par Markar, Sheraz;Cappeliez, Serge ; [et al.]
Référence European journal of cancer, 75, page (179-189)
Publication Publié, 2017-04
Article révisé par les pairs
Résumé : Background The objectives of this study were to compare peri-operative and long-term outcomes from oesophageal cancer (EC) (i) that arose in a previously radiated field (ECRF) versus primary (PEC) and among ECRF patients and (ii) radiotherapy-induced (RIEC) versus non-radiotherapy–induced EC (NRIEC). Methods Data were collected from 30 European centres from 2000 to 2010. Two thousand four hundred eighty nine EC patients surgically treated were included in the PEC group and 136 in the ECRF group, NRIEC group (n = 61) and RIEC group (n = 75). Propensity score matching analyses were used to compensate for differences in baseline characteristics. Results Compared to the PEC group, the ECRF group was characterised by less use of neoadjuvant chemoradiotherapy (0% versus 29.5%; P < 0.001), less pathological stage III/IV (31.6% versus 39.2%, P = 0.036), greater incidence of R1/2 margins (21.3% versus 10.9%; P < 0.001), increased in-hospital mortality (14.0% versus 7.1%; P = 0.003) and overall morbidity (68.4% versus 56.4%, P = 0.006). After matching, 5-year overall (28.8% versus 50.5%; hazard ratio [HR] = 1.53, 95% confidence interval [CI]: 1.15–2.04; P = 0.003) and event-free (32.2% versus 42.5%; HR = 1.56, 95% CI: 1.18–2.05; P = 0.002) survivals were significantly reduced in the ECRF group. There were no significant differences in incidence or pattern of tumour recurrence. Comparing RIEC and NRIEC groups, there were no significant differences in short- or long-term outcomes before and after matching. Conclusions ECRF is associated with poorer long-term survival related to a reduced utilisation of neoadjuvant chemoradiotherapy and an increased incidence of tumour margin involvement at surgery. Outcomes appear to be dictated by the limitations related to previous radiotherapy administration more than the radiotherapy-induced carcinogenesis.