Liquid Biopsy Assessment of Circulating Tumor Cell PD-L1 and IRF-1 Expression in Patients with Advanced Solid Tumors Receiving Immune Checkpoint Inhibitor.

BACKGROUND: Reliable biomarkers that can be serially monitored to predict treatment response to immune checkpoint inhibitors (ICIs) are still an unmet need. Here, we present a multiplex immunofluorescence (IF) assay that simultaneously detects circulating tumor cells (CTCs) and assesses CTC expression of programmed death ligand-1 (PD-L1) and interferon regulatory factor 1 (IRF-1) as a candidate biomarker related to ICI use. OBJECTIVE: To assess the potential of CTC PD-L1 and IRF-1 expression as candidate biomarkers for patients with advanced epithelial solid tumors receiving ICIs. PATIENTS AND METHODS: We tested the IF CTC assay in a pilot study of 28 patients with advanced solid tumors who were starting ICI. Blood for CTC evaluation was obtained prior to starting ICI, after a single cycle of therapy, and at the time of radiographic assessment or treatment discontinuation. RESULTS: At baseline, patients with 0-1 CTCs had longer progression-free survival (PFS) compared to patients with ≥ 2 CTCs (4.3 vs 1.3 months, p = 0.01). The presence of any PD-L1+ CTCs after a single dose of ICI portended shorter PFS compared to patients with no CTCs or PD-L1- CTCs (1.2 vs 4.2 months, p = 0.02); the presence of any PD-L1+ or IRF-1+ CTCs at time of imaging assessment or treatment discontinuation also was associated with shorter PFS (1.9 vs 5.5 months, p < 0.01; 1.6 vs 4.7 months, p = 0.05). CTC PD-L1 and IRF-1 expression did not correlate with tumor tissue PD-L1 or IRF-1 expression. Strong IRF-1 expression in tumor tissue was associated with durable (≥ 1 year) radiographic response (p = 0.02). CONCLUSIONS: Based on these results, CTC PD-L1 and IRF-1 expression is of interest in identifying ICI resistance and warrants further study.

Clinical Validation of a Circulating Tumor Cell Assay Using Density Centrifugation and Automated Immunofluorescence Microscopy.

OBJECTIVES: The US Food and Drug Administration (FDA)-approved CELLSEARCH assay (Menarini Silicon Biosystems) for circulating tumor cells (CTCs) relies on expression of an epithelial cell adhesion molecule to enrich for CTCs. We sought to validate a CTC assay (RareCyte) for clinical use that instead collects a buffy coat preparation enriched for CTCs. METHODS: Normal peripheral blood specimens spiked with cultured breast and prostate cancer cells and 47 clinical samples were used to validate assay performance. Specimens were enriched for buffy coat cells and applied onto 8 glass slides. The slides were immunofluorescently stained and imaged by automated microscopy and computer-aided image analysis. RESULTS: The assay was 100% specific for detecting spiked tumor cells. For samples spiked with 25, 50, and 125 cells, the percentage coefficients of variation were 42%, 21%, and 3.7%, respectively. Linearity studies demonstrated a slope of 0.99, an intercept of 1.6, and R2 of 0.96. Recoveries at the 25-, 50-, and 125-cell levels were 92%, 111%, and 100%, respectively. Clinical samples run on both CELLSEARCH and RareCyte correlated with an R2 of 0.8 after log-transformation and demonstrated 87.5% concordance using the CELLSEARCH criteria for predicting adverse outcomes. CONCLUSIONS: The RareCyte CTC assay has comparable performance to the FDA-cleared method and is ready for further clinical validation studies.

Beyond Circulating Tumor Cell Enumeration: Cell-Based Liquid Biopsy to Assess Protein Biomarkers and Cancer Genomics Using the RareCyte® Platform.

The practice of medicine has steadily employed less invasive methods to obtain information derived from the tumor to guide clinical management of patients. Liquid biopsy-the sampling of blood-is a non-invasive method for generating information previously only available from tissue biopsies of the tumor mass. Analysis of fragmented circulating tumor DNA in the plasma is clinically used to identify actionable mutations and detect residual or recurrent disease. Plasma analysis cannot, however, assess cancer phenotypes, including the expression of drug targets and protein biomarkers. Circulating tumor cells (CTCs) are intact cancer cells that have entered the blood that have the potential for distant metastasis. While enumeration of CTCs is prognostic of outcome, recently developed technology allows for the interrogation of protein biomarkers on CTCs that could be predictive of response. Furthermore, since CTCs contain intact whole cancer genomes, isolating viable CTCs detected during therapy could provide a rational approach to assessing mutational profiles of resistance. Identification, characterization and molecular analysis of CTCs together will advance the capacity of liquid biopsy to meet the requirements of twenty-first century medicine.

