Temporal validation of metabolic nodal response of esophageal cancer to neoadjuvant chemotherapy as an independent predictor of unresectable disease, survival, and recurrence

OBJECTIVES: We recently described metabolic nodal stage (mN) and response (mNR) of cancer of the esophagus and gastro-esophageal junction (GEJ) to neoadjuvant chemotherapy (NAC) using (18)F-FDG PET-CT as new markers of disease progression, recurrence, and death. We aimed to validate our findings. METHODS: Our validation cohort comprised all patients consecutive to our discovery cohort, staged before and after NAC using PET-CT from 2014 to 2017. Multivariate binary logistic and Cox regression were performed. RESULTS: Fifty-one of the 200 patients had FDG-avid nodes after NAC (25.5%; i.e., lack of complete mNR), and were more likely to progress during NAC to incurable disease on PET-CT or at surgery: odds ratio 3.84 (1.46–10.1; p = 0.006). In 176 patients undergoing successful resection, patients without complete mNR had a worse prognosis: disease-free survival hazard ratio 2.46 (1.34–4.50); p = 0.004. These associations were independent of primary tumor metabolic, pathological response, and stage. In a hybrid pathological/metabolic nodal stage, avid nodal metastases conferred a worse prognosis than non-avid metastases. Lack of complete mNR predicted recurrence or death at 1 and 2 years: positive predictive values 44.4% (31.7–57.8) and 74.1% (56.6–86.3) respectively. CONCLUSIONS: This study provides temporal validation for mNR as a new and independent predictive and prognostic marker of esophageal and GEJ cancer treated with NAC and surgery, although external validation is required to assess generalizability. mNR may provide surrogate information regarding the phenotype of metastatic cancer clones beyond the mere presence of nodal metastases, and might be used to better inform patients, risk stratify, and personalize management, including adjuvant therapy. KEY POINTS: • We previously described metabolic nodal response (mNR) of esophageal cancer to neoadjuvant chemotherapy using(18)F-FDG PET-CT as a predictor of unresectable disease, early recurrence, and death. • We report the first validation of these findings. In an immediately consecutive cohort, we found consistent proportions of patients with and without mNR, and associations with abandoned resection, early recurrence, and death. • This supports mNR as a new and actionable biomarker in esophageal cancer. Although external validation is required, mNR may provide surrogate information about the chemosensitivity of metastatic subclones, and the means to predict treatment success, guide personalized therapy, and follow-up. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (10.1007/s00330-019-06310-9) contains supplementary material, which is available to authorized users.