Identification of genes whose expression patterns differ in benign lymphoid tissue and follicular, mantle cell, and small lymphocytic lymphoma

Quantitative comparison of immunohistochemical staining measured by digital image analysis versus pathologist visual scoring

Abstract

Immunohistochemical (IHC) assays performed on formalin-fixed paraffin-embedded (FFPE) tissue sections traditionally have been semi-quantified by pathologist visual scoring of staining. IHC is useful for validating biomarkers discovered through genomics methods as large clinical repositories of FFPE specimens support the construction of tissue microarrays (TMAs) for high throughput studies. Due to the ubiquitous availability of IHC techniques in clinical laboratories, validated IHC biomarkers may be translated readily into clinical use. However, the method of pathologist semi-quantification is costly, inherently subjective, and produces ordinal rather than continuous variable data. Computer-aided analysis of digitized whole slide images may overcome these limitations. Using TMAs representing 215 ovarian serous carcinoma specimens stained for S100A1, we assessed the degree to which data obtained using computer-aided methods correlated with data obtained by pathologist visual scoring. To evaluate computer-aided image classification, IHC staining within pathologist annotated and software-classified areas of carcinoma were compared for each case. Two metrics for IHC staining were used: the percentage of carcinoma with S100A1 staining (%Pos), and the product of the staining intensity (optical density [OD] of staining) multiplied by the percentage of carcinoma with S100A1 staining (OD*%Pos). A comparison of the IHC staining data obtained from manual annotations and software-derived annotations showed strong agreement, indicating that software efficiently classifies carcinomatous areas within IHC slide images. Comparisons of IHC intensity data derived using pixel analysis software versus pathologist visual scoring demonstrated high Spearman correlations of 0.88 for %Pos (p < 0.0001) and 0.90 for OD*%Pos (p < 0.0001). This study demonstrated that computer-aided methods to classify image areas of interest (e.g., carcinomatous areas of tissue specimens) and quantify IHC staining intensity within those areas can produce highly similar data to visual evaluation by a pathologist.

Virtual slides

The virtual slide(s) for this article can be found here: http://www.diagnosticpathology.diagnomx.eu/vs/1649068103671302

Measuring molecular biomarkers in epidemiologic studies: laboratory techniques and biospecimen considerations

The future of personalized medicine depends on the ability to efficiently and rapidly elucidate a reliable set of disease-specific molecular biomarkers. High-throughput molecular biomarker analysis methods have been developed to identify disease risk, diagnostic, prognostic, and therapeutic targets in human clinical samples. Currently, high throughput screening allows us to analyze thousands of markers from one sample or one marker from thousands of samples and will eventually allow us to analyze thousands of markers from thousands of samples. Unfortunately, the inherent nature of current high throughput methodologies, clinical specimens, and cost of analysis is often prohibitive for extensive high throughput biomarker analysis. This review summarizes the current state of high throughput biomarker screening of clinical specimens applicable to genetic epidemiology and longitudinal population-based studies with a focus on considerations related to biospecimens, laboratory techniques, and sample pooling.

Development of multigene expression signature maps at the protein level from digitized immunohistochemistry slides

Molecular classification of diseases based on multigene expression signatures is increasingly used for diagnosis, prognosis, and prediction of response to therapy. Immunohistochemistry (IHC) is an optimal method for validating expression signatures obtained using high-throughput genomics techniques since IHC allows a pathologist to examine gene expression at the protein level within the context of histologically interpretable tissue sections. Additionally, validated IHC assays may be readily implemented as clinical tests since IHC is performed on routinely processed clinical tissue samples. However, methods have not been available for automated n-gene expression profiling at the protein level using IHC data. We have developed methods to compute expression level maps (signature maps) of multiple genes from IHC data digitized on a commercial whole slide imaging system. Areas of cancer for these expression level maps are defined by a pathologist on adjacent, co-registered H&E slides, allowing assessment of IHC statistics and heterogeneity within the diseased tissue. This novel way of representing multiple IHC assays as signature maps will allow the development of n-gene expression profiling databases in three dimensions throughout virtual whole organ reconstructions.

