Circulating Tumor Cell Enrichment and Single-Cell Isolation Combining the CellSearch® and DEPArray™ Systems

The analysis of circulating tumor cells (CTCs) has shown potential for detection of cancer spread, prognosis, therapeutic target selection, and monitoring of treatment response. CTCs can be obtained repeatedly by simple blood draws as so-called "liquid biopsy." Thus, they can serve as a surrogate material for primary or metastatic tissue biopsies. In addition, isolation of CTCs provides the possibility to investigate those cells which may hold the (molecular) traits responsible for metastatic progression and ultimately patient death. As such, CTCs represent a target of utmost importance in cancer research and therapy. In this chapter, we describe a workflow for the enrichment of CTCs with the FDA-cleared CellSearch® system followed by the isolation of single CTCs using the DEPArray™ technology enabling further molecular single-cell analyses.

Ex vivo expansion of lung cancer-derived disseminated cancer cells from lymph nodes identifies cells associated with metastatic progression

The cellular basis of the apparent aggressiveness in lung cancer is poorly understood but likely associated with functional or molecular features of disseminated cancer cells (DCCs). DCCs from epithelial cancers are mostly detected by antibodies directed against histogenetic markers such as cytokeratin or EpCAM. It has been argued that marker-negative metastatic founder cells might escape detection. We therefore used ex vivo sphere formation for functional detection of candidate metastasis founders. We generated cell suspensions from 199 LN samples of 131 lung cancer patients and placed them into non-adherent cell culture. Sphere formation was associated with detection of DCCs using EpCAM immunocytology and with significantly poorer prognosis. The prognostic impact of sphere formation was strongly associated with high numbers of EpCAM-positive DCCs and aberrant genotypes of expanded spheres. We also noted sphere formation in patients with no evidence of lymphatic spread, however such spheres showed infrequent expression of signature genes associated with spheres from EpCAM-positive samples and displayed neither typical lung cancer mutations (KRAS, TP53, ERBB1) nor copy number variations, but might be linked to disease progression >5 years post curative surgery. We conclude that EpCAM identifies relevant disease-driving DCCs, that such cells can be expanded for model generation and that further research is needed to clarify the functional and prognostic role of rare EpCAM-negative sphere forming cells.

Dynamics of CXCR4 positive circulating tumor cells in prostate cancer patients during radiotherapy

Ablative radiotherapy is a highly efficient treatment modality for patients with metastatic prostate cancer (PCa). However, a subset of patients does not respond. Currently, this subgroup with bad prognosis cannot be identified before disease progression. We hypothesize that markers indicative of radioresistance, stemness and/or bone tropism may have a prognostic potential to identify patients profiting from metastases-directed radiotherapy. Therefore, circulating tumor cells (CTCs) were analyzed in patients with metastatic PCa (n = 24) during radiotherapy with CellSearch, multicolor flow cytometry and imaging cytometry. Analysis of copy-number alteration indicates a polyclonal CTC population that changes after radiotherapy. CTCs were found in 8 out of 24 patients (33.3%) and were associated with a shorter time to biochemical progression after radiotherapy. Whereas the total CTC count dropped after radiotherapy, a chemokine receptor CXCR4-expressing subpopulation representing 28.6% of the total CTC population remained stable up to 3 months. At once, we observed higher chemokine CCL2 plasma concentrations and proinflammatory monocytes. Additional functional analyses demonstrated key roles of CXCR4 and CCL2 for cellular radiosensitivity, tumorigenicity and stem-like potential in vitro and in vivo. Moreover, a high CXCR4 and CCL2 expression was found in bone metastasis biopsies of PCa patients. In summary, panCK⁺ CXCR4⁺ CTCs may have a prognostic potential in patients with metastatic PCa treated with metastasis-directed radiotherapy.

Abstract PD9-07: Mdm2 gene amplification in estrogen receptor-positive breast cancer cells is associated with enhanced solid tumor growth and pronounced metastatic potential in humanized tumor mice (HTM) and a poor outcome of patients with luminal breast cancer

Background: Luminal, i.e., estrogen receptor (ER)-positive breast cancer (BC) is a heterogeneous disease in terms of tumor progression, therapy response, and relapse. Additional biomarkers with a prognostic and predictive impact would facilitate advanced patient stratification and can reveal advanced therapeutic options for individual patients suffering from luminal BC subtype. First objective of this study was the hypothesis driven validation and identification of biomarkers associated with successful engraftment, augmented tumor growth, and enhanced metastasis upon xenotransplantation into HTM and non-humanized tumor mice. Second objective was the retrospective validation and correlation of aforementioned markers with clinical outcomes (disease free-survival [DFS] and overall survival [OS]) of luminal BC patients after neoadjuvant chemotherapy within the GeparTrio trial. Methods: Immunodeficient NSG mice with and without immunological humanization were transplanted with primary, ER-positive BC tissues and cells. Engraftment rates, tumor progression, and metastasis were monitored as a function of marker expression and genomic alterations, considered to be associated with tumor cell stemness and tumor initiating capacity. Functional assays were applied to demonstrate the impact of identified markers on tumor cell viability, growth and migration in-vitro. Dual color fluorescence-in-situ-hybridization to monitor mdm2 gene and the cen12 region was applied to 502 pretherapeutic ER-positive tissue specimens of BC patients treated with anthracycline/taxane based neoadjuvant chemotherapy within the GeparTrio trial. Mdm2 gene expression was analyzed in the entire luminal BC cohort as well as in luminal-A and luminal-B samples classified based on a Ki-67 cut-off of 20% (St. Gallen guideline). Associations with survival outcomes were studied by Cox regression models. Results: We observed an elevated CD44 and c-MET expression in metastatic cells compared to the primarily growing xenotransplantants. Moreover, we found mdm2 gene amplification was associated with tumor growth and pronounced metastatic potential in NSG mice. Functional assays unveiled a reduced viability, proliferation, and migration capacity of inherently mdm2 positive breast cancer cells upon mdm2 knock-down or anti-mdm2 targeting. Validation of mdm2 gain in luminal BC cohort within the GeparTrio trial revealed a significant association of mdm2 amplification with worse DFS (HR=1.80 [95%CI 1.16-2.79], log rank p=0.008) and OS (HR=1.75 [95%CI 1.00-3.05], log-rank p=0.047). This association was even stronger in luminal-A BC: DFS (HR=2.56 [95%CI 1.40-4.71], log rank p=0.002 and OS (HR=3.27 [95%CI 1.51-7.09], log-rank p=0.001). Conclusions: Mdm2 gene amplification facilitates ER-positive BC engraftment and progression in a preclinical in-vivo xenograft humanized NSG mouse model. Targeting mdm2 in-vitro reduced malignant cell propagation and growth. In addition, an increased mdm2 gene dose is strongly associated with an unfavorable outcome of luminal BC. Prospective studies are required to verify the suitability of mdm2 for advanced luminal BC stratification and therapeutic targeting.

