Circulating Tumor Cell-Free DNA Genes as Prognostic Gene Signature for Platinum Resistant Ovarian Cancer Diagnosis

Clinical management of gynecological cancer begins by optimal debulking with first-line platinum-based chemotherapy. However, in ~80% patients, ovarian cancer will recur and is lethal. Prognostic gene signature panel identifying platinum-resistance enables better patient stratification for precision therapy. Retrospectively collected serum from 11 "poor" (<6 months progression free interval [PFI]) and 22 "favorable" (>24 months PFI) prognosis patients, were evaluated using circulating cell-free DNA (cfDNA). DNA from both groups showed 50 to 10 000 bp fragments. Pairwise analysis of sequenced cfDNA from patients showed that gene dosages were higher for 29 genes and lower for 64 genes in poor than favorable prognosis patients. Gene ontology analysis of higher dose genes predominantly grouped into cytoskeletal proteins, while lower dose genes, as hydrolases and receptors. Higher dosage genes searched for cancer-relatedness in Reactome database indicated 15 genes were referenced with cancer. Among them 3 genes, TGFBR2, ZMIZ2, and NRG2, were interacting with more than 4 cancer-associated genes. Protein expression analysis of tumor samples indicated that TGFBR2 was downregulated and ZMIZ2 was upregulated in poor prognosis patients. Our results indicate that the cfDNA gene dosage combined with protein expression in tumor samples can serve as gene signature panel for prognosis determination amongst ovarian cancer patients.

Evaluation of Machine Learning Algorithm Utilization for Lung Cancer Classification Based on Gene Expression Levels

BACKGROUND

Lung cancer remains one of the most common cancers in the world, both in terms of new cases (about 13% of total per year) and deaths (nearly one cancer death in five), because of the high case fatality. Errors in lung cancer type or malignant growth determination lead to degraded treatment efficacy, because anticancer strategy depends on tumor morphology.

MATERIALS AND METHODS

We have made an attempt to evaluate effectiveness of machine learning algorithms in the task of lung cancer classification based on gene expression levels. We processed four publicly available data sets. The Dana-Farber Cancer Institute data set contains 203 samples and the task was to classify four cancer types and sound tissue samples. With the University of Michigan data set of 96 samples, the task was to execute a binary classification of adenocarcinoma and non-neoplastic tissues. The University of Toronto data set contains 39 samples and the task was to detect recurrence, while with the Brigham and Women's Hospital data set of 181 samples it was to make a binary classification of malignant pleural mesothelioma and adenocarcinoma. We used the k-nearest neighbor algorithm (k=1, k=5, k=10), naive Bayes classifier with assumption of both a normal distribution of attributes and a distribution through histograms, support vector machine and C4.5 decision tree. Effectiveness of machine learning algorithms was evaluated with the Matthews correlation coefficient.

RESULTS

The support vector machine method showed best results among data sets from the Dana-Farber Cancer Institute and Brigham and Women's Hospital. All algorithms with the exception of the C4.5 decision tree showed maximum potential effectiveness in the University of Michigan data set. However, the C4.5 decision tree showed best results for the University of Toronto data set.

CONCLUSIONS

Machine learning algorithms can be used for lung cancer morphology classification and similar tasks based on gene expression level evaluation.