Study of the revisited, revised, and expanded Silva pattern system for Chinese endocervical adenocarcinoma patients.

As a new pattern-based classification, the Silva pattern system has been recently developed to evaluate invasive lymph node metastasis and the prognosis of endocervical adenocarcinoma (EAC). Therefore, our study was conducted to explore the reproducibility and prognostic significance of this system in a multi-institutional Chinese cohort, with the goal of revising and expanding its application. The clinicopathological data of 191 EAC patients from 3 medical centers were examined in a retrospective manner. The Silva pattern system demonstrated great prognostic value, significance in guiding treatment selection, and acceptable reproducibility in 191 patients that included additional histologic variants and 124 usual-type EAC patients. Collectively, compared with usual-type EAC, the whole cohort demonstrated similar statistical significance for relevant clinicopathological parameters, such as International Federation of Gynecology and Obstetrics stage (R = 0.612 versus R = 0.600), tumor thickness (P < .0001 versus P < .0001), lymphovascular invasion (P < .0001 versus P < .0001), lymph node metastasis (P = .033 versus P = .018), perineural invasion (P = .003 versus P = .001), and recurrence-free survival (P = .047 versus P = .020). Moreover, perineural invasion was significantly correlated (P = .001) with the Silva pattern system and appeared in most Silva C tumors. In conclusion, the Silva pattern system is consistent with the biological behavior of EAC and has acceptable reproducibility. Compared with International Federation of Gynecology and Obstetrics stage, it can predict patient prognosis before surgery. We suggest revising the Silva C criteria by adding perineural invasion as a factor and propose expanding the Silva pattern system to include more histologic variants. It seems that the Silva pattern system can be applied in routine clinical practice to guide EAC therapeutic strategies in the near future.