Objective: Most molecular analyses of high-grade serous ovarian, peritoneal and fallopian tube carcinomas (HGSC) require ≥70% tumor (neoplastic) cell nuclei. We characterized the distribution of the percentage of neoplastic nuclei (PNN) in a large cohort of HGSC and correlated PNN with clinical outcomes to determine the fraction of cases outside this range and whether this cut-off introduces selection bias.
Methods: Subjects were prospectively enrolled and normal and neoplastic tissues were snap-frozen. All subjects had grade 2 to 3 HGSC. Subjects that received neoadjuvant chemotherapy were excluded. PNN was determined by estimating the fraction of neoplastic nuclei relative to non-neoplastic nuclei on a representative hematoxylin and eosin stained frozen section from the primary neoplasm. Germline BRCA mutation status was determined with Sanger or BROCA sequencing.
Results: PNN was <70% in 101 (33%) of 306 cases. PNN was significantly higher among subjects without optimal cytoreduction (P=0.018). 55 subjects had germline BRCA1/BRCA2 mutations. HGSC associated with BRCA2 but not BRCA1 mutations had significantly lower PNN compared to HGSC in non-carriers (54% vs. 70%, P=0.018). Overall survival was not significantly different between subjects with <70% or ≥70% PNN (median survival 51.8 vs. 46.6months, P=0.858).
Conclusions: One-third of HGSC has PNN <70%. Higher PNN is associated with suboptimal cytoreduction, while lower PNN is associated with inherited BRCA2 mutations. Our findings suggest a nonrandom distribution of PNN that may reflect cancer biology. Further studies exploring the stromal microenvironment are needed. Molecular analyses of HGSC selected for high PNN exclude a significant fraction of patients.
Keywords: BRCA1; BRCA2; Cancer histology; Ovarian cancer.
Copyright © 2016 Elsevier Inc. All rights reserved.