Preoperative multiparameter MRI-based prediction of Ki-67 expression in primary central nervous system lymphoma

Abstract

Background: Ki-67 is a key marker of tumor proliferation. This study aimed to develop machine learning models using single- and multi-parameter MRI radiomic features for the preoperative prediction of Ki-67 expression in primary central nervous system lymphoma (PCNSL), aiding prognosis and individualized treatment planning.

Methods: A retrospective analysis of 74 patients was conducted using MRI scans, including T1, contrast-enhanced T1, T2, T2-FLAIR, DWI, and ADC sequences. Patients were categorized into high-expression (Ki-67 > 70%) and low-expression (Ki-67 ≤ 70%) groups. Tumor volumes of interest (VOIs) were manually delineated by radiologists, and 851 radiomic features were extracted using 3DSlicer. After preprocessing, including bias field correction and normalization, feature selection was performed using SelectKBest and ANOVA. Eight machine learning classifiers, including Logistic Regression, Random Forest, and SVM, were applied to single- and multi-parameter datasets.

Results: Multiparameter models, particularly Naive Bayes and Logistic Regression, demonstrated superior predictive performance (AUC: 0.78, 0.73; AP: 0.90, 0.83) compared to single-parameter models. Decision curve analysis highlighted that Logistic Regression provides the highest net benefit, followed by Naive Bayes.

Conclusion: Multiparameter MRI models are more accurate and stable for predicting Ki-67 expression in PCNSL, supporting clinical decision-making.

Keywords

Ki-67 / Lymphoma / Machine Learning / Magnetic Resonance Imaging / Radiomics

References

[1] Liu Y, Yao Q, Zhang F. Diagnosis, prognosis and treatment of primary central nervous system lymphoma in the elderly population (Review). Int J Oncol. 2021; 58(3): 371-387.

[2] Grommes C, DeAngelis LM. Primary CNS Lymphoma. J Clin Oncol. 2017; 35(21): 2410-2418.

[3] Lukas RV, Stupp R, Gondi V, Raizer JJ. Primary central nervous system lymphoma—part 1: epidemiology, diagnosis, staging, and prognosis. Oncology (Williston Park). 2018; 32(1): 17-22.

[4] Houillier C, Soussain C, Ghesquieres H, et al. Management and outcome of primary CNS lymphoma in the modern era: an LOC network study. Neurology. 2020; 94(10): e1027-1039.

[5] Schaff LR, Grommes C. Primary central nervous system lymphoma. Blood. 2022; 140(9): 971-979.

[6] Fu Q, Liu SL, Hao DP, et al. CT radiomics model for predicting the ki67 index of lung cancer an exploratory study. Front Oncol.2021; 11: 743490

[7] Zheng YM, Chen J, Zhang M, et al. CT radiomics nomogram for prediction of the ki67 index in head and neck squamous cell carcinoma. Eur Radiol. 2023; 33(3): 2160-2170.

[8] Lashen A, Toss MS, Green AR, Mongan NP, Rakha E. Ki67 assessment in invasive luminal breast cancer: a comparative study between different scoring methods. Histopathology. 2022; 81(6): 786-798.

[9] Wen Y, Song Z, Li Q, et al. Development and validation of a model for predicting the expression of ki67 in pancreatic ductal adenocarcinoma with radiological features and dual-energy computed tomography quantitative parameters. Insights Imaging. 2024; 15(1): 41.

[10] Li Q, Song Z, Li X, et al. Development of a CT radiomics nomogram for preoperative prediction of ki67 index in pancreatic ductal adenocarcinoma: a two-center retrospective study. Eur Radiol.2024; 34: 2934-2943.

[11] Li ZM, Huang JJ, Xia Y, et al. High ki67 expression in diffuse large B-cell lymphoma patients with non-germinal center subtype indicates limited survival benefit from R-CHOP therapy. Eur J Haematol. 2012; 88(6): 510-517.

[12] Han Y, Yang J, Liu P, et al. Prognostic Nomogram for Overall Survival in Patients with Diffuse Large B-Cell Lymphoma. Oncologist. 2019; 24(11): e1251-e1261.

[13] Bi S, Li J, Wang T, et al. Multi-parametric MRI-based radiomics signature for preoperative prediction of ki67 proliferation status in sinonasal malignancies: a two-centre study. Eur Radiol. 2022; 32(10): 6933-6942.

[14] Wu C, Chen J, Fan Y, et al. Nomogram Based on CT Radiomics Features Combined With Clinical Factors to Predict ki67 Expression in Hepatocellular Carcinoma. Front Oncol. 2022; 12: 943942.

[15] Zhao Y, Xu J, Chen B, Cao L, Chen C. Efficient Prediction of ki67 Proliferation Index in Meningiomas on MRI: From Traditional Radiological Findings to a Machine Learning Approach. Cancers (Basel). 2022; 14(15): 3637.

[16] Li C, Song L, Yin J. Intratumoral and Peritumoral Radiomics Based on Functional Parametric Maps from Breast DCE-MRI for Prediction of HER-2 and ki67 Status. J Magn Reson Imaging. 2021; 54(3): 703-714.

[17] Louis DN, Perry A, Wesseling P, et al. The 2021 WHO classification of tumors of the central nervous system: a summary. Neuro Oncol. 2021; 23(8): 1231-1251.

[18] Haga A, Takahashi W, Aoki S, et al. Standardization of imaging features for radiomics analysis. J Med Invest. 2019; 66(1.2): 35-37.

[19] Park BE, Jang WS, Yoo SK. Texture Analysis of Supraspinatus Ultrasound Image for Computer Aided Diagnostic System. Healthc Inform Res. 2016; 22(4): 299-304.

