Gene Profiles As Biomarkers For The Diagnosis And Classification Of Human Cancers

Gene profiling has proven quite valuable in identifying genes whose expression can be used to diagnose specific cancers, including many that are difficult to distinguish by other methods. For example, Kohlman et al. in 2003 used gene profiling to differentiate between acute lymphoblastic leukemia (ALL) and acute myeloid leukemia (AML) [18]. Bone marrow samples were isolated from 90 patients, 25 having been diagnosed with ALL, and 65 with AML. RNA was extracted from the samples and labeled cDNAs hybridized to Affymetrix U95Av2 and U133 microarrays. The WV algorithm and the leave-one-out cross-validation approach were then used to identify a subset of genes that were expressed differentially between the two leukemias. The expression of 24 genes was then used to accurately identify patients with AML, while 19

genes were sufficient to diagnose ALL. However, upon comparing the genes identified by the two types of microarrays, it was observed that the 24 genes identified by the U133 arrays were different from those identified by U95Av2 chips. Furthermore, only five genes were observed to be in common within the 19-gene set when the two types of array tools were used. These results show the importance of using the same array platforms when comparing across experiments.

The utility of the U133 arrays was studied further in the diagnosis of leukemia using 937 patients (892 with clinically relevant leukemia subtypes and 45 nonleukemic patients) [27]. Patients were divided into 13 subgroups of leukemia type, and each subgroup was split equally into training and validation sets. Class - specific gene expression was determined using an SVM approach, with an overall classification accuracy of 95.1% when the top 100 genes for leukemia class discrimination were used. The studies cited above clearly demonstrate the utility of gene profiling for the diagnosis of cancers and suggest that this approach can be used further in the classification of cancers into various subgroups. S0rlie et al.'s study in 2001 further illustrates the latter utility by demonstrating that breast cancers could be classified into a basal epithelial-iike group, an ERBB2. overexpressing group, a normal breastlike group, and a luminal epithelial/estrogen receptor-positive group, which could be further divided into at least two additional subgroups, each with a distinct gene expression profile. Interestingly, survival analyses using a subset of uniformly treated patients with locally advanced breast cancer revealed clear differences in outcomes among the various groups. For example, the basal-like group had a poor prognosis, and the two estrogen receptor-positive groups had clear differences in treatment outcome [ 28]. These findings have been well corroborated in additional experiments performed across array platforms and using RNA extracted from paraffin-embedded tissues [29-31]. Similar approaches and findings have been observed for endometrial cancer, hepatocellular carcinoma, sarcomas, neuroblastoma, and other cancers [32-35].

10 Ways To Fight Off Cancer

10 Ways To Fight Off Cancer

Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.

Get My Free Ebook

Post a comment