Application value of synthetic magnetic resonance imaging in predicting cervical lymph node metastasis of oral cancer

doi:10.3969/j.issn.1674-8115.2025.07.012

Abstract

Abstract:

Objective ·To explore the value of synthetic magnetic resonance imaging (SyMRI) technology in the diagnosis of cervical lymph node metastasis in patients with oral cancer. Methods ·A retrospective analysis was conducted on patients admitted to Shanghai Ninth People′s Hospital, Shanghai Jiao Tong University School of Medicine, from November 2023 to April 2024, who were pathologically diagnosed with oral cancer and had a clear lymph node metastasis status. Pre-operative maxillofacial magnetic resonance images of these patients were collected. Histogram features were extracted and screened from the quantitative maps generated by SyMRI, including synthetic T1, T2, and proton density (PD) maps, apparent diffusion coefficient (ADC) maps, and volumes of interest (VOIs) from contrast-enhanced images. The optimal range was determined by comparing the histogram parameters of different peritumoral regions. On this basis, habitat analysis was further carried out by combining the quantitative maps and ADC maps, and the habitat features of the tumor invasive sub-regions were extracted to construct a prediction model. The performance of the model was comprehensively evaluated using receiver operator characteristic curves (ROC curves), net reclassification improvement (NRI), integrated discrimination improvement (IDI), and decision curve analysis (DCA). Results ·A total of 61 patients with oral cancer were included in the study. The area under the curve (AUC) value for predicting cervical lymph node metastasis based on the intratumoral histogram features extracted from SyMRI was 0.798 (95%CI 0.673‒0.924). Adding the ADC map could increase the AUC value to 0.818 (95%CI 0.635‒0.861). By combining the histogram features from the 12-mm peritumoral region with habitat features, the AUC value for predicting lymph node metastasis could be further increased to 0.907 (95%CI 0.812‒0.993). The analysis results of NRI, IDI, and DCA all showed that the predictive performance of the model was better than that of clinical diagnosis. Conclusion ·Based on SyMRI, combining the histogram features from the intratumoral and peritumoral regions with habitat features shows high performance in the prediction of lymph node metastasis in oral cancer, providing a feasible, contrast agent-free approach for rapid imaging and prediction of metastatic lymph nodes.

Key words: oral cancer, synthetic magnetic resonance imaging, lymph node metastasis, histogram features, peritumor

CLC Number:

R781.9

WANG Rui, YUAN Ying, TAO Xiaofeng. Application value of synthetic magnetic resonance imaging in predicting cervical lymph node metastasis of oral cancer[J]. Journal of Shanghai Jiao Tong University (Medical Science), 2025, 45(7): 900-909.