Safety, Feasibility, and Merits of Longitudinal Molecular Testing of Multiple Metastatic Sites to Inform mTNBC Patient Treatment in the Intensive Trial of Omics in Cancer.

PURPOSE: Patients with metastatic triple-negative breast cancer (mTNBC) have poor outcomes. The Intensive Trial of Omics in Cancer (ITOMIC) sought to determine the feasibility and potential efficacy of informing treatment decisions through multiple biopsies of mTNBC deposits longitudinally over time, accompanied by analysis using a distributed network of experts. METHODS: Thirty-one subjects were enrolled and 432 postenrollment biopsies performed (clinical and study-directed) of which 332 were study-directed. Molecular profiling included whole-genome sequencing or whole-exome sequencing, cancer-associated gene panel sequencing, RNA-sequencing, and immunohistochemistry. To afford time for analysis, subjects were initially treated with cisplatin (19 subjects), or another treatment they had not received previously. The results were discussed at a multi-institutional ITOMIC Tumor Board, and a report transmitted to the subject's oncologist who arrived at the final treatment decision in conjunction with the subject. Assistance was provided to access treatments that were predicted to be effective. RESULTS: Multiple biopsies in single settings and over time were safe, and comprehensive analysis was feasible. Two subjects were found to have lung cancer, one had carcinoma of unknown primary site, tumor samples from three subjects were estrogen receptor-positive and from two others, human epidermal growth factor receptor 2-positive. Two subjects withdrew. Thirty-four of 112 recommended treatments were accessed using approved drugs, clinical trials, and single-patient investigational new drugs. After excluding the three subjects with nonbreast cancers and the two subjects who withdrew, 22 of 26 subjects (84.6%) received at least one ITOMIC Tumor Board-recommended treatment. CONCLUSION: Further exploration of this approach in patients with mTNBC is merited.

Circulating tumor cell investigation in breast cancer patient-derived xenograft models by automated immunofluorescence staining, image acquisition, and single cell retrieval and analysis.

BACKGROUND: Breast cancer patient-derived xenograft (BC-PDX) models represent a continuous and reproducible source of circulating tumor cells (CTCs) for studying their role in tumor biology and metastasis. We have previously shown the utility of BC-PDX models in the study of CTCs by immunohistochemistry (IHC) on serial paraffin sections and manual microscopic identification of cytokeratin-positive cells, a method that is both low-throughput and labor-intensive. We therefore aimed to identify and characterize CTCs from small volume mouse blood samples and examined its practical workflow in a study of BC-PDX mice treated with chemotherapy using an automated imaging platform, the AccuCyte®-CyteFinder® system. METHODS: CTC analysis was conducted using blood from non-tumor bearing SCID/Beige mice spiked with human breast cancer cells, BC-PDX-bearing mice, and BC-PDX mice treated with vehicle or chemotherapeutic agent(s). After red blood cell lysis, nucleated cells were mixed with transfer solution, processed onto microscope slides, and stained by immunofluorescence. The CyteFinder automated scanning microscope was used to identify CTCs, defined as nucleated cells that were human cytokeratin-positive, and mouse CD45-negative. Disaggregated primary BC-PDX tumors and lung metastatic nodules were processed using the same immunostaining protocol. Collective expression of breast cancer cell surface markers (EpCAM, EGFR, and HER2) using a cocktail of target-specific antibodies was assessed. CTCs and disaggregated tumor cells were individually retrieved from slides using the CytePicker® module for sequence analysis of a BC-PDX tumor-specific PIK3CA mutation. RESULTS: The recovery rate of human cancer cells spiked into murine blood was 83 ± 12%. CTC detection was not significantly different from the IHC method. One-third of CTCs did not stain positive for cell surface markers. A PIK3CA T1035A mutation present in a BC-PDX tumor was confirmed in isolated single CTCs and cells from dissociated metastatic nodules after whole genome amplification and sequencing. CTC evaluation could be simply implemented into a preclinical PDX therapeutic study setting with substantial improvements in workflow over the IHC method. CONCLUSIONS: Analysis of small volume blood samples from BC-PDX-bearing mice using the AccuCyte-CyteFinder system allows investigation of the role of CTCs in tumor biology and metastasis independent of surface marker expression.

