists. (e establish a predictive model for preoperative prediction of 2014 World Health Organization (WHO) classification [16] PLNM by combining the ADC value, MRI-based 3D and the 2018 revised FIGO staging criteria [17] for EC were radiomics features of EC primary lesion, and clinical pa- used for histological diagnosis, grading, and pathological rameters of patients. staging. 2. Materials and Methods 2.4. Imaging Analyses. EC lesions and pelvic lymph node 2.1. Patients. (is retrospective study with anonymous data status (PLNS) were determined independently from MRI was approved by the Ethics Committee of our hospital, and images by two attending physicians with 8 and 10years of the informed consent requirement was waived. Data of 162 experience in gynecologic radiology, respectively. ADC patientswithECconfirmedbypostoperativepathologyfrom value of the tumor were measured using GE Advantage Workstation 4.6, Function Tool software by the two at- May 2014 to November 2021 were obtained by searching the picture archiving and communication system of our hos- tending physicians, respectively. (e region of interest pital. Figure 1 showed the patient recruitment pathway. (ROI) was outlined at the largest tumor cross section and Inclusion criteria were (1) conventional MRI and DWI scan its two adjacent levels, and three ADC values were obtained were performed one week before surgery, (2) radical hys- and finally averaged, avoiding the areas of cystic change, terectomy, bilateral adnexectomy, and pelvic lymphade- hemorrhage, necrosis, and calcification as much as pos- nectomy were performed in our hospital, (3) absence of any sible. (e process of measuring the ADC value of EC tumor-relatedtreatmentpriortoimaging,and(4)maximum primary lesion was shown in Figure 2. High-resolution diameter of tumor (Mdot) ≥10.0mm. Exclusion criteria axial T WI, axial DWI (b �1000s/mm ), and axial 2 Journal of Oncology 3 Patients with EC confirmed by postoperative pathology between May 2014 and November 2021 (n=162) Exclusion criteria (1) Combination of other malignant tumors (n=2) (2) Combined with other pelvic diseases or a history of pelvic surgery (n=16) (3) Poor picture quality due to motion artifacts caused by respiration, intestinal Patients were peristalsis, etc (n=8) enrolled (n=136) Training set (n=95) Test set (n=41) PLNM (−) (n=64) PLNM (−) (n=28) PLNM (+) (n=31) PLNM (+) (n=13) Figure 1: Flowchart of patient recruitment pathway. Table 1: MRI scanning protocols. Sequence TR/TE FOV Matrix Slice gap (mm) Slice thickness (mm) Axial T WI 500/7.2 38cm ×26cm 352 ×192 2 6 Axial T WI 4600/68 24cm ×24cm 320 ×256 2 6 Axial (FS)-T WI 4600/68 24cm ×24cm 320 ×256 2 6 Sagittal T WI 4600/68 26cm ×24cm 320 ×256 2 6 Axial T WI LAVA-FLEX 36/1.3 38cm ×36cm 320 ×224 0 4 Sagittal T WI LAVA-FLEX 36/1.3 28cm ×26cm 320 ×224 0 4 Axial DWI 4000/65 38cm ×26cm 96 ×130 1 4 (a) (b) Figure 2: DWI image of a 57-year-old patient with EC of stage IB. (a) ROI was outlined at the largest tumor cross section on DWI 2 −3 2 (b �1000s/mm ). (b) ADC pseudocolor image showed that the ADC was 0.966 ×10 mm /s. contrast-enhanced T -weighted (CE-T WI) delayed-phase morphological signs, and PLNM was diagnosed if one of 1 1 images of all patients were imported into ITK-SNAP the following criteria was met: (1) short diameter of LNs in (Version 3.6.0, http://www.itksnap.org) software in the axial plane ≥10.0mm; (2) central necrosis or circum- DICOM format, and the two attending physicians outlined ferential enhancement of LNs, (3) extra-peripheral inva- the whole picture of the tumor layer by layer, and the ROIs sion of LNs, including irregular enhancement of LN should include degeneration, necrosis, and hemorrhage margins,blurringofsurroundingfattyspaces,andfusionof areas.(eoriginalimagesandROIswereimportedintoAK LNs with each other. Two attending physicians performed software to extract radiomics features. 828 radiomics the above steps twice separately, one week apart. Inter- and features were extracted from each sequence separately. (e intragroup correlation coefficients and Cohen’s kappa radiomics feature extraction and screening process was coefficient were calculated based on the two measurements shown in Figure 3. (e PLNS was evaluated based on MRI for assessing consistency. 