This study aimed to develop and optimize an analytical method for profiling 21 amino acids in cerebrospinal fluid and plasma, addressing the need for improved diagnostic tools in leukemia research. Using high-performance liquid chromatography coupled with electrospray ionization mass spectrometry, the method achieved enhanced resolution, sensitivity, and specificity. Rigorous sample preparation, including liquid-liquid extraction, ensured high recovery rates, while validation confirmed the method's accuracy and reproducibility. Clinical application in pediatric leukemia patients revealed significant variations in amino acid concentrations across treatment stages, providing insights into disease progression and therapeutic response. Statistical analysis with IBM SPSS Statistics 25 compared amino acid levels in patients to healthy controls, identifying distinct patterns on day 1, day 15, and day 33 of treatment. Correlation analysis highlighted relationships between amino acid levels and factors such as treatment duration, sex, age, and blood test results. Key amino acids, including proline, leucine, and hydroxyproline, emerged as significant predictors of white blood cell count, effectively distinguishing between patient and control groups. This method demonstrates robust potential for broader leukemia research applications, pending further validation on larger cohorts.