T1N0 oesophageal cancer: patterns of care and outcomes over 25 years

Abstract

Objectives: Historically, surgical resection has been the mainstay of treatment for T1N0 oesophageal cancer (OC). More recently, oesophageal sparing endoscopic techniques have shown value for local control in a large institutional series. However, the effect of their utilization upon survival rates in large population series is largely unknown.

Methods: The surveillance, epidemiology, and end results (SEER) database was queried for T1N0M0-OC patients (1988-2013). Patients with multiple treatment types were excluded. Time periods were divided by 5-year increments. Overall survival and cancer-specific survival (CSS) were compared in the group as a whole and in propensity-matched subgroups. Independent predictors of cancer-specific mortality were studied by the Cox proportional hazard models.

Results: We identified 5497 patients with cT1N0M0 OC. Treatment modalities used were changed significantly over time. The ratio of oesophagectomy when compared with local therapy decreased from 15:1 in 1998-92 to 1.4:1 in 2008-13. The proportion of patients treated with radiation slightly increased (35% vs 41%) between 1988-92 and 2008-13. In the propensity-matched groups, 5-year CSS was similar in patients treated with oesophagectomy and local therapy (81% vs 89%; P = 0.257) (n = 216 in each group), whereas oesophagectomy had superior 5-year CSS compared with radiation alone (73% vs 38%; P < 0.001) (n = 497 in each group). In multivariable analysis, significant predictors of cancer-specific mortality included age [hazard ratio (HR) 1.022], tumour size (HR 1.005), radiation therapy (HR 3.67), tumour Grade III/IV (HR 1.25) and early time period of diagnosis (HR 1.75).

Conclusions: Oesophageal sparing endoscopic techniques have been increasingly utilized in the treatment of cT1N0-OC but without compromising CSS. Local therapy, either endoscopic techniques or surgery, remains superior to radiation therapy.