Hybrid model for the classification of carcinoma subtypes

摘要

A two-tiered classification system that can be integrated with the current algorithm used by pathologists for identification of the site of origin for ‘malignancy with unknown primary’ is presented. In use, morphology, immunohistochemical (IHC) studies, and microarray-based top tier gene expression classifiers first subclassify cytokeratin positive carcinomas into adenocarcinoma, squamous cell carcinoma, neuroendocrine carcinoma and urothelial carcinoma. Subsequently, organ-specific IHC-markers, if available, are used in conjunction with microarray-based second tier gene expression classifiers to assign the primary site of origin to the sample. This new hybrid approach combines IHC with a hierarchy of quantitative gene expression based classifiers into an algorithmic method that can assist pathologists to further refine and support their decision making process.

主权项

1. A method of identifying the origin of a neoplasm of unknown primary comprising:
obtaining a sample of a neoplasm; analyzing morphology of the sample to distinguish between solid and liquid tumor types; testing the solid sample for cytokeratin expression to determine if the solid sample is cytokeratin positive or negative; performing immunohistochemistry on the cytokeratin positive sample to differentiate between carcinoma, mesothelioma and germ cell tumors; performing immunohistochemistry on the carcinoma sample to differentiate between subclasses of carcinomas wherein the subclasses of carcinoma are selected from the group consisting of adenocarcinoma, squamous cell carcinoma, urothelial carcinoma, and neuroendocrine carcinoma; applying a hierarchal microarray gene expression classifier to gene expression data from the carcinoma sample to verify differentiation between the subclasses of the carcinomas wherein the hierarchal gene expression classifier is comprised of top tier classifiers and second tier classifiers; assigning the carcinoma sample to one of the subclasses of carcinoma based on comparing the immunohistochemistry and the top tier classifier results; applying the second tier classifier to gene expression data from the carcinoma sample to assign a primary site of origin to the carcinoma sample; and administering a treatment to the patient according to the subclass of carcinoma assigned to the carcinoma sample; wherein the top tier classifiers differentiate between the subclasses of the carcinomas wherein the subclasses of carcinoma are selected from the group consisting of adenocarcinoma, squamous cell carcinoma, urothelial carcinoma, and neuroendocrine carcinoma; wherein the top tier classifiers for adenocarcinoma use gene expression data selected from the group of genes consisting of hexokinase domain containing 1 (HKDC1), malectin (KIAA0152), calmodulin-like 4 (CALML4), amiloride binding protein 1 (ABP1), tripartite motif-containing 15 (TRIM15), hepatocyte nuclear factor 4 gamma (HNF4G), and crystallin lambda 1 (CRYL1).