Figures/Tables 8

TrendMD

References 34

[1]	BRAV F, LAVERSANNE M, SUNG H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J]. CA Cancer J Clin, 2024, 74: 229-263.
[2]	RADBRUCH A. Gadolinium deposition in the brain: we need to differentiate between chelated and dechelated gadolinium[J]. Radiology, 2018, 288(2): 434-435.
[3]	崔亚东, 李春媚, 陈敏. 合成MRI技术临床应用进展[J]. 中华放射学杂志, 2021, 55(6): 677-681.
	CUI Y D, LI C M, CHEN M. Progress in clinical application of synthetic MRI technology [J]. Chinese Journal of Radiology, 2021, 55(6): 677-681.
[4]	MAI J, ABUBRIG M, LEHMANN T, et al. T2 mapping in prostate cancer[J]. Invest Radiol, 2019, 54(3): 146-152.
[5]	WANG F, YANG Q, ZHANG Y P, et al. 3D variable flip angle T1 mapping for differentiating benign and malignant liver lesions at 3T: comparison with diffusion weighted imaging[J]. BMC Med Imaging, 2022, 22(1): 146.
[6]	GE Y X, HU S D, WANG Z, et al. Feasibility and reproducibility of T2 mapping and DWI for identifying malignant lymph nodes in rectal cancer[J]. Eur Radiol, 2021, 31(5): 3347-3354.
[7]	WANG P, HU S D, WANG X Y, et al. Synthetic MRI in differentiating benign from metastatic retropharyngeal lymph node: combination with diffusion-weighted imaging[J]. Eur Radiol, 2023, 33(1): 152-161.
[8]	ZHAO L, LIANG M, SHI Z, et al. Preoperative volumetric synthetic magnetic resonance imaging of the primary tumor for a more accurate prediction of lymph node metastasis in rectal cancer[J]. Quant Imaging Med Surg, 2021, 11(5): 1805-1816.
[9]	JARDIM J F, GONDAK R, GALVIS M M, et al. A decreased peritumoral CD1a+ cell number predicts a worse prognosis in oral squamous cell carcinoma[J]. Histopathology, 2018, 72(6): 905-913.
[10]	赵楠楠, 朱芸, 汤晓敏, 等. 基于瘤内及瘤周MRI影像组学列线图预测乳腺癌腋窝淋巴结转移[J]. 磁共振成像, 2023, 14(3): 81-87, 94.
	ZHAO N N, ZHU Y, TANG X M, et al. Prediction of axillary lymph node metastasis in breast cancer based on intratumoral and peritumoral MRI radiomics nomogram[J]. Chinese Journal of Magnetic Resonance Imaging, 2023, 14(3): 81-87, 94.
[11]	CUI Y, THA K K, TERASAKA S, et al. Prognostic imaging biomarkers in glioblastoma: development and independent validation on the basis of multiregion and quantitative analysis of MR images[J]. Radiology, 2016, 278(2): 546-553.
[12]	PARK J E, KIM H S, KIM N, et al. Spatiotemporal heterogeneity in multiparametric physiologic MRI is associated with patient outcomes in IDH-wildtype glioblastoma[J]. Clin Cancer Res, 2021, 27(1): 237-245.
[13]	傅旖, 马辰莺, 张露, 等. 生境分析在恶性肿瘤影像组学中的研究进展[J]. 国际肿瘤学杂志, 2024, (5): 292-297.
	FU Y, MA C Y, ZHANG L, et al. Research progress of habitat analysis in radiomics of malignant tumors[J]. Journal of International Oncology, 2024, (5): 292-297.
[14]	GOURTSOYIANNI S, DOUMOU G, PREZZI D, et al. Primary rectal cancer: repeatability of global and local-regional MR imaging texture features[J]. Radiology, 2017, 284(2): 552-561.
[15]	CUI Y F, YANG X T, DU X S, et al. Whole-tumour diffusion kurtosis MR imaging histogram analysis of rectal adenocarcinoma: correlation with clinical pathologic prognostic factors[J]. Eur Radiol, 2018, 28(4): 1485-1494.
[16]	YANG L Q, LIU D, FANG X, et al. Rectal cancer: can T2WI histogram of the primary tumor help predict the existence of lymph node metastasis?[J]. Eur Radiol, 2019, 29(12): 6469-6476.
[17]	REN J L, YUAN Y, TAO X F. Histogram analysis of diffusion-weighted imaging and dynamic contrast-enhanced MRI for predicting occult lymph node metastasis in early-stage oral tongue squamous cell carcinoma[J]. Eur Radiol, 2022, 32(4): 2739-2747.
[18]	REN J, LI Y, LIU X Y, et al. Diagnostic performance of ADC values and MRI-based radiomics analysis for detecting lymph node metastasis in patients with cervical cancer: a systematic review and meta-analysis[J]. Eur J Radiol, 2022, 156: 110504.
