Predictive value of a combined model for lymph node metastasis in NSCLC based on primary lesion radiomics from <sup>18</sup>F-FDG PET/CT

Abstract

Abstract:

ObjectiveTo evaluate the value of a combined model based on primary lesion¹⁸F-fluorodeoxyglucose（¹⁸F-FDG） PET/CT radiomics for predicting lymph node metastasis in non-small cell lung cancer（NSCLC）.MethodsA retrospective analysis was conducted on the clinical data of 203 NSCLC patients who underwent pre-treatment PET/CT imaging at Nanjing Drum Tower Hospital from June 2013 to July 2023. Patients were randomly assigned to the training set（n=142） and the validation set（n=61） at a ratio of 7∶3. A predictive model was developed in the training set， and its predictive performance and clinical application value were assessed in both the training and validation sets. Traditional PET/CT parameters and PET/CT radiomics features of the primary lesion were obtained by 3D-slicer software. Least absolute shrinkage and selection operator（LASSO）， random forest， and extreme gradient boosting were performed to extract features. Support vector machine was used to construct a radiomics score（Radscore）. Univariate and multivariate logistic regression analysis was used to predict the influencing factors of lymph node metastasis in NSCLC patients and to establish models. Predictive performance of the models was evaluated by receiver operator characteristic（ROC） curves and clinical application value was assessed by calibration curves and decision curve analysis（DCA）.ResultsAmong 203 NSCLC patients， 116 had lymph node metastasis， with 64 cases in the training set and 52 cases in the validation set. Three complementary classical machine learning methods were used for feature screening， and finally 10 radiomics features were obtained. The optimal threshold for Radscore-PET was 0.43 and the optimal threshold for Radscore-CT was 0.39. Univariate analysis showed that， sex（OR=0.48， 95%CI：0.24-0.95，P=0.036）， tumor marker levels（OR=3.81， 95%CI：1.84-7.91，P<0.001）， long diameter of tumor（OR=2.56， 95%CI：1.27-5.16，P=0.009）， short diameter of tumor（OR=3.73， 95%CI：1.75-7.92，P=0.001）， vacuolar sign（OR=0.32， 95%CI：0.12-0.86，P=0.024）， ring-like metabolism（OR=3.67， 95%CI：1.33-10.13，P=0.012）， maximum standardized uptake value（SUV_max）（OR=6.57， 95%CI：3.03-14.25，P<0.001）， metabolic tumor volume（MTV）（OR=2.91， 95%CI：1.43-5.92，P=0.003）， total lesion glycolysis（TLG）（OR=4.23， 95%CI：2.08-8.59，P<0.001）， Radscore-PET（OR=21.93， 95%CI：9.04-53.20，P<0.001） and Radscore-CT（OR=13.72， 95%CI：6.12-30.76，P<0.001） were all influencing factors for predicting lymph node metastasis in NSCLC patients. Multivariate analysis showed that， tumor marker levels（OR=2.55， 95%CI：1.11-5.90，P=0.028）， vacuolar sign（OR=0.26， 95%CI：0.08-0.83，P=0.023）， SUV_max（OR=5.94， 95%CI：1.99-17.75，P=0.001）， Radscore-PET（OR=25.51， 95%CI：5.92-110.22，P<0.001）， and Radscore-CT（OR=8.68， 95%CI：2.73-27.61，P<0.001） were independent influencing factors for predicting lymph node metastasis in patients with NSCLC. Based on the above independent influencing factors， models were constructed：the traditional model（tumor marker levels， vacuolar sign， SUV_max）， the PET model（SUV_max， Radscore-PET）， the CT model（vacuolar sign， Radscore-CT）， and the combined model（tumor marker levels， vacuolar sign， SUV_max， Radscore-PET， Radscore-CT）. ROC curve analysis showed that， the area under curve（AUC） of the traditional， PET， CT， and combined models in the training set were 0.75（95%CI：0.67-0.82）， 0.90（95%CI：0.84-0.95）， 0.85（95%CI：0.78-0.90）， and 0.94（95%CI：0.88-0.97）， respectively. The predictive value of the combined model was higher than that of the traditional model（Z=5.01，P<0.001）， the PET model（Z=1.99，P=0.047）， and the CT model（Z=3.25，P=0.001）. In the validation set， the AUCs for the traditional model， PET model， CT model， and combined model were 0.65（95%CI：0.52-0.77）， 0.86（95%CI：0.74-0.93）， 0.85（95%CI：0.73-0.93）， and 0.90（95%CI：0.80-0.96）， respectively. The predictive value of the combined model was superior to that of the traditional model（Z=3.23，P=0.001）. The sensitivity and specificity of the combined model in the training set were 84.37% and 91.03%， while in the validation set， the sensitivity and specificity were 82.61% and 94.74%， respectively. Calibration curves showed a good agreement between the predicted and actual probabilities in both the training and validation sets. DCA showed that the combined models had good discriminative ability in both the training and validation sets.ConclusionsTumor marker levels， vacuolar sign， SUV_max， Radscore-PET， and Radscore-CT are all independent influencing factors for predicting lymph node metastasis in patients with NSCLC. The combined model based on these factors demonstrates excellent predictive performance and clinical application value for predicting lymph node metastasis in NSCLC.

