Virtual reality (VR) simulation has become a useful tool for students to develop clinical skills in medical schools. To investigate the effectiveness of VR simulation teaching thyroid ultrasonography skills in medical students. A total of 106 medical students who had finished basic medical courses were recruited and randomized to VR group and control group. All participants received 2 theoretical lessons, followed by a practical class. The control group had traditional PowerPoint (PPT) teaching to learn the thyroid ultrasonography process, while VR group studied the complete process via the VR simulation platform after PPT teaching. All students took a skill examination to evaluate their performance after two weeks. They were scored by 2 examiners who were blinded to group assignment on the global rating scale of ultrasonography, with a total score of 40 points for 8 components. A month later, all participants were required to complete the thyroid ultrasound examination on the VR platform. Ninety-one students completed the program (VR group, n=47; control group, n=44). The score of VR group (median, 32.5; interquartile range [IQR], 29.5-35.0) outperformed the control group (median, 31.0; IQR, 27.5-33.0, p=0.028) on the global rating scale of thyroid ultrasound examination. Meanwhile, students in the VR group (mean score, 51.5; time, 16.2min) demonstrated better performance and less time than the control group (mean score, 44.3; time, 23.7min) when assessed in the VR platform. The VR simulation platform of thyroid ultrasound examination showed promising results in improving medical students’ skills, making it an alternative or additional approach to traditional teaching patterns in medical education. ChiCTR2200064833

Artificial intelligence (AI) holds enormous potential for noninvasively identifying patients with rectal cancer who could achieve pathological complete response (pCR) following neoadjuvant chemoradiotherapy (nCRT). We aimed to conduct a meta-analysis to summarize the diagnostic performance of image-based AI models for predicting pCR to nCRT in patients with rectal cancer. This study followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. A literature search of PubMed, Embase, Cochrane Library, and Web of Science was performed from inception to July 29, 2023. Studies that developed or utilized AI models for predicting pCR to nCRT in rectal cancer from medical images were included. The Quality Assessment of Diagnostic Accuracy Studies-AI was used to appraise the methodological quality of the studies. The bivariate random-effects model was used to summarize the individual sensitivities, specificities, and areas-under-the-curve (AUCs). Subgroup and meta-regression analyses were conducted to identify potential sources of heterogeneity. Protocol for this study was registered with PROSPERO (CRD42022382374). Thirty-four studies (9933 patients) were identified. Pooled estimates of sensitivity, specificity, and AUC of AI models for pCR prediction were 82

Objective To explore the potential of pre-therapy computed tomography (CT) parameters in predicting the treatment response to initial conventional TACE (cTACE) in intermediate-stage hepatocellular carcinoma (HCC) and develop an interpretable machine learning model. Methods This retrospective study included 367 patients with intermediate-stage HCC who received cTACE as first-line therapy from three centers. We measured the mean attenuation values of target lesions on multi-phase contrast-enhanced CT and further calculated three CT parameters, including arterial (AER), portal venous (PER), and arterial portal venous (APR) enhancement ratios. We used logistic regression analysis to select discriminative features and trained three machine learning models via 5-fold cross-validation. The performance in predicting treatment response was evaluated in terms of discrimination, calibration, and clinical utility. Afterward, a Shapley additive explanation (SHAP) algorithm was leveraged to interpret the outputs of the best-performing model. Results The mean diameter, ECOG performance status, and cirrhosis were the important clinical predictors of cTACE treatment response, by multiple logistic regression. Adding the CT parameters to clinical variables showed significant improvement in performance (net reclassification index, 0.318, P < 0.001). The Random Forest model (hereafter, RF-combined model) integrating CT parameters and clinical variables demonstrated the highest performance on external validation dataset (AUC of 0.800). The decision curve analysis illustrated the optimal clinical benefits of RF-combined model. This model could successfully stratify patients into responders and non-responders with distinct survival ( P = 0.001). Conclusion The RF-combined model can serve as a robust and interpretable tool to identify the appropriate crowd for cTACE sessions, sparing patients from receiving ineffective and unnecessary treatments.

