Constructing an ovarian cancer metastasis index by dissecting medical records

Abstract

Globally, ovarian cancer (OC) is the leading cause of gynecological cancer-associated deaths. Metastasis, especially multi-organ metastasis, determines the speed of disease progression. A multicenter retrospective study was performed to identify the factors that drive metastasis, from medical records of 534 patients with OC. The average number of target organs per patient was 3.66, indicating multi-organ metastasis. The most common sites of metastasis were large intestine and greater omentum, which were prone to co-metastasis. Results indicated that ascites and laterality, rather than age and menopausal status, were the potential drivers for multi-organ metastasis. Cancer antigen (CA) 125 (CA-125) and nine other blood indicators were found to show a significant, but weak correlation with multi-organ metastasis. A neural network cascade-multiple linear regression hybrid model was built to create an ovarian cancer metastasis index (OCMI) by integration of six multi-organ metastasis drivers including CA-125, blood platelet count, lymphocytes percentage, prealbumin, ascites, and laterality. In an independent set of 267 OC medical records, OCMI showed a moderate correlation with multi-organ metastasis (Spearman ρ = 0.67), the value being 0.72 in premenopausal patients, and good performance in identifying multi-organ metastasis (area under the receiver operating characteristic curve = 0.832), implying a potential prognostic marker for OC.

Early identification of recurrence in ovarian cancer: a comparison between the ovarian cancer metastasis index and CA-125 levels

“…For each patient, the information in her medical record was use to confirm recurrence and to calculate the OCMI, following the method established previously ( Qu et al, 2018 ). The information included the patient’s age, pharmacotherapeutic regimen, number of metastatic organs confirmed during surgery, ascites, laterality, imageologically-diagnosed or pathologically-confirmed recurrence, and four blood test indicators (CA-125, lymphocyte percentage, prealbumin, and blood platelet count) that were measured at the time of diagnosis, before surgery, and before each cycle of chemotherapy.…”

“…OCMI values were calculated for each patient when the four blood test indicators mentioned above were available, typically at the time of diagnosis, before surgery, and before any cycle of chemotherapy. An established neural network cascade (NNC)-multiple linear regression (MLR) hybrid model ( Qu et al, 2018 ) was directly used for this calculation. The hybrid model contained an NNC and an MLR formula.…”

“…The hybrid model contained an NNC and an MLR formula. The NNC, an artificial neural network model of a serial data input architecture ( Li et al, 2015 ; Hou et al, 2016 ; Cui et al, 2018 ), was previously built using a training dataset containing 534 patients with OC ( Qu et al, 2018 ). Briefly, the four blood test indicators were normalized into a 0 to 1 digital number before further use, as previously described ( Zhu & Kan, 2014 ).…”

“…The training dataset containing 534 patients with OC ( Qu et al, 2018 ) was used to determine thresholds for CA-125 levels, the NNC, and the OCMI for discriminating whether metastasis occurs in a patient with OC ( Table S1 ). MedCalc version 15.8 (MedCalc, Mariakerke, Belgium) was used, and receiver operating characteristic (ROC) curve analysis was performed to obtain the optimal thresholds of CA-125 levels, the NNC, and the OCMI.…”

“…Multi-organ metastasis often indicates a more serious disease status and lower chance of survival ( Bast Jr, Hennessy & Mills, 2009 ). Based on the inner logic of this definition, we developed a novel indicator, the ovarian cancer metastasis index (OCMI), which integrates CA-125 levels with six routine clinical examination indicators by a neural network cascade (NNC)-multiple linear regression hybrid model ( Qu et al, 2018 ). NNC is an artificial neural network (ANN) with a serial data input architecture, in which multiple small ANN units that handle single input parameter are connected in tandem to complete the overall prediction task ( Li et al, 2015 ; Hou et al, 2016 ; Cui et al, 2018 ).…”