In Response to “Impact of Neoadjuvant Chemoradiation on Lymph Node Status in Esophageal Cancer”
We read with great interest the recently published post hoc analysis of the FFCD 9901 randomized trial by Robb et al.1 Because the data were collected as part of a multicenter randomized control trial of high-quality surgical resection and pathologic evaluation, it provides important observations on the relationship among neoadjuvant therapy, lymph node count in the resected specimen (lymph node yield), and overall survival in esophageal cancer.
In the study, 98 patients were assigned to neoadjuvant chemoradiation (nCRT) followed by surgical resection, and nCRT was found to be an independent predictor of reduced lymph node yield. The influence of neoadjuvant therapy on lymph node yield has been well studied in rectal cancer,2,3 where there is evidence that lymph node yield is selectively reduced in patients with favorable tumor regression grade (TRG), a pathologic measurement of response to nCRT.4 Although there is accumulating evidence that parallels may exist in esophageal cancer,5–7 the relationship between treatment-related tumor regression and lymph node yield has not been widely studied. This is important because the quality of surgical treatment of esophageal cancer is judged in part by lymph node yield. In the study reported by Robb et al, TRG, dichotomized as either favorable (TRG 1/2) or neutral/unfavorable (TRG3–5), was not associated with lymph node yield, although there was a nonsignificant trend toward lower lymph node yield in patients with more substantial tumor regression [TRG 1/2: median = 16.5 (range, 4–39) vs TRG3–5: median = 20.5 (range, 8–33), P = 0.23]. Favorable TRG was associated with a significant reduction in disease-positive lymph nodes.
At our institution, we recently retrospectively reviewed 50 consecutive cases of locally advanced (EUS T1N1 or T2–3N0–1) esophageal carcinoma (46 adenocarcinoma; 4 squamous cell carcinoma) treated with neoadjuvant chemoradiation followed by transhiatal esophagectomy (manuscript in preparation). Neoadjuvant therapy was associated with a nonsignificant reduction in mean lymph node counts compared with an institution surgery-only cohort (14.3 ± 6.1 vs 15.8 ± 6.7, P = 0.38). Interestingly, the patients with complete pathologic response after nCRT (14%) had a significantly reduced median number of lymph nodes identified in the surgical specimen [5.0 (IQR = 4–14) vs 15.0 (IQR = 12–18), P = 0.02]. In contrast, there were no significant differences in median lymph node yield in patients with residual cancer post-nCRT who were nonetheless downstaged (ie, postoperative pathologic stage lower than preoperative clinical stage) by T category [14.0 (IQR = 8.75–18.0) vs 16.0 (IQR = 12.75–18.0), P = 0.37], N category [14.0 (IQR = 9.0–18.0) vs 15.0 (IQR = 12.5–18.0), P = 0.55], or AJCC stage [14.5 (IQR = 8.75–18.0) vs 15.0 (IQR = 12.0–18.0), P = 0.47]. These data suggest that nCRT in esophageal cancer decreases lymph node yield in patients with complete pathologic response but not in those patients with partial response to treatment. Low lymph node yield in esophagectomy for esophageal cancer cannot be attributed to nCRT except in patients with favorable pathologic response. Whether the latter group represents a unique subgroup of esophageal cancer patients identifiable by pretreatment tissue analysis (genetic, proteomic, immunologic, metabolic, etc) remains largely speculative.