Key words:Carcinoma,non-small-cell lung,Positron emission tomography computed tomography,Radiomics,Lymph node metastasis

Lai Ruihe, Teng Yue, Rong Jian, Sheng Dandan, Geng Yuzhi, Chen Jianxin, Jiang Chong, Ding Chongyang, Zhou Zhengyang. Predictive value of a combined model for lymph node metastasis in NSCLC based on primary lesion radiomics from¹⁸F-FDG PET/CT[J]. Journal of International Oncology, 2025, 52(3): 144-151.

Figures/Tables8

项目	训练集（n=142）	验证集（n=61）	χ²值	P值	项目	训练集（n=142）	验证集（n=61）	χ²值	P值
年龄（岁）					钙化
≤60	45（31.69）	23（37.70）	0.69	0.405	无	136（95.77）	58（95.08）	<0.01	>0.999
>60	97（68.31）	38（62.30）	0.69	0.405	有	6（4.23）	3（4.92）	<0.01	>0.999
性别					毛刺征
男	84（59.15）	35（57.38）	0.06	0.814	无	14（9.86）	6（9.84）	<0.01	0.996
女	58（40.85）	26（42.62）	0.06	0.814	有	128（90.14）	55（90.16）	<0.01	0.996
吸烟史					分叶征
无	101（71.13）	47（77.05）	0.76	0. 384	无	24（16.90）	8（13.11）	0.46	0.497
有	41（28.87）	14（22.95）	0.76	0. 384	有	118（83.10）	53（86.89）	0.46	0.497
饮酒史					血管集束征
无	122（85.92）	56（91.80）	1.37	0.242	无	99（69.72）	41（67.21）	0.13	0.724
有	20（14.08）	5（8.20）	1.37	0.242	有	43（30.28）	20（32.79）	0.13	0.724
肿瘤标志物水平					空泡征
正常	57（40.14）	20（32.79）	0.98	0.322	无	117（82.39）	48（78.69）	0.39	0.535
升高^a	85（59.86）	41（67.21）	0.98	0.322	有	25（17.61）	13（21.31）	0.39	0.535
病理类型					环形代谢
鳞状细胞癌	105（73.94）	45（73.77）	<0.01	0.979	无	121（85.21）	57（93.44）	2.68	0.102
腺癌	37（26.06）	16（26.23）	<0.01	0.979	有	21（14.79）	4（6.56）	2.68	0.102
肿瘤位置					SUV_max
右叶	80（56.34）	45（73.77）	5.48	0.019	≤7.83	59（41.55）	25（40.98）	0.01	0.940
左叶	62（43.66）	16（26.23）	5.48	0.019	>7.83	83（58.45）	36（59.02）	0.01	0.940
外周型/中央型					MTV（cm³）
外周型	127（89.44）	55（90.16）	0.02	0.876	≤4.64	57（40.14）	25（40.98）	0.01	0.911
中央型	15（10.56）	6（9.84）	0.02	0.876	>4.64	85（59.86）	36（59.02）	0.01	0.911
肿瘤长径（cm）					TLG（g）
≤2.35	57（40.14）	27（44.26）	0.30	0.585	≤24.81	69（48.59）	30（49.18）	0.01	0.939
>2.35	85（59.86）	34（55.74）	0.30	0.585	>24.81	73（51.41）	31（50.82）	0.01	0.939
肿瘤短径（cm）
≤1.77	51（35.92）	32（52.46）	4.83	0.028
>1.77	91（64.08）	29（47.54）	4.83	0.028