Citation Format: Anja Kathrin Wege, Valentina Vladimirova, Christine Solbach, Eva-Maria Rom-Jurek, Jens-Uwe Blohmer, Paul Jank, Bruno Sinn, Andreas Trumpp, Elisabetta Marangoni, Knut Engels, Wilko Weichert, Nicole Pfarr, Christoph Irlbeck, Bernhard Polzer, Olaf Ortmann, Marion van Mackelenbergh, Carsten Denkert, Sibylle Loibl, Gero Brockhoff. Mdm2 gene amplification in estrogen receptor-positive breast cancer cells is associated with enhanced solid tumor growth and pronounced metastatic potential in humanized tumor mice (HTM) and a poor outcome of patients with luminal breast cancer [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr PD9-07.

ViBiBa: Virtual BioBanking for the DETECT multicenter trial program - decentralized storage and processing

•

ViBiBa is an open-source sample banking tool.

•

ViBiBa was purpose built for liquid biopsy specimen.

•

ViBiBa allows for decentralized storage while promoting collaboration.

•

ViBiBa's plugin support requires no change in existing data structures.

•

ViBiBa empowers translational research projects and cohort formation.

Background

Liquid Biopsy (LB) in the form of e.g., circulating tumor cells (CTCs) is a promising non-invasive approach to support current therapeutic cancer management. However, the proof of clinical utility of CTCs in informing therapeutic decision-making for e.g., breast cancer in clinical trials and associated translational research projects is facing the issues of low CTC positivity rates and low CTC numbers – even in the metastasized situation. To compensate for this dilemma, clinical CTC trials are designed as large multicenter endeavors with decentralized sample collection, processing and storage of products, making data management highly important to enable high-quality translational CTC research.

Aim

In the DETECT clinical CTC trials we aimed at developing a custom-made, browser-based virtual database to harmonize and organize both decentralized processing and storage of LB specimens and to enable the collection of clinically meaningful LB sample.

Methods

ViBiBa processes data from various sources, harmonizes the data and creates an easily searchable multilayered database.

Results

An open-source virtual bio-banking web-application termed ViBiBa was created, which automatically processes data from multiple non-standardized sources. These data are automatically checked and merged into one centralized databank and are providing the opportunity to extract clinically relevant patient cohorts and CTC sample collections.

Summary

ViBiBa, which is a highly flexible tool that allows for decentralized sample storage of liquid biopsy specimens, facilitates a solution which promotes collaboration in a user-friendly, federalist and highly structured way. The source code is available under the MIT license from https://vibiba.com or https://github.com/asperciesl/ViBiBa

Detection of Genomically Aberrant Cells within Circulating Tumor Microemboli (CTMs) Isolated from Early-Stage Breast Cancer Patients

Simple Summary

Distant metastases derive from the shedding and dissemination of single cancer cells (CTCs) or circulating tumor emboli (CTMs) into circulation. Previous studies on CTMs were mainly run in patients with metastatic disease; however, we observed that CTMs are more frequently detected in patients with early-stage breast cancer. Here, we collected single CTMs and their relative primary tumor tissue samples in early-stage patients. By studying genomic aberrations, present in tumors cells and absent in normal cells, we predicted the tumor fraction thanks to a statistical model developed from a calibration curve with breast cancer cell lines. The tumor fraction ranged from 8% to 48% and CTMs contained specific and shared alterations with respect to tissue. Thus, CTMs may derive from different regions of the primary tumor or from occult micrometastases. Moreover, CTM-private mutations may inform us about specific metastasis-associated functions of involved genes that should be further explored in follow-up and mechanistic studies.

Abstract

Circulating tumor microemboli (CTMs) are clusters of cancer cells detached from solid tumors, whose study can reveal mechanisms underlying metastatization. As they frequently comprise unknown fractions of leukocytes, the analysis of copy number alterations (CNAs) is challenging. To address this, we titrated known numbers of leukocytes into cancer cells (MDA-MB-453 and MDA-MB-36, displaying high and low DNA content, respectively) generating tumor fractions from 0–100%. After low-pass sequencing, ichorCNA was identified as the best algorithm to build a linear mixed regression model for tumor fraction (TF) prediction. We then isolated 53 CTMs from blood samples of six early-stage breast cancer patients and predicted the TF of all clusters. We found that all clusters harbor cancer cells between 8 and 48%. Furthermore, by comparing the identified CNAs of CTMs with their matched primary tumors, we noted that only 31–71% of aberrations were shared. Surprisingly, CTM-private alterations were abundant (30–63%), whereas primary tumor-private alterations were rare (4–12%). This either indicates that CTMs are disseminated from further progressed regions of the primary tumor or stem from cancer cells already colonizing distant sites. In both cases, CTM-private mutations may inform us about specific metastasis-associated functions of involved genes that should be explored in follow-up and mechanistic studies.

Abstract LB-239: A novel workflow for isolation and multi-omic profiling of DCCs derived from cerebrospinal fluid of patients with pediatric brain cancer

Background: Clinical management of cancers of the central nevus system (CNS) is very challenging as they often exhibit low responsiveness to radiation and chemotherapy resulting in overall poor survival. Moreover, analysis of mechanisms driving cancer progression and selection of targeted therapies in CNS tumors is hindered by the limited availability to tumor tissues accessible only though surgical biopsies. A potential source of cancer material in patients with CNS is cerebrospinal fluid (CSF). Analysis of CSF-derived disseminated cancer cells (csfDCCs) holds a promise for improvement of diagnostics and monitoring of CNS tumors. For this reason, we developed a novel workflow allowing detection, isolation and multi-omic analysis of csfDCCs.