4 Journal of Oncology ROI Segmentation Feature Extraction Screening Features Heatmap of training samples for corellation model Lasso paths 3.0 7.5 1.5 5.0 0.0 −1.5 2.5 −3.0 0.0 −2.5 −5.0 −7.5 −10.0 −10 −8−6−4−2 ln(alpha) Heatmap of testing samples for variance model 3.0 1.5 0.0 −1.5 −3.0 828 Radiomics features 130 2 first order: 198 Shap: 14 GLCM: 264 GLRLM: 176 GLSZM: 176 Figure 3: Radiomics workflow. 2.5. Statistical Analysis. Statistical analyses were implemented Adopting a calibration curve to measure the predictive with R software (version 3.5.2; R Foundation for Statistical performance of the model. Decision curve analysis was used Computing, Vienna, Austria) and SPSS (version 24.0; IBM to assess the clinical practicality of the model. Finally, the Corporation, Armonk, NY, USA). Agreement between two model performance was validated in the testing set. attending physicians in assessing PLN based on MRI images wasassessedbycalculatingCohen’skappacoefficient.(efinal 3. Results consensus of two attending physicians was used for data analysis.(eintraclasscorrelationcoefficient(ICC)wasusedto 3.1. Intraobserver and Interobserver Agreement. (e ranges evaluate the intra- and interobserver agreement of ADC value, of ICC (inter), ICC (intra), and Cohen’s kappa coefficient radiomicsparameters,andMR-reportedMdot.ADCvalueand were 0.791–0.939, 0.834–0.954, and 0.715, respectively, MR-reported Mdot with higher intraobserver ICC were indicating that the intra- and interreproducibility is retained for subsequent data analysis. Radiometric features good. with intra- and interobserver ICC>0.75 were retained for subsequent features screening. Quantitative variables were tested for normality. Data 3.2. General Clinicopathological Data, ADC Value, and were expressed as mean±standard deviation (SD) when the Radiomics Features of Primary Lesion. (e general clinico- distribution was normal and analyzed by independent pathological data of the patients was shown in Table 2. A samples t-test. Data conforming to nonnormal distribution total of 136 patients with EC, aged 36–76years, were in- were expressed as median and quartile and analyzed by cluded in thisstudy. (ere were 44 patients inthe PLNM (+) Mann–Whitney U test. Categorical variables were expressed group and 92 patients in the PLNM (−) group. In the as composition ratios and analyzed by chi-square test or training and test sets, there was no statistical difference in Fisher’s exact test. A two-tailed p value <0.05 indicated age (p � 0.196,0.227, resp.) and menopausal status statistical significance. (e receiver operating characteristic (p � 0.675, 0.460, resp.) between the PLNM (+) and PLNM (ROC) curves are plotted and the area under the curve (−) groups. Adnexal metastasis, deep myometrial invasion, (AUC) is calculated to assess the predictive power of the lymphatic vascular space infiltration (LVSI), and MR-re- prediction model. DeLong’s test was used to compare the portedPLNSweresignificantlymoreinthePLNM(+)group AUCs between the prognostic models. (e Hos- than in the PLNM (−) group. A significantly higher FIGO mer–Lemeshow test was used to verify the goodness-of-fit of staging, CA125, and Ki-67 were found in PLNM (+) group, the prediction model. A nomogram for PLNM was created when compared to PLNM (−) group. MR-reported Mdot based on the multivariate logistic regression model. and pathological type were significantly different in the coefficients Journal of Oncology 5 Table 2: Selected clinicopathological data of patients with endometrial cancer. Training set Test set Characteristics p p p PLNM (+) PLNM (−) PLNM (+) PLNM (−) Age at onset (years) 55 (51, 63) 53 (49, 61.8) 0.196 54 (49.5, 59) 53 (49, 57.5) 0.227 0.132 CA125 (U/mol) <0.001 0.007 <0.001 ≥35 18 (58.1%) 6 (9.4%) 76.9 (%) 9 (32.1%) <35 13 (41.9%) 58 (90.6%) 3 (23.1%) 19 (67.9%) MR-reported Mdot (cm) 5 (3.5, 5.5) 4 (3, 5) 0.034 5 (2.9, 6) 4.3 (2.9, 5.8) 0.555 0.034 MR-reported PLNS 0.009 0.069 <0.001 MR-reported PLNM (+) 20 (64.5%) 8 (12.5%) 6 (46.2%) 4 (14.3%) MR-reported PLNM (−) 11 (35.5%) 56 (87.5%) 7 (53.8%) 24 (85.7%) LVSI <0.001 0.002 <0.001 Present 17 (54.8%) 4 (6.20%) 8 (61.5%) 3 (10.7%) Absent 14 (45.2%) 60 (93.8%) 5 (38.5%) 25 (89.3%) Ki-67(%) 60 (50, 70) 50 (30, 70) 0.014 70 (60, 75) 45 (32.5, 67.5) 0.005 <0.001 Adnexal metastasis <0.001 0.024 <0.001 Present 13 (41.9%) 5 (7.80%) 5 (38.5%) 2 (7.10%) Absent 18 (58.1%) 59 (92.2%) 8 (61.5%) 26 (92.9%) Dmi <0.001 <0.001 <0.001 Present 21 (67.7%) 14 (21.9%) 10 (76.9%) 8 (28.6%) Absent 10 (32.3%) 50 (78.1%) 3 (23.1%) 20 (71.4%) Menopause 0.675 0.460 0.912 Present 17 (54.8%) 38 (59.4%) 9 (69.2%) 16 (57.1%) Absent 14 (45.2%) 26 (40.6%) 4 (30.8%) 12 (42.9%) FIGO stage <0.001 <0.001 <0.001 I–II 0 56 (87.5%) 0 25 (89.3%) IIIA 0 8 (12.5%) 0 3 (10.7%) IIIC 25 (80.6%) 0 9 (69.2%) 0 IVA 6 (19.4%) 0 2 (15.4%) 0 IVB 0 0 2 (15.4%) 0 Pathological types 0.036 0.443 0.017 Endometrioid 19 (61.3%) 52 (81.2%) 8 (61.5%) 22 (78.6%) Nonendometrioid 12 (38.7%) 12 (18.8%) 5 (38.5%) 6 (21.4%) Note. CA125 �carcinoma antigen 125; MR-reported Mdot �MR-reported maximum diameter of tumor; MR-reported; PLNS �MR-reported pelvic lymph node status; LVSI �lymphatic vascular space infiltration; and Dmi �deep myometrial invasion. training set, while they were not significantly different in the MR-reported Mdot was divided into ≥4cm and <4cm test set (p � 0.555,0.443, resp.). groups according to the median. (e ADC value and Rads-score were divided into two groups according to (eradiomicsfeaturesextractionprocessandresultsareas follows: (1) 828 radiomics features were extracted from each the cut-off values. We investigated the association of image sequence (T WI, DWI, and CE-T WI) of each patient, CA125, ADC value, MR-reported Mdot, MR-reported 2 1 and a total of 2484 features were extracted from each patient. PLNS, menopausal status, age at onset, and Rads-score (2)1890featureswithgoodstability(ICC>0.75)wereretained. with PLNM using univariate analysis. CA125, ADC (3) A z-score standardization was applied to each feature. (4) value, MR-reported PLNS, MR-reported Mdot, and (e least absolute shrinkage and selection operator (LASSO) Rads-score were significantly associated with PLNM. A algorithm was used for filtering and 10 radiomics parameters multivariate analysis of these factors showed that were screened with |c|<0.7. (5) Four radiomics features were CA125(OR �6.971; 95% CI �1.545–31.443; p � 0.012), ADC value (OR �0.004; 95% CI �0.000–0.427; obtained by multivariate logistic regression as independent discriminant features. (6) Rads-score was calculated based on p � 0.020), and Rads-score (OR �1.721; 95% CI �1.139–2.601; p � 0.010) were independent risk fac- linearcombinationsofregressioncoefficients.Comparedtothe PLNM (−) group, the PLNM (+) group had significantly lower tors for PLNM. (e higher the ADC value, the higher the ADC value and higher Rads-score. (e ADC value, radiomics likelihood of not developing PLNM in EC patients. (e features, and Rads-score for the training and test sets were higher the CA125 and Rads-score, the higher the risk of shown in Table 3. PLNM (Table 4). In addition, the MRI-reported PLNS (OR �9.126; 95% CI �2.043–40.759; p � 0.004) also re- flected the risk of PLNM to some extent. 3.3. Diagnostic Efficacy of the Predictive Model. Age at onset (e following cut-off values were obtained by ROC was divided into ≥54years and <54years groups and the curve analysis and used to distinguish PLNM (+) from 6 Journal of Oncology Table 3: (e ADC value, radiomics features, and Rads-score of the training and test sets. Training set Test set Characteristics p p p PLNM (+) PLNM (−) PLNM (+) PLNM (−) 0.810 (0.780, −3 2 ADC (×10 mm /s) 1.001 (0.878, 1.205) <0.001 0.884 (0.783, 0.906) 0.930(0.889,1.099) 0.006 <0.001 0.901) glcm_MCC @ CE-T WI 0.736±0.110 0.661±0.102 0.001 0.767±0.095 0.684±0.087 0.008 <0.001 shape_MajorAxisLength@ 49.972 (38.648, 37.884 (31.643, 57.149 (45.914, 39.518 (34.102, 0.001 0.002 <0.001 T WI 62.845) 46.827) 103.431) 48.604) HLL_firstorder_Kurtosis@ 3.563 (2.946, 4.056(3.388,4.805) 0.011 3.991 (3.387, 4.727) 5.122 (4.534, 5.671) 0.002 0.009 DWI 4.140) 0.334 (0.190, glcm_Correlation@DWI 0.284 (0.118, 0.346) 0.002 0.297 (0.275, 0.399) 0.263 (0.143, 0.323) 0.019 0.008 0.387) Rads-score 0.741±2.323 −2.290±2.190 <0.001 