Decentral gene expression analysis for ER+/Her2- breast cancer: results of a proficiency testing program for the EndoPredict assay.

Gene expression profiles provide important information about the biology of breast tumors and can be used to develop prognostic tests. However, the implementation of quantitative RNA-based testing in routine molecular pathology has not been accomplished, so far. The EndoPredict assay has recently been described as a quantitative RT-PCR-based multigene expression test to identify a subgroup of hormone-receptor-positive tumors that have an excellent prognosis with endocrine therapy only. To transfer this test from bench to bedside, it is essential to evaluate the test-performance in a multicenter setting in different molecular pathology laboratories. In this study, we have evaluated the EndoPredict (EP) assay in seven different molecular pathology laboratories in Germany, Austria, and Switzerland. A set of ten formalin-fixed paraffin-embedded tumors was tested in the different labs, and the variance and accuracy of the EndoPredict assays were determined using predefined reference values. Extraction of a sufficient amount of RNA and generation of a valid EP score was possible for all 70 study samples (100%). The EP scores measured by the individual participants showed an excellent correlation with the reference values, respectively, as reflected by Pearson correlation coefficients ranging from 0.987 to 0.999. The Pearson correlation coefficient of all values compared to the reference value was 0.994. All laboratories determined EP scores for all samples differing not more than 1.0 score units from the pre-defined references. All samples were assigned to the correct EP risk group, resulting in a sensitivity and specificity of 100%, a concordance of 100%, and a kappa of 1.0. Taken together, the EndoPredict test could be successfully implemented in all seven participating laboratories and is feasible for reliable decentralized assessment of gene expression in luminal breast cancer.

Single-cell immunohistochemical mutation load assay (SCIMLA) using human paraffin-embedded tissues.

It would be advantageous to measure mutation load in situ in order to determine the relationship between a high mutation load and increased risk for cancer or other diseases and to evaluate sources of possible mutagen exposure. Previously, in situ mutation detection assays have been plagued with multiple rounds of amplification and high rates of false-positives and false-negatives. The single cell immunohistochemical mutation load assay (SCIMLA) was developed to measure somatic mutation frequency, pattern, and spectrum in normal tissues with a single round of amplification. The P53 gene was utilized as a mutation reporter because of the unusual property that missense mutations often cause P53 protein to accumulate in the cell, allowing the mutant proteins to be detected by immunohistochemical staining. Alternative reporter genes with stabilized mutant proteins may be envisioned. Single cells that stain positively for P53 protein overabundance (red cells) were microdissected from ethanol-fixed and paraffin-embedded tissues. A novel stimulated-PCR (S-PCR) protocol permitted successful amplification of a 1.8-kb segment of the P53 gene (i.e., exons 5-9) in 87% of single mammary cells. Subsequent sequence analysis demonstrated that 35% of the amplified red-stained epithelial cells from normal breast tissue have missense mutations at evolutionarily conserved amino acids. Jackpot mutations, presumably due to clonal expansion, were common. False-positive missense mutations at conserved residues were observed in 3% of the clear cells (i.e., without red stain), presumably due to DNA polymerase error in early PCR cycles. The allele dropout rate was measured at 40% of the amplified cells. SCIMLA is applicable to a variety of tissues, utilizes a single amplification of an endogenous gene, displays mutant cells in situ, and may be adapted to other species.

Absence of somatic ATM missense mutations in 58 mammary carcinomas.

Accumulating evidence indicates that germline missense mutations in the ATM gene predispose to breast cancer. To investigate the potential role of somatic ATM mutations in the tumorigenesis of breast cancer, the ATM gene was scanned in 58 mammary carcinomas using DOVAM-S (detection of virtually all mutations-SSCP [single-strand conformation polymorphism]), a robotically enhanced, highly redundant form of SSCP that detects virtually all mutations. A total of 1.65 megabases of tumor DNA sequence was scanned and 16 structural variants were identified, including one novel nonsense mutation, four novel missense mutations, and a common missense change in African-Americans. Sequencing from microdissected normal cells reveals that all variants were present in the germline. Thus, the ATM gene may be similar to the BRCA1/BRCA2 genes in that germline mutations are important in cancer predisposition, but somatic mutations are seldom present in tumors. Loss of heterozygosity (LOH) is common in these tumors, but ATM missense mutations occur with similar frequencies when LOH is present or absent (P=0.73). If germline ATM missense mutations predispose to breast cancer, the unmasking of a recessive missense allele by LOH does not seem to be a critical step in breast neoplasia.

Toward efficient analysis of mutations in single cells from ethanol-fixed, paraffin-embedded, and immunohistochemically stained tissues.

Only a few studies have demonstrated successful molecular analysis after whole genome amplification using single cells dissected from paraffin-embedded tissues. The results in these studies were limited by low-amplification efficiency and high rates of allele dropout. In the present study, the amplification rate using a thoroughly modified primer extension and preamplification-PCR protocol was improved significantly for single cells microdissected from paraffin-embedded and immunohistochemically stained tissues. Tissue fixation with ethanol (85%) and the addition of 0.2 mmol/L EDTA helped to achieve an amplification rate between 67% (segments 200 to 400 bp) and 72% (segments <200 bp). Normal tissue sections were immunohistochemically double stained for overabundance of p53 protein and proliferating cell nuclear antigen. Microdissection of single cells was performed with a manual micromanipulator equipped with a Tungsten needle. Sequence analysis of the TP53 gene was performed after improved primer extension preamplification-PCR and multiplex PCR from single microdissected cells. The rate of allele dropout was at least 68%. These technical advances facilitate routine mutation analysis using a single cell or a few cells microdissected from routinely processed paraffin-embedded normal and tumor tissues. Allele dropout still represents a serious problem in single-cell mutation analysis, especially in samples with limited template DNA and prone to DNA damage.

Occurrence of chromosome 9 and p53 alterations in multifocal dysplasia and carcinoma in situ of human urinary bladder.

To define the genetic changes of flat urothelial lesions, carcinoma in situ (CIS) and moderate dysplasias (DII) were investigated for alterations in the two chromosomal regions most frequently involved in bladder cancer. Overall, 33 CIS and 16 DII from 21 patients were used to microdissect urothelium. Dual color fluorescence in situ hybridization (FISH) using gene locus probes of 9q22 (FACC), 9p21 (CDK), 17p13 (p53), and related centromeric probes was applied on interphase nuclei. In parallel, preamplified DNA of these samples was used for loss of heterozygosity (LOH) analyses with eight microsatellite markers on chromosomes 9p, 9q and 17p, and for sequencing of exons 5-9 of p53. Data indicated nearly identical deletion frequencies for chromosomes 9 and 17 for CIS (chromosome 9, 86%; p53, 84%). DII showed a lower deletion rate in comparison with CIS (chromosome 9, 75%; p53, 53%). A very high correlation between the results of FISH and LOH analyses was found. p53 mutations were detected in 12 of 15 patients (CIS, 72%; DII, 67%). In three of 16 patients with multifocal tumors, oligoclonal lesions were identified by LOH analyses, a finding further supported by sequencing of p53, by which two different p53 deletions were detected in two cases. In conclusion, data from microdissected flat urothelial lesions indicate that chromosome 9 deletions cannot be regarded as indicators of papillary growth, because they are found frequently in both types of flat lesions of the urothelium: those associated with papillary tumors and those that are not. The similar distribution and lower amount of genetic changes in DII render DII a possible precursor lesion of CIS.