Circulating tumor cells and circulating nucleic acids in oncology

Strategies for circulating tumor cell enrichment

CTCs can be infrequent, depending on the cancer stage. One CTC may be present among 10 ⁶ to 10 ⁸ peripheral blood cells. CTC isolation and enrichment from peripheral blood are thus extremely challenging and very demanding. It is not only that these cells circulate at very low numbers, but they are heterogeneous even within the same patient. Highly standardized and robust isolation protocols are necessary for downstream CTC analysis and molecular characterization. Toward this goal, a lot of effort has focused on developing novel technologies for the isolation and enrichment of CTCs from peripheral blood.

A large spectrum of technologies is available to enrich CTCs from surrounding normal hematopoietic cells (see Fig. 71.6 ). These enrichment methods rely on different properties of CTCs that can be positively or negatively enriched on (i) label-dependent systems based on biological properties (e.g., surface protein expression) and (ii) label-independent systems based on physical properties (e.g., size, density, electric charges, and deformability). In vivo methods for increasing CTC yield have also been developed. Moreover, CTC recovery from blood samples collected close to the tumor (i.e., in vessels located in the drainage area of the primary tumor or metastases) can considerably increase the chance to collect more CTCs ( Fig. 71.7 ). ^, In colorectal cancer (CRC), the comparative analysis of CTCs obtained from the mesenteric and peripheral blood demonstrated a cascade of genomic events related to metastatic progression ( Fig. 71.8 ).

The most important systems for the isolation/enrichment of CTCs are discussed in the following text. In many cases, these technologies are complementary to each other because they target different properties of CTCs and may define different CTC populations.

Positive selection.

Positive selection is the most widely used CTC isolation/enrichment system. This approach captures CTCs through specific monoclonal antibodies against epithelial cell surface markers that are expressed on CTCs but are absent from normal leukocytes. CTCs can be tagged using antibody-conjugated magnetic microbeads (diameter: 0.5 to 5 μm) or nanoparticles (diameter: 50 to 250 nm) that bind to a specific surface antigen. Intracellular antigens like cytokeratins can also be used as targets. Immunomagnetic assays require a short incubation (∼30 minutes) for antigen/antibody binding that couples the cells to magnetic beads, followed by isolation of the cells using a magnetic field.

Various antigens have been exploited for the positive immunomagnetic isolation of CTCs. Among these, EpCAM is the most common. This approach is well established in terms of proven clinical significance of the captured cells. However, capture with EpCAM has the disadvantage of missing some cells that are undergoing epithelial-mesenchymal transition (EMT; Fig. 71.9 ). We also now know that CTCs are highly heterogeneous, but one approach to partially overcome this issue is to use a “cocktail” of antibodies that targets multiple antigens. Along these lines, several organ- or tumor-specific markers, such as carcinoembryonic antigen (CEA), EGFR, prostate-specific antigen, HER-2, cell surface–associated mucin 1 (MUC-1), ephrin receptor B4 (EphB4), insulin-like growth factor 1 receptor (IGF-1R), cadherin-11 (CAD11), and tumor-associated glycoprotein 72 (TAG-72), have been used to isolate CTCs. The expression of vimentin by CTCs may also be a good marker for epithelial cancers undergoing EMT. By using a specific monoclonal antibody against vimentin, EMT-CTCs were detected in patients undergoing postsurgery adjuvant chemotherapy for metastatic colon cancer. These isolated EMT-CTCs were characterized further using EMT-specific markers, fluorescent in situ hybridization, and single-cell mutation analysis. This antibody exhibited high specificity and sensitivity toward different epithelial cancer cells and was used to detect and enumerate EMT-CTCs from patients. The number of EMT-CTCs detected correlated with the therapeutic outcome of the disease. According to these results, cell surface vimentin is a promising marker for the isolation of EMT-CTCs from a wide variety of tumor types.