0.0746±2.612 −1.751±2.308 0.029 <0.001 Table4:Univariateand multivariatelogisticregressionanalyses forADCvalueof ECprimary lesion,clinical parameters,and Rads-scorein estimating PLNM. Univariate analysis Multivariate analysis Parameters OR 95% CI p OR 95% CI p CA125 (≥35U/mol/<35U/mol) 13.385 4.444–40.308 <0.001 6.971 1.545–31.443 0.012 Age at onset (≥54/<54) 1.037 0.982–1.095 0.192 Menopause (present/absent) 0.831 0.350–1.974 0.675 MR-reported Mdot (≥4cm/<4cm) 1.324 1.022–1.715 0.033 −3 2 −3 2 ADC (≥0.908 ×10 mm /s/<0.908 ×10 mm /s) 0.001 0.000–0.030 <0.001 0.004 0.000–0.427 0.020 MR-reported PLNS (present/absent) 12.727 4.480–35.155 <0.001 9.126 2.043–40.759 0.004 Rads-score (≥−1.161/<−1.161) 2.215 1.536–3.195 <0.001 1.721 1.139–2.601 0.010 Table 5: Diagnostic performance of single parameters and predictive model for predicting PLNM in EC. Training set Test set Parameters AUC Sensitivity (%) Specificity (%) Accuracy (%) AUC Sensitivity (%) Specificity (%) Accuracy (%) CA125 0.743 58.1 90.6 80.0 0.724 76.9 67.9 70.7 ADC 0.791 80.7 72.6 71.6 0.772 84.6 64.3 68.3 MR-reported PLNS 0.760 64.5 87.5 80.0 0.659 46.2 85.7 73.1 Rads-score 0.855 93.6 67.2 74.7 0.728 61.54 82.1 73.2 Model 0.940 87.1 90.6 88.4 0.918 92.3 89.3 85.4 −3 2 PLNM (−): ADC �0.908 ×10 mm /s, Rads-score- 4. Discussion � −1.161. (eAUC of theRads-score was 0.855, the highest Independent risk factors for EC lymph node metastasis in- among all predictive parameters. In the training set, the clude positive peritoneal cytology, deep myometrial invasion, AUC, sensitivity, and specificity of the prediction model LVSI, and FIGO staging [18, 19]. However, this data was established by combining ADC value, clinical parameters, usually available from the postoperative pathology, so the andradiomicsfeatures(0.940,87.1%,and90.6%resp.)were conditionsforpreoperativeevaluationofPLNMwerelimited. allhigher than MRI morphological signs (0.760, 64.5%,and In this study, we developed a predictive model based on 87.5%, resp.) (Table 5 and Figure 4(a)). DeLong’s test accessible and available clinical parameters (CA125 and MR- showed that the diagnostic efficacy of the prediction model reported PLNS), ADC value of the primary lesion, and Rads- was better than that of the other single parameter in both score.Itshowedagooddiagnosticefficacy(AUC �0.940)and the training and test sets (Table 6). (e nomogram of the the sensitivity (87.1%) and specificity (90.6%) were greatly prediction model based on CA125, ADC value of EC improved compared with the MRI morphological signs. primary lesion, MR-reported PLNS, and Rads-score in the Furthermore, we developed and validated a nomogram for training group was shown in Figure 4(b). (e nonsignif- this prediction model to facilitate individualized and non- icant statistic of the Hosmer–Lemeshow test (p � 0.815) invasive preoperative assessment of the PLNM in EC. indicated that the training set prediction model was well Conventional MRI remained the main tool for preop- fitted; calibration curves for nomogram in the training set erative assessment of LNM in EC, but daily diagnosis were shown in Figure 5(a). (e decision curve analysis remained at the macromorphology stage. ADC value was showedthat model couldadda netbenefittothetreat-allor negatively correlated with the cell density of tumors. treat-none scheme (Figure 5(b)). Journal of Oncology 7 0 20406080 100 100 – specificity ADC Rads-score CA125 Model MR-reported PLNS (a) 0 102030405060708090 100 Points Rads-score –8 –6 –4 –20 2468 ADC 1.7 1.5 1.3 1.1 0.9 0.7 0.5 ≥ 35 CA125 < 35 (+) MR-reported PLNS (–) Total Points 0 20 40 60 80 100 120 140 160 180 Probability of PLNM 0.1 0.3 0.5 0.7 0.9 (b) Figure 4: (a) ROC curves of the ADC, CA125, MR-reported PLNS, Rads-score, and model in the training set. AUCs of ADC, CA125, MR- reported PLNS, Rads-score, and model were 0.791, 0.743, 0.760, 0.855, and 0.940 in the training set, respectively. (b) Nomogram of the model for PLNM in EC in the training set. –3 2 Investigators found ADC value of LNs was unable to dif- diagnostic threshold for ADC value was (0.908 ×10 mm / ferentiate benign LNs from malignant LNs [20, 21]. How- s) in our study. Its sensitivity, specificity, and