Methods: In a proof of concept study a new workflow was used to analyze CSF samples from two patients with medulloblastoma and pineoblastoma. CSF-derived cells were stained for CD45 to allow identification of infiltrating immune cells. Putative csfDCCs (CD45-negative) and control cells (CD45-positive) were subjected to a multi-omic workflow allowing parallel sequencing of genomes and transcriptomes of the same cells. Single-cell mRNA was physically separated from DNA, amplified by means of whole transcriptome amplification (Ampli1 WTA) and analyzed using endpoint PCR and a proprietary single-cell RNA-Seq approach. In parallel, DNA was subjected to whole genome amplification (Ampli1 WGA) and analyzed for the presence of copy number variations as well as point mutations (Ampli1 LowPass Kit and targeted sequencing of actionable hot-spots).

Results: We analyzed seven CNS-derived single cells and five cell clusters from the medulloblastoma patient and further eight cell clusters and eight single cells obtained from the pineoblastoma patient. Transcriptome analysis revealed that expression of neural lineage markers (e.g. CD133, SYP, OTX2, MSI1, MAP2, NEUROG1 and NEUROD1) is present almost exclusively in CD45-negative cells. Only two samples collected from medulloblastoma patient co-expressed CD45 and GFAP. However, DNA analysis revealed that CD45-positive and CD45/GFAP double-positive samples showed non-aberrant genomic profiles, thus these cells were classified as non-malignant. In contrast, all CD45-negative cells harbored genetic alterations confirming their malignant origin.

Conclusion: Our proof of concept study shows a novel workflow allowing identification, isolation as well as parallel genome and transcriptome analysis csfDCCs of patients with pediatric CNS tumors.

Citation Format: Zbigniew T. Czyz, Giancarlo Feliciello, Miodrag Guzvic, Thomas Schamberger, Sandra Grunewald, Selim Corbacioglu, Markus J. Riemenschneider, Bernhard Polzer, Christoph A. Klein. A novel workflow for isolation and multi-omic profiling of DCCs derived from cerebrospinal fluid of patients with pediatric brain cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr LB-239.

Abstract 1625: Challenges of screening applications for early detection of cancer using liquid biopsy

In the recent years, early detection of cancer using next-generation sequencing (NGS) has been en vogue, due to its potential to drastically improve the survival rate of cancer patients. Liquid biopsy is often viewed as the modus operandi for early detection, as taking biopsies from the actual organs is not practical for widespread screening tests. Among others, two main approaches attempted for early detection are based on cell-free tumor DNA (ctDNA) and circulating tumor cells (CTCs). The rarity of ctDNA and CTC is often debated as a major potential challenge for detection of cancer at early stages. In this work, we will discuss several other challenges that are necessary to address for developing a practical and high-quality screening application for early-stage detection of cancer. We will also provide a framework in which much of these challenges can be addressed. While a reasonable sensitivity is needed for such test to be of value, the most prominent matter regarding screening for cancer is specificity. Due to low prevalence of cancer, what may appear as a high specificity (e.g., 99%) could actually be insufficient, as it translates to a low positive predictive value (PPV) (e.g., in the 10% range). A low PPV would result in a large number of false positives in the test. These false positives would cause follow-up tests that are not only costly, but also could trigger unnecessary anxiety for the patients. A comprehensive validation process is another critical element in early detection. While it is tempting to perform validation based, solely, on groups of Healthy and Cancer patients, it can be shown that some signals caused by Benign conditions could highly resemble those of Cancer patients, and therefore result in false positives. We will show that distinguishing Benign from Cancer necessitates detection of more complex patterns, as opposed to simplistic measures such as genome-wide copy number variation (CNV). Yet another challenge in a liquid biopsy test is identification of the tissue-of-origin. Due to the mixed-signal problem inherent in ctDNA-based technologies, we argue that they face a higher challenge, as compared to CTC-based counterparts, for obtaining this goal. In conclusion, we will show how a system based on CTCs, whole genome sequencing (WGS), and artificial intelligence (AI) could provide an appropriate framework for addressing much of the challenges in early detection of cancer for a true screening application with the specificity in the 99.99% range.

Citation Format: Bahram G. Kermani, Sarah Huecker, Stefan Kirsch, Bernhard Polzer. Challenges of screening applications for early detection of cancer using liquid biopsy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 1625.

Abstract LB-312: Interleukin 6 transsignaling is a candidate mechanism to drive progression of human DCCs during periods of clinical latency

While thousands of breast cancer cells disseminate and home to bone marrow (BM) until primary surgery, usually less than a handful will succeed in establishing manifest metastases months to years later. Signals and mechanisms determining failure or success of disseminated cancer cells (DCCs) are largely unknown and there is no in vivo model available to study the spontaneous progression and genomic evolution from early bone marrow infiltration to manifestation of bone metastasis, as spontaneous or transgenic mouse models do not generate bone metastases. We therefore profiled DCCs from BM of breast cancer patients long before manifestation of metastasis by RNAseq to identify signals supporting survival or outgrowth of DCCs and identified IL6/PTEN/PI3K signaling as candidate pathway for DCC activation. Since early DCCs often display close-to-normal genomes we used mammary epithelial cells ex vivo isolated from reduction mammoplasties and immortalized pre-malignant breast cancer cell lines as model for functional testing in vitro. Using specific activators and inhibitors of IL6 signaling revealed that IL6 trans, but not classical signaling, regulates stemness of mammary epithelial cells. Moreover, knock-down of PTEN revealed that PI3K/PTEN pathway activation renders cells independent of IL6 trans-signaling. Interestingly, gp130 expression, a pre-requisite for IL6 trans-signaling was found to be down-regulated by bone marrow stromal and endosteal, but not vascular niche cells, and as a consequence the number of cells with stem-like ability was significantly reduced. Consistent with a bottleneck function of microenvironmental DCC control, we found PIK3CA mutations highly associated with late-stage metastatic DCCs and CTCs while generally absent in early DCCs. Our data suggest that the initial steps of metastasis formation depend on microenvironmental signals and are not cancer cell-autonomous.

Citation Format: Melanie Werner-Klein, Ana Grujovic, Milan Obradovic, Martin Hoffmann, Xin Lu, Stefan Kirsch, Steffi Treitschke, Cäcilia Köstler, Kathrin Weidele, Christoph Irlbeck, Catherine Botteron, Christian Werno, Bernhard Polzer, Miodrag Guzvic, Stefan Buchholz, Petra Rümmele, Norbert Heine, Stefan Rose-John, Christoph A. Klein. Interleukin 6 transsignaling is a candidate mechanism to drive progression of human DCCs during periods of clinical latency [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr LB-312.

