Role of barium esophagography in patients with locally advanced esophageal cancer: evaluation of response to neoadjuvant chemoradiotherapy

Abstract

Purpose. This retrospective study examined the usefulness of barium esophagography, focusing on the luminal stenosis, in the response evaluation of neoadjuvant chemoradiotherapy (NACRT) in patients with esophageal cancer. Materials and Methods. Thirty-four patients with primary advanced esophageal cancer (≥T2) who were treated with NACRT before surgical resection were analyzed. All patients underwent barium esophagography before and after NACRT. The tumor length, volume, and percent esophageal stenosis (PES) before and after NACRT were measured. These values and their changes were compared between histopathologic responders (n = 22) and nonresponders (n = 12). Results. Posttreatment tumor length and PES in responders (4.5 cm ± 1.1 and 33.0% ± 18.5) were significantly smaller than those in nonresponders (5.8 cm ± 1.9 and 48.0% ± 12.9) (P = 0.018). Regarding posttherapeutic changes, the decrease in PES in responders (31.5% ± 13.9) was significantly greater than that in nonresponders (14.4% ± 10.7) (P < 0.001). The best decrease in PES cutoff with which to differentiate between responders and nonresponders was 18.8%, which yielded a sensitivity of 91% and a specificity of 75%. Conclusions. Decrease in PES is a good parameter to differentiate responders from nonresponders for NACRT. Barium esophagography is useful in response evaluation to NACRT in patients with locally advanced esophageal cancer.

Figure 1

Measuring methods of esophageal stenosis.

Figure 2

76-year-old male with esophageal cancer who obtained a grade 3 pathologic response. (a) Before NACRT, esophagography shows irregular wall stenosis in the middle esophagus with the PES of 55.6%. (b) After NACRT, esophageal wall stenosis has improved result in the PES of 33.8%, which indicates the decrease in PES of 18.8%. (c), (d) Pathological specimen of the resected esophagus shows no carcinoma cells (grade 3). Many degenerative cells with keratinization and diffuse fibrosis are seen in the submucosa and muscularis propria.

Figure 3

ROC analysis. The best decrease in PES cutoff with which to differentiate between responders and nonresponders is 18.8%, which yields a sensitivity of 91% and a specificity of 75%.