Candidate genes contributing to the aggressive phenotype of mantle cell lymphoma

Mantle cell lymphoma and small lymphocytic lymphoma are lymphocyte cancers that have similar morphologies and a common age of onset. Mantle cell lymphoma is generally an aggressive B cell lymphoma with a short median survival time, whereas small lymphocytic lymphoma is typically an indolent B cell lymphoma with a prolonged median survival time. Using primary tumor samples in bi-directional suppression subtractive hybridization, we identified genes with differential expression in an aggressive mantle cell lymphoma versus an indolent small lymphocytic lymphoma. "Virtual" Northern blot analyses of multiple lymphoma samples confirmed that a set of genes was preferentially expressed in aggressive mantle cell lymphoma compared to indolent small lymphocytic lymphoma. These analyses identified mantle cell lymphoma-specific genes that may be involved in the aggressive behavior of mantle cell lymphoma and possibly other aggressive human lymphomas. Interestingly, most of these differentially expressed genes have not been identified using other techniques, highlighting the unique ability of suppression subtractive hybridization to identify potentially rare or low expression genes.

Gene expression profiling in follicular lymphoma and its implication for clinical practice

Follicular lymphoma (FL) is characterized by an indolent and relapsing course. Recently, the clinical outcome of FL has been distinguished by immune microenvironment-associated gene signatures. In our study, gene expression profiling (GEP) was performed in 31 non-selected patients with follicular lymphoma (FL), 12 of whom were in relapse and the remaining 19 newly diagnosed. A custom oligonucleotide microarray (Agilent 8 × 15K) was used which contained probes for about 3500 genes, including those that had been previously published as demonstrating significant prognostic value. An unsupervised approach was not able to recognize clinically different FLs. As the previously published prognostically relevant gene signatures could not be properly verified, probably due to microarray platform differences, template matching was therefore used in order to define two gene sets with differential gene expression among our samples. These gene sets shared an overrepresentation of genes with similar biological functions and were termed 'T-CELL' and 'PROLIFERATION' profiles. The 'poor profile' was then defined by a high PROLIFERATION score (upper tertile) and/or low T-CELL score (lower tertile). The 'poor profile' cohort contained a significantly higher proportion of relapsed cases (p < 0.05, Fisher's exact test). Additionally, a comparison of samples from initial diagnosis and from relapse showed significant differences mainly in the T-CELL profile (p = 0.036; χ(2)). This supports the hypothesis that the number of T-cells and their expression pattern play a major role in FL development.

Gene knockdown studies revealed CCDC50 as a candidate gene in mantle cell lymphoma and chronic lymphocytic leukemia

The two B-cell non-Hodgkin's lymphoma entities, chronic lymphocytic leukemia (CLL) and mantle cell lymphoma (MCL), show recurrent chromosomal gains of 3q25-q29, 12q13-q14 and 18q21-q22. The pathomechanisms affected by these aberrations are not understood. The aim of this study was to identify genes, located within these gained regions, which control cell death and cell survival of MCL and CLL cancer cells. Blood samples collected from 18 patients with CLL and 6 patients with MCL, as well as 6 cell lines representing both malignancies were analyzed by gene expression profiling. By a comparison of genomic DNA and gene expression, 72 candidate genes were identified. We performed a limited RNA interference screening with these candidates to identify genes affecting cell survival. CCDC50 (coiled coil domain containing protein 50), SERPINI2 and SMARCC2 mediated a reduction of cell viability in primary CLL cells as well as in cell lines. Gene knockdown and a nuclear factor kappa B (NFkappaB) reporter gene assay revealed that CCDC50 is required for survival in MCL and CLL cells and controls NFkappaB signaling.

The critical role of histology in an era of genomics and proteomics: a commentary and reflection

The role of histologic examination in lymphoma diagnosis has been called into question by proponents of new technologies, such as genomics and proteomics. We review the history and salient features of morphologic evaluation in lymphoid diseases, and discuss the general and specific limitations of mature ancillary techniques, such as immunohistochemistry, flow cytometry, and molecular studies. We then speculate on the future relationship between morphology and the new genomic and proteomic technologies as they become integrated into clinical practice.