accuracy were ever, several reports showed that the ADC value of breast 80.7%, 72.6%, and 71.6%, respectively. In fact, the reason for lesions in patients with positive axillary LN metastasis was the lower ADC value of EC primary lesions in PLNM (+) significantly lower than those with negative axillary LNM group is not clear. However, in our study, the Ki-67 indexes [22]. (erefore, our study focused on the ADC value of EC of PLNM (+) group were higher than those of PLNM (−) primary lesion in different LN states and found that ADC group. Ki-67 is a marker of cell proliferation, suggesting that value showed inverse correlation with PLNM. (e optimal EC tumor cells in the PLNM (+) group had higher Sensitivity 8 Journal of Oncology Table 6: AUCs of the four parameters and the predictive model were compared. Training set Test set ADC vs.CA125 0.474 0.686 ADC vs. MR-reported PLNS 0.625 0.358 ADC vs. Rads-score 0.286 0.678 ADC vs. Model <0.001 0.080 CA125 vs. MR-reported PLNS 0.816 0.525 CA125 vs. Rads-score 0.047 0.973 CA125 vs. Model <0.001 0.007 MR-reported PLNS vs. Rads-score 0.138 0.637 MR-reported PLNS vs. model <0.001 0.002 Rads-score vs. model 0.005 0.025 Decision Curve of logistic model in training samples 1.0 0.5 0.4 0.8 0.3 0.6 0.2 0.4 0.1 0.2 0.0 −0.1 0.0 0.2 0.4 0.6 0.8 1.0 0.0 probability thresholds 0.0 0.2 0.4 0.6 0.8 1.0 DCA Nomogram Predicted Probability Treat all Apparent Treat None Bias-corrected Ideal (a) (b) Figure 5: (a) Calibration curves for nomogram in the training set. (e x-axis is the predicted probability of the nomogram; the y-axis is the actual probability of occurrence of PLNM. Perfect prediction corresponds to the “Ideal” line, the “Apparent” line represents the entire cohort (n �95), and the “Bias-corrected” line represents the performance of the nomogram obtained by bootstrapping (B �1000 repe- titions). (b) (e decision curve of nomogram of the model in the training set. proliferative activity, stronger invasiveness, and higher cell and glcm_Correlation@DWI). (e glcm_MCC reflected the density. It may be the reason for the lower ADC value in the complexity of textures. (e shape_MajorAxisLength de- PLNM (+) group. scribed the difference in the geometric shape of the two Metastasis is closely related to the heterogeneity of groups of lesions, and it was independent of the gray level malignant tumors, and radiomics features can quantify the intensity distribution in the ROI. HLL_firstorder_Kurtosis microscopic heterogeneity at the histopathological level was a measure of the “peakness” of the distribution of values [23,24].MRI-basedradiomicshasbeenconsideredareliable in the image ROI. (e glcm_Correlation showed the linear tool for accurate preoperative assessment of LNM in several dependency of gray level values to their respective voxels in cancers, such as cervical cancer [25], rectal cancer [15], and the GLCM. Multifactorial logistic regression analysis was oral tongue squamous carcinoma [26]. From this, we de- performed on the 4 obtained characteristics, and Rads-score veloped a Rads-score based on 3D radiomics features from was calculated based on linear combinations of regression axialT WI,axialDWI(b �1000s/mm ),andaxialCE-T1WI coefficients. In our study, the AUC, sensitivity, and speci- ficity of Rads-score in predicting PLNM were 0.855, 93.6%, delay period toimprove diagnostic efficiency. Inourstudy, 4 features with predictive value for PLNM were screened from and 67.2%, respectively, suggesting that radiomics had 2484 radiomics features (glcm_MCC@CE-T1WI, shape_- promising clinical applications in the preoperative evalua- MajorAxisLength@T WI, HLL_firstorder_Kurtosis@DWI, tion of PLNM in EC. Actual Probability Net_benifit Journal of Oncology 9 It is undeniable that the most intuitive way to assess efficiency compared to MR-reported PLNS based on MRI PLNM was based on LN morphology. However, the present morphological signs and contributed to clinical decision- study found a high specificity of 87.5% but low sensitivity of making and improving prognosis further. 