The RareCyte® platform for next-generation analysis of circulating tumor cells.

Circulating tumor cells (CTCs) can reliably be identified in cancer patients and are associated with clinical outcome. Next-generation "liquid biopsy" technologies will expand CTC diagnostic investigation to include phenotypic characterization and single-cell molecular analysis. We describe here a rare cell analysis platform designed to comprehensively collect and identify CTCs, enable multi-parameter assessment of individual CTCs, and retrieve single cells for molecular analysis. The platform has the following four integrated components: 1) density-based separation of the CTC-containing blood fraction and sample deposition onto microscope slides; 2) automated multiparameter fluorescence staining; 3) image scanning, analysis, and review; and 4) mechanical CTC retrieval. The open platform utilizes six fluorescence channels, of which four channels are used to identify CTC and two channels are available for investigational biomarkers; a prototype assay that allows three investigational biomarker channels has been developed. Single-cell retrieval from fixed slides is compatible with whole genome amplification methods for genomic analysis. © 2018 The Authors. Cytometry Part A published by Wiley Periodicals, Inc. on behalf of International Society for Advancement of Cytometry.

Reliable detection of subchromosomal deletions and duplications using cell-based noninvasive prenatal testing.

OBJECTIVE: To gather additional data on the ability to detect subchromosomal abnormalities of various sizes in single fetal cells isolated from maternal blood, using low-coverage shotgun next-generation sequencing for cell-based noninvasive prenatal testing (NIPT). METHOD: Fetal trophoblasts were recovered from approximately 30 mL of maternal blood using maternal white blood cell depletion, density-based cell separation, immunofluorescence staining, and high-resolution scanning. These trophoblastic cells were picked as single cells and underwent whole genome amplification for subsequent genome-wide copy number analysis and genotyping to confirm the fetal origin of the cells. RESULTS: Applying our fetal cell isolation method to a series of 125 maternal blood samples, we detected on average 4.17 putative fetal cells/sample. The series included 15 cases with clinically diagnosed fetal aneuploidies and five cases with subchromosomal abnormalities. This method was capable of detecting findings that were 1 to 2 Mb in size, and all were concordant with the microarray or karyotype data obtained on a fetal sample. A minority of fetal cells showed evidence of genome degradation likely related to apoptosis. CONCLUSION: We demonstrate that this cell-based NIPT method has the capacity to reliably diagnose fetal chromosomal abnormalities down to 1 to 2 Mb in size.

RareCyte® CTC Analysis Step 1: AccuCyte® Sample Preparation for the Comprehensive Recovery of Nucleated Cells from Whole Blood.

The RareCyte platform addresses important technology limitations of current circulating tumor cell (CTC) collection methods, and expands CTC interrogation to include advanced phenotypic characterization and single-cell molecular analysis. In this respect, it represents the "next generation" of cell-based liquid biopsy technologies. In order to identify and analyze CTCs, RareCyte has developed an integrated sample preparation, imaging and individual cell retrieval process. The first step in the process, AccuCyte®, allows the separation, collection, and transfer to a slide the nucleated cell fraction of the blood that contains CTCs. Separation and collection are based on cell density-rather than size or surface molecular expression-and are performed within a closed system, without wash or lysis steps, enabling high CTC recovery. Here, we describe our technique for nucleated cell collection from a blood sample, and the spreading of these nucleated cells onto glass slides permitting immunofluorescent staining, cell identification, and individual cell picking described in subsequent chapters. In addition to collection of rare cells such as CTCs, AccuCyte also collects cells of the circulating immune system onto archivable slides as well as plasma from the same sample.

RareCyte® CTC Analysis Step 2: Detection of Circulating Tumor Cells by CyteFinder® Automated Scanning and Semiautomated Image Analysis.

The RareCyte CyteFinder instrument is an automated scanner that allows rapid identification of circulating tumor cells (CTCs) on microscope slides prepared by the AccuCyte process (see Chapter 13 ) and stained by immunofluorescence. Here, we present the workflow for CyteFinder scanning, analysis, and CyteMapper scan review which includes CTC confirmation and report generation.

RareCyte® CTC Analysis Step 3: Using the CytePicker® Module for Individual Cell Retrieval and Subsequent Whole Genome Amplification of Circulating Tumor Cells for Genomic Analysis.

