RESUMO
BACKGROUND: The prognostic value of traditional clinical indicators for locally recurrent nasopharyngeal carcinoma is limited because of their inability to reflect intratumor heterogeneity. We aimed to develop a radiomic signature to reveal tumor immune heterogeneity and predict survival in locally recurrent nasopharyngeal carcinoma. METHODS: This multicenter, retrospective study included 921 patients with locally recurrent nasopharyngeal carcinoma. A machine learning signature and nomogram based on pretreatment magnetic resonance imaging features were developed for predicting overall survival in a training cohort and validated in 2 independent cohorts. A clinical nomogram and an integrated nomogram were constructed for comparison. Nomogram performance was evaluated by concordance index and receiver operating characteristic curve analysis. Accordingly, patients were classified into risk groups. The biological characteristics and immune infiltration of the signature were explored by RNA-sequencing analysis. RESULTS: The machine learning signature and nomogram demonstrated comparable prognostic ability to a clinical nomogram, achieving concordance indexes of 0.729, 0.718, and 0.731 in the training, internal, and external validation cohorts, respectively. Integration of the signature and clinical variables statistically improved the predictive performance. The proposed signature effectively distinguished patients between risk groups with statistically distinct overall survival rates. Subgroup analysis indicated the recommendation of local salvage treatments for low-risk patients. Exploratory RNA-sequencing analysis revealed differences in interferon response and lymphocyte infiltration between risk groups. CONCLUSIONS: A magnetic resonance imaging-based radiomic signature predicted overall survival more accurately. The proposed signature associated with tumor immune heterogeneity may serve as a valuable tool to facilitate prognostic stratification and guide individualized management for locally recurrent nasopharyngeal carcinoma patients.
Assuntos
Carcinoma Nasofaríngeo , Neoplasias Nasofaríngeas , Recidiva Local de Neoplasia , Nomogramas , Radiômica , Adulto , Idoso , Feminino , Humanos , Masculino , Pessoa de Meia-Idade , Aprendizado de Máquina , Imageamento por Ressonância Magnética , Carcinoma Nasofaríngeo/mortalidade , Carcinoma Nasofaríngeo/imunologia , Carcinoma Nasofaríngeo/diagnóstico por imagem , Carcinoma Nasofaríngeo/patologia , Neoplasias Nasofaríngeas/mortalidade , Neoplasias Nasofaríngeas/diagnóstico por imagem , Neoplasias Nasofaríngeas/imunologia , Neoplasias Nasofaríngeas/patologia , Estudos Retrospectivos , Taxa de SobrevidaRESUMO
OBJECTIVE: To establish a nasopharyngeal carcinoma-specific big data platform based on electronic health records (EHRs) to provide data support for real-world study of nasopharyngeal carcinoma. METHODS: A multidisciplinary expert team was established for this project. Based on industry standards and practical feasibility, the team designed the nasopharyngeal carcinoma data element standards including 14 modules and 640 fields. Data from patients diagnosed with nasopharyngeal carcinoma who visited Southern Hospital after 1999 were extracted from 15 EHRs systems and were cleaned, structured, and standardized using information technologies such as machine learning and natural language processing. In addition, a series of measures such as quality control and data encryption were taken to ensure data quality and patient privacy. At the platform application level, 10 functional modules were designed according to the needs of nasopharyngeal carcinoma research. RESULTS: As of 1 October 2022, the Big Data platform has included 11,617patients, of whom 8228 (70.83 %) were male and 3389 (29.17 %) were female, with a median age of 48 years (interquartile range, 40 years). The data in the platform were validated to have a high level of completeness and accuracy, especially for key variables such as social demographics, laboratory tests and vital signs. Currently, six projects involving risk factors, early diagnosis, treatment efficacy and prevention of treatment-related toxic reactions have been conducted on the platform. CONCLUSIONS: We have established a high-quality NPC-specific big data platform by integrating heterogeneous data from multiple sources in the EHR. The platform provides an effective tool and strong data support for real-world studies of nasopharyngeal carcinoma, which helps to improve research efficiency, reduce costs, and improve the quality of research results. We expect to promote multicenter nasopharyngeal carcinoma data sharing in the future to facilitate the generation of high-quality real-world evidence in nasopharyngeal carcinoma. This article may provide some reference value for other comprehensive hospitals to establish a big data platform for nasopharyngeal carcinoma.
Assuntos
Big Data , Registros Eletrônicos de Saúde , Carcinoma Nasofaríngeo , Neoplasias Nasofaríngeas , Humanos , Carcinoma Nasofaríngeo/terapia , Carcinoma Nasofaríngeo/diagnóstico , Masculino , Feminino , Pessoa de Meia-Idade , Adulto , Aprendizado de Máquina , Processamento de Linguagem NaturalRESUMO
BACKGROUND: Post-radiation nasopharyngeal necrosis (PRNN) is a severe adverse event following re-radiotherapy for patients with locally recurrent nasopharyngeal carcinoma (LRNPC) and associated with decreased survival. Biological heterogeneity in recurrent tumors contributes to the different risks of PRNN. Radiomics can be used to mine high-throughput non-invasive image features to predict clinical outcomes and capture underlying biological functions. We aimed to develop a radiogenomic signature for the pre-treatment prediction of PRNN to guide re-radiotherapy in patients with LRNPC. METHODS: This multicenter study included 761 re-irradiated patients with LRNPC at four centers in NPC endemic area and divided them into training, internal validation, and external validation cohorts. We built a machine learning (random forest) radiomic signature based on the pre-treatment multiparametric magnetic resonance images for predicting PRNN following re-radiotherapy. We comprehensively assessed the performance of the radiomic signature. Transcriptomic sequencing and gene set enrichment analyses were conducted to identify the associated biological processes. RESULTS: The radiomic signature showed discrimination of 1-year PRNN in the training, internal validation, and external validation cohorts (area under the curve (AUC) 0.713-0.756). Stratified by a cutoff score of 0.735, patients with high-risk signature had higher incidences of PRNN than patients with low-risk signature (1-year PRNN rates 42.2-62.5% vs. 16.3-18.8%, P < 0.001). The signature significantly outperformed the clinical model (P < 0.05) and was generalizable across different centers, imaging parameters, and patient subgroups. The radiomic signature had prognostic value concerning its correlation with PRNN-related deaths (hazard ratio (HR) 3.07-6.75, P < 0.001) and all causes of deaths (HR 1.53-2.30, P < 0.01). Radiogenomics analyses revealed associations between the radiomic signature and signaling pathways involved in tissue fibrosis and vascularity. CONCLUSIONS: We present a radiomic signature for the individualized risk assessment of PRNN following re-radiotherapy, which may serve as a noninvasive radio-biomarker of radiation injury-associated processes and a useful clinical tool to personalize treatment recommendations for patients with LANPC.