High Fluorescent Cells on Automated Body Fluid Analysis as Discriminator for Malignant Cell Detection

The automated examination of body fluids (BF) serves as a valuable screening tool for the presence of malignant cells in such samples. Malignant cells are identified as high fluorescence cells (HFC) when analyzed using the Sysmex XN-1000 automated analyzer. This study aimed to assess the correlation between HFC cell counts generated by the automated analyzer and manual cytological examination for detecting malignant cells. Additionally, it sought to establish reliable cutoff values for malignant cells since there is a lack of literature on this subject. Conducted at the department of pathology hematology and cytology laboratory in a tertiary care hospital in India from January 2019 to May 2020, this hospital-based comparative study analyzed 120 BF samples, each subjected to cytological evaluation. The mean age of the study population was 52 years, with 70 male and 50 female patients (male-to-female ratio of 1.4:1). The samples consisted of 53 ascitic fluids (44.17%), 46 pleural fluids (38.33%), and 21 cerebrospinal fluids (CSF; 17.50%). Cytopathological examination revealed malignant cells in 50 (41.67%) of the BF samples, with 70 (58.33%) samples classified as nonmalignant. Specifically, among the ascitic fluids, 24 (48%) were malignant, while 29 (41.43%) were nonmalignant. For pleural fluids, 24 (48%) were malignant, and 22 (31.43%) were nonmalignant. In CSF, 2 (4%) samples were malignant, and 19 (27.14%) were nonmalignant. The total white blood cell counts provided by automated hematology analyzers were significantly higher in malignant samples, ranging from a minimum of 100 cells to a maximum of 60,000, with a median count of 800. Nonmalignant samples had white blood cell counts ranging from 2 to 12,000, with a median count of 100. Subgroup analysis for ascitic, pleural, and CSF samples revealed significantly higher median HFC counts in malignant samples. Receiver operating characteristic curve analysis indicated that the HF-BF parameter could effectively distinguish between benign and malignant fluids. For HF#, the area under the curve (AUC) was 0.844, with a sensitivity of 82% and specificity of 81%, while HF% had an AUC of 0.706, with sensitivity and specificity values of 72% and 72.9%, respectively. This study highlights that the HFC count in the BF mode of Sysmex XN-1000 can be a valuable tool for predicting the presence of malignant cells in serous fluids and for selecting samples for further microscopic examination. Based on this study, cutoff values of 15.70/µL for absolute HFC count and 5.05% for relative HFC count can be applied to screen BF samples for malignancy, offering good sensitivity and specificity.