RESUMEN
CONTEXT: Oral cancers represent a disparate group of tumors with diverse clinical behavior and chemosensitivity profile. Currently, it is difficult to predict whether a tumor will respond to chemotherapy and which drug(s) will achieve the maximum clinical response. AIMS: To study in vitro chemosensitivity profile of oral cancers and to correlate the in vitro chemosensitivity of oral cancer to clinical response to chemotherapy. SETTINGS AND DESIGN: Prospective study in a tertiary cancer care center. METHODS AND MATERIAL: We prospectively studied the chemosensitivity profile of 57 untreated, advanced, unresectable oral cancers to cisplatin, methotrexate, 5-fluorouracil and their combinations by using histoculture drug response assay (HDRA) and correlated them to the clinical response to chemotherapy. STATISTICAL ANALYSIS USED: Chi Square test. RESULTS: Biopsy samples were successfully histocultured in 52/57 (91%) cases. Of these 52 evaluable patients, 47 had primary gingivo-buccal cancers and five had tongue / floor of mouth cancers. Based on the assay, 27 (52%) tumors were sensitive to cisplatin, 27 (52%) to methotrexate, 24 (46%) to 5-fluorouracil, 38 (73%) to combination of cisplatin and methotrexate and 36 (69%) to combination of cisplatin and 5-fluorouracil. Of these, 31 patients with good performance status received two cycles of chemotherapy using one or more of these test drugs. There was a significant correlation (p=0.03) between the in vitro chemosensitivity and the clinical response. Negative predictive value of the test was 80%, positive predictive value-69%, sensitivity-79% and specificity -71%. The overall accuracy of the assay was 74%. CONCLUSIONS: We found HDRA to be a fairly good predictor of chemo-response of oral cancer.