Abstract Postoperative acute kidney injury requiring dialysis (PO‐AKID) is a serious adverse event that not only affects acute morbidity and mortality, but also long‐term prognosis. Here, we developed a practical and explainable web‐based calculator (PO‐AKID‐teller) to detect patients who might experience PO‐AKID after acute type A aortic dissection (ATAAD) surgery. This retrospective study reviewed 549 patients undergoing ATAAD surgery from October 2016 to June 2021. PO‐AKID frequency was 19.7% (108 of 549 patients). The initial dataset was split into an 80% training cohort (n = 439) and a 20% test cohort (n = 110). There were seven predictors that could indicate PO‐AKID, including prior cardiovascular surgery, platelet, serum creatinine, the terminal site of dissection involvement, right coronary artery involvement, estimated blood loss, and urine output. Among six machine learning classifiers, the random forest model exhibited the best predictive performance, with an area under the curve of 0.863 in the training cohort and 0.763 in the test cohort. This model was translated into a web‐based risk calculator PO‐AKID‐teller to estimate an individual's probability of PO‐AKID. The PO‐AKID‐teller can accurately estimate an individual's risk for PO‐AKID in an interpretable manner, which may aid in informed decision‐making, patient counseling, perioperative optimization, and longer‐term care provision.

Deep learning models have been developed for various predictions in glioma; yet, they were constrained by manual segmentation, task-specific design, or a lack of biological interpretation. Herein, we aimed to develop an end-to-end multi-task deep learning (MDL) pipeline that can simultaneously predict molecular alterations and histological grade (auxiliary tasks), as well as prognosis (primary task) in gliomas. Further, we aimed to provide the biological mechanisms underlying the model's predictions. We collected multiscale data including baseline MRI images from 2,776 glioma patients across two private (FAHZU and HPPH, n=1,931) and three public datasets (TCGA, n=213; UCSF, n=410; and EGD, n=222). We trained and internally validated the MDL model using our private datasets, and externally validated it using the three public datasets. We used the model-predicted deep prognosis score (DPS) to stratify patients into low-DPS and high-DPS subtypes. Additionally, a radio-multiomics analysis was conducted to elucidate the biological basis of the DPS. In the external validation cohorts, the MDL model achieved average areas under the curve of 0.892-0.903, 0.710-0.894, and 0.850-0.879 for predicting IDH mutation status, 1p/19q co-deletion status, and tumor grade, respectively. Moreover, the MDL model yielded a C-index of 0.723 in the TCGA and 0.671 in the UCSF for the prediction of overall survival. The DPS exhibits significant correlations with activated oncogenic pathways, immune infiltration patterns, specific protein expression, DNA methylation, tumor mutation burden, and tumor-stroma ratio. Accordingly, our work presents an accurate and biologically meaningful tool for predicting molecular subtypes, tumor grade, and survival outcomes in gliomas, which provides personalized clinical decision-making in a global and non-invasive manner.

PURPOSETumor stage is crucial for prognostic evaluation and therapeutic decisions in locally advanced nasopharyngeal carcinoma (NPC) but is imprecise. We aimed to propose a new prognostic system by integrating quantitative imaging features and clinical factors.MATERIALS AND METHODSThis retrospective study included 1,319 patients with stage III-IVa NPC between April 1, 2010, and July 31, 2019, who underwent pretherapy magnetic resonance imaging (MRI) and received concurrent chemoradiotherapy with or without induction chemotherapy. The hand-crafted and deep-learned features were extracted from MRI for each patient. After feature selection, the clinical score, radiomic score, deep score, and integrative scores were constructed via Cox regression analysis. The scores were validated in two external cohorts. The predictive accuracy and discrimination were measured by the area under the curve (AUC) and risk group stratification. The end points were progression-free survival (PFS), overall survival (OS), and distant metastasis-free survival (DMFS).RESULTSBoth radiomics and deep learning were complementary to clinical variables (age, T stage, and N stage; all P < .05). The clinical-deep score was superior or equivalent to clinical-radiomic score, whereas it was noninferior to clinical-radiomic-deep score (all P > .05). These findings were also verified in the evaluation of OS and DMFS. The clinical-deep score yielded an AUC of 0.713 (95% CI, 0.697 to 0.729) and 0.712 (95% CI, 0.693 to 0.731) in the two external validation cohorts for predicting PFS with good calibration. This scoring system could stratify patients into high- and low-risk groups with distinct survivals (all P < .05).CONCLUSIONWe established and validated a prognostic system integrating clinical data and deep learning to provide an individual prediction of survival for patients with locally advanced NPC, which might inform clinicians in treatment decision making.