[19]	YANG X, LU Z, TAN X Y, et al. Evaluating the added value of synthetic magnetic resonance imaging in predicting sentinel lymph node status in breast cancer[J]. Quant Imaging Med Surg, 2024, 14(6): 3789-3802.
[20]	ROONEY W D, JOHNSON G, LI X, et al. Magnetic field and tissue dependencies of human brain longitudinal ¹H₂O relaxation in vivo[J]. Magn Reson Med, 2007, 57(2): 308-318.
[21]	IORDANISHVILI E, SCHALL M, LOUÇÃO R, et al. Quantitative MRI of cerebral white matter hyperintensities: a new approach towards understanding the underlying pathology[J]. Neuroimage, 2019, 202: 116077.
[22]	DENG S Z, WANG S M, SHI X K, et al. Microenvironment in oral potentially malignant disorders: multi-dimensional characteristics and mechanisms of carcinogenesis[J]. Int J Mol Sci, 2022, 23(16): 8940.
[23]	谢佳培, 张卫东, 朱婧怡, 等. 磁共振T1、T2值在脑胶质瘤分级及细胞增殖活性预测中的临床价值[J]. 磁共振成像, 2021, 12(1):15-20.
	XIE J P, ZHANG W D, ZHU J Y, et al. The clinical value of MRI T1 and T2 values in predicting glioma grading and cell proliferation activity [J]. Chinese Journal of Magnetic Resonance Imaging, 2021, 12(1): 15-20.
[24]	JOHNSON D E, BURTNESS B, LEEMANS C R, et al. Head and neck squamous cell carcinoma[J]. Nat Rev Dis Primers, 2020, 6(1): 92.
[25]	章梓郁, 杜楠楠, 李源茂, 等. 肿瘤淋巴结转移: 肿瘤细胞与免疫系统的博弈[J]. 中华转移性肿瘤杂志, 2022, 5(4): 357-361.
	ZHANG Z Y, DU N N, LI Y M, et al. Lymph node metastasis of tumor: the battle between tumor cells and immune system[J]. Chinese Journal of Metastatic Cancer, 2022, 5(4): 357-361.
[26]	GRACIEN R M, REITZ S C, HOF S M, et al. Changes and variability of proton density and T1 relaxation times in early multiple sclerosis: MRI markers of neuronal damage in the cerebral cortex[J]. Eur Radiol, 2016, 26(8): 2578-2586.
[27]	CHO Y A, YOON H J, LEE J I, et al. Relationship between the expressions of PD-L1 and tumor-infiltrating lymphocytes in oral squamous cell carcinoma[J]. Oral Oncol, 2011, 47(12): 1148-1153.
[28]	MERMOD M, BONGIOVANNI M, PETROVA T V, et al. Prediction of occult lymph node metastasis in squamous cell carcinoma of the oral cavity and the oropharynx using peritumoral Prospero homeobox protein 1 lymphatic nuclear quantification[J]. Head Neck, 2016, 38(9): 1407-1415.
[29]	WU Q X, WANG S, CHEN X, et al. Radiomics analysis of magnetic resonance imaging improves diagnostic performance of lymph node metastasis in patients with cervical cancer[J]. Radiother Oncol, 2019, 138: 141-148.
[30]	WANG Y X, SHANG Y Y, GUO Y X, et al. Clinical study on the prediction of ALN metastasis based on intratumoral and peritumoral DCE-MRI radiomics and clinico-radiological characteristics in breast cancer[J]. Front Oncol, 2024, 14: 1357145.
[31]	BAI H L, XIA W, JI X F, et al. Multiparametric magnetic resonance imaging-based peritumoral radiomics for preoperative prediction of the presence of extracapsular extension with prostate cancer[J]. J Magn Reson Imaging, 2021, 54(4): 1222-1230.
[32]	WANG F, TAN R K, FENG K, et al. Magnetic resonance imaging-based radiomics features associated with depth of invasion predicted lymph node metastasis and prognosis in tongue cancer[J]. J Magn Reson Imaging, 2022, 56(1): 196-209.
[33]	WANG S X, LIU X W, WU Y, et al. Habitat-based radiomics enhances the ability to predict lymphovascular space invasion in cervical cancer: a multi-center study[J]. Front Oncol, 2023, 13: 1252074.
[34]	HUANG H Z, CHEN H, ZHENG D Z, et al. Habitat-based radiomics analysis for evaluating immediate response in colorectal cancer lung metastases treated by radiofrequency ablation[J]. Cancer Imaging, 2024, 24(1): 44.

