Objectives: Occupational noise-induced hearing loss (ONIHL) represents a prevalent occupational health condition, traditionally necessitating multiple pure-tone audiometry assessments. We have developed and validated a machine learning model leveraging routine haematological and biochemical parameters, thereby offering novel insights into the risk prediction of ONIHL. Design, setting and participants: This study analysed data from 3297 noise-exposed workers in Shenzhen, including 160 ONIHL cases, with the data set divided into D1 (2868 samples, 107 ONIHL cases) and D2 (429 samples, 53 ONIHL cases). The inclusion criteria were formulated based on the GBZ49-2014 Diagnosis of Occupational Noise-Induced Hearing Loss . Model training was performed using D1, and model validation was conducted using D2. Routine blood and biochemical indicators were extracted from the case data, and a range of machine learning algorithms including extreme gradient boosting (XGBoost) were employed to construct predictive models. The model underwent refinement to identify the most representative variables, and decision curve analysis was conducted to evaluate the net benefit of the model across various threshold levels. Primary outcome measures: Model creation data set and validation data sets: ONIHL. Results: The prediction model, developed using XGBoost, demonstrated exceptional performance, achieving an area under the receiver operating characteristic curve (AUC) of 0.942, a sensitivity of 0.875 and a specificity of 0.936 on the validation data set. On the test data set, the model achieved an AUC of 0.990. After implementing feature selection, the model was refined to include only 16 features, while maintaining strong performance on a newly acquired independent data set, with an AUC of 0.872, a balanced accuracy of 0.798, a sensitivity of 0.755 and a specificity of 0.840. The analysis of feature importance revealed that serum albumin (ALB), platelet distribution width (PDW), coefficient of variation in red cell distribution width (RDW-CV), serum creatinine (Scr) and lymphocyte percentage (LYMPHP) are critical factors for risk stratification in patients with ONIHL. Conclusion: The analysis of feature importance identified ALB, PDW, RDW-CV, Scr and LYMPHP as pivotal factors for risk stratification in patients with ONIHL. The machine learning model, using XGBoost, effectively distinguishes patients with ONIHLamong individuals exposed to noise, thereby facilitating early diagnosis and intervention.

Background: Chronic benzene poisoning (CBP) is one of the most common chronic occupational poisoning which is associated with mitochondrial oxidative damage, and lead to increasing risk of respiratory diseases such as lung cancer. Cytochrome c oxidase subunit I (COI) is one of the key enzymes that plays an important role in oxidative damage regulation by eliminating reactive oxygen species (ROS). This study investigated the relationship between COI gene variants and the risk of CBP. Methods: We investigated 44 non-smoking patients who were diagnosed with CBP and 57 unexposed non-smoking controls between the ages of 23 and 60 with their background including work experience, lifestyle and medical records. Peripheral blood (2 mL) was collected in EDTA tube and the platelet was purified from the collected blood. Variants of COI were analyzed by PCR and sequencing. Multivariable linear regression analysis was used to assess the association between CBP exposure and variants. Results: The frequency of the mitochondrial DNA (mtDNA) T6392C, G6962 variants were 10, 7 out of 44 CBP group patients, which was higher when compared to that of 4, 2 out of 57 in the control group, suggesting these variants could be the risk factor for CBP [odds ratio (OR) 3.897, 95% CI: 1.131–13.425, P=0.023; OR 5.203, 95% CI: 1.024–26.442, P=0.034]. There was a significant difference (P<0.05) of COI variants, including T6392C and G6962A, in platelet mtDNA between patients and control samples. Meanwhile, the frequency of the mtDNA C7196A variant were 13 out of 44 control group, which was higher when compared to that of 2 of 57 in the CBP group patients, suggesting this variant could be the protective factor for CBP (OR 6.205, 95% CI: 1.320–29.162, P=0.010). Conclusions: Our study suggests that T6392C, G6962A and C7196A from platelet mtDNA variants play a significant role in the etiology of CBP and facilitate the development of molecular biomarker on CBP diagnosis.