Abstract 431: Semi-automated detection, isolation and molecular analysis of single disseminated melanoma cells from lymph nodes

We recently showed that lymph node disaggregation followed by immunocytology enables precise quantification of disseminated cancer cells (DCCs) and demonstrated that this approach has a 20-fold higher sensitivity to detect melanoma DCCs than routine histopathology (Ulmer et al., PLoS Medicine, 2014). Moreover, genetic profiling of single melanoma DCCs identified a colonization signature consisting of specific copy number alterations and point mutations that identify patients with high risk of progression (Werner-Klein et al., Nature Communications 2018). Here, we present the adaptation of this method to a semi-automated workflow for detection, isolation and molecular analysis of single melanoma DCCs. The developed workflow includes a mechanical disaggregation of lymph node tissue and collection of the mononuclear cells, immunofluorescence staining against melanoma-associated markers gp100 and MCSP and depletion of CD45-positive cells. Individual melanoma cells are then detected and isolated by DEPArrayTM technology enabling single cell whole genome amplification (Ampli1TM) for subsequent molecular analysis. In total, we processed 20 lymph nodes of melanoma patients and detected melanoma DCCs in 11/20 samples (55%). The quality of isolated cells was checked by Ampli1TM QC and 174 isolated single cells were further analyzed by Sanger sequencing for specific point mutations (BRAF, NRAS and cKIT) and Ampli1TM low pass kit for Illumina for copy number variation (CNV). Successful molecular analysis was correlated with genome integrity score (GII) as determined by Ampli1TM QC, with more than 95% of cells with GII 3-4 showing good performance in low pass sequencing. In conclusion, a new DepArrayTM based application for marker-dependent single cell isolation from malignant melanoma lymph nodes was successfully established and tested on a cohort of 20 melanoma patients. Molecular analyses of isolated single cells confirmed the tumor origin by CNV profiling and mutational analysis of melanoma-associated mutations. In the future, this approach could help to select individualized therapies for melanoma patients.

Citation Format: Barbara Alberter, Sebastian Scheitler, Giancarlo Feliciello, Alberto Ferrarini, Melanie Werner-Klein, Sebastian Haferkamp, Christoph A. Klein, Bernhard Polzer. Semi-automated detection, isolation and molecular analysis of single disseminated melanoma cells from lymph nodes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 431.

Microfluidic enrichment, isolation and characterization of disseminated melanoma cells from lymph node samples

For the first time in melanoma, novel therapies have recently shown efficacy in the adjuvant therapy setting, which makes companion diagnostics to guide treatment decisions a desideratum. Early spread of disseminated cancer cells (DCC) to sentinel lymph nodes (SLN) is indicative of poor prognosis in melanoma and early DCCs could therefore provide important information about the malignant seed. Here, we present a strategy for enrichment of DCCs from SLN suspensions using a microfluidic device (Parsortix™, Angle plc). This approach enables the detection and isolation of viable DCCs, followed by molecular analysis and identification of genetic changes. By optimizing the workflow, the established protocol allows a high recovery of DCC from melanoma patient-derived lymph node (LN) suspensions with harvest rates above 60%. We then assessed the integrity of the transcriptome and genome of individual, isolated DCCs. In LNs of melanoma patients, we detected the expression of melanoma-associated transcripts including MLANA (encoding for MelanA protein), analyzed the BRAF and NRAS mutational status and confirmed the malignant origin of isolated melanoma DCCs by comparative genomic hybridization. We demonstrate the feasibility of epitope-independent isolation of LN DCCs using Parsortix™ for subsequent molecular characterization of isolated single DCCs with ample application fields including the use for companion diagnostics or subsequent cellular studies in personalized medicine.

Isolated metastasis of an EGFR-L858R-mutated NSCLC of the meninges: the potential impact of CXCL12/CXCR4 axis in EGFRmut NSCLC in diagnosis, follow-up and treatment

Brain and leptomeningeal metastasis (LMM) of non-small cell lung cancer is still associated with poor prognosis. Moreover, the current diagnostic standard for LMM often yields false negative results and the scientific progress in this field is still unsatisfying. We present a case of a 71-year old patient with an isolated LMM. While standard diagnostics could only diagnose a cancer of unknown primary, the use of [⁶⁸Ga]-Pentixafor-PET/CT (CXCR4-PET/CT, a radiotracer targeting CXCR4) and a liquid biopsy of the cerebrospinal fluid revealed the primary NSCLC. The detection of L858R-EGFR, a common driver mutation in NSCLC, enabled us to treat the patient with Afatinib and monitor treatment using [⁶⁸Ga]-Pentixafor PET/CT. To estimate the impact of CXCR4 signaling and its ligands in NSCLC brain metastasis we looked at their expression and correlation with EGFR mutations in a primary and brain metastasis data set and investigated the previously described binding of extracellular ubiquitin to CXCR4. In conclusion, we describe a novel approach to improve diagnostics towards LMM and underline the impact of the CXCL12/CXCR4 axis in brain metastasis in a subset of NSCLC patients. We cannot confirm a correlation of CXCR4 expression with EGFR mutations or the binding of extracellular ubiquitin as previously reported.

Diagnostic pathology of early systemic cancer: ERBB2 gene amplification in single disseminated cancer cells determines patient survival in operable esophageal cancer

Early metastatic dissemination and evolution of disseminated cancer cells (DCCs) outside the primary tumor is one reason for the failure of adjuvant therapies because it generates molecular genotypes and phenotypes different from primary tumors, which still underlie therapy decisions. Since ERBB2 amplification in esophageal DCCs but not in primary tumor cells predict outcome, we aimed to establish an assay with diagnostic reliability for single DCCs or circulating tumor cells. For this, we evaluated copy number alterations of more than 600 single DCCs from multiple cancer types to define reference regions suitable for quantification of target regions, such as ERBB2. We then compared ERBB2 quantitative PCR (qPCR) measurements with fluorescent in situ hybridization (FISH) data of various breast cancer cell lines and identified the aberration-calling threshold. The method was applied to two independent cohorts of esophageal cancer patients from Hamburg (n = 59) and Düsseldorf (n = 53). We found a high correlation between the single cell qPCR assay and the standard FISH assay (R = 0.98) and significant associations between amplification and survival for both patient cohorts (Hamburg (HH), p = 0.033; Düsseldorf (D), p = 0.052; pooled HH + D, p = 0.002) when applied to DCCs of esophageal cancer patients. Detection of a single ERBB2-amplified DCC was the most important risk factor for death from esophageal cancer (relative risk = 4.22; 95% CI = 1.91-9.32; p < 0.001). In our study, we detected ERBB2-amplified cells in 7% of patients. These patients could benefit from anti-ERBB2 targeting therapies.

