Possibility of Conversion Surgery and Histological Evaluation in Definite cT4 Patients of Esophageal Cancer Treated with Preoperative Chemoradiation Followed by Surgery

Abstract

Patients of cT4NxM0 have been regarded as appropriate candidates for definitive chemoradiation (CRT) but as controversial for surgery following preoperative CRT. There is wide variation of mixed rate of cT3 in cT4 among institutions despite of best diagnostic technology. Objectives of this study are (1) to select definite cT4 (d-cT4) patients with consistent histological evidence and to compare d-cT4 with marginal cT3 from view of prognosis, (2) to discuss the possibility of conversion surgery in d-cT4. Marginally resectable 126 patients with cT3-4NxM0 were enrolled to preoperative CRT regimen. The regimen included cisplatin (70 mg/m2 on day 1 and 28), 5-FU (700 mg/m2 on days 1-5 and 28-32) and radiotherapy (RT, total 50.4 Gy/28F). Within 6 weeks after the accumulated RT dose reached 50.4 Gy, the feasibility of esophagectomy was evaluated in all patients. Histological evidence of disappeared tumors by CRT are coagulation necrosis with/without replacement by fibrosis, xantho-granulomatous region with/without foreign body giant cells, and deposition of keratinized tissue without viable tumor cells. Three stages of surgical curability were defined as Cur A; Stage 0∼lll, D>N and R0, as Cur B; neither Cur A nor C, and as Cur C; residual tumor and R2. Out of categorized 4 Grades (Gr 0 to 3) of histological effectiveness of CRT, no viable cancer cells were defined Gr 3. Conclusions were drawn from a retrospective analysis of phase II study for including marginal cT3 and cT4. The reasons which esophagectomy was not feasible were CR evaluation after CRT in 9 patients, PD after CRT in 6, and toxicities of CRT in 3. After all, 108 patients (average 64 years old, 43-80) were treated with CRT followed by surgery. By final examination of the differences between pretreatment T stages and pathological findings of surgical specimens, d-cT4 were 40 (37%) patients (between 11 of cT4a and 29 of cT4b, there was no statistical difference of survivals). As other 68 operated patients were not verified as d-cT4, they were defined as “cT3/<4” in this study. R0 surgeries were done in 93 (86%) patients. In d-cT4, 22 (55%) were evaluated in Cur A, and 9 (22.5%) in Cur C, on the other hand in cT3/<4, 60 (88%) in Cur A and 1 (1.5%) in Cur C, respectively. Gr 3 histological effectiveness was presented in 11 (28%) patients of d-cT4, and 30 (44%) in cT3/<4. In univariate analyses for OS, age younger than 75, R0 and Gr 3 were not significant (p=0.92, p=0.24, p=0.16, respectively), and Cur A and cT3/<4 were significant (p=0.002, p<0.001, respectively). In a multivariate analysis for OS, only cT3/<4 alone was significant (Exp.β 3.0, 95%CI 1.22-7.30, p=0.17). In 40 patients of d-cT4, 2 and 4-year survival rates were 56% and 42%, respectively. Although we presented possibility of conversion surgery for definite cT4 and a fair evaluation of preoperative CRT, more effective multimodality therapy is expected because of the significant prognostic impact of depth of tumor invasion.