Magnetic resonance imaging in categorization of ovarian epithelial cancer and survival analysis with focus on apparent diffusion coefficient value: correlation with Ki-67 expression and serum cancer antigen-125 level

Abstract

Background To determine whether magnetic resonance (MR) imaging features combined with apparent diffusion coefficient (ADC ) values could be used as a tool for categorizing ovarian epithelial cancer (OEC) and predicting survival, as well as correlating with laboratory tests (serum cancer antigen 125, serum CA-125) and tumor proliferative index (Ki-67 expression). Methods and materials MRI examination was undertaken before invasive procedures. MRI features were interpreted and recorded on the picture archive communication system (PACS). ADC measurements were manually performed on post-process workstation. Clinical characteristics were individually retrieved and recorded through the hospital information system (HIS). Cox hazard model was used to estimate the effects of both clinical and MRI features on overall survival. Results Both clinical and MRI features differed significantly between Type I and Type II cancer groups ( p < 0.05). The mean ADC value was inversely correlated with Ki-67 expression in Type I cancer ( ρ = − 0.14, p < 0.05). A higher mean ADC value was more likely to suggest Type I ovarian cancer (Odds Ratio (OR) = 16.80, p < 0.01). Old age and an advanced International Federation of Gynecology and Obstetrics (FIGO) stage were significantly related to Type II ovarian cancer (OR = 0.22/0.02, p < 0.05). An advanced FIGO stage, solid components, and old age were significantly associated with poor survival (Hazard Ratio (HR) = 23.54/3.69/2.46, p < 0.05). Clear cell cancer type had a poorer survival than any other pathological subtypes of ovarian cancer (HR = 13.6, p < 0.01). Conclusions MR imaging features combined with ADC value are helpful in categorizing OEC. ADC values can reflect tumor proliferative ability. A solid mass may predict poor prognosis for OEC patients. Electronic supplementary material The online version of this article (10.1186/s13048-019-0534-0) contains supplementary material, which is available to authorized users.

Factors associated with surgical morbidity of primary debulking in epithelial ovarian cancer

“…Epithelial ovarian cancer (EOC) represents nearly 70% of ovarian cancers and is the second most common cancer among gynecological cancers [1]. EOC is usually diagnosed at the advanced stage (up to 75%).…”

MRI-based radiomics and ADC values are related to recurrence of endometrial carcinoma: a preliminary analysis

“…Surov et al [ 30 ] conducted a multicenter study of breast cancer and found that ADC values were significantly correlated with Ki-67 expression in breast cancer. Similarly, Zhang et al [ 31 ] reported that lower mean ADC values were associated with higher Ki-67 expression in Type I ovarian epithelial cancer. ADC values can reflect the diffusion degree of water molecules and is closely related to cell density.…”

An Application of Machine Learning That Uses the Magnetic Resonance Imaging Metric, Mean Apparent Diffusion Coefficient, to Differentiate between the Histological Types of Ovarian Cancer

“…Since ECs have a similar structure, showing a solid mass of cells, immunohistochemical analysis is crucial for differentiating between the two types of EOC. Our results showed that HGSOC exhibited denser cellularity than EC did [ 22 , 24 , 26 , 27 ]. In contrast, the lower cellularity of MC was found to be accompanied by an abundant mucinous component, which resulted in a higher mean ADC value than HGSOC [ 28 ].…”

“…Tanaka et al found that bilateral disease, small tumours, and increased signal intensity on DWI were representative of serous carcinomas, while multilocular cysts and increased CA 19-9 levels were representative of mucinous tumours [ 23 ]. Zhang et al found that an increased mean ADC value was more likely to suggest type I EOC (odds ratio (OR) = 16.80, p < 0.01); however, the study was limited by selection bias, as patients with borderline epithelial tumours comprised 51% of the study patients with type I ovarian cancer [ 24 ]. Therefore, to the best of our knowledge, ours is the first study to evaluate the correlations between five histological types of EOC, and we believe that the ADC findings correlated with the histology of EOCs are an important contribution to the findings of previous studies.…”