OBJECTIVE

To identify serum-based biomarkers predicting response to neoadjuvant chemoradiotherapy (neo-CRT) in esophageal cancer.

PURPOSE

Increasingly, the standard of care for esophageal cancer involves neo-CRT followed by surgery. The identification of biomarkers predicting response to therapy may represent a major advance, enabling clinical trials and improved outcomes.

BACKGROUND DATA

Patients with esophageal cancer (n = 31) received a standard neo-CRT regimen. Histopathologic response to therapy was assessed by using the Mandard tumor regression grade (TRG) classification. Serum was collected pretreatment and at 24-hour and 48-hour time points into treatment. Serum samples were analyzed by using Surface-enhanced laser desorption/ionization time-of-flight mass spectrometry and enzyme-linked immunosorbent assay. A leave-one-out cross-validation predictive algorithm assessed the ability of validated biomarkers to correctly predict therapeutic outcome.

RESULTS

Fifty-one percent (16) of patients were poor responders (TRG 3-5), whereas 49% (15) responded well (TRG 1-2). On CM10 biochips, serum expression of 9 protein peaks was significantly different between the response groups. Two differential spectrum peaks were identified as complement C4a and C3a and were subsequently analyzed by enzyme-linked immunosorbent assay. Pretreatment serum C4a and C3a levels were significantly higher in poor responders versus good responders. Subdivision of the response groups by TRG indicated an inverse correlation between levels of C4a and C3a and pathological response to neo-CRT. The leave-one-out cross-validation analysis revealed that these serum proteins could predict response to neo-CRT with a sensitivity and specificity of 78.6% and 83.3%, respectively.

CONCLUSIONS

This translational application of proteomics technology identifies pretreatment serum levels of C4a and C3a as predictive biomarkers of response. Large validation studies in an independent cohort are merited.