Another example of positive selection is the MagSweeper, an immunomagnetic cell separator. In this device, magnetic beads are coated with an antibody targeting epithelial cell surface markers. These immunomagnetic beads are added into blood samples, and cancer cells are attached to the beads. This device gently enriches target cells and eliminates cells that are not bound to magnetic particles by using centrifugal forces. The isolated cells are easily accessible and can be extracted individually based on their physical characteristics to deplete any cells nonspecifically bound to beads. The same group has recently developed the magnetic sifter, a miniature microfluidic chip with a dense array of magnetic pores; tumor cells are labeled with magnetic nanoparticles and are captured from whole blood with high efficiency. The use of isolation technologies that take advantage of magnetic fields may lead the way to routine preparation and characterization of liquid biopsies from cancer patients.

Negative selection.

This isolation approach is completely independent of the phenotype of CTCs and is based on the depletion of noncancerous peripheral blood cells. A first step is often the lysis of red blood cells (RBCs) and the second step may use specific markers for white blood cells (WBCs) like CD45 or CD61 to magnetically remove them from the sample. ^, Another variation of CTC enrichment by negative depletion is the commercially available RosetteSep system (StemCell technologies, Canada): tetrameric antibody complexes target unwanted cells for removal. This approach is independent of the expression of EpCAM on CTCs and has better recoveries than the conventional density gradient method for the isolation of CTCs from normal leukocytes in blood.

Analysis of high blood volumes to increase circulating tumor cell capture rate

To analyze a high volume of blood, an alternative to in vivo CTC detection is analysis by leukapheresis. Indeed, leukapheresis is a laboratory procedure in which WBCs or peripheral blood stem cells are separated from blood. During leukapheresis, a patient’s blood is passed through a machine that removes the WBCs or peripheral blood stem cells and then returns the balance of the blood back to the patient. Eifler and colleagues first showed that isolation of CTCs via leukapheresis was feasible. Screening leukapheresis products generated from up to 2.5 L of processed blood per patient demonstrated that CTCs can be detected in more than 90% of nonmetastatic breast cancer patients. Label-free enrichment and molecular characterization of viable CTCs can be performed from diagnostic leukapheresis (DLA) products, and DLA has enabled transcriptomic profiling of single CTCs demonstrating intercellular heterogeneity of endocrine resistance in estrogen receptor (ER)-positive breast cancer.

Strategies for circulating tumor cell detection

After the enrichment step, the samples still contain a substantial number of contaminating leukocytes; thus CTCs need to be specifically identified at the single-cell level by a robust reproducible method that can distinguish them from normal blood cells. CTC detection is achieved by (i) immunofluorescence, (ii) molecular nucleic acid analyses (multiplex RT-qPCR), and (iii) functional assays (EPISPOT and EPIDROP). An outline of the main approaches for CTC detection is presented in Fig. 71.6 .

Immunologic technologies

Immunologic technologies are the most frequent methods employed for CTC detection, and they use a combination of membrane and/or intracytoplasmic anti-epithelial, anti-mesenchymal, and anti–tissue-specific markers and anti–tumor-associated antibodies. Over the past 20 years, keratins as constituents of the cytoskeleton of epithelial cells have been established as detection markers for CTCs in patients with various kinds of carcinomas. Many CTC assays follow principles that are similar to the FDA-cleared CellSearch system, which has been the “gold standard” for many years and will be therefore discussed in more detail. After EpCAM–based immunomagnetic enrichment of CTCs, cells are fluorescently stained for epithelial keratins (CK8, 18, and 19) as markers for CTCs, the common leukocyte antigen CD45 as an exclusion marker, and a nuclear dye (4′,6-diamidino-2-phenylindole [DAPI]) to access cellular integrity ( Fig. 71.10 ), and suspicious events are listed in a photo gallery using automated digital microscopy. The main advantages of the CellSearch system are its high reproducibility and robustness including defined preanalytical steps of blood sampling that ensure stability for 96 hours at room temperature. ^, Most importantly, there is a wealth of data from large-scale clinical studies demonstrating the correlation of CellSearch-based CTC counts with clinical outcome in breast cancer and may other solid tumors. However, limitations to this system are that the enrichment and detection steps depend entirely on the expression of the epithelial markers EpCAM and keratins. Thus CTCs that have undergone a complete EMT and do not express EpCAM and keratins (see below) cannot be detected. Therefore there is an ongoing search for additional markers that detect EMT CTCs and are not expressed on contaminating blood cells to ensure both high sensitivity and specificity of CTC detection.

