In 2010, World Health Organization (WHO) reclassified pancreatic neuroendocrine tumors (p-NETs) into 4 main groups: neuroendocrine tumor G1 (NET G1), neuroendocrine tumor G2 (NET G2), neuroendocrine carcinoma G3 (NEC G3), mixed adeno and neuroendocrine carcinoma (MANEC). Clinical value of these newly updated WHO grading criteria has not been rigorously validated. The authors aimed to evaluate the clinical consistency of the new 2010 grading classifications by WHO and the 2010 tumor-node metastasis staging systems by American Joint Committee on Cancer (AJCC) on survivals for patients with surgically resected p-NETs. Moreover, the authors would validate the prognostic value of both criteria for p-NETs.
The authors retrospectively collected the clinicopathologic data of 120 eligible patients who were all surgically treated and histopathologically diagnosed as p-NETs from January 2004 to February 2014 in our single institution.
The new WHO criteria were assigned to 4 stratified groups with a respective distribution of 62, 35, 17, and 6 patients. Patients with NET G1 or NET G2 obtained a statistically better survival compared with those with NEC G3 or MANEC (P < 0.001). Survivals of NET G1 was also better than those of NET G2 (P = 0.023), whereas difference of survivals between NEC G3 and MANEC present no obvious significance (P = 0.071). The AJCC 2010 staging systems were respectively defined in 61, 36, 12, and 11 patients for each stage. Differences of survivals of stage I with stage III and IV were significant (P < 0.001), as well as those of stage II with III and IV (P < 0.001); whereas comparisons of stage I with stage II and stage III with IV were not statistically significant (P = 0.129, P = 0.286; respectively). Together with radical resection, these 2 systems were both significant in univariate and multivariate analysis (P < 0.05).
The newly updated WHO 2010 grading classifications and the AJCC 2010 staging systems could consistently reflect the clinical outcome of patients with surgically resected p-NETs. Meanwhile, both criteria could be independent predictors for survival analysis of p-NETs.