Indicator	Non-metastasis (n=42)	Metastasis (n=19)	P value
Age/year	56.3±16.2	54.9±12.5	0.486
Female/n (%)	15 (35.7)	2 (10.5)	0.066
Anatomical site/n (%)			0.306
Tongue	32 (76.2)	18 (94.7)
Buccal mucosa	10 (23.8)	1 (5.3)
T stage/n (%)			0.094
T1/T2	22 (52.4)	5 (26.3)
T3/T4	20 (47.6)	14 (73.7)
DOI/n (%)			0.094
≤10 mm	22 (52.4)	5 (26.3)
>10 mm	20 (47.6)	14 (73.7)
MRI diagnostic report/n (%)
Positive	9 (21.4)	14 (73.7)	<0.001
Negative	33 (78.6)	5 (26.3)

Indicator	Non-metastasis (n=42)	Metastasis (n=19)	P value
Age/year	56.3±16.2	54.9±12.5	0.486
Female/n (%)	15 (35.7)	2 (10.5)	0.066
Anatomical site/n (%)			0.306
Tongue	32 (76.2)	18 (94.7)
Buccal mucosa	10 (23.8)	1 (5.3)
T stage/n (%)			0.094
T1/T2	22 (52.4)	5 (26.3)
T3/T4	20 (47.6)	14 (73.7)
DOI/n (%)			0.094
≤10 mm	22 (52.4)	5 (26.3)
>10 mm	20 (47.6)	14 (73.7)
MRI diagnostic report/n (%)
Positive	9 (21.4)	14 (73.7)	<0.001
Negative	33 (78.6)	5 (26.3)

Variable		Metastasis	Non- metastasis	P value		AUC (95%CI)	Sensitivity/%	Specificity/%	Accuracy/%
Variable		Metastasis	Non- metastasis	Univariate	Multivariate	AUC (95%CI)	Sensitivity/%	Specificity/%	Accuracy/%
Tumor in situ						0.798 (0.673‒0.924)	78.95	83.33	73.8
T1	P₁₀/ms	906.88±146.33	768.07±156.07	0.001	0.007
T1	Total energy/(×10¹⁰ ms^-1)	2.01±4.53	5.25±6.83	0.004	0.077
PD	Total energy/(×10⁸ ms^-1)	1.68±2.01	0.65±0.13	0.003	0.048
5 mm						0.812 (0.689‒0.935)	78.95	80.95	72.1
T1	Mean/ms	1 149.42±143.60	996.79±160.04	0.001	0.701
	Median/ms	1 149±143.6	996.8±160.0	<0.001	0.131
	Total energy/(×10¹⁰ ms^-1)	7.53±7.18	3.61±6.14	0.001	0.049
PD	Total energy/(×10⁸ ms^-1)	2.69±2.28	1.42±1.71	0.004	0.049
7 mm						0.822 (0.701‒0.944)	84.21	78.57	73.8
T1	Mean/ms	1 208.57±158.09	1064.47±184.70	0.001	0.916
	Median/ms	1 114±132.8	961.1±147.1	0.001	0.041
	Total energy/(×10¹⁰ ms^-1)	8.58±7.89	4.34±6.89	0.001	0.023
PD	Total energy/(×10⁸ ms^-1)	3.21±2.71	1.80±1.96	0.005	0.026
10 mm						0.891 (0.790‒0.992)	78.95	92.86	86.9
T1	Mean/ms	1 175±152.8	1040±178.4	0.006	0.780
	Median/ms	1 079.24±127.72	931.57±156.00	0.013	0.019
	Total energy/(×10¹⁰ ms^-1)	1.02±7.64	5.64±7.66	0.005	0.007
PD	Total energy/(×10⁸ ms^-1)	4.00±2.51	2.48±2.20	0.002	0.008
12 mm						0.897 (0.800‒0.995)	78.95	92.86	85.2
T1	Mean/ms	1 164±164.5	1033±177.1	0.007	0.733
	Median/ms	1 068.05±121.88	922.69±155.50	0.001	0.015
	Total energy/(×10¹⁰ ms^-1)	10.88±7.80	6.16±7.89	0.001	0.005
PD	Total energy/(×10⁸ ms^-1)	4.34±2.60	2.76±2.27	0.003	0.005
15 mm						0.746 (0.612‒0.870)	84.21	57.14	73.8
T1	Median/ms	982.02±245.12	886.61±203.35	0.008	0.657
T1	Total energy/(×10¹⁰ ms^-1)	13.0±10.53	7.60±8.74	0.004	0.023
PD	Total energy/(×10¹⁰ ms^-1)	239.61±149.49	60.69±26.65	0.006	0.037

