METHODS: We did a genome-wide association study of 189 patients with extranodal NKTCL, nasal type (WHO classification criteria; cases) and 957 controls from Guangdong province, southern China. We validated our findings in four independent case-control series, including 75 cases from Guangdong province and 296 controls from Hong Kong, 65 cases and 983 controls from Guangdong province, 125 cases and 1110 controls from Beijing (northern China), and 60 cases and 2476 controls from Singapore. We used imputation and conditional logistic regression analyses to fine-map the associations. We also did a meta-analysis of the replication series and of the entire dataset.
FINDINGS: Associations exceeding the genome-wide significance threshold (p<5 × 10(-8)) were seen at 51 single-nucleotide polymorphisms (SNPs) mapping to the class II MHC region on chromosome 6, with rs9277378 (located in HLA-DPB1) having the strongest association with NKTCL susceptibility (p=4·21 × 10(-19), odds ratio [OR] 1·84 [95% CI 1·61-2·11] in meta-analysis of entire dataset). Imputation-based fine-mapping across the class II MHC region suggests that four aminoacid residues (Gly84-Gly85-Pro86-Met87) in near-complete linkage disequilibrium at the edge of the peptide-binding groove of HLA-DPB1 could account for most of the association between the rs9277378*A risk allele and NKTCL susceptibility (OR 2·38, p value for haplotype 2·32 × 10(-14)). This association is distinct from MHC associations with Epstein-Barr virus infection.
INTERPRETATION: To our knowledge, this is the first time that a genetic variant conferring an NKTCL risk is noted at genome-wide significance. This finding underlines the importance of HLA-DP antigen presentation in the pathogenesis of NKTCL.
FUNDING: Top-Notch Young Talents Program of China, Special Support Program of Guangdong, Specialized Research Fund for the Doctoral Program of Higher Education (20110171120099), Program for New Century Excellent Talents in University (NCET-11-0529), National Medical Research Council of Singapore (TCR12DEC005), Tanoto Foundation Professorship in Medical Oncology, New Century Foundation Limited, Ling Foundation, Singapore National Cancer Centre Research Fund, and the US National Institutes of Health (1R01AR062886, 5U01GM092691-04, and 1R01AR063759-01A1).
OBJECTIVE: Our study aimed to evaluate the discriminative and predictive ability of unimodal, bimodal, and multimodal approaches in a total of seven machine learning (ML) models-clinical, demographic, functional near-infrared spectroscopy (fNIRS), combinations of two unimodal models, as well as a combination of all three-for MDD.
METHODS: We recruited 65 adults with MDD and 68 matched healthy controls, who provided both sociodemographic and clinical information, and completed the HAM-D questionnaire. They were also subject to fNIRS measurement when participating in the verbal fluency task. Using the nested cross validation procedure, the classification performance of each ML model was evaluated based on the area under the receiver operating characteristic curve (ROC), balanced accuracy, sensitivity, and specificity.
RESULTS: The multimodal ML model was able to distinguish between depressed patients and healthy controls with the highest balanced accuracy of 87.98 ± 8.84% (AUC = 0.92; 95% CI (0.84-0.99) when compared with the uni- and bi-modal models.
CONCLUSIONS: Our multimodal ML model demonstrated the highest diagnostic accuracy for MDD. This reinforces the biological and clinical heterogeneity of MDD and highlights the potential of this model to improve MDD diagnosis rates. Furthermore, this model is cost-effective and clinically applicable enough to be established as a robust diagnostic system for MDD based on patients' biosignatures.