Results of a phase II trial with iodine 131-labeled murine antitenascin monoclonal antibody 81C6 (m81C6) via surgically created resection cavities in the treatment of patients with recurrent malignant brain tumors

5238 Adequate quality and quantity of genomic DNA is a bottleneck in genetic analysis of clinical tumor samples. In brain tumor patients, most often only tiny amounts of tissue are available from stereotactic needle biopsies. The yield of genomic template from these specimens by common methods of DNA isolation does frequently not allow high-throughput genetic analysis in these patients, which however may be important for treatment stratification in targeted therapy. Much effort has been invested in developing methods for whole genome amplification (WGA). Substantial variation in the extent of amplification occurring between different markers, incomplete coverage, and inadequate average DNA size has limited the use of existing WGA methods, making them particularly unsuitable for diagnostic testing. Most recently, a novel technique for WGA has been described, termed multiple displacement amplification (MDA), which provides a highly uniform representation across the human genome. MDA employs the Phi29 DNA polymerase and random primers to generate high-molecular weight DNA. The utility of MDA-based WGA has been shown for plasmid and bacteriophage DNA as well as whole blood, buccal swabs, buffy coats, and cultured cell lines. Here we have extended the use of MDA to the WGA of fresh-frozen (FF) as well as formalin-fixed/paraffin-embedded (FP) clinical solid brain tumor samples. Genomic DNA was isolated from snap-frozen glioblastoma specimens and corresponding archival formalin-fixed/ paraffin-embedded specimens and varying amounts of genomic DNA were subjected to overnight MDA. We demonstrate that MDA of nanograms of input template DNA generated microgram amounts of high-fidelity output DNA. Three nanograms of FF and FP input DNA resulted in 1610 ± 96-fold and 1591 ± 72-fold WGA, respectively, demonstrating that the yield of genomic DNA is comparable for the same amounts of input DNA from FF and FP samples. When the DNA input was increased, amplification efficiency as measured by the fold-change of input to output DNA decreased steadily, suggesting saturation of either the enzyme or the primers. Fifty nanograms of DNA input generated only 110 ± 1-fold and 81 ± 7-fold amplification of DNA template from FF and FP tissue, respectively. Our results suggest that MDA-based WGA may be an invaluable tool for DNA probe preparation for genotyping and DNA sequencing analysis from limited patient DNA sources such as fresh-frozen stereotactic biopsy material of brain tumors as well as archival formalin-fixed/paraffin-embedded, and laser microdissected brain tumor tissues. We will further report on the suitability of MDA-based WGA in quantitative assessment of genome-wide gene dosages, utilizing comparative genomic hybridization via 42,000-element cDNA microarrays (array-CGH) and real-time PCR.