AUTHOR=Yang Xue , Yang Jing , Li Rui , Dong Hui , Liu Yaling 
  
TITLE=Peripheral immune cells and glycation indices as potential diagnostic biomarkers in amyotrophic lateral sclerosis
  
JOURNAL=Experimental Biology and Medicine
  
VOLUME=Volume 251 - 2026
  
YEAR=2026
  
URL=https://www.ebm-journal.org/journals/experimental-biology-and-medicine/articles/10.3389/ebm.2026.10987
  
DOI=10.3389/ebm.2026.10987
  
ISSN=1535-3699
  
ABSTRACT=The diagnosis of amyotrophic lateral sclerosis (ALS) mainly relies on clinical symptoms and the exclusion of other diseases, with a lack of specific biomarkers, leading to delayed diagnosis and a high rate of misdiagnosis. This study aims to explore the utility of peripheral immune cells and glycosylation indices as potential diagnostic biomarkers for ALS to enhance the accuracy and efficiency of early ALS diagnosis. This retrospective study included 54 ALS patients diagnosed in our hospital from June 2023 to October 2024, along with 54 healthy controls. Blood samples and laboratory data, including levels of peripheral immune cells and glycosylation indices, were collected from both groups. Through logistic regression, random forest models, receiver operating characteristic (ROC) curve analysis, and SHAP interpretability analysis, the predictive abilities and clinical significance of each candidate indicator were screened and evaluated. Notable disparities were detected in age, leukocyte count, monocyte levels, glycated haemoglobin A1c (HbA1c), and haemoglobin glycation index (HGI) between the control and ALS groups (all P < 0.05). Logistic regression analysis revealed that age (OR = 1.114) and monocyte (OR = 3.174) were risk factors for ALS, while leukocyte (OR = 0.533) and HbA1c (OR = 0.069) were protective factors. The random forest algorithm, ranked by decreasing importance, showed that leukocyte, HGI, monocyte, and HbA1c level all influenced ALS. Using these indicators to predict ALS resulted in a false-positive rate of 18% and a false-negative rate of 6%. ROC curve analysis indicated that the combined use of leukocyte, monocyte, HbA1c level, and HGI provided the highest diagnostic value for ALS (AUC = 0.774), which was higher than that of any individual indicator (all P < 0.05). SHAP analysis visualization demonstrated that increased monocyte and decreased leukocyte, HGI, and HbA1c level were all associated with an increased risk of ALS onset, ranked in descending order of feature importance as monocyte, leukocyte, HGI, and HbA1c. Peripheral blood white blood cells, monocytes, HbA1c, and HGI can serve as potential diagnostic biomarkers for ALS. Combined detection can improve the diagnostic accuracy of ALS, facilitating early diagnosis and intervention, and ultimately improving patient prognosis. Further validation in cohorts including disease controls is required to confirm specificity.