Strategies for circulating tumor cell characterization

Serial CTC measurement can follow the evolution of tumor subclones during treatment and disease progression and hold the key to understand the biology of the metastatic cascade. Improvements in technologies to yield purer CTC populations improve cellular and molecular investigations. Characterization of single CTCs allows better insight into tumor heterogeneity within assays, including immunofluorescence, array comparative genomic hybridization (CGH), massively parallel sequencing (MPS) of both DNA and RNA, and fluorescence in situ hybridization (FISH).

Immunologic detection and characterization allow isolation of stained CTCs for subsequent molecular characterization. While manual isolation by micromanipulation of CTCs is possible, it is rather arduous and time-consuming. An alternative automated single-cell selection device is the DEPArray, which is based on DEP that traps single cells in DEP cages and is designed for single-cell recovery of CTCs. Multiple clinical studies have used this technology to detect and isolate single CTCs for subsequent genetic analyses.

Another approach is FISH analysis of single CTCs identified by immunocytochemistry. Recently, padlock probe technology, which enables in situ analysis of AR-V7 in CTCs, showed that 71% (22 of 31) of castration-resistant prostate cancer (CRPC) patients had detectable AR-V7 expression ranging from low to high expression. Patients with AR-V7-positive CTCs respond better to taxane-based chemotherapy than novel hormonal therapies, indicating a treatment-selection biomarker.

Circulating tumor cells as a “window to metastasis”

In 1889, in the very first issue of Lancet , Steve Paget described “the seed and soil hypothesis,” in which “metastasis depends on the cross talk between selected cancer cells (the seed) and specific organ microenvironments (the soil),” a hypothesis revisited many years later by Fidler. Detailing the mechanism of metastasis remains a very hot topic in cancer research today ( Fig. 71.13 ). ^, Analysis of disease course, tumor growth rates, autopsy studies, clinical trials, and molecular genetic analyses of primary and DTCs all contribute to our understanding of systemic cancer. ^, Molecular characterization of CTCs provides a level of detail not previously possible and may be the key to our further understanding of metastatic progression.

CTC biology can be viewed as a “window to metastasis” because CTCs play a critical role in the metastatic spread of carcinomas (see Fig. 71.13 ). ^, If CTCs are effectively targeted or kept in a dormant state, the cancer may be prevented from progressing to metastatic disease. Molecular characterization of CTCs from patients may be the shortest path to determine which and when patients might relapse and to identify specific mechanisms to target these cells. Dormancy gene signatures that identify individuals with dormant disease have also been explored. ^, In contrast, CTCs with stemness and EMT features display enhanced malignant and metastatic potential. The role of CTCs in treatment failure and disease progression is likely explained by their biological processes, such as EMT, stemness features, dormancy, and heterogeneity ( Fig. 71.14 ).

Mentioning the evolution of dispersal, it is of utmost interest to evaluate whether similar ecologic and evolutionary principles can be applied to metastasis, and how these processes may shape the spatiotemporal dynamics of CTCs. Indeed, CTCs have themselves been observed to disperse alone and in groups of up to around 100 cells, called clusters or microemboli, and cancer cells reproduce asexually, so it is uncertain which of the aforementioned evolutionary outcomes is the most likely to explain a role of kin selection in metastasis ( Fig. 71.15 ).

