Pulmonary carcinoid tumors are relatively uncommon and have an indolent clinical course. The role of histologic grading and cell proliferation as measured by a Ki-67 index in predicting long-term recurrence in carcinoid tumors of the lung is not defined. We report the largest single-institution study of carcinoid tumors and correlate histologic grade and Ki-67 index with clinical outcome. We reviewed all surgical lung resection cases from 1995 to 2016 with a diagnosis of primary carcinoid tumor. We collected clinicopathologic parameters, including tumor size, nodal status, histologic pattern, presence of lymphovascular invasion, mitotic count, %Ki-67 positive cells (Ki-67 index) using a digital algorithm, time to tumor recurrence, and staged these tumors based on the 8th edition of TNM Staging. The final cohort consists of 176 carcinoid tumor cases with complete data: 165 (94%) were typical carcinoids and 11 (6%) were atypical carcinoids. The Ki-67 index is significantly increased in atypical versus typical carcinoids and in higher stage disease. Only the Ki-67 index and not the histologic patterns or lymphovascular invasion status was a significant predictor of tumor recurrence on multivariate analysis among all pulmonary carcinoid tumors and within typical carcinoid tumors alone. A Ki-67 index cutoff of 5% offered the optimal combination of sensitivity and specificity in predicting long-term recurrence based on the receiver operating characteristic curve. In addition, stratifying pulmonary carcinoid tumors based on a 3-tier histologic grading system (grade 1: typical carcinoids with Ki-67 index ≤5%, grade 2: typical carcinoids with Ki-67 index >5%, and grade 3: atypical carcinoids regardless of Ki-67 index) significantly correlated with likelihood of tumor recurrence. Finally, we propose an integrated staging system unique to pulmonary carcinoid tumors by keeping the original TNM stage for grade 1 tumors, but upstaging grade 2 tumors to stage II, and grade 3 tumors to stage III.
Department of Pathology, Robert J. Tomsich Pathology and Laboratory Medicine Institute, Cleveland Clinic, Cleveland, OH
Conflicts of Interest and Source of Funding: The authors have disclosed that they have no significant relationships with, or financial interest in, any commercial companies pertaining to this article.
Correspondence: Carol F. Farver, MD, Department of Pathology, The University of Michigan, NCRC Building 35, Room 30-1531, 2800 Plymouth Road, SPC 2800, Ann Arbor, MI 48109-2800 (e-mail: firstname.lastname@example.org).