Neural decoding is useful for understanding brain functions and developing neural interface applications. However, neural interfaces based on rigid electronics often suffer from recording instability due to the foreign body responses caused by their mechanical mismatch with soft tissues, limiting the longitudinal accuracy of neural decoding methods. Herein, it is reported that flexible electronics can be integrated with machine learning algorithms to achieve long-term reliable neural decoding. Wet-spun conductive polymer microfibers showed mechanical robustness and flexibility, low impedance, and chronic biocompatibility, enabling intracerebral neural recordings in epileptic mice at a high signal-to-noise ratio eight weeks after implantation. When the signals recorded by the flexible electrodes were used in machine learning analyses with diverse complex algorithms, they consistently showed higher prediction accuracy for epileptic seizures than stiff metal electrode signals, particularly in the case of using long-term recordings for testing or small-sample datasets for training. A real-time warning system based on the flexible neural electrodes was built that predicted seizures eight minutes in advance with a low false alarm rate. Our work bridges flexible electronics and artificial intelligence for neural decoding applications such as long-term treatment of chronic neurological disorders.

RATIONALE AND OBJECTIVES:Chinese Thyroid Imaging Reporting and Data Systems (C-TIRADS) was developed to provide a more simplified tool for stratifying thyroid nodules. Here we aimed to validate the efficacy of C-TIRADS in distinguishing benign from malignant and in guiding fine-needle aspiration biopsies in comparison with the American College of Radiology TIRADS (ACR-TIRADS) and European TIRADS (EU-TIRADS). MATERIALS AND METHODS:This study retrospectively included 3438 thyroid nodules (≥10 mm) in 3013 patients (mean age, 47.1 years ± 12.9) diagnosed between January 2013 and November 2019. Ultrasound features of the nodules were evaluated and categorized according to the lexicons of the three TIRADS. We compared these TIRADS by using the area under the receiver operating characteristic curve (AUROC), the area under the precision-recall curve (AUPRC), sensitivity, specificity, net reclassification improvement (NRI), and unnecessary fine-needle aspiration biopsy (FNAB) rate. RESULTS:Of the 3438 thyroid nodules, 707 (20.6%) were malignant. C-TIRADS showed higher discrimination performance (AUROC, 0.857; AUPRC, 0.605) than ACR-TIRADS (AUROC, 0.844; AUPRC, 0.567) and EU-TIRADS (AUROC, 0.802; AUPRC, 0.455). The sensitivity of C-TIRADS (85.3%) was lower than that of ACR-TIRADS (89.1%) but higher than that of EU-TIRADS (78.4%). The specificity of C-TIRADS (76.9%) was similar to that of EU-TIRADS (78.9%) and higher than that of ACR-TIRADS (69.5%). The unnecessary FNAB rate was lowest with C-TIRADS (21.2%), followed by ACR-TIRADS (41.7%) and EU-TIRADS (58.3%). C-TIRADS obtained significant NRI for recommending FNAB over ACR-TIRADS (19.0%, P < 0.001) and EU-TIRADS (25.5%, P < 0.001). CONCLUSION:C-TIRADS may be a clinically applicable tool to manage thyroid nodules, which warrants thorough tests in other geographic settings.

Artificial intelligence (AI) has unparalleled potential to unlock useful information from real-world data to innovate trial design. Here, we discuss how AI can be used to optimize clinical trial design and potentially boost the success rate of clinical trials. Zhang et al. discuss how artificial intelligence (AI) can be used to optimize clinical trial design and potentially boost the success rate of clinical trials. AI has unparalleled potential to leverage real-world data and unlock valuable insights for innovative trial design.

Background The efficacy of adjuvant chemotherapy (AC) on survival outcomes of patients with stage I gastric cancer (GC) after curative resection remains controversial. We aimed to determine whether these patients would benefit from AC.Methods This retrospective study included patients with pathologically confirmed stage I GC who underwent curative resection between November 2010 and December 2020. Patients were divided into AC and non-AC groups, then a 1:1 propensity score matching (PSM) analysis was performed to minimize the selection bias. Potential risk factors including age, pN stage, pT stage, lymphovascular invasion, perineural invasion, tumor size, histological type, and carcinoembryonic antigen level were used as matching covariates. The recurrence-free survival (RFS) and disease-specific survival (DSS) were compared between groups using the Kaplan-Meier method.Results A total of 902 consecutive patients were enrolled and 174 (19.3%) patients were treated with AC. PSM created 123 pairs of patients. Before PSM, patients receiving AC had lower 10-year RFS rates (90% vs 94.6%, P = 0.035) than those who did not receive AC; the two groups had similar 10-year DSS rates (93.8% vs 95.0%, P = 0.240). After PSM, there were no statistical differences in the 10-year RFS (90.9% vs 93.0%, P = 0.507) or DSS rates (93.5% vs 93.6%, P = 0.811) between the two groups. Similar results were found in the stage IA and IB subgroups. Moreover, these findings were not affected by AC cycles.Conclusions The addition of AC could not provide survival benefits for patients with stage I GC after surgery and follow-up is thus recommended. However, large-scale randomized clinical trials are required.