Previously, we had cross-sectionally explored the characteristics of T cell receptor (TCR) repertoires from occupational medicamentosa-like dermatitis due to trichloroethylene (OMDT) patients, now we further analyzed the dynamic features of OMDT TCR repertoires. Peripheral blood TCR β-chain complementarity-determining region 3 (CDR3) genes were detected with the high throughput sequencing in 24 OMDT cases in their acute, chronic and recovery stages, respectively, and in 24 trichloroethylene-exposed healthy controls. The TCR repertoire diversity, TRBV/TRBD/TRBJ gene usage and combination, frequencies of CDR3 nucleotide (nt) and amino acid (aa) sequences in the cases in different stages and in the controls were analyzed. TRBV6-4 and TRBV7-9 frequencies significantly differed between the cases and controls (both P < 6.1 × 10^–4). TRBV6-4 combination with TRBJ2-1, TRBJ2-2, TRBJ2-3, and TRBJ2-6, and TRBV7-9 combination with TRBJ2-1 were associated with the stage by OMDT severity (all P < 0.001). Ten CDR3-nt and 7 CDR3-aa sequences in TRBV7-9-TRBJ2-1 combination and 1 CDR3-nt and 1 CDR3-aa sequences in TRBV6-4-TRBJ2-1 combination were identified as associated with the severity of OMDT (all P < 0.001). We revealed further how TCR repertoires vary with the severity in the development of OMDT, and severity-related TCRs may provide important therapeutic targets for OMDT in clinical practice.

Introduction: As a hematopoietic carcinogen, benzene induces human leukemia through its active metabolites such as benzoquinone, which may cause oxidative damage to cancer-related nuclear genes by increasing reactive oxygen species (ROS). Mitochondrion is the main regulatory organelle of ROS, genetic abnormality of mitochondrion can impede its regulation of ROS, leading to more severe oxidative damage. Mutations have been related to certain types of cancer in several mitochondrial genes, but they have never been completely analyzed genome-wide in leukemia. Patient concerns: The patient was a 52-year-old female who had chronic exposure to benzene for several years. Her symptoms mainly included recurrent dizziness, fatigue, and they had lasted for nearly 8 years and exacerbated in recent weeks before diagnosis. Diagnosis: Samples of peripheral blood were taken from the patient using evacuated tubes with EDTA anticoagulant on the second day of her hospitalization. At the same time blood routine and BCR/ABL genes of leukemic phenotype were tested. Platelets were isolated for mitochondrial DNA (mtDNA) extraction. The genetic analysis of ATP synthase Fo subunit 8 (complex V), ATP synthase Fo subunit 6 (complex V), cytochrome c oxidase subunit 1 (complex IV), cytochrome c oxidase subunit 2 (complex IV), cytochrome c oxidase subunit 3, Cytb, NADH dehydrogenase subunit 1 (complex I) (ND) 1, ND2, ND3, ND4, ND5, ND6, 12S-RNA, 16S-RNA, tRNA-Cysteine, A, N, tRNA-Leucine, E, displacement loop in platelet mtDNA were performed. All the detected gene mutations were validated using the conventional Sanger sequencing method. Interventions: The patient received imatinib, a small molecule kinase inhibitor, and symptomatic treatments. Outcomes: After 3 months treatment her blood routine test indicators were restored to normal. Conclusion: A total of 98 mutations were found, and 25 mutations were frame shift. The ND6 gene mutation rate was the highest among all mutation points. Frame shifts were identified in benzene-induced leukemia for the first time. Many mutations in the platelet mitochondrial genome were identified and considered to be potentially pathogenic in the female patient with benzene-induced leukemia. The mutation rate of platelet mitochondrial genome in the benzene-induced leukemia patient is relatively high, and the complete genome analysis is helpful to fully comprehend the disease characteristics.