Serum squamous cell carcinoma antigen level and magnetic resonance imaging for the prognosis of locally advanced cervical cancer

OBJECTIVE: Squamous cell carcinoma antigen (SCC-ag) and magnetic resonance imaging (MRI) were explored to serve as biomarkers to predict the prognosis of cervical cancer (CC) patients treated with neoadjuvant chemotherapy (NACT) prior to radical surgery, with the aim of identifying the subgroup that least benefits from the combined therapy. PATIENTS AND METHODS: All patients were treated with NACT prior to radical surgery and received MRI and SCC-ag examinations before and after NACT. For these three cycles of NACT, patients were treated with intravenous paclitaxel at 150 mg/m2 over a period of 3 hours and carboplatin, with the area under the sera concentration-time curve of 5 over a period of 30 minutes on the first day of each cycle. Meanwhile, the blood pressure, ECG, and blood oxygen saturation of the patients were observed during the infusion. A discovery cohort and a validation cohort were applied to examine the prognostic performance of SCC-ag, MRI, and their combination. The endpoints of our study were overall survival (OS) and progression-free survival (PFS). RESULTS: A total of 384 patients diagnosed between August 2006 and December 2010 were enrolled in our research, with 206 patients in the discovery cohort and 178 patients in the validation cohort. The high -risk group identified by MRI had a worse OS [hazard ratio (HR), 3.567; 95% confidence interval (CI), 1.466-8.677; logrank p=0.0027) and PFS (HR, 4.062; 95% CI, 2.171-7.6; log -rank p<0.0001) than the low-risk group. Meanwhile, the SCC-RC could serve as a strong prognostic factor to predict OS (HR, 5.614; 95% CI, 2.473-12.744; log -rank p<0.0001) and PFS (HR, 7.481; 95% CI, 4.194-13.344; logrank p<0.0001) for CC. In addition, the combined MRI and SCC-ag had greater prognostic efficiency and were used to divide the whole patient population into three groups. Compared with patients in the low-risk group, patients in the high -risk group had a worse OS (HR, 8.216; 95% CI, 2.98-22.651; log -rank p<0.0001) and PFS (HR, 11.757; 95% CI, 5.735-24.104; log -rank p<0.0001). Multivariate analyses revealed that MRI, SCC-ag, and their combination were independent prognostic factors. CONCLUSIONS: SCC-ag and MRI, individually or in combination, were bound up with OS and PFS in CC. Additionally, the predictive efficiency improved when SCC-ag and MRI were combined in a risk model that predicted the OS and PFS of SCC compared with the predictive efficiency of either SCC-ag or MRI alone, revealing that the combination of these two biomarkers could help to ameliorate prognostic stratification and to guide personalized therapy for SCC patients.