• d-dimer;
  • esophageal cancer;
  • neoadjuvant chemotherapy

SUMMARY.  Neoadjuvant chemotherapy may improve survival of responders in esophageal cancer patients but is useless and harmful in non-responders. Thus, it is important to predict the effect of the chemotherapy, and that any predictor must be applicable clinically. The aim of this study is to examine the correlation between pretherapeutic hypercoagulopathy as determined by plasma d-dimer levels and response to chemotherapy. In 71 patients with esophageal cancer who underwent neoadjuvant chemotherapy (cisplatin, adriamycin and 5-fluorouracil) followed by surgery, plasma d-dimer levels were measured before chemotherapy and the clinical and pathological responses to chemotherapy were assessed at 4 weeks after therapy (after surgery). Pretherapeutic plasma d-dimer level was significantly lower in clinical responders (complete response/partial response [CR/PR]; 0.62 ± 1.10 µg/mL, mean ± SD) than in non-responders (no change/progressive disease [NC/PD]; 1.15 ± 1.08 µg/mL, P = 0.0491), and in pathological responders (Grade 1b-3; 0.62 ± 1.11 µg/mL) and non-responders (Grade 0–1a; 1.15 ± 1.05 µg/mL, P = 0.0107). The optimal cut-off level of the plasma d-dimer levels for predicting clinical and pathological responses was 0.6 µg/mL. Then, sensitivity and specificity for the prediction of CR/PR were 68% and 73%, and those for Grade 1b–3 were 91% and 69%, respectively. Our results suggested that pretherapeutic plasma d-dimer level correlated significantly with clinical and pathological responses to chemotherapy. Pretherapeutic plasma d-dimer level can be used as a predictor for chemosensitivity.