Incorporating assessments of tumor heterogeneity as part of a diagnostic may facilitate better prediction of a patient’s potential for response, and provide rationale and selection for therapy regimens that increase the response rates of refractory patients.
Recent clearances in companion diagnostics for specific mutations in melanoma and lung cancer demonstrate the importance of evaluating heterogeneity in tissue. Despite having a driving mutation in these tumors, 40-50% of selected patients will not show a response to drugs which target these mutations. This could be due to variability in mutated tumor cells, or other independent factors causing insensitivity to the drug in sub-populations of cells. Determining mutation penetration and the variance in the expression of specific proteins may help in understanding the underlying cause for refractory disease.
In this HER2 stained breast cancer sample, the lower right displays classic +3 staining pattern. However, other areas of the tumor have DCIS and much lower stained tumor levels. It also illustrates that diagnostic heterogeneity is linked to IHC heterogeneity
The DNA heterogeneity complements the IHC heterogeneity measurement. The penetrance of a particular mutation can be correlated to specific factors assessed by IHC. While immunohistochemisty is a semiquantitative technique, it has the exquisite advantage of cell-level protein expression analysis, with cellular details and contextual information about the expression of a protein in nests of cells.
Heterogeneity reflects plasticity of tumor cells, which is influenced by both gene mutations and epigenetic gene product regulation. These processes have the potential to create sub-populations of cells within a tumor which are inherently resistant or acquire resistance to a targeted therapy. The current techniques for assessing a diagnostic marker to select patients for targeted therapies fail to account for the heterogeneity seen in some types of cancer.
Flagship has developed novel, patent-pending approaches to evaluating heterogeneity in tumor samples. The steps include the following:
1. Evaluate percent tumor in consecutive sections of a tissue block
2. Evaluate DNA mutations in a given gene (e.g. EGFR, KRAS, BRAF, etc). The specific genes involved are customized for each project and each client. The DNA analysis includes both RT-PCR and Sanger Sequencing. With Sanger sequencing, a proprietary bioinformatics approach normalizes the trace profiles to yield a better fraction estimation of a mutation. Both novel and anticipated mutations can be identified.
3. Brightfield IHC or Immunofluorescence is applied to consecutive sections using FACTS with a set of protein antibodies. Whole slide scanning is followed by image analysis supervised by board-certified pathologists. Histology pattern recognition identifies tumor tissue in the sample, and then image analysis quantifies protein expression level on each slide (cell-based image analysis is generally best for nuclear or membrane markers). Heterogeneity is calculated both for cell-level (variability from cell to cell) and tumor level (variability across tumor nests).
4. Multivariate statistics is utilized to compare DNA and protein IHC results. An n-dimensional mapping and visualization approach can be used to view each sample in the context of the entire population. Samples with higher levels of heterogeneity are hypothesized to perform more poorly in response to therapy and have higher levels of eventual resistance. In addition, high heterogeneity samples will more likely result in higher interpathologist variability in evaluating these samples clinically.
The image analysis and pathology techniques behind HetMap will be presented at the Pathology Visions 2011 conference, the premier conference on Digital Pathology. The use of immunohistochemistry to analyze heterogeneity was presented at the San Antonio Breast Cancer meeting 2011.