Multiplex Analysis of Cavitronic Ultrasonic Surgical Aspiration (CUSA) Specimens Can Rapidly Detect High Level Oncogene Amplifications
Long N Truong, Shashikant Patel, Sherry S Martin, Jay F LeBlanc, Anil Nanda, Mary L Nordberg, Marie E Beckner. Louisiana State University Shreveport, Shreveport, LA; Louisiana State University Health Sciences Center - Shreveport, Shreveport, LA; Delta Pathology, Shreveport, LA; Louisiana State University Health Sciences Center-NO, Baton Rouge, LA; Delta Pathology Molecular Diagnostics, Shreveport, LA
Background: Detection of amplified oncogenes in tumors is potentially influenced by specimen source. Two sources are commonly available from brain tumors, i.e. Cavitronic Ultrasonic Surgical Aspirations (CUSAs) and formalin-fixed, paraffin-embedded (FFPE) sections. Also, multiplex assay results are hindered by multiple comparisons when determining significance.
Design: This study evaluated CUSAs and FFPE specimens from the same brain tumors. Four glioblastomas, two lung carcinoma metastases, and an ependymoma were tested. Multiplex ligation dependent probe amplification (MLPA) assays to quantify seventy eight oncogenes were normalized by DNA from non-neoplastic brain (NB), [tumor DNA]/[NB DNA], to yield copy number (CN) ratios. Data were adjusted additionally by normal DNA values provided in MLPA kits. Also, FFPE and CUSA data were combined for probability calculations regarding amplifications.
Results: Purification of DNA from CUSAs was rapid (<2 days) versus FFPE (weeks) and yielded greater amounts in 6 of 7 tumors. Although normalized CN ratios ≥2.0 were more numerous in FFPE specimens, some were found only in CUSAs, consistent with tumor heterogeneity. Additionally, CN ratios adjusted further by normal DNA values provided in MLPA kits revealed 25 amplifications (≥ 2.0 CN ratio), with the nine high level (≥ 6.0) amplifications in FFPE also detected in CUSAs at increased levels. Only 2 of 6 mid level (≥ 3.0 & < 6.0 CN ratio) and 1 of 10 low level (≥ 2.0 & < 3.0 CN ratio) amplifications were detected in CUSAs. In t tests of amplified genes versus genes with normal CN ratios (≥0.75 to ≤1.5), corrections for multiple comparisons negatively impacted p values in both specimens. However, combined use of corrected CUSA and FFPE values yielded statistical significance in 64% of p values for amplifications.
Conclusions: CUSA specimens contained abundant brain tumor DNA, permitted detection of high level oncogene amplifications, and strengthened probabilities for amplifications. Possibly, whenever tumors in any body site are debulked as CUSA specimens, the DNA can be rapidly assessed for high-level oncogene amplifications and results can be used to aid statistical analyses of FFPE data from the same tumors.
Monday, March 19, 2012 1:00 PM
Poster Session II # 316, Monday Afternoon