Characterization of Rare, Dormant, and Therapy-Resistant Cells in Acute Lymphoblastic Leukemia

Tumor relapse is associated with dismal prognosis, but responsible biological principles remain incompletely understood. To isolate and characterize relapse-inducing cells, we used genetic engineering and proliferation-sensitive dyes in patient-derived xenografts of acute lymphoblastic leukemia (ALL). We identified a rare subpopulation that resembled relapse-inducing cells with combined properties of long-term dormancy, treatment resistance, and stemness. Single-cell and bulk expression profiling revealed their similarity to primary ALL cells isolated from pediatric and adult patients at minimal residual disease (MRD). Therapeutically adverse characteristics were reversible, as resistant, dormant cells became sensitive to treatment and started proliferating when dissociated from the in vivo environment. Our data suggest that ALL patients might profit from therapeutic strategies that release MRD cells from the niche.

Abstract 472: A digital sorting-based detection and isolation assay for single disseminated cancer cells from lymph nodes of melanoma patients

We recently showed that sentinel lymph node (SLN) disaggregation followed by immunocytology enables precise quantification of disseminated cancer cells (DCC). We demonstrated that this approach has a 20-fold higher sensitivity to detect melanoma DCCs than routine histopathology and that the DCC density (DCCD, defined as the number of DCCs per million lymph node cells) together with ulceration state and tumor thickness enables individual risk prediction superior to current AJCC guidelines (Ulmer et al., PLoS Medicine, 2014). Here, we present the adaptation of this method to a semi-automated workflow, including automated single cell detection and isolation by DEPArrayTM technology for subsequent molecular analysis of DCCs from SLN of melanoma patients.

The developed workflow includes a mechanical disaggregation of lymph node tissue and collection of the mononuclear cells after Percoll density centrifugation. Several tumor cell enrichment methods were tested (CellSearch® and size-based enrichment technologies), however, failed to process cell suspensions generated from lymph nodes due to clumping of cells during the enrichment procedure. Tumor cell enrichment using MACS depletion of CD45+ cells delivered approx. 50% of spiked-in cell line cells with a mean depletion rate of lymphocytes and erythrocytes &gt; 98%. To prevent limitations in DCC detection caused by phenotypic heterogeneity of tumor cells, we established a double staining against two melanoma-associated markers gp100 and MCSP. Individual melanoma cells are then detected and isolated by DEPArrayTM technology enabling single cell whole genome amplification (Ampli1TM) for subsequent molecular analysis as assessment of copy number alterations (e.g. by arrayCGH) or sequence analysis for melanoma specific point mutations (e.g. BRAF or NRAS). We successfully applied the workflow to first SLN samples from melanoma patients.

The established workflow enables a standardized detection of single DCCs from lymph nodes of melanoma patients applicable in clinical diagnostic labs. Automated single cell isolation and subsequent molecular analysis of DCCs is feasible within short time after SLN biopsy. In the future, this approach could help to select individualized therapies for melanoma patients.

Citation Format: Sebastian Scheitler, Kathrin Weidele, Barbara Alberter, Christian Werno, Christoph A. Klein, Bernhard Polzer. A digital sorting-based detection and isolation assay for single disseminated cancer cells from lymph nodes of melanoma patients. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 472.

Abstract 1551: High-resolution parallel analysis of genome and transcriptome of single disseminated prostate cancer cells

Introduction: Manifestation of bone metastases is the most frequent cause of death among prostate cancer patients. Since the progression from non-metastatic (M0) to metastatic (M1) is highly unpredictable, molecular analysis of disseminated cancer cells (DCCs) detected before overt metastasis may provide an opportunity to dissect the early stages of systemic cancer and enable detection of critical therapy targets. However, we previously showed that early stage prostate DCCs acquire a bone marrow-associated transcriptomic phenotype (Guzvic et al. Cancer Res. 2014, 15;74(24):7383-94) impeding the definite identification of DCCs. To address this problem we developed a workflow of combined single-cell RNA-Seq and single-cell aCGH allowing high-resolution analysis of genomes and transcriptomes of individual cells. Here, we present a proof-of-concept study using DCCs isolated from bone marrow of prostate cancer patients.

Materials and methods: We isolated 24 EPCAM-positive DCCs from bone marrow of prostate cancer patients. In addition, we analyzed two VCaP prostate cancer cells and two peripheral blood lymphocytes of healthy individuals. All isolated cells were subjected to combined whole genome and whole transcriptome amplification using Ampli1TM WGA Kit and a WTA approach (Klein CA et al., Nat Biotechnol. 2002, 20(4):387-92), respectively. The suitability of WTA and WGA products for downstream analyses were assessed using PCR-based QC-assays. Subsequently, WGA products were hybridized on high-resolution SurePrint aCGH arrays and analyzed with Agilent Genomic Workbench Software. WTA products were sequenced using Roche 454 GS FLX+ or Illumina HiSeq 2000 systems. Single-cell RNA-Seq data were evaluated using a bioinformatic pipeline adjusted to the needs of the single cell WTA method.

Results: We provide two experimental protocols for single-cell RNA-Seq: (i) allowing detection of low-abundant transcripts and (ii) enabling comprehensive gene-expression analysis. The protocols were validated using VCaP cells, in which we could successfully detect low expression levels of TMPRSS2-ERG fusion transcripts. The established analysis allowed also detection of novel fusion transcripts in a M1-stage prostate cancer patient. Comprehensive gene expression analysis revealed presence of 4.000 to 13.000 transcripts expressed in bone marrow-derived cells. The comparison of RNA-Seq and aCGH data allowed to simultaneously detecting copy number alterations and corresponding changes in gene expression dosage. Thereby, the combined genome and transcriptome analysis of single cells enables to uncover the impact of genomic gains and losses on cellular transcriptomes.

Conclusions: Combined genome and transcriptome analysis of patient-derived single DCCs is feasible and enables unambiguous identification and genotypic and phenotypic characterization of DCCs.