[20] Boutahar K, Laghmati S, Moujahid H, Gannour OE, Cherradi B, Raihani A. Exploring Machine Learning Approaches for Breast Cancer Prediction: A Comparative Analysis with ANOVA-Based Feature Selection. 2024 4th International Conference on Innovative Research in Applied Science, Engineering and Technology (IRASET), FEZ, Morocco.2024: 1-7.

[21] Bi Q, Wang Y, Deng Y, et al. Different multiparametric MRI-based radiomics models for differentiating stage IA endometrial cancer from benign endometrial lesions: A multicenter study. Front Oncol. 2022; 12: 939930.

[22] Guo Y, Wu J, Wang Y, Jin Y. Development and Validation of an Ultrasound-Based Radiomics Nomogram for Identifying HER2 Status in Patients with Breast Carcinoma. Diagnostics (Basel). 2022; 12(12): 3130.

[23] Rui W, Qiao N, Wu Y, et al. Radiomics analysis allows for precise prediction of silent corticotroph adenoma among non-functioning pituitary adenomas. Eur Radiol. 2022; 32(3): 1570-1578.

[24] Huber F, Zwickl-Traxler E, Pecherstorfer M, Singer J. Evaluation of Ki-67 as a Prognostic Marker in Diffuse Large B-Cell Lymphoma-A Single-Center Retrospective Cohort Study. Curr Oncol. 2021; 28(6): 4521-4529.

[25] Liu J, Tu J, Xu L, et al. MRI-based radiomics signatures for preoperative prediction of Ki-67 index in primary central nervous system lymphoma. Eur J Radiol. 2024; 178: 111603.

[26] Broyde A, Boycov O, Strenov Y, Okon E, Shpilberg O, Bairey O. Role and prognostic significance of the ki67 index in non-Hodgkin's lymphoma. Am J Hematol. 2009; 84(6): 338-343.

[27] Cho U, Oh WJ, Hong YK, Lee YS. Prognostic Significance of High ki67 Index and Histogenetic Subclassification in Primary Central Nervous System Lymphoma. Appl Immunohistochem Mol Morphol. 2018; 26(4): 254-262.

[28] Liu Y, He C, Fang W, et al. Prediction of ki67 expression in gastrointestinal stromal tumors using radiomics of plain and multiphase contrast-enhanced CT. Eur Radiol. 2023; 33(11): 7609-7617.

[29] Ni J, Zhang H, Yang Q, et al. Machine-Learning and Radiomics-Based Preoperative Prediction of Ki-67 Expression in Glioma Using MRI Data. Acad Radiol. 2024; 31(8): 3397-3405.

[30] Zhang L, Shen M, Zhang D, et al. Radiomics Nomogram Based on Dual-Sequence MRI for Assessing Ki-67 Expression in Breast Cancer. J Magn Reson Imaging. 2024; 60(3): 1203-1212.

[31] Mohammadi M, Banisharif S, Moradi F, Zamanian M, Tanzifi G, Ghaderi S. Brain diffusion MRI biomarkers after oncology treatments. Rep Pract Oncol Radiother. 2024; 28(6): 823-834.

[32] Tang VH, Duong STM, Nguyen CDT, et al. Wavelet radiomics features from multiphase CT images for screening hepatocellular carcinoma: analysis and comparison. Sci Rep. 2023; 13(1): 19559.

[33] Chen J, Lu S, Mao Y, et al. An MRI-based radiomics-clinical nomogram for the overall survival prediction in patients with hypopharyngeal squamous cell carcinoma: a multi-cohort study. Eur Radiol. 2022; 32(3): 1548-1557.

[34] van Griethuysen JJM, Fedorov A, Parmar C, et al. Computational Radiomics System to Decode the Radiographic Phenotype. Cancer Res. 2017; 77(21): e104-e107.

[35] Zhu Y, Dou Y, Qin L, Wang H, Wen Z. Prediction of Ki-67 of Invasive Ductal Breast Cancer Based on Ultrasound Radiomics Nomogram. J Ultrasound Med. 2023; 42(3): 649-664.

[36] Ni J, Zhang H, Yang Q, et al. Machine-Learning and Radiomics-Based Preoperative Prediction of Ki-67 Expression in Glioma Using MRI Data. Acad Radiol. 2024; 31(8): 3397-3405.

[37] Zhang T, Ming Y, Xu J, et al. Radiomics and Ki-67 index predict survival in clear cell renal cell carcinoma. Br J Radiol. 2023; 96(1150): 20230187.

[38] Zhou L, Peng H, Ji Q, et al. Radiomic signatures based on multiparametric MR images for predicting Ki-67 index expression in medulloblastoma. Ann Transl Med. 2021; 9(22): 1665.

[39] Zheng YM, Chen J, Zhang M, et al. CT radiomics nomogram for prediction of the Ki-67 index in head and neck squamous cell carcinoma. Eur Radiol. 2023; 33(3): 2160-2170.

[40] Liu J, Tu J, Xu L, et al. MRI-based radiomics signatures for preoperative prediction of Ki-67 index in primary central nervous system lymphoma. Eur J Radiol. 2024; 178: 111603.

[41] Zhao E, Yang YF, Bai M, et al. MRI radiomics-based interpretable model and nomogram for preoperative prediction of Ki-67 expression status in primary central nervous system lymphoma. Front Med (Lausanne). 2024; 11: 1345162.

RIGHTS & PERMISSIONS

2025 The Author(s). Precision Radiation Oncology published by John Wiley & Sons Australia, Ltd on behalf of Shandong Cancer Hospital & Institute.