To develop and validate an MRI texture-based machine learning model for the noninvasive assessment of renal function. A retrospective study of 174 diabetic patients (training cohort, n = 123; validation cohort, n = 51) who underwent renal MRI scans was included. They were assigned to normal function (n = 71), mild or moderate impairment (n = 69), and severe impairment groups (n = 34) according to renal function. Four methods of kidney segmentation on T2-weighted images (T2WI) were compared, including regions of interest covering all coronal slices (All-K), the largest coronal slices (LC-K), and subregions of the largest coronal slices (TLCO-K and PIZZA-K). The speeded-up robust features (SURF) and support vector machine (SVM) algorithms were used for texture feature extraction and model construction, respectively. Receiver operating characteristic (ROC) curve analysis was used to evaluate the diagnostic performance of models. The models based on LC-K and All-K achieved the nonsignificantly highest accuracy in the classification of renal function (all p values > 0.05). The optimal model yielded high performance in classifying the normal function, mild or moderate impairment, and severe impairment, with an area under the curve of 0.938 (95

目的建立基于多参数MRI影像组学及临床特征的机器学习预测模型,并评价其治疗前预测鼻咽癌(nasopharyngeal carcinoma,NPC)远处转移风险的效能及临床应用价值.材料与方法回顾性分析2010年6月至2017年9月来自三家医院的1393例经病理证实的NPC患者的临床资料及MRI图像(训练队列1049例、外部验证队列344例).用ITK-SNAP勾画感兴趣区并用Pyradiomics包逐层提取特征.使用相关性分析、单因素分析和递归特征消除法筛选特征,最后通过梯度提升机(Gradient Boosting Machine,GBM)算法构建模型.通过受试者工作特征(receiver operating characteristic,ROC)曲线比较模型的预测效能,以及决策曲线分析评估临床实用性.利用SHAP(SHapley Additive exPlanation)算法赋予最佳预测模型可解释性.结果经筛选后最终保留10个影像组学特征.基于影像组学特征、临床特征、影像组学+临床特征三种特征组合构建了GBM_R、GBM_C和GBM_RC模型.三者在训练集上的ROC曲线下面积(area under the curve,AUC)值分别为0.938、0.724和0.938;GBM_RC(命名为NPC-Wise)在外部验证集中取得了最高的AUC值,为0.775.N分期是NPC-Wise预测过程中最重要的特征.SHAP模型预测力图能直观可视化特征影响NPC-Wise预测远处转移风险的过程.结论基于多参数MRI影像组学及临床特征的可解释性机器学习预测模型NPC-Wise在预测NPC远处转移风险方面具有较高效能,SHAP算法为其提供了个体水平的可解释性,能为个性化治疗提供有价值的决策依据.

To reveal a radiogenomic correlation between the presence of the T2-fluid-attenuated inversion recovery resection (T2-FLAIR) mismatch sign on MR images and isocitrate dehydrogenase (IDH) mutation status in adult patients with lower-grade gliomas (LGGs). A web-based systemic search for eligible literature up to April 13, 2021, was conducted on PubMed, Embase, and the Cochrane Library databases by two independent reviewers. This study was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines. We included studies evaluating the accuracy of the T2-FLAIR mismatch sign in diagnosing the IDH mutation in adult patients with LGGs. The T2-FLAIR mismatch sign was defined as a T2-hyperintense lesion that is hypointense on FLAIR except for a hyperintense rim. Fourteen studies (n = 1986) were finally identified. The mean age of patients in the included studies ranged from 38.5 to 56 years. The pooled area under the curve (AUC), sensitivity, and specificity were obtained for each molecular profile: IDHmut-Codel: 0.46 (95% confidence interval [CI]: 0.42–0.50), 1% (95%CI: 0–7%), and 69% (95%CI: 62–75%), respectively; IDHmut-Noncodel: 0.75 (95%CI: 0.71–0.79), 42% (95%CI: 34–50%), and 99% (95%CI: 96–100%), respectively; IDH-Mutation regardless of 1p/19q codeletion status: 0.77 (95%CI: 0.73–0.80), 29% (95%CI: 21–40%), and 99% (95%CI: 92–100%), respectively. The T2-FLAIR mismatch sign was an insensitive but highly specific marker for IDHmut-Noncodel and IDH-Mutation LGGs, whereas it was not a useful marker for IDHmut-Codel LGGs. The findings might identify the T2-FLAIR mismatch sign as a non-invasive imaging biomarker for the selection of patients with IDH-mutant LGGs. • The T2-FLAIR mismatch sign was not a sensitive sign for IDH mutation in LGGs. • The T2-FLAIR mismatch sign was related to IDHmut-Noncodel with a specificity of 99%. • The pooled specificity (69%) of the T2-FLAIR mismatch sign for IDHmut-Codel was low.

