Authors
Yick Fu Wong, Zachariah E. Selvanayagam, Nien Wei, Joseph Porter, Ragini Vittal, Rong Hu, Yong Lin, Jason Liao, Joe Weichung Shih, Tak Hong Cheung, Keith Wing Kit Lo, So Fan Yim, Shing Kai Yip, Danny Tse Ngong, Nelson Siu, Loucia Kit Ying Chan, Chun Sing Chan, Tony Kong, Elena Kutlina, Randall D. McKinnon, David T. Denhardt, Khew-Voon Chin, and Tony Kwok Hung Chung
Purpose
The incidence and mortality rates of cervical cancer are declining in the United States; however, worldwide, cervical cancer is still one of the leading causes of death in women, second only to breast cancer. This disparity is at least partially explained by the absence of or comparatively ineffective screening programs in the developing world. Recent advances in expression genomics have enabled the use of DNA microarray to profile gene expression of various cancers. These expression profiles may be suitable for molecular classification and prediction of disease outcome and treatment response. We envision that expression genomics applied in cervical cancer may provide a more rational approach to the classification and treatment of the disease.
Experimental Design
In this report, we examined the expression profiles of cervical cancer compared with normal cervical tissues in DNA microarrays that contained approximately 11,000 features that correspond to either human transcripts with known function or anonymous expressed sequence tags.
Results
Our results showed that normal cervical tissues were completely segregated from the cancer samples using about 40 genes whose expressions were significantly different between these specimens. In addition, clinical stage IB and stage IIB tumors could also be classified based on their signature expression patterns. Most importantly, some of the tumor samples were further stratified into two major groups based on their response to radiotherapy, and we were able to predict the response of these patients to radiotherapy from their expression profiles.
Conclusions
Gene expression profiling by DNA microarray may be used for further molecular classification of disease stages and prediction of treatment response in cervical cancer.