The CytePicker module built into the RareCyte CyteFinder instrument allows researchers to easily retrieve individual cells from microscope slides for genomic analyses, including array CGH, targeted sequencing, and next-generation sequencing. Here, we describe the semiautomated retrieval of CTCs from the blood processed by AccuCyte (see Chapter 13) and amplification of genomic DNA so that molecular analysis can be performed.

Evidence for feasibility of fetal trophoblastic cell-based noninvasive prenatal testing.

OBJECTIVE: The goal was to develop methods for detection of chromosomal and subchromosomal abnormalities in fetal cells in the mother's circulation at 10-16 weeks' gestation using analysis by array comparative genomic hybridization (CGH) and/or next-generation sequencing (NGS). METHOD: Nucleated cells from 30 mL of blood collected at 10-16 weeks' gestation were separated from red cells by density fractionation and then immunostained to identify cytokeratin positive and CD45 negative trophoblasts. Individual cells were picked and subjected to whole genome amplification, genotyping, and analysis by array CGH and NGS. RESULTS: Fetal cells were recovered from most samples as documented by Y chromosome PCR, short tandem repeat analysis, array CGH, and NGS including over 30 normal male cells, one 47,XXY cell from an affected fetus, one trisomy 18 cell from an affected fetus, nine cells from a trisomy 21 case, three normal cells and one trisomy 13 cell from a case with confined placental mosaicism, and two chromosome 15 deletion cells from a case known by CVS to have a 2.7 Mb de novo deletion. CONCLUSION: We believe that this is the first report of using array CGH and NGS whole genome sequencing to detect chromosomal abnormalities in fetal trophoblastic cells from maternal blood. © 2016 The Authors. Prenatal Diagnosis published by John Wiley & Sons, Ltd.

A Distributed Network for Intensive Longitudinal Monitoring in Metastatic Triple-Negative Breast Cancer.

Accelerating cancer research is expected to require new types of clinical trials. This report describes the Intensive Trial of OMics in Cancer (ITOMIC) and a participant with triple-negative breast cancer metastatic to bone, who had markedly elevated circulating tumor cells (CTCs) that were monitored 48 times over 9 months. A total of 32 researchers from 14 institutions were engaged in the patient's evaluation; 20 researchers had no prior involvement in patient care and 18 were recruited specifically for this patient. Whole-exome sequencing of 3 bone marrow samples demonstrated a novel ROS1 variant that was estimated to be present in most or all tumor cells. After an initial response to cisplatin, a hypothesis of crizotinib sensitivity was disproven. Leukapheresis followed by partial CTC enrichment allowed for the development of a differential high-throughput drug screen and demonstrated sensitivity to investigational BH3-mimetic inhibitors of BCL-2 that could not be tested in the patient because requests to the pharmaceutical sponsors were denied. The number and size of CTC clusters correlated with clinical status and eventually death. Focusing the expertise of a distributed network of investigators on an intensively monitored patient with cancer can generate high-resolution views of the natural history of cancer and suggest new opportunities for therapy. Optimization requires access to investigational drugs.

High-recovery visual identification and single-cell retrieval of circulating tumor cells for genomic analysis using a dual-technology platform integrated with automated immunofluorescence staining.

BACKGROUND: Circulating tumor cells (CTCs) are malignant cells that have migrated from solid cancers into the blood, where they are typically present in rare numbers. There is great interest in using CTCs to monitor response to therapies, to identify clinically actionable biomarkers, and to provide a non-invasive window on the molecular state of a tumor. Here we characterize the performance of the AccuCyte®--CyteFinder® system, a comprehensive, reproducible and highly sensitive platform for collecting, identifying and retrieving individual CTCs from microscopic slides for molecular analysis after automated immunofluorescence staining for epithelial markers. METHODS: All experiments employed a density-based cell separation apparatus (AccuCyte) to separate nucleated cells from the blood and transfer them to microscopic slides. After staining, the slides were imaged using a digital scanning microscope (CyteFinder). Precisely counted model CTCs (mCTCs) from four cancer cell lines were spiked into whole blood to determine recovery rates. Individual mCTCs were removed from slides using a single-cell retrieval device (CytePicker™) for whole genome amplification and subsequent analysis by PCR and Sanger sequencing, whole exome sequencing, or array-based comparative genomic hybridization. Clinical CTCs were evaluated in blood samples from patients with different cancers in comparison with the CellSearch® system. RESULTS: AccuCyte--CyteFinder presented high-resolution images that allowed identification of mCTCs by