Pathway-specific apoptotic gene expression profiling in chronic lymphocytic leukemia and follicular lymphoma

Defects in the apoptotic pathway are pathogenetically important in chronic lymphocytic leukemia and follicular lymphoma. To further understand these defects, we profiled the apoptotic gene expression of these two neoplasms. Oligonucleotide arrays with 112 apoptotic genes were used, and data analysis was performed on seven chronic lymphocytic leukemia and 10 follicular lymphoma frozen tumor samples from six and seven patients, respectively. The overall gene expression pattern was strikingly similar among all 17 samples, regardless of the type of lymphoma and history of chemotherapy exposure. MCL1, TNFRSF1B and TNFRSF7 were highly expressed in most cases. The apoptotic gene expression between the groups of untreated chronic lymphocytic leukemia (n=3) and untreated follicular lymphoma (n=6) was also similar (Pearson correlation coefficient, 0.94). Comparison between the groups of untreated chronic lymphocytic leukemia (n=3) and postchemotherapy chronic lymphocytic leukemia (n=4) revealed six genes with >2-fold changes, including BIRC5/Survivin that was higher in the postchemotherapy samples. This finding was validated by immunohistochemistry. Similar analysis of follicular lymphoma cases did not identify any significant differences. To conclude, our findings suggest that chronic lymphocytic leukemia and follicular lymphoma share common apoptotic defects, and highlight the importance of MCL1 and the TNF pathway. Upregulation of survivin may be one of the mechanisms by which chronic lymphocytic leukemia becomes desensitized to chemotherapy.

Acute myeloid leukemia subgroups identified by pathway-restricted gene expression signatures

Acute myeloid leukemia (AML) is a heterogeneous group of disorders characterized by abnormal proliferation of myeloid precursors and a maturation block. Underlying genetic lesions determine an altered expression program (transcriptosome) that can be studied in depth by massive technologies. Alternatively, we selected a pathway profiling strategy based on the current knowledge in order to stratify de novo AML patients and identify those cases which would potentially benefit from the use of new chemotherapeutic agents. One hundred and thirty-two RNA samples obtained from de novo adult AML patients were tested for FLT3, FLT3-LG, NDST1, HDAC2, ATRX, FOS, DNMT1, DNMT3A, DNMT3B, NBS1, RAD50, MRE11A, Meis1 and Meis2 expression using quantitative PCR (qPCR) assays. Clinical and biologic findings were correlated with expression results. Cluster analysis was also performed. FLT3 expression defined three subgroups of patients. The best outcome was found in the group with the lowest FLT3 expression. Intermediate levels of FLT3 were associated with the worst outcome. Patients with low levels of ATRX more frequently presented an adverse karyotype whereas cases with preserved ATRX levels showed an excellent outcome. In accordance with previous results, Meis1 downregulation is a useful surrogate marker indicating a good prognosis in AML patients. Simple qPCR platforms may help to identify different biologic subgroups in AML.

Gene expression profiling data in lymphoma and leukemia: review of the literature and extrapolation of pertinent clinical applications

CONTEXT

Gene expression (GE) analyses using microarrays have become an important part of biomedical and clinical research in hematolymphoid malignancies. However, the methods are time-consuming and costly for routine clinical practice.

OBJECTIVES

To review the literature regarding GE data that may provide important information regarding pathogenesis and that may be extrapolated for use in diagnosing and prognosticating lymphomas and leukemias; to present GE findings in Hodgkin and non-Hodgkin lymphomas, acute leukemias, and chronic myeloid leukemia in detail; and to summarize the practical clinical applications in tables that are referenced throughout the text.

DATA SOURCE

PubMed was searched for pertinent literature from 1993 to 2005.

CONCLUSIONS

Gene expression profiling of lymphomas and leukemias aids in the diagnosis and prognostication of these diseases. The extrapolation of these findings to more timely, efficient, and cost-effective methods, such as flow cytometry and immunohistochemistry, results in better diagnostic tools to manage the diseases. Flow cytometric and immunohistochemical applications of the information gained from GE profiling assist in the management of chronic lymphocytic leukemia, other low-grade B-cell non-Hodgkin lymphomas and leukemias, diffuse large B-cell lymphoma, nodular lymphocyte-predominant Hodgkin lymphoma, and classic Hodgkin lymphoma. For practical clinical use, GE profiling of precursor B acute lymphoblastic leukemia, precursor T acute lymphoblastic leukemia, and acute myeloid leukemia has supported most of the information that has been obtained by cytogenetic and molecular studies (except for the identification of FLT3 mutations for molecular analysis), but extrapolation of the analyses leaves much to be gained based on the GE profiling data.

Molecular biology of lymphoma in the microarray era

This review will focus on the molecular biology of lymphoproliferative disorders with emphasis on lymphomas. The spectrum of known recurrent gene rearrangements found in lymphomas will be outlined and their relevance to diagnosis and subclassification of disease will be discussed. Finally, a survey of the current trends in gene expression profiling of lymphomas by microarray technology will be presented with reference to implications for diagnosis, classification, prognosis and treatment.