Current risk stratification systems for thyroid nodules suffer from low specificity and high biopsy rates. Recently, machine learning (ML) is introduced to assist thyroid nodule diagnosis but lacks interpretability. Here, we developed and validated ML models on 3965 thyroid nodules, as compared to the American College of Radiology Thyroid Imaging, Reporting and Data System (ACR TI‐RADS). Subsequently, a SHapley Additive exPlanation (SHAP) algorithm was leveraged to interpret the results of the best‐performing ML model. Clinical characteristics including thyroid‐function tests were collected from medical records. Five ACR TI‐RADS ultrasonography (US) categories plus nodule size were assessed by experienced radiologists. Random forest (RF), support vector machine (SVM) and extreme gradient boosting (XGBoost) were used to build US‐only and US‐clinical ML models. The ML models and ACR TI‐RADS were compared in terms of diagnostic performance and unnecessary biopsy rate. Among the ML models, the US‐only RF model (hereafter, Thy‐Wise) achieved the optimal performance. Compared to ACR TI‐RADS, Thy‐Wise showed higher accuracy (82.4% vs 74.8% for the internal validation; 82.1% vs 73.4% for external validation) and specificity (78.7% vs 68.3% for internal validation; 78.5% vs 66.9% for external validation) while maintaining sensitivity (91.7% vs 91.2% for internal validation; 91.9% vs 91.1% for external validation), as well as reduced unnecessary biopsies (15.3% vs 32.3% for internal validation; 15.7% vs 47.3% for external validation). The SHAP‐based interpretation of Thy‐Wise enables clinicians to better understand the reasoning behind the diagnosis, which may facilitate the clinical translation of this model.

OBJECTIVES:Radiomic features derived from routine medical images show great potential for personalized medicine in gastric cancer (GC). We aimed to evaluate the current status and quality of radiomic research as well as its potential for identifying biomarkers to predict therapy response and prognosis in patients with GC.METHODS:We performed a systematic search of the PubMed and Embase databases for articles published from inception through July 10, 2021. The phase classification criteria for image mining studies and the radiomics quality scoring (RQS) tool were applied to evaluate scientific and reporting quality.RESULTS:Twenty-five studies consisting of 10,432 patients were included. 96% of studies extracted radiomic features from CT images. Association between radiomic signature and therapy response was evaluated in seven (28%) studies; association with survival was evaluated in 17 (68%) studies; one (4%) study analyzed both. All results of the included studies showed significant associations. Based on the phase classification criteria for image mining studies, 18 (72%) studies were classified as phase II, with two, four, and one studies as discovery science, phase 0 and phase I, respectively. The median RQS score for the radiomic studies was 44.4% (range, 0 to 55.6%). There was extensive heterogeneity in the study population, tumor stage, treatment protocol, and radiomic workflow amongst the studies.CONCLUSIONS:Although radiomic research in GC is highly heterogeneous and of relatively low quality, it holds promise for predicting therapy response and prognosis. Efforts towards standardization and collaboration are needed to utilize radiomics for clinical application.KEY POINTS:• Radiomics application of gastric cancer is increasingly being reported, particularly in predicting therapy response and survival. • Although radiomics research in gastric cancer is highly heterogeneous and relatively low quality, it holds promise for predicting clinical outcomes. • Standardized imaging protocols and radiomic workflow are needed to facilitate radiomics into clinical use.

