OBJECTIVE: To examine the use of squamous cell carcinoma antigen (SCCA) as a biomarker of chemotherapy response in patients who underwent chemotherapy for metastatic cervical carcinoma. STUDY DESIGN: The study population consisted of patients who underwent first-line chemotherapy for metastatic cervical carcinoma between 1999 and 2009. SCCA levels were serially measured before, during and after chemotherapy. Radiographic responses were evaluated according to the criteria of the World Health Organization. A logistic model was used to determine the best prediction model, and internal and external validation of the prediction model were performed to compare the areas under the receiver operating characteristic curves (AUCs). RESULTS: In total, 55 patients were included in the analysis. Data for 32 patients enrolled in various clinical trials were used to develop the prediction model. Patients who achieved a radiographic response showed a significant decline in SCCA levels between the second and third cycles of chemotherapy, whereas patients who did not achieve a radiographic response showed constant SCCA levels over the same period. The prediction model was developed on the basis of changes in the SCCA level between the second and third cycles of chemotherapy (AUC=0.832) and the baseline SCCA level. The AUC after external validation, calculated using the data of the clinical practice population (n=22), was 0.871. CONCLUSIONS: A response to chemotherapy was possible for patients in whom SCCA levels declined between the second and third cycles of chemotherapy.
OBJECTIVE: To examine the use of squamous cell carcinoma antigen (SCCA) as a biomarker of chemotherapy response in patients who underwent chemotherapy for metastatic cervical carcinoma. STUDY DESIGN: The study population consisted of patients who underwent first-line chemotherapy for metastatic cervical carcinoma between 1999 and 2009. SCCA levels were serially measured before, during and after chemotherapy. Radiographic responses were evaluated according to the criteria of the World Health Organization. A logistic model was used to determine the best prediction model, and internal and external validation of the prediction model were performed to compare the areas under the receiver operating characteristic curves (AUCs). RESULTS: In total, 55 patients were included in the analysis. Data for 32 patients enrolled in various clinical trials were used to develop the prediction model. Patients who achieved a radiographic response showed a significant decline in SCCA levels between the second and third cycles of chemotherapy, whereas patients who did not achieve a radiographic response showed constant SCCA levels over the same period. The prediction model was developed on the basis of changes in the SCCA level between the second and third cycles of chemotherapy (AUC=0.832) and the baseline SCCA level. The AUC after external validation, calculated using the data of the clinical practice population (n=22), was 0.871. CONCLUSIONS: A response to chemotherapy was possible for patients in whom SCCA levels declined between the second and third cycles of chemotherapy.