Epithelial and mesenchymal transitions of circulating tumor cells

EMT is an essential process in the metastatic cascade. This biological process is highly associated with an invasive phenotype and enables detachment of tumor cells from the primary site and migration ( Fig. 71.16 ). The reverse process of mesenchymal epithelial transition (MET) might play a crucial role in the further steps of metastasis when CTCs seed distant organs and establish metastasis. The mechanisms and the interplay of EMT and MET have been intensively studied, and current data suggest the existence of the EMT process in CTCs. ^, It is now clear that CTCs from MBC patients exhibit heterogeneous epithelial and mesenchymal phenotypes and display higher frequencies of partial or full-blown mesenchymal phenotype than carcinoma cells within primary tumors. Mesenchymal-like CTCs are also elevated in patients who are refractory to therapy.

Currently, most systems that detect CTCs, including the CellSearch system, are based on the expression of the epithelial marker EpCAM and do not specifically identify CTC subtypes with EMT. Over the past years, several EMT-related markers have been applied in CTC studies. Three EMT markers ( TWIST1, AKT2, and PI3K α) and the stem cell marker ALDH1 were evaluated in CTCs from 502 primary breast cancer patients by a multiplex RT-PCR assay. A subset of CTCs showed EMT and stem cell characteristics. The expression levels of EMT-inducing transcription factors ( TWIST1, SNAIL1, SLUG, ZEB1, and FOXC2 ) were also determined in CTCs from primary breast cancer patients. In another study, rare primary tumor cells simultaneously expressed mesenchymal and epithelial markers, but mesenchymal cells were highly enriched in CTCs, and serial monitoring suggested an association of mesenchymal CTCs with disease progression. Mesenchymal CTCs occurred as both single cells and multicellular clusters, expressed known EMT regulators, including transforming growth factor (TGF)-β pathway components and the FOXC1 transcription factor. These data support a role for EMT in the blood-borne dissemination of human breast cancer. When the EMT phenotype of CTCs was studied through the expression of two important EMT-connected genes—namely, VIM and SNAIL —using cytokeratin-negative CTCs from nonmetastatic breast cancer patients, the simultaneous detection of both EGFR and EMT markers may improve prognostic or predictive information. A differential expression pattern of ALDH1 (a stemness marker) and TWIST (an EMT marker) on single CTCs was observed both in early and MBC. CTCs expressing high ALDH1 along with nuclear TWIST were more frequently found in patients with MBC, suggesting that CTCs undergoing EMT may prevail during disease progression.

Counter to expectation, Gorges and colleagues reported low CTC numbers in some patients with late metastatic cancers. These results prompted the search for new markers, including those for mesenchymal-like subpopulations. Plastin-3 is an EMT marker in CTCs in CRC. Aberrant expression of plastin-3 was associated with increased CTCs and poor prognosis in CRC and may be involved in the regulation of EMT. Cell-surface vimentin is specifically expressed on the surface of CTCs from epithelial cancers undergoing EMT, and the number of CTCs undergoing EMT correlates with disease outcome.

Cytokeratins are widely used for the identification of CTCs by immunocytochemistry, but even these established markers might be modulated during EMT. Breast cancer cells display a complex pattern of cytokeratin expression with potential biological relevance. Individual cytokeratin antibodies may recognize only certain cytokeratins, and important subsets of biologically relevant CTCs in cancer patients may be missed.

EMT and MET transitions are central to the metastatic potential of CTCs; the elucidation of CTC plasticity on the clinical outcome of cancer patients may shed more light on this important topic. Thus far, a large number of prognostic studies in patients with breast cancer and other solid tumors and recent experimental studies on CTCs ^, indicate that CTCs with epithelial attributes are relevant for the onset of metastasis. Thus one might postulate that EMT is important for the release of CTCs from tumor tissues, while MET and the resulting epithelial phenotype of CTCs might allow them to colonize distant organs. ^, However, the most important feature might be the plasticity of CTCs to change their phenotype depending on the actual requirements of their changing microenvironments (i.e., blood or diverse organ sites such as bone, liver, or brain). This plasticity is not easily assessable by snapshot analysis but requires longitudinal analysis or experimental models such as cell lines or xenografts established from CTCs. ^{, ,}