Introduction: Nasopharyngeal carcinoma (NPC) is one of the common head and neck malignancies found in Indonesia. Diagnosis was based on histopathological result of tumor tissues taken from the nasopharynx. Histopathological from nasopharyngeal blind biopsy remains a golden standard of NPC diagnosis. Blind biopsy causes many problems for patients including pain sensation, risk of bleeding and traumatic effect. Another technique for obtaining specimen for evaluation malignant cells of nasopharynx was by cytological investigation. Cytology evaluation could detect malignant cells from smear specimen. Cytological specimen could be obtained from nasopharynx by brushing method. Cells are collected by brushing then processed with liquid-based technique; specimens were smeared and stained for cytology evaluation. Since nasopharynx location is invisible, then nasopharyngoscopy guidance was used to obtain a brushing specimen. Objective: To analyze an agreement between cytological results from nasopharyngeal brushing under nasopharyngoscopy and histopathological results from blind biopsy as a golden standard for the detection of malignant cells in NPC patients. Method: Diagnostic test with the cross-sectional comparative study. Population of this study was NPC patients who came to outpatient Department of Otorhinolaryngology Head and Neck Surgery Dr. Soetomo Hospital Surabaya. Samples were collected with consecutive sampling method during February until May 2011. During that period we obtained 36 patients as study samples. First we performed brushing method under nasopharyngoscopy guidance and then blind biopsy of nasopharynx on each patient. Cytology preparation processed with liquid-based technique liquiPREP and Papanicolaou stain. Histopathologic preparation used hematoxylin and eosin (H&E) stain. Sample data was tested by McNemar test and Kappa association test. Result: 36 samples were obtained consisted of 27 male (75%) and 9 female (25%). The result for cytological evaluation in detecting malignant cells of NPC patients was sensitivity 87.10%, specificity 80%, positive predictive value 96.43%, negative predictive value 50% and accuracy 86.11%. Statistically, analysis result with McNemar test was no significant difference and from Kappa association test there was a significant association between brushing method under nasopharyngoscopy guidance and blind biopsy. Conclusion: There was an agreement between brushing method under nasopharyngoscopy guidance and blind biopsy in detecting malignant cells of NPC.