变量	β值	SE值	Waldχ²值	OR值	95%CI	P值
年龄	-0.36	0.36	0.97	0.70	0.34~1.42	0.325
性别	-0.74	0.35	4.38	0.48	0.24~0.95	0.036
吸烟史	0.49	0.37	1.70	1.63	0.78~3.38	0.192
饮酒史	0.46	0.49	0.92	1.59	0.62~4.12	0.338
肿瘤标志物水平^a	1.34	0.37	12.91	3.81	1.84~7.91	<0.001
病理类型	0.49	0.39	1.62	1.63	0.78~3.47	0.203
肿瘤位置	0.35	0.34	1.08	1.42	0.73~2.78	0.299
外周型/中央型	0.68	0.56	1.47	1.96	0.66~5.85	0.225
肿瘤长径	0.94	0.36	6.85	2.56	1.27~5.16	0.009
肿瘤短径	1.32	0.39	11.71	3.73	1.75~7.92	0.001
钙化	1.88	1.11	2.86	6.53	0.74~57.36	0.091
毛刺征	-0.22	0.56	0.15	0.80	0.27~2.42	0.697
分叶征	-0.24	0.45	0.28	0.79	0.33~1.90	0.595
血管集束征	-0.46	0.37	1.53	0.63	0.30~1.31	0.216
空泡征	-1.14	0.50	5.09	0.32	0.12~0.86	0.024
环形代谢	1.30	0.52	6.33	3.67	1.33~10.13	0.012
SUV_max	1.88	0.40	22.70	6.57	3.03~14.25	<0.001
MTV	1.07	0.36	8.68	2.91	1.43~5.92	0.003
TLG	1.44	0.36	15.93	4.23	2.08~8.59	<0.001
Radscore-PET	3.02	0.45	46.62	21.93	9.04~53.20	<0.001
Radscore-CT	2.62	0.41	40.44	13.72	6.12~30.76	<0.001

变量	β值	SE值	Waldχ²值	OR值	95%CI	P值
性别	-0.65	0.44	2.24	0.52	0.22~1.22	0.134
肿瘤标志物水平^a	0.94	0.43	4.81	2.55	1.11~5.90	0.028
肿瘤长径	-0.75	0.83	0.82	0.47	0.09~2.39	0.365
肿瘤短径	0.31	0.76	0.17	1.37	0.31~6.11	0.681
空泡征	-1.37	0.60	5.20	0.26	0.08~0.83	0.023
环形代谢	0.01	0.63	<0.01	1.01	0.29~3.50	0.983
SUV_max	1.78	0.56	10.19	5.94	1.99~17.75	0.001
MTV	0.69	0.69	1.00	1.20	0.52~7.71	0.316
TLG	0.14	0.77	0.03	1.15	0.26~5.19	0.856
Radscore-PET	3.24	0.75	18.87	25.51	5.92~110.22	<0.001
Radscore-CT	2.16	0.59	13.39	8.68	2.73~27.61	<0.001

分组	模型	AUC	95%CI	敏感性(%)	特异性(%)	准确性(%)
训练集	传统模型	0.75^a	0.67~0.82	57.81	84.62	63.38
	PET模型	0.90^b	0.84~0.95	89.06	78.21	79.58
	CT模型	0.85^c	0.78~0.90	90.62	67.95	77.46
	联合模型	0.94	0.88~0.97	84.37	91.03	84.51
验证集	传统模型	0.65^d	0.52~0.77	91.30	47.37	62.30
	PET模型	0.86^e	0.74~0.93	78.26	76.32	78.69
	CT模型	0.85^f	0.73~0.93	78.26	94.74	81.97
	联合模型	0.90	0.80~0.96	82.61	94.74	83.61

