Grading of well-differentiated pancreatic neuroendocrine tumors is improved by the inclusion of both Ki67 proliferative index and mitotic rate.

The grading system for pancreatic neuroendocrine tumors (PanNETs) adopted in 2010 by the World Health Organization (WHO) mandates the use of both mitotic rate and Ki67/MIB-1 index in defining the proliferative rate and assigning the grade. In cases when these measures are not concordant for grade, it is recommended to assign the higher grade, but specific data justifying this approach do not exist. Thus, we counted mitotic figures and immunolabeled, using the Ki67 antibody, 297 WHO mitotic grade 1 and 2 PanNETs surgically resected at a single institution. We quantified the Ki67 proliferative index by marking at least 500 cells in "hot spots" and by using digital image analysis software to count each marked positive/negative cell and then compared the results with histologic features and overall survival. Of 264 WHO mitotic grade 1 PanNETs, 33% were WHO grade 2 by Ki67 proliferative index. Compared with concordant grade 1 tumors, grade-discordant tumors were more likely to have metastases to lymph node (56% vs. 34%) (P<0.01) and to distant sites (46% vs. 12%) (P<0.01). Discordant mitotic grade 1 PanNETs also showed statistically significantly more infiltrative growth patterns, perineural invasion, and small vessel invasion. Overall survival was significantly different (P<0.01), with discordant mitotic grade 1 tumors showing a median survival of 12 years compared with 16.7 years for concordant grade 1 tumors. Conversely, mitotic grade 1/Ki67 grade 2 PanNETs showed few significant differences from tumors that were mitotic grade 2 and either Ki67 grade 1 or 2. Our data demonstrate that mitotic rate and Ki67-based grades of PanNETs are often discordant, and when the Ki67 grade is greater than the mitotic grade, clinical outcomes and histopathologic features are significantly worse than concordant grade 1 tumors. Patients with discordant mitotic grade 1/Ki67 grade 2 tumors have shorter overall survival and larger tumors with more metastases and more aggressive histologic features. These data strongly suggest that Ki67 labeling be performed on all PanNETs in addition to mitotic rate determination to define more accurately tumor grade and prognosis.

Survival benefit of liver transplantation and the effect of underlying liver disease.

BACKGROUND: The benefit of liver transplantation relative to initial degree of underlying liver disease and time on the waiting list remains poorly defined. We sought to examine the survival benefit attributable to liver transplantation across a wide range of Model for End-Stage Liver Disease (MELD) scores. METHODS: The study population included patients with end-stage liver disease enlisted in Rio Grande do Sul, Brazil, between 2001 and 2005. Survival and hazard function for enlisted and transplanted patients were estimated using parametric and nonparametric methods. MELD score was utilized to account for underlying liver disease. RESULTS: Of 1,130 eligible patients, 520 (46.0%) were transplanted, 266 (23.5%) died on the waiting list, 141 (12.5%) were excluded from the waiting list, and 203 (18.0%) remained enlisted and were awaiting transplantation at the time of last observation. At 1 year after transplantation, a MELD score of 15 represented a transition point in terms of overall survival benefit (MELD 10, 90% vs 83%; MELD 15, 81% vs 80%; MELD 20, 63% vs 78%; MELD 25, 42% vs 74%; MELD 30, 21% vs71%; enlisted vs transplant patients, respectively). MELD scores at which transplantation seemed to be beneficial relative to the amount of follow-up time was MELD 23, 17, 15, and 12 at 6 months, and 1, 2, and 5 years, respectively, from time of transplantation/enlistment. CONCLUSION: Although patients with greater MELD scores enjoy a pronounced and early benefit from transplantation, patients with lesser MELD scores do gain from transplantation, although a greater period of time is needed to realize the survival benefit.