Citation Format: Stefan Kirsch, Urs Lahrmann, Miodrag Guzvic, Zbigniew T. Czyz, Giancarlo Feliciello, Bernhard Polzer, Christoph A. Klein. High-resolution parallel analysis of genome and transcriptome of single disseminated prostate cancer cells. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 1551.

Single-cell analysis of CTCs with diagnostic precision: opportunities and challenges for personalized medicine

The generation of variant cancer cells is the major cause of acquired resistance against systemic therapies and consequently, of our inability to cure advanced cancer patients. Circulating tumor cells are gaining increasing clinical attention because they may enable the monitoring cancer progression and adjustment of treatment. In recent years multiple technologies for enrichment, isolation as well as molecular and functional analysis of circulating tumor cells have been developed. Implementation of these technologies in standardized and automated workflows in clinical diagnostics could provide valuable information for real-time monitoring of cancer and eventually new therapeutic strategies for the benefit of patients.

Abstract 371: Longitudinal genetic characterization of circulating tumor cells in metastatic breast cancer patients

Background:

Little is known about the evolution of genetic aberrations during metastatic cancer progression and in response to systemic treatment. Obtaining repeated tissue biopsies is often impractical. On the other hand, it has been shown that circulating tumor cells (CTCs) can be easily followed during disease course and genetic characterization at the single cell level is possible with high reliability [1].

Methods:

Individual CTCs of a ER+ and HER2- de novo metastatic breast cancer patient treated with weekly paclitaxel/gemcitabine as first line therapy, were collected at three different time points (before start, after one and two cycles of treatment). Whole peripheral blood was enriched using the CellSearch® system and CTCs were sorted by DEPArray™ device. The whole genome of single CTCs was amplified with Ampli1™ WGA kit and WGA quality was assessed by Ampli1™ QC Kit. Genome wide single cell copy number variation (CNV) profile was evaluated with Agilent SurePrint 180k array comparative genomic hybridization (aCGH).

Results:

CTCs count at CellSearch was 22, 75 and 15 at three time points respectively.

A total of 25 single CTCs were collected and 23 (92%) showed high Genome Integrity Index (GII) as measured by Ampli1™ QC kit (GII = 3 or 4). For each time point multiple CTCs (3, 6 and 3 respectively) were selected for single cell aCGH analysis. The genomic profile was strikingly similar (1q gain, 12p, 13p, 16q and 17p losses) across individual cells of the same blood sample and throughout different time points evaluated. After 3 cycles of therapy a disease progression was documented by CAT-scan.

Discussion:

The observed high GII, low genetic heterogeneity and stable genome across different time points suggests the presence of an aggressive clone resistant to the treatment and cancer-associated genes analysis for sequence variants by NGS targeted sequencing is ongoing. Further patients with longitudinal follow-up will be enrolled in order to evaluate if the heterogeneity between aCGH profile at a given time point and longitudinal evolution of aCGH profiles can be associated with early treatment response. Results will be presented at the conference.

[1] Polzer B, Medoro G, et al, EMBO Mol Med. 2014 Oct 30;6(11):1371-86.

Citation Format: Valeria Sero, Francesca De Luca, Anna Doffini, Francesca Galardi, Marta Pestrin, Zbignew T. Czyz, Genny Buson, Giulia Bregola, Chiara Bolognesi, Francesca Fontana, Gianni Medoro, Bernhard Polzer, Angelo Di Leo, Christoph A. Klein, Nicolo Manaresi. Longitudinal genetic characterization of circulating tumor cells in metastatic breast cancer patients. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 371. doi:10.1158/1538-7445.AM2015-371

Molecular profiling of single circulating tumor cells with diagnostic intention

Several hundred clinical trials currently explore the role of circulating tumor cell (CTC) analysis for therapy decisions, but assays are lacking for comprehensive molecular characterization of CTCs with diagnostic precision. We therefore combined a workflow for enrichment and isolation of pure CTCs with a non-random whole genome amplification method for single cells and applied it to 510 single CTCs and 189 leukocytes of 66 CTC-positive breast cancer patients. We defined a genome integrity index (GII) to identify single cells suited for molecular characterization by different molecular assays, such as diagnostic profiling of point mutations, gene amplifications and whole genomes of single cells. The reliability of > 90% for successful molecular analysis of high-quality clinical samples selected by the GII enabled assessing the molecular heterogeneity of single CTCs of metastatic breast cancer patients. We readily identified genomic disparity of potentially high relevance between primary tumors and CTCs. Microheterogeneity analysis among individual CTCs uncovered pre-existing cells resistant to ERBB2-targeted therapies suggesting ongoing microevolution at late-stage disease whose exploration may provide essential information for personalized treatment decisions and shed light into mechanisms of acquired drug resistance.

Laser Microdissection of FFPE Tissue Areas and Subsequent Whole Genome Amplification by Ampli1™

Laser microdissection (LMD) and whole genome amplification (WGA) are valuable tools to isolate, purify, and genetically analyze cancer cells from tissue sections. In this chapter, we describe a workflow for microdissecting small regions of interest from cancer tissue, i.e. formalin-fixed paraffin-embedded (FFPE) and cryo-conserved specimens, and subsequent whole genome amplification by a deterministic WGA approach (Ampli1™ WGA).

Principles of Whole-Genome Amplification

Modern molecular biology relies on large amounts of high-quality genomic DNA. However, in a number of clinical or biological applications this requirement cannot be met, as starting material is either limited (e.g., preimplantation genetic diagnosis (PGD) or analysis of minimal residual cancer) or of insufficient quality (e.g., formalin-fixed paraffin-embedded tissue samples or forensics). As a consequence, in order to obtain sufficient amounts of material to analyze these demanding samples by state-of-the-art modern molecular assays, genomic DNA has to be amplified. This chapter summarizes available technologies for whole-genome amplification (WGA), bridging the last 25 years from the first developments to currently applied methods. We will especially elaborate on research application, as well as inherent advantages and limitations of various WGA technologies.