References20

[1]	Sun J, Wu S, Jin Z, et al. Lymph node micrometastasis in non-small cell lung cancer[J].Biomed Pharmacother,2022,149: 112817. DOI:10.1016/j.biopha.2022.112817.
[2]	Siegel RL, Miller KD, Wagle NS, et al. Cancer statistics, 2023[J].CA Cancer J Clin,2023,73(1): 17-48. DOI:10.3322/caac.21763.
[3]	Hua J, Li L, Liu L, et al. The diagnostic value of metabolic, morphological and heterogeneous parameters of¹⁸F-FDG PET/CT in mediastinal lymph node metastasis of non-small cell lung cancer[J].Nucl Med Commun,2021,42(11): 1247-1253. DOI:10.1097/MNM.0000000000001456.
[4]	Moon Y, Choi SY, Park JK, et al. Risk factors for occult lymph node metastasis in peripheral non-small cell lung cancer with invasive component size 3 cm or less[J].World J Surg,2020,44(5): 1658-1665. DOI:10.1007/s00268-019-05365-5.
[5]	Seol HY, Kim YS, Kim SJ. Predictive value of¹⁸F-fluorodeoxyglucose positron emission tomography or positron emission tomography/computed tomography for assessment of occult lymph node metastasis in non-small cell lung cancer[J].Oncology,2021,99(2): 96-104. DOI:10.1159/000509988.
[6]	Zhong Y, She Y, Deng J, et al. Deep learning for prediction of N₂metastasis and survival for clinical stage Ⅰ non-small cell lung cancer[J].Radiology,2022,302(1): 200-211. DOI:10.1148/radiol.2021210902.
[7]	Xie Y, Zhao H, Guo Y, et al. A PET/CT nomogram incorporating SUVmax and CT radiomics for preoperative nodal staging in non-small cell lung cancer[J].Eur Radiol,2021,31(8): 6030-6038. DOI:10.1007/s00330-020-07624-9. pmid:33560457
[8]	Vansteenkiste J, De Ruysscher D, Eberhardt WE, et al. Early and locally advanced non-small-cell lung cancer (NSCLC): ESMO clinical practice guidelines for diagnosis, treatment and follow-up[J].Ann Oncol,2013,24 Suppl 6: vi89-vi98. DOI:10.1093/annonc/mdt241.
[9]	Nicholson AG, Chansky K, Crowley J, et al. The international association for the study of lung cancer lung cancer staging project: proposals for the revision of the clinical and pathologic staging of small cell lung cancer in the forthcoming eighth edition of the TNM classification for lung cancer[J].J Thorac Oncol,2016,11(3): 300-311. DOI:10.1016/j.jtho.2015.10.008. pmid:26723244
[10]	Postmus PE, Kerr KM, Oudkerk M, et al. Early and locally advanced non-small-cell lung cancer (NSCLC): ESMO clinical practice guidelines for diagnosis, treatment and follow-up[J].Ann Oncol,2017,28(suppl_4): iv1-iv21. DOI:10.1093/annonc/mdx222.
[11]	Fang C, Xiang Y, Han W. Preoperative risk factors of lymph node metastasis in clinical N₀lung adenocarcinoma of 3 cm or less in diameter[J].BMC Surg,2022,22(1): 153. DOI:10.1186/s12893-022-01605-z.
[12]	Zheng Y, Ju S, Huang R, et al. Lymph node metastasis risk factors in clinical stage ⅠA3 lung adenocarcinoma[J].J Cancer Res Ther,2023,19(1): 34-38. DOI:10.4103/jcrt.jcrt_1370_22.
[13]	王清北, 朱林, 吴正国. NSCLC肺内叶段淋巴结转移规律及全胸腔镜肺叶切除术中淋巴结清扫方式的选择[J].国际肿瘤学杂志,2024,51(9): 569-577. DOI:10.3760/cma.j.cn371439-20240422-00095.
[14]	Zhai X, Guo Y, Qian X. Combination of fluorine-18 fluorodeoxyglucose positron-emission tomography/computed tomography (¹⁸F-FDG PET/CT) and tumor markers to diagnose lymph node metastasis in non-small cell lung cancer (NSCLC): a retrospective and prospective study[J].Med Sci Monit,2020,26: e922675. DOI:10.12659/MSM.922675.
[15]	Lederlin M, Puderbach M, Muley T, et al. Correlation of radio- and histomorphological pattern of pulmonary adenocarcinoma[J].Eur Respir J,2013,41(4): 943-951. DOI:10.1183/09031936.00056612. pmid:22835610
[16]	Cong M, Yao H, Liu H, et al. Development and evaluation of a venous computed tomography radiomics model to predict lymph node metastasis from non-small cell lung cancer[J].Medicine (Baltimore),2020,99(18): e20074. DOI:10.1097/MD.0000000000020074.
[17]	Xu ZQ, Xie LJ, Fan W, et al. Risk factors for mediastinal lymph node metastasis in non-small-cell lung cancer by PET/CT[J].Nucl Med Commun,2014,35(5): 466-471. DOI:10.1097/MNM.0000000000000077.
[18]	Nakanishi K, Nakamura S, Sugiyama T, et al. Diagnostic utility of metabolic parameters on FDG PET/CT for lymph node metastasis in patients with cN₂non-small cell lung cancer[J].BMC Cancer,2021,21(1): 983. DOI:10.1186/s12885-021-08688-6. pmid:34474680
[19]	Huang Y, Jiang X, Xu H, et al. Preoperative prediction of mediastinal lymph node metastasis in non-small cell lung cancer based on¹⁸F-FDG PET/CT radiomics[J].Clin Radiol,2023,78(1): 8-17. DOI:10.1016/j.crad.2022.08.140.
[20]	周见远.¹⁸F-FDG PET/CT影像组学在非小细胞肺癌的临床研究[D]. 武汉: 华中科技大学,2021. DOI:10.27157/d.cnki.ghzku.2021.000423.