64.5% based solely on morphological characteristics of LN. (is may be due to the difficulty in distinguishing benign Data Availability enlarged LN (e.g., infection, granulomatous disease, and (e data used to support the findings of this study are reactive hyperplasia) from metastatic LN based on mor- available from the corresponding author upon request. phology alone [27]. (e independent diagnostic efficacy of MRI-reported PLNS in this study was low (AUC �0.760). In Ethical Approval clinical practice, CA125 has been routinely used to diagnose avariety of malignancies, especiallyepithelialovarian cancer (e Ethics Committee of Anhui Provincial Cancer Hospital [28]. Quan et al. [29] demonstrated that serum CA125 approvedthisretrospectivestudy(EthicsApprovalno.2022- significantly correlated with clinicopathological risk factors, YXK-05). such as depth of myometrial invasion, LN status, and FIGO stageand thatCA125 is apromising prognostic indicator for Consent EC [30, 31]. In our study, CA125 could be used as an in- dependent predictor of PLNM in EC patients, which is (e informed consent was waived. consistent with the results of Lei et al. [32]. However, its sensitivity is not high, which may be related to the vul- Conflicts of Interest nerability of CA125 to other factors, such as menstruation, pregnancy, endometriosis, and peritonitis [33]. (e authors declare that they have no conflicts of interest Significantly,wefoundthatageatonset,menopause,and regarding the publication of this paper. MR-reported Mdot were not risk factors for PLNM. (e age of patients and the MR-reported Mdot in the PLNM (+) Authors’ Contributions group were slightly greater than those in the PLNM (−) JD, JB, and HJ conceptualized and designed the study. JB, group, but the difference was not statistically significant. EC usually occurs after menopause [34]. (e patients in this HJ, YZ, BF, and XJ collected and arranged the data. JB, HJ, YC, and BS analyzed the data and wrote the manuscript. XF: study were relatively concentrated inage, mainlybetween 50 implemented MR examination. All authors contributed to and 60 years. Most of these patients have reached the age of the article and approved the submitted version. JB and HJ menopause;menopause was not associated with PLNM also. are co-first authors and contributed equally to this study. In this study, the nomogram model combining ADC value, MRI-based 3D radiomics features of the EC primary lesion, and clinical parameters had a good predictive per- Acknowledgments formance for PLNM in EC. Probably, because ADC value (isstudywassupportedby2020SKYImageResearchFund indirectly reflected the proliferative status of cells in the EC (no. Z-2014-07-2003-11), National Cancer Center Climbing primary lesion, clinical parameters such as CA125 reflected Foundation for Clinical Research (no. NCC201912B03), and the biological behavior of EC to some extent, and radiomics Key Research and Development Projects of Anhui Province features reflected the microscopic characteristics and het- (no. 2022e07020008) granted to JD. (e authors are grateful erogeneity of EC, so the nomogram model could predict to all involved cancer patients for their participation in the PLNM in EC from different perspectives and had a better study. performance. (e limitations of this study include three aspects. First, References inthisretrospectivestudy,themodelwasonlyvalidatedwith an internal independent test group, and external validation [1] R. L. Siegel, K. D. Miller, H. E. Fuchs, and A. Jemal, “Cancer and prospective validation were not performed. Second, the statistics, 2021,” CA: A Cancer Journal for Clinicians, vol. 71, relationship between the microcirculatory status of EC no. 1, pp. 7–33, 2021. primaryfociandPLNMwasnotanalyzedinthisstudy,while [2] J. Casarin, G. Bogani, E. Piovano et al., “Survival implication the microangiogenic status of EC primary lesion may be of lymphadenectomy in patients surgically treated for ap- relatedtoPLNM,whichwillbefurtherexploredinthefuture parent early-stage uterine serous carcinoma,” Journal of Gynecologic Oncology, vol. 31, no. 5, p. e64, 2020. by combining intravoxel incoherent motion diffusion- [3] A. B. Jacques, D. Cantu-de-Le ´ on, ´ D. P. Montiel et al., “Di- weighted imaging (IVIM-DWI) and dynamic contrast-en- agnostic performance of intraoperative assessment in grade 2 hanced MRI (DCE-MRI). endometrioid endometrial carcinoma,” World Journal of Surgical Oncology, vol. 18, no. 1, p. 284, 2020. 