Combined genome and transcriptome analysis of single disseminated cancer cells from bone marrow of prostate cancer patients reveals unexpected transcriptomes

Bone is the most frequent site of metastasis in prostate cancer and patients with bone metastases are deemed incurable. Targeting prostate cancer cells that disseminated to the bone marrow before surgery and before metastatic outgrowth may therefore prevent lethal metastasis. This prompted us to directly analyze the transcriptome of disseminated cancer cells (DCC) isolated from patients with nonmetastatic (UICC stage M0) prostate cancer. We screened 105 bone marrow samples of patients with M0-stage prostate cancer and 18 bone marrow samples of patients without malignancy for the presence of EpCAM(+) single cells. In total, we isolated 270 cells from both groups by micromanipulation and globally amplified their mRNA. We used targeted transcriptional profiling to unambiguously identify DCCs for subsequent in-depth analysis. Transcriptomes of all cells were examined for the expression of EPCAM, KRT8, KRT18, KRT19, KRT14, KRT6a, KRT5, KLK3 (PSA), MAGEA2, MAGEA4, PTPRC (CD45), CD33, CD34, CD19, GYPC, SCL4A1 (band 3), and HBA2. Using these transcripts, we found it impossible to reliably identify true DCCs. We then applied combined genome and transcriptome analysis of single cells and found that EpCAM(+) cells from controls expressed transcripts thought to be epithelial-specific, whereas true DCCs may express hematopoietic transcripts. These results point to an unexpected transcriptome plasticity of epithelial cancer cells in bone marrow and question common transcriptional criteria to identify DCCs.

Genomic High-Resolution Profiling of Single CKpos/CD45neg Flow-Sorting Purified Circulating Tumor Cells from Patients with Metastatic Breast Cancer

BACKGROUND

Circulating tumor cells (CTCs) are promising surrogate markers for systemic disease, and their molecular characterization might be relevant to guide more individualized cancer therapies. To enable fast and efficient purification of individual CTCs, we developed a work flow from CellSearchTM cartridges enabling high-resolution genomic profiling on the single-cell level.

METHODS

Single CTCs were sorted from 40 CellSearch samples from patients with metastatic breast cancer using a MoFlo XDP cell sorter. Genomes of sorted single cells were amplified using an adapter–linker PCR. Amplification products were analyzed by array-based comparative genomic hybridization, a gene-specific quantitative PCR (qPCR) assay for cyclin D1 (CCND1) locus amplification, and genomic sequencing to screen for mutations in exons 1, 9, and 20 of the phosphatidylinositol-4,5-bisphosphate 3-kinase, catalytic subunit alpha (PIK3CA) gene and exons 5, 7, and 8 of the tumor protein p53 (TP53) gene.

RESULTS

One common flow-sorting protocol was appropriate for 90% of the analyzed CellSearch cartridges, and the detected CTC numbers correlated positively with those originally detected with the CellSearch system (R2 = 0.9257). Whole genome amplification was successful in 72.9% of the sorted single CTCs. Over 95% of the cells displayed chromosomal aberrations typical for metastatic breast cancers, and amplifications at the CCND1 locus were validated by qPCR. Aberrant CTCs from 2 patients harbored mutations in exon 20 of the PIK3CA gene.

CONCLUSIONS

This work flow enabled effective CTC isolation and provided insights into genomic alterations of CTCs in metastatic breast cancer. This approach might facilitate further molecular characterization of rare CTCs to increase understanding of their biology and as a basis for their molecular screening in the clinical setting.

Quantitative measurement of melanoma spread in sentinel lymph nodes and survival

BACKGROUND

Sentinel lymph node spread is a crucial factor in melanoma outcome. We aimed to define the impact of minimal cancer spread and of increasing numbers of disseminated cancer cells on melanoma-specific survival.

METHODS AND FINDINGS

We analyzed 1,834 sentinel nodes from 1,027 patients with ultrasound node-negative melanoma who underwent sentinel node biopsy between February 8, 2000, and June 19, 2008, by histopathology including immunohistochemistry and quantitative immunocytology. For immunocytology we recorded the number of disseminated cancer cells (DCCs) per million lymph node cells (DCC density [DCCD]) after disaggregation and immunostaining for the melanocytic marker gp100. None of the control lymph nodes from non-melanoma patients (n = 52) harbored gp100-positive cells. We analyzed gp100-positive cells from melanoma patients by comparative genomic hybridization and found, in 45 of 46 patients tested, gp100-positive cells displaying genomic alterations. At a median follow-up of 49 mo (range 3-123 mo), 138 patients (13.4%) had died from melanoma. Increased DCCD was associated with increased risk for death due to melanoma (univariable analysis; p<0.001; hazard ratio 1.81, 95% CI 1.61-2.01, for a 10-fold increase in DCCD + 1). Even patients with a positive DCCD ≤3 had an increased risk of dying from melanoma compared to patients with DCCD = 0 (p = 0.04; hazard ratio 1.63, 95% CI 1.02-2.58). Upon multivariable testing DCCD was a stronger predictor of death than histopathology. The final model included thickness, DCCD, and ulceration (all p<0.001) as the most relevant prognostic factors, was internally validated by bootstrapping, and provided superior survival prediction compared to the current American Joint Committee on Cancer staging categories.

CONCLUSIONS

Cancer cell dissemination to the sentinel node is a quantitative risk factor for melanoma death. A model based on the combined quantitative effects of DCCD, tumor thickness, and ulceration predicted outcome best, particularly at longer follow-up. If these results are validated in an independent study, establishing quantitative immunocytology in histopathological laboratories may be useful clinically.

Abstract A63: Identification and molecular characterization of single disseminated metastasis progenitor cells in prostate cancer: The significance of EpCAM-positive cells

