Survival of patients with neuroendocrine carcinoma of the colon and rectum: a population-based analysis.

BACKGROUND

High-grade neuroendocrine carcinoma is a rare colorectal pathology described in a case series. The role of surgery in this disease has been questioned.

OBJECTIVE

The purpose of this work was to describe the incidence, management, and outcomes of neuroendocrine carcinoma in comparison with high-grade adenocarcinoma.

DESIGN

This was a retrospective, population-based outcomes research study.

SETTINGS

The Survey of Epidemiology and End Results database was used.

PATIENTS

A total of 1367 patients with colorectal neuroendocrine carcinoma (distinguishing small-cell and non-small-cell subtypes) and 72,533 with high-grade adenocarcinoma diagnosed between 2000 and 2011 were included in this study.

INTERVENTIONS

Resection of the primary tumor was the main intervention.

MAIN OUTCOME MEASURES

Median overall and 5-year relative survival were measured. Trends were expressed as the annual percent change in incidence and relative survival.

RESULTS

The incidence rate increased for neuroendocrine carcinoma (annual percent change, +2.2%; p =0.035) and decreased for high-grade adenocarcinoma (annual percent change, -3.1%; p < 0.00001) during the study period. Relative survival at 5 years in neuroendocrine carcinoma was 16.3% overall and 57.4%, 56.4%, 26.3%, and 3.0% for stages I, II, III, and IV cancer. Small-cell tumors had worse survival (10% versus 19% for non-small cell). There was no improvement in the relative survival for neuroendocrine carcinoma (annual percent change, -1.1%; p =0.06) in contrast to adenocarcinoma (annual percent change, +0.7%; p < 0.00001). Patients with localized non-small-cell neuroendocrine carcinoma had better overall survival with surgery (median, 21 months) than without (6 months; log-rank, p < 0.0001), whereas those with small-cell neuroendocrine carcinoma did not (18 versus 14 months; p = 0.95). Prognosis in resected neuroendocrine carcinoma was worse with an increasing number of metastatic lymph nodes.

LIMITATIONS

Histology and grade assignments were not centrally verified. Data on chemotherapy use, patient performance status, and comorbidities were unavailable.

CONCLUSIONS

Neuroendocrine carcinoma did not benefit from advances in the prevention and treatment of colorectal adenocarcinoma over the past decade. Relatively poor survival in early stage neuroendocrine carcinoma warrants studies of adjuvant systemic therapy. The differences in outcomes between small-cell and non-small-cell neuroendocrine carcinomas indicate a need for histology-specific management.