5. Conclusion [4] H. Eggemann, T. Ignatov, K. Kaiser, E. Burger, S. D. Costa, and A. Ignatov, “Survival advantage of lymphadenectomy in In conclusion, the predictive model based on ADC value, endometrial cancer,” Journal of Cancer Research and Clinical MRI 3D radiomics features of primary lesion, and clinical Oncology, vol. 142, no. 5, pp. 1051–1060, 2016. parameters showed promising performance for preoperative [5] Y. Ito, Y. Terao, S. Noma et al., “Nanopore sequencing reveals evaluation of PLNM in EC. It had higher diagnostic TACC2 locus complexity and diversity of isoforms 10 Journal of Oncology transcribed from an intronic promoter,” Scientific Reports, Journal of Gynecological Cancer, vol. 28, no. 6, pp.1145–1152, vol. 11, no. 1, p. 9355, 2021. 2018. [6] E. Yoshida, Y. Terao, N. Hayashi et al., “Promoter-level [20] C.Roy,G.Bierry,A.Matau,G.Bazille,andR.Pasquali,“Value transcriptome in primary lesions of endometrial cancer of diffusion-weighted imaging to detect small malignant pelviclymph nodes at3T,” European Radiology, vol.20, no.8, identified biomarkers associated with lymph node metasta- pp. 1803–1811, 2010. sis,” Scientific Reports, vol. 7, no. 1, Article ID 14160, 2017. [21] G. Nakai, M. Matsuki, Y. Inada et al., “Detection and eval- [7] F. Teng, Y. F. Zhang, Y. M. Wang et al., “Contrast-enhanced uation of pelvic lymph nodes in patients with gynecologic MRI in preoperative assessment of myometrial and cervical malignancies using body diffusion-weighted magnetic reso- invasion, and lymph node metastasis: diagnostic value and nance imaging,” Journal of Computer Assisted Tomography, error analysis in endometrial carcinoma,” Acta Obstetricia et vol. 32, no. 5, pp. 764–768, 2008. Gynecologica Scandinavica, vol. 94, no. 3, pp. 266–273, 2015. [22] F. Kato, K. Kudo, H. Yamashita et al., “Predicting metastasis [8] B. Geppert, C. Lonnerfors, ¨ M. Bollino, A. Arechvo, and in clinically negative axillary lymph nodes with minimum J. Persson, “A study on uterine lymphatic anatomy for apparent diffusion coefficient value in luminal a-like breast standardization of pelvic sentinel lymph node detection in cancer,” Breast Cancer, vol. 26, no. 5, pp. 628–636, 2019. endometrial cancer,” Gynecologic Oncology, vol. 145, no. 2, [23] X. Hua, W. Zhao, A. C. Pesatori et al., “Genetic and epigenetic pp. 256–261, 2017. intratumor heterogeneity impacts prognosis of lung adenocar- [9] M. Plante, J. Stanleigh, M. C. Renaud, A. Sebastianelli, cinoma,” Nature Communications, vol. 11, no. 1, p. 2459, 2020. K. Grondin, and J. Gregoire, “Isolated tumor cells identified [24] E. Sala, E. Mema, Y. Himoto et al., “Unravelling tumour by sentinel lymph node mapping in endometrial cancer: does heterogeneity using next-generation imaging: radiomics, adjuvant treatment matter?” Gynecologic Oncology, vol. 146, radiogenomics, and habitat imaging,” Clinical Radiology, no. 2, pp. 240–246, 2017. vol. 72, no. 1, pp. 3–10, 2017. [10] J. A. R. P´erez, Y. V. Navarro, M. E. Jimenez ´ de Los Santos, [25] Y.Zhang,K.Y.Zhang,H.D.Jiaetal.,“Feasibilityofpredicting I. P. Bravo, M. C. Loja, and I. S. Dupont, “(e apparent pelvic lymph node metastasis based on IVIM-DWI and texture diffusion coefficient (ADC) on 3-T MRI differentiates myo- parameters of the primary lesion and lymph nodes in patients metrial invasion depth and histological grade in patients with with cervical cancer,” Academic Radiology, vol. 29, 2021. endometrial cancer,” Acta Radiologica, vol. 61, no. 9, [26] Y. Yuan, J. Ren, and X. Tao, “Machine learning–based MRI pp. 1277–1286, 2020. texture analysis to predict occult lymph node metastasis in [11] X. Jiang, H. Jia, Z. Zhang, C. Wei, C. Wang, and J. Dong, “(e early-stage oral tongue squamous cell carcinoma,” European feasibility of combining ADC value with texture analysis of Radiology, vol. 31, no. 9, pp. 6429–6437, 2021. T WI, DWI and CE-T WI to preoperatively predict the ex- 2 1 [27] G. Lai and A. G. Rockall, “Lymph node imaging in gyne- pression levels of ki-67 and p53 of endometrial carcinoma,” cologic malignancy,” Seminars