We discuss the pros and cons of radiomics as a novel non-invasive tool to acquire phenotypic, prognostic and molecular information derived from standard-of-care high-resolution computed tomography in patients with ILD.https://bit.ly/36vYyl2

Tumour recurrence and drug resistance in hepatocellular carcinoma remain challenging. Cancer stem cells (CSCs) are responsible for tumour initiation because of their stemness characteristics. CSCs accounting for drug resistance and tumour relapse are promising therapeutic targets. We report that Abelson interactor 2 (ABI2) is a novel therapeutic target of HCC CSCs. First, ABI2 was upregulated in HCC tissues compared with liver tissues and was associated with tumour size, pathological grade, liver cirrhosis, worse prognosis and a high recurrence rate. Functional studies illustrate that ABI2 knockdown suppresses cell growth, migration, invasion and sorafenib resistance in vitro. Furthermore, ABI2 knockdown inhibited HCC sphere formation and decreased the CD24(+), CD133(+) and CD326(+) CSCs populations, suggesting the suppression of HCC stemness characteristics. A tumour xenograft model and limiting dilution assay demonstrated the inhibition of tumorigenicity and tumour initiation. Moreover, molecular mechanism studies showed that ABI2 recruits and directly interacts with the transcription factor MEOX2, which binds to the KLF4 and NANOG promoter regions to activate their transcription. Furthermore, overexpression of MEOX2 restored HCC malignant behaviour and the CSC population. The ABI2-mediated transcriptional axis MEOX2/KLF4-NANOG promotes HCC growth, metastasis and sorafenib resistance by maintaining the CSC population, suggesting that ABI2 is a promising CSC target in HCC treatment.

鼻咽解剖位置深在，病变种类繁多，CT及MRI是鼻咽病变治疗前主要评估手段。不同鼻咽病变影像表现不同，但也存在较多重叠，临床工作中需从病变位置、形态、边界、信号及密度特征、强化方式、与邻近组织结构关系、有无骨质破坏或淋巴结转移等方面进行综合分析，并结合临床鼻咽镜及实验室检查进一步明确诊断，从而提升诊断准确性。笔者系统总结鼻咽病变影像诊断思路，并归纳分析鼻咽病变的影像诊断要点，在治疗前明确诊断及病变范围，从而为临床治疗决策提供重要支持证据。

Background: Hyperprogressive disease (HPD) is a new progressive pattern in patients with advanced hepatocellular carcinoma (HCC) treated with programmed cell death 1 (PD-1) inhibitors. We aimed to investigate risk factors associated with HPD in advanced HCC patients undergoing anti-PD-1 therapy. Methods: A total of 69 patients treated with anti-PD-1 therapy between March 2017 and January 2020 were included. HPD was determined according to the time to treatment failure, tumour growth rate, and tumour growth rate ratio. Univariate and multivariate analyses were performed to identify clinical variables significantly associated with HPD. A risk model was constructed based on clinical variables with prognostic significance for HPD. Findings: Overall, 10 (14·49%) had HPD. Haemoglobin level, portal vein tumour thrombus, and Child-Pugh score were significantly associated with HPD. The risk model had an area under the curve of 0·931 (95% confidence interval, 0·844–1·000). Patients with HPD had a significantly shorter overall survival (OS) than that of the patients with non-HPD (p < 0·001). However, there was no significant difference in OS between PD (progressive disease) patients with and without HPD (p = 0·05). Interpretation: We identified three clinical variables as risk factors for HPD, providing an opportunity to aid the pre-treatment evaluation of the risk of HPD in patients treated with immunotherapy. Funding: This study was funded by the National Natural Science Foundation of China (81571664, 81871323, and 81801665); National Natural Science Foundation of Guangdong Province (2018B030311024); Scientific Research General Project of Guangzhou Science Technology and Innovation Commission (201707010,328); and China Postdoctoral Science Foundation (2016M600145).

OBJECTIVE:To determine the benefits of adding induction chemotherapy (IC) and adjuvant chemotherapy (AC) to concurrent chemoradiotherapy (CCRT) for nasopharyngeal carcinoma (NPC) based on propensity score-matching (PSM) studies. METHODS:Eligible PSM studies were searched in the PubMed, Web of Science, and Embase databases from inception to September 1, 2020. The primary endpoints included overall survival (OS), distant metastasis-free survival (DMFS), and locoregional recurrence-free survival (LRFS). RESULTS:A total of 14 trials consisting of 4086 participants were included. Significant benefits were observed between IC + CCRT and CCRT for OS (hazard ratio [HR], 0.76; 95% confidence interval [CI]: 0.64-0.91) and DMFS (HR, 0.77; 95% CI: 0.64-0.94) with the exception of LRFS (HR, 1.14; 95% CI: 0.90-1.43). However, CCRT + AC did not achieve significant improvements. CONCLUSIONS:IC with CCRT yields significant survival benefits in terms of OS and DMFS, whereas CCRT with AC fails to achieve any additional benefit in all endpoints.