In:
Cancer Research, American Association for Cancer Research (AACR), Vol. 71, No. 8_Supplement ( 2011-04-15), p. LB-268-LB-268
Kurzfassung:
Formalin-fixed archival tissues represent an invaluable resource for cancer research, since they are widely available and afford correlations of historical clinical data with gene expression signatures thereby facilitating better disease classification and prediction. While FFPE tissue is commonly available in block form, it is also standard practice to archive sectioned slices (AS-FFPE), often at room temperature, which further degrades genetic material due to oxidation. Because of these limitations, gene expression studies using FFPE material are usually performed on material derived from freshly cut blocks and not AS-FFPE samples. Previous mRNA data generated on freshly cut FFPE blocks for two independent liver cancer sample sets yielded gene expression signatures which classified hepatocellular carcinoma (HCC) samples into three subclasses (Cancer Res 2009, 69: 7385) and cirrhotic liver (CL) samples into two (good and poor survival) groups (N Engl J Med 2008, 359:1995). Here, using the WG-DASL HT assay (29K probes), we report on our efforts to use RNA derived from 5-year old AS-FFPE samples that were obtained from a subset of the previously assayed liver cancer sample sets, to generate clinically informative gene expression profiles. With a passing quality control filter of & gt;80% probes detected (p & lt; 0.01) we were able to obtain good quality data for 64/83 (77%) and 37/48 (77%) of the HCC and CL samples, respectively. On average, for the two sample sets, the sample reproducibility across all 29K probe intensities was R2 ∼ 0.85. We next performed prediction analysis using a nearest template prediction method and compared the results with that originally predicted for each of the two sample set. With a confidence threshold of p & lt; 0.05, 38/64 (59%) of the passing HCC samples yielded a prediction with an accuracy of 95% and an error rate of 5%. At the same confidence threshold (p & lt; 0.05), 13/37 (35%) of the passing LC samples yielded a prediction with an accuracy of 69% and an error rate of 31%. Together our results suggest that, despite degradative events such as oxidation and hydrolysis that further fragment archived FFPE sections, gene expression signatures derived from AS-FFPE samples can accurately classify disease states and provide clinically relevant prognoses. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr LB-268. doi:10.1158/1538-7445.AM2011-LB-268
Materialart:
Online-Ressource
ISSN:
0008-5472
,
1538-7445
DOI:
10.1158/1538-7445.AM2011-LB-268
Sprache:
Englisch
Verlag:
American Association for Cancer Research (AACR)
Publikationsdatum:
2011
ZDB Id:
2036785-5
ZDB Id:
1432-1
ZDB Id:
410466-3
