Tumour budding and a low host inflammatory response are associated with a poor prognosis in oesophageal and gastro-oesophageal junction cancers.

AIMS

Tumour budding and host inflammatory response are parameters easily assessed histologically that have prognostic significance in many cancers. There have been few studies examining these parameters in oesophageal or gastro-oesophageal cancers. This study aims to address that deficiency.

METHODS AND RESULTS

A two-centre, retrospective study was carried out on 356 patients. Tumour budding and host inflammatory response at the invasive front were assessed histologically. Statistical analysis was performed to determine the prognostic significance of these factors. The median number of tumour buds was four (range 0-50) with 172 of 356 cases having five or more buds at the invasive front. The presence of five or more buds was associated with a poor prognosis on univariate analysis (P = 0.0001), as was a sparse or moderate host inflammatory response (P = 0.001). Tumour budding retained prognostic significance when tumours were separated into adenocarcinomas (n = 287) and squamous cell carcinomas (n = 69), but host inflammatory response was a significant prognostic factor only for adenocarcinomas. On multivariate analysis the presence of five or more buds retained significance (P = 0.002).

CONCLUSIONS

Tumour budding and host inflammatory response are important prognostic factors in patients with oesophageal/gastro-oesophageal cancer and can be used to identify high-risk patients who would benefit from closer follow-up and adjuvant therapies.