Gene expression signature for classification of tissue of origin of tumor samples

摘要

The present invention provides a process for classification of cancers and tissues of origin through the analysis of the expression patterns of specific microRNAs and nucleic acid molecules relating thereto. Classification according to a microRNA tree-based expression framework allows optimization of treatment, and determination of specific therapy.

主权项

1. A method of identifying a tissue of origin of a cancer sample, the method comprising:
(a) obtaining a cancer sample from a subject, wherein said sample is of a primary cancer, a metastatic cancer, or a cancer of unknown primary origin; (b) determining an expression profile in said sample of nucleic acids comprising SEQ ID NOS: 1-48, wherein said determining an expression profile comprises contacting the sample with a primer comprising a sequence selected from the group consisting of SEQ ID NOs: 73, 74, and 75, or a sequence at least 90% identical thereto; (c) comparing said expression profile to a reference expression profile by using a classifier algorithm selected from the group consisting of a decision tree classifier, logistic regression classifier, nearest neighbor classifier, neural network classifier, Gaussian mixture model (GMM), Support Vector Machine (SVM) classifier, nearest centroid classifier, linear regression classifier and random forest classifier; whereby the altered expression of SEQ ID NOs: 1-48 relative to the reference expression profile allows the identification of the tissue of origin of said sample; and (d) identifying the tissue of origin of said sample based on the altered expression profile of SEQ ID NOs: 1-48 in said sample.