in Ultrasound, CT and MRI, Frontiers in Oncology, vol. 11, Article ID 805545, 2022. vol. 31, no. 5, pp. 363–376, 2010. [12] J. Song, J. Shi, D. Dong et al., “A new approach to predict [28] K. Lakatos, G. Gonzalez, ´ J. Hoballah et al., “Application of a progression-free survival in stage IV EGFR-mutant NSCLC novel microscopic technique for quantifying CA125 binding patients with EGFR-TKI therapy,” Clinical Cancer Research, to circulating mononuclear cells in longitudinal specimens vol. 24, no. 15, pp. 3583–3592, 2018. during treatment for ovarian cancer,” Journal of Ovarian [13] K. Zhang, Y. Zhang, X. Fang, J. Dong, and L. Qian, “MRI- Research, vol. 15, no. 1, 2022. based radiomics and ADC values are related to recurrence of [29] Q. Quan, Q. Liao, W. Yin, S. Zhou, S. Gong, and X. Mu, endometrial carcinoma: a preliminary analysis,” BMC Cancer, “Serum HE4 and CA125 combined to predict and monitor vol. 21, no. 1, p. 1266, 2021. recurrence of type II endometrial carcinoma,” Scientific Re- [14] X. Zheng, W. Guo, J. Dong, and L. Qian, “Prediction of early ports, vol. 11, no. 1, Article ID 21694, 2021. [30] W.Zhong,C.Zhou,L.Chenetal.,“(ecoefficientofvariation response to concurrent chemoradiotherapy in cervicalcancer: of red blood cell distribution width combined with cancer value of multi-parameter MRI combined with clinical antigen 125 predicts postoperative overall survival in endo- prognostic factors,” Magnetic Resonance Imaging, vol. 72, metrial cancer,” International Journal of General Medicine, pp. 159–166, 2020. vol. 14, pp. 5903–5910, 2021. [15] H. Jia, X. Jiang, K. Zhang et al., “A nomogram of combining [31] S. Kakimoto, M. Miyamoto, T. Einama et al., “Significance of IVIM-DWI and MRI radiomics from the primary lesion of mesothelin and CA125 expression in endometrial carcinoma: rectal adenocarcinoma to assess nonenlarged lymph node a retrospective analysis,” Diagnostic Pathology, vol. 16, no. 1, metastasis preoperatively,” Journal of Magnetic Resonance p. 28, 2021. Imaging: JMRI, 2022. [32] H. Lei, S. Xu, X. Mao et al., “Systemic immune-inflammatory [16] R. J. Kurman, M. L. Carcangiu, C. S. Herrington, and index as a predictor of lymph node metastasis in endometrial R. H. Young, WHO Classification of Tumors of Female Re- cancer,” Journal of Inflammation Research, vol. 14, productive Organs, 4th edition, 2014. pp. 7131–7142, 2021. [17] F. Amant, M. R. Mirza, M. Koskas, and C. L. Creutzberg, [33] P. Buamah, “Benign conditions associated with raised serum “Cancer of the corpus uteri,” International Journal of Gyne- CA-125 concentration,” Journal of Surgical Oncology, vol. 75, cology & Obstetrics, vol. 131, pp. S96–S104, 2015. no. 4, pp. 264-265, 2000. [18] Y. Li, P. Cong, P. Wang, C. Peng, M. Liu, and G. Sun, “Risk [34] E. H. Cappelletti, J. Humann, R. Torrejon, ´ and factors for pelvic lymph node metastasis in endometrial P. Gambadauro, “Chances of pregnancy and live birth among cancer,” Archives of Gynecology and Obstetrics, vol. 300, no. 4, women undergoing conservative management of early-stage pp. 1007–1013, 2019. endometrial cancer: a systematic review and meta-analysis,” [19] P. Widschwendter, E. Bauer, N. De Gregorio et al., “Influence Human Reproduction Update, vol. 28, no. 2, pp. 282–295, of prognostic factors on lymph node involvement in endo- metrial cancer: a single-center experience,” International
Journal of Oncology – Hindawi Publishing Corporation
Published: Jun 30, 2022
You can share this free article with as many people as you like with the url below! We hope you enjoy this feature!
Read and print from thousands of top scholarly journals.
Already have an account? Log in
Bookmark this article. You can see your Bookmarks on your DeepDyve Library.
To save an article, log in first, or sign up for a DeepDyve account if you don’t already have one.
Copy and paste the desired citation format or use the link below to download a file formatted for EndNote
Access the full text.
Sign up today, get DeepDyve free for 14 days.
All DeepDyve websites use cookies to improve your online experience. They were placed on your computer when you launched this website. You can change your cookie settings through your browser.