Identification of specific subtypes of circulating tumor cells in peripheral blood of cancer patients can provide information about the biology of metastasis and improve patient management. However, to be effective, the method used to identify circulating tumor cells must detect all tumor cell types. We investigated whether the five subtypes of human breast cancer cells that have been defined by global gene expression profiling—normal-like, basal, HER2-positive, and luminal A and B—were identified by CellSearch, a US Food and Drug Administration–approved test that uses antibodies against the cell surface–expressed epithelial cell adhesion molecule (EpCAM) to isolate circulating tumor cells. We used global gene expression profiling to determine the subtypes of a well-defined panel of 34 human breast cancer cell lines (15 luminal, nine normal-like, five basal-like, and five Her2-positive). We mixed 50-150 cells from 10 of these cell lines with 7.5 mL of blood from a single healthy human donor, and the mixtures were subjected to the CellSearch test to isolate the breast cancer cells. We found that the CellSearch isolation method, which uses EpCAM on the surface of circulating tumor cells for cell isolation, did not recognize, in particular, normal-like breast cancer cells, which in general have aggressive features. New tests that include antibodies that specifically recognize normal-like breast tumor cells but not cells of hematopoietic origin are needed.

Breast tumors often display extreme genetic heterogeneity characterized by hundreds of gross chromosomal aberrations and tens of thousands of somatic mutations. Tumor evolution is thought to be ongoing and driven by multiple mutagenic processes. A major outstanding question is whether primary tumors have preexisting mutations for therapy resistance or whether additional DNA damage and mutagenesis are necessary. Drug resistance is a key measure of tumor evolvability. If a resistance mutation preexists at the time of primary tumor presentation, then the intended therapy is likely to fail. However, if resistance does not preexist, then ongoing mutational processes still have the potential to undermine therapeutic efficacy. The antiviral enzyme APOBEC3B (apolipoprotein B mRNA-editing enzyme, catalytic polypeptide-like 3B) preferentially deaminates DNA C-to-U, which results in signature C-to-T and C-to-G mutations commonly observed in breast tumors. We use clinical data and xenograft experiments to ask whether APOBEC3B contributes to ongoing breast tumor evolution and resistance to the selective estrogen receptor modulator, tamoxifen. First, APOBEC3B levels in primary estrogen receptor–positive (ER⁺) breast tumors inversely correlate with the clinical benefit of tamoxifen in the treatment of metastatic ER⁺ disease. Second, APOBEC3B depletion in an ER⁺ breast cancer cell line results in prolonged tamoxifen responses in murine xenograft experiments. Third, APOBEC3B overexpression accelerates the development of tamoxifen resistance in murine xenograft experiments by a mechanism that requires the enzyme’s catalytic activity. These studies combine to indicate that APOBEC3B promotes drug resistance in breast cancer and that inhibiting APOBEC3B-dependent tumor evolvability may be an effective strategy to improve efficacies of targeted cancer therapies.

Current normalization methods for RNA-sequencing data allow either for intersample comparison to identify differentially expressed (DE) genes or for intrasample comparison for the discovery and validation of gene signatures. Most studies on optimization of normalization methods typically use simulated data to validate methodologies. We describe a new method, GeTMM, which allows for both inter- and intrasample analyses with the same normalized data set. We used actual (i.e. not simulated) RNA-seq data from 263 colon cancers (no biological replicates) and used the same read count data to compare GeTMM with the most commonly used normalization methods (i.e. TMM (used by edgeR), RLE (used by DESeq2) and TPM) with respect to distributions, effect of RNA quality, subtype-classification, recurrence score, recall of DE genes and correlation to RT-qPCR data. We observed a clear benefit for GeTMM and TPM with regard to intrasample comparison while GeTMM performed similar to TMM and RLE normalized data in intersample comparisons. Regarding DE genes, recall was found comparable among the normalization methods, while GeTMM showed the lowest number of false-positive DE genes. Remarkably, we observed limited detrimental effects in samples with low RNA quality. We show that GeTMM outperforms established methods with regard to intrasample comparison while performing equivalent with regard to intersample normalization using the same normalized data. These combined properties enhance the general usefulness of RNA-seq but also the comparability to the many array-based gene expression data in the public domain.

Published prognostic gene signatures in breast cancer have few genes in common. Here we provide a rationale for this observation by studying the prognostic power and the underlying biological pathways of different gene signatures. Gene signatures to predict the development of metastases in estrogen receptor-positive and estrogen receptor-negative tumors were identified using 500 re-sampled training sets and mapping to Gene Ontology Biological Process to identify over-represented pathways. The Global Test program confirmed that gene expression profilings in the common pathways were associated with the metastasis of the patients. The apoptotic pathway and cell division, or cell growth regulation and G-protein coupled receptor signal transduction, were most significantly associated with the metastatic capability of estrogen receptor-positive or estrogen-negative tumors, respectively. A gene signature derived of the common pathways predicted metastasis in an independent cohort. Mapping of the pathways represented by different published prognostic signatures showed that they share 53% of the identified pathways. We show that divergent gene sets classifying patients for the same clinical endpoint represent similar biological processes and that pathway-derived signatures can be used to predict prognosis. Furthermore, our study reveals that the underlying biology related to aggressiveness of estrogen receptor subgroups of breast cancer is quite different.