Carcinoma of the prostate is the most frequently diagnosed cancer and the second leading cause of cancer deaths in males. While it can be detected early, its progression is variable and patients that will benefit from aggressive therapy are hard to identify, leading to considerable overtreatment. Nevertheless, a fraction of patients will develop metastasis years after surgery and finally succumb to the disease. The detection of early disseminated cancer cells (DCCs) in bone marrow (BM) of carcinoma patients without distant metastasis is associated with poor prognosis. DCCs comprise precursor cells of later arising metastasis, and they are direct targets for adjuvant therapies, aiming to delay or prevent outgrowth of macrometastasis. DCCs are extremely rare and can be detected in BM by using antibodies against epithelial markers, such as cytokeratins (CKs) and epithelial cell adhesion molecule (EpCAM). We have previously observed that CK+ DCCs in BM detected in samples taken before surgery are associated with poor survival of prostate cancer patients. In contrast, we found that the detection of persisting CK+ cells in BM samples up to 8 years after surgery had no impact on the survival of prostate cancer patients, indicating that prostate cancer can become a chronic, non-progressive disease. This result prompted us to search for cells that persist after surgery and are associated with metastatic progression. We therefore tested the EpCAM in a cohort of 220 prostate cancer patients. We directly compared the prognostic impact of CK+ and EpCAM+ DCCs detected in BM samples taken months to years after surgery. Strikingly, while CK+ cells were, as previously shown, not informative about progression, the detection of EpCAM+ DCCs after surgery was associated with poor survival. Furthermore, double staining against both epithelial markers identified different subsets of DCCs. The analysis of single DCCs from BM by single cell comparative genomic hybridization (SCOMP) demonstrated that both CK+ and EpCAM+ cells are tumor cells. Interestingly, EpCAM+ cells harbored significantly more genomic aberrations than CK+ cells and are defined by characteristic changes. Since staining for EpCAM enables isolation of single cell's mRNA, we have isolated EpCAM+ cells from an additional cohort of 108 patients and from 8 healthy donors. Total genomic DNA and mRNA from these single cells were subjected to combined analysis of genome and transcriptome. PCR analysis of epithalial markers confirmed the existence of different subpopulations of EpCAM+ cells. Furthermore, using microarray analysis, we have obtained gene expression signatures of different subpopulations of DCCs. Finally, using prostate epithelial cell lines, we have investigated the functional differences of different subpopulations of EpCAM+ cells. Our results indicate an unexpected diversity of early-disseminated cancer cells displaying different phenotypes and genotypes and divergent metastatic potential. Gene expression analysis revealed great heterogeneity of the expression of analyzed markers within cells isolated from patients' BM samples. This suggests that different subpopulations of EpCAM+ DCCs may exist in diverse functional states, and may have different functional relevance.

Citation Format: Miodrag Gužvić, Bernhard Polzer, Roman Ganzer, Dorothea Weckermann, Bernhard Braun, Inka Appel, Matthias Maneck, Wolfgang Wieland, Christoph A. Klein. Identification and molecular characterization of single disseminated metastasis progenitor cells in prostate cancer: The significance of EpCAM-positive cells [abstract]. In: Proceedings of the AACR Special Conference on Advances in Prostate Cancer Research; 2012 Feb 6-9; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2012;72(4 Suppl):Abstract nr A63.

Perioperative activation of disseminated tumor cells in bone marrow of patients with prostate cancer

PURPOSE

The outcome of prostate cancer is highly unpredictable. To assess the dynamics of systemic disease and to identify patients at high risk for early relapse we followed the fate of disseminated tumor cells in bone marrow for up to 10 years and genetically analyzed such cells isolated at various stages of disease.

PATIENTS AND METHODS

Nine hundred bone marrow aspirates from 384 patients were stained using the monoclonal antibody A45-B/B3 directed against cytokeratins 8, 18, and 19. Log-rank statistics and Cox regression analysis were applied to determine the prognostic impact of positive cells detected before surgery (244 patients) and postoperatively (214 patients). Samples from primary tumors (n = 55) and single disseminated tumor cells (n = 100) were analyzed by comparative genomic hybridization.

RESULTS

Detection of cytokeratin-positive cells before surgery was the strongest independent risk factor for metastasis within 48 months (P < .001; relative risk [RR], 5.5; 95% CI, 2.4 to 12.9). In contrast, cytokeratin-positive cells detected 6 months to 10 years after radical prostatectomy were consistently present in bone marrow with a prevalence of approximately 20% but had no influence on disease outcome. Characteristic genotypes of cytokeratin-positive cells were selected at manifestation of metastasis.

CONCLUSION

Cytokeratin-positive cells in the bone marrow of prostate cancer patients are only prognostically relevant when detected before surgery. Because we could not identify significant genetic differences between pre- and postoperatively isolated tumor cells before manifestation of metastasis, we postulate the existence of perioperative stimuli that activate disseminated tumor cells. Patients with cytokeratin-positive cells in bone marrow before surgery may therefore benefit from adjuvant therapies.

Genomic analysis of single cytokeratin-positive cells from bone marrow reveals early mutational events in breast cancer

Chromosomal instability in human breast cancer is known to take place before mammary neoplasias display morphological signs of invasion. We describe here the unexpected finding of a tumor cell population with normal karyotypes isolated from bone marrow of breast cancer patients. By analyzing the same single cells for chromosomal aberrations, subchromosomal allelic losses, and gene amplifications, we confirmed their malignant origin and delineated the sequence of genomic events during breast cancer progression. On this trajectory of genomic progression, we identified a subpopulation of patients with very early HER2 amplification. Because early changes have the highest probability of being shared by genetically unstable tumor cells, the genetic characterization of disseminated tumor cells provides a novel rationale for selecting patients for targeted therapies.

Genetic heterogeneity of single disseminated tumour cells in minimal residual cancer

BACKGROUND

Because cancer patients with small tumours often relapse despite local and systemic treatment, we investigated the genetic variation of the precursors of distant metastasis at the stage of minimal residual disease. Disseminated tumour cells can be detected by epithelial markers in mesenchymal tissues and represent targets for adjuvant therapies.

METHODS

We screened 525 bone-marrow, blood, and lymph-node samples from 474 patients with breast, prostate, and gastrointestinal cancers for single disseminated cancer cells by immunocytochemistry with epithelial-specific markers. 71 (14%) of the samples contained two or more tumour cells whose genomic organisation we studied by single cell genomic hybridisation. In addition, we tested whether TP53 was mutated. Hierarchical clustering algorithms were used to determine the degree of clonal relatedness of sister cells that were isolated from individual patients.

FINDINGS

Irrespective of cancer type, we saw an unexpectedly high genetic divergence in minimal residual cancer, particularly at the level of chromosomal imbalances. Although few disseminated cells harboured TP53 mutations at this stage of disease, we also saw microheterogeneity of the TP53 genotype. The genetic heterogeneity was strikingly reduced with the emergence of clinically evident metastasis.

INTERPRETATION

Although the heterogeneity of primary tumours has long been known, we show here that early disseminated cancer cells are genomically very unstable as well. Selection of clonally expanding cells leading to metastasis seems to occur after dissemination has taken place. Therefore, adjuvant therapies are confronted with an extremely large reservoir of variant cells from which resistant tumour cells can be selected.