We have designed a drug pathway identification system, which we called Drug Pathway Decipherer, to generate hypotheses on drug treatment responsive pathway. Decipherer takes in both pre- and post- treatment gene expression data, and evaluates known biological pathways against the data. We applied Decipherer to two gene expression datasets of human nasopharyngeal carcinoma treated with CYC202. Results show that the identified RAS-ERK pathway and PI3K-NFκB-IAP pathway are closely associated with treatment outcome. Decipherer is implemented in Java, and it is available together with supplementary material at http://www.comp.nus.edu.sg/~wongls/projects/drug-pathway