Predictive value of a combined model for lymph node metastasis in NSCLC based on primary lesion radiomics from¹⁸F-FDG PET/CT

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables8

References20

Related Articles15

Recommended Articles

Metrics

Comments

[1]	Yang Shengjun, Ren Jiang, Yang Dan, Long Yu, Shang Qunxian.Expression levels and clinical significance of miR-4262，NRG1 in non-small cell lung cancer tissues[J]. Journal of International Oncology, 2025, 52(3): 129-135.
[2]	Han Shuang.CT feature analysis and predictive value of visceral pleural invasion in stage Ⅰ lung adenocarcinoma with peripheral solid nodules[J]. Journal of International Oncology, 2025, 52(3): 136-143.
[3]	Wang Zhiying, Sheng Lijun.Research progress of peripheral blood biomarkers in immunotherapy of non-small cell lung cancer[J]. Journal of International Oncology, 2025, 52(3): 180-185.
[4]	Ye Yongying, Zou Yan, Chen Tianming, Wu Weili.Research progress of clock gene Period family in head and neck squamous cell carcinoma[J]. Journal of International Oncology, 2025, 52(2): 113-118.
[5]	Chen Ruyan, Fu Zhenming.Current status and advances in immunotherapy for advanced renal cell carcinoma[J]. Journal of International Oncology, 2025, 52(2): 124-128.
[6]	Ma Peihan, Zhang Lingmin, Li Qian, Lu Ning, Wen Hua, Zhang Mingxin.Effects of ALKBH5 on the malignant biological behavior of esophageal squamous cell carcinoma and the related mechanism[J]. Journal of International Oncology, 2025, 52(2): 79-88.
[7]	Wu Xiaowei, Hu Ge, Chen Li, Qian Xiaotao, Cui Xiangli, Zhu Fengqin.Predictive value of pre-radiotherapy maximum tumor diameter and peripheral blood NLR for esophageal fistula in esophageal squamous carcinoma patients[J]. Journal of International Oncology, 2025, 52(1): 38-42.
[8]	Yu Yang, Tang Shimin, Yang Lu, Li Na.Research progress in treatment strategies and prognostic factors for stage pT_2-3N₀M₀thoracic esophageal squamous cell carcinoma[J]. Journal of International Oncology, 2025, 52(1): 43-47.
[9]	Yuan Shengfang, Ren Jie, Lin Weijia, Ji Zexuan, Zhang Changhong, Wang Bu.Prognostic value of EGFR co-mutation status in patients with advanced lung adenocarcinoma[J]. Journal of International Oncology, 2024, 51(9): 556-562.
[10]	Huang Bo, Wang Peng.Early predictive value of serum sCD163， IFN-γ combined with TGF-β1 in the development of radiation-induced pneumonia in NSCLC patients[J]. Journal of International Oncology, 2024, 51(9): 563-568.
[11]	Wang Qingbei, Zhu Lin, Wu Zhengguo.Pattern of lymph node metastasis in the lung lobe of NSCLC and selection of lymph node dissection methods in complete video-assisted thoracoscopic lobectomy surgery[J]. Journal of International Oncology, 2024, 51(9): 569-577.
[12]	Ni Guoying, Huang Qian, Liang Hongxiang, Yang Zhiyong, Ding Yingli.Analysis of changes in serum miR-499 and miR-362 levels and their relationship with prognosis in advanced NSCLC patients[J]. Journal of International Oncology, 2024, 51(8): 487-492.
[13]	Lyu Jun, Xiong Hao, Zheng Yanqiu, Dong Li.Risk factors and prediction model construction of pulmonary IFD in patients with NSCLC after radiotherapy[J]. Journal of International Oncology, 2024, 51(8): 493-497.
[14]	Meng Kexin, Lu Haijun.Oral microbiota：a biomarker for the diagnosis and prognosis of oral squamous cell carcinoma[J]. Journal of International Oncology, 2024, 51(8): 515-519.
[15]	Mo Huimin, Cai Yusen, Zhang Zengrui, Zhu Wentian.Research progress of the combined application of immune checkpoint inhibitors in the treatment of hepatocellular carcinoma with portal vein tumor thrombus[J]. Journal of International Oncology, 2024, 51(8): 520-525.

Predictive value of a combined model for lymph node metastasis in NSCLC based on primary lesion radiomics from18F-FDG PET/CT

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables8

References20

Related Articles15

Recommended Articles

Metrics

Comments

Predictive value of a combined model for lymph node metastasis in NSCLC based on primary lesion radiomics from¹⁸F-FDG PET/CT