Variable		Metastasis	Non- metastasis	P value		AUC (95%CI)	Sensitivity/%	Specificity/%	Accuracy/%
Variable		Metastasis	Non- metastasis	Univariate	Multivariate	AUC (95%CI)	Sensitivity/%	Specificity/%	Accuracy/%
Tumor in situ						0.798 (0.673‒0.924)	78.95	83.33	73.8
T1	P₁₀/ms	906.88±146.33	768.07±156.07	0.001	0.007
T1	Total energy/(×10¹⁰ ms^-1)	2.01±4.53	5.25±6.83	0.004	0.077
PD	Total energy/(×10⁸ ms^-1)	1.68±2.01	0.65±0.13	0.003	0.048
5 mm						0.812 (0.689‒0.935)	78.95	80.95	72.1
T1	Mean/ms	1 149.42±143.60	996.79±160.04	0.001	0.701
	Median/ms	1 149±143.6	996.8±160.0	<0.001	0.131
	Total energy/(×10¹⁰ ms^-1)	7.53±7.18	3.61±6.14	0.001	0.049
PD	Total energy/(×10⁸ ms^-1)	2.69±2.28	1.42±1.71	0.004	0.049
7 mm						0.822 (0.701‒0.944)	84.21	78.57	73.8
T1	Mean/ms	1 208.57±158.09	1064.47±184.70	0.001	0.916
	Median/ms	1 114±132.8	961.1±147.1	0.001	0.041
	Total energy/(×10¹⁰ ms^-1)	8.58±7.89	4.34±6.89	0.001	0.023
PD	Total energy/(×10⁸ ms^-1)	3.21±2.71	1.80±1.96	0.005	0.026
10 mm						0.891 (0.790‒0.992)	78.95	92.86	86.9
T1	Mean/ms	1 175±152.8	1040±178.4	0.006	0.780
	Median/ms	1 079.24±127.72	931.57±156.00	0.013	0.019
	Total energy/(×10¹⁰ ms^-1)	1.02±7.64	5.64±7.66	0.005	0.007
PD	Total energy/(×10⁸ ms^-1)	4.00±2.51	2.48±2.20	0.002	0.008
12 mm						0.897 (0.800‒0.995)	78.95	92.86	85.2
T1	Mean/ms	1 164±164.5	1033±177.1	0.007	0.733
	Median/ms	1 068.05±121.88	922.69±155.50	0.001	0.015
	Total energy/(×10¹⁰ ms^-1)	10.88±7.80	6.16±7.89	0.001	0.005
PD	Total energy/(×10⁸ ms^-1)	4.34±2.60	2.76±2.27	0.003	0.005
15 mm						0.746 (0.612‒0.870)	84.21	57.14	73.8
T1	Median/ms	982.02±245.12	886.61±203.35	0.008	0.657
T1	Total energy/(×10¹⁰ ms^-1)	13.0±10.53	7.60±8.74	0.004	0.023
PD	Total energy/(×10¹⁰ ms^-1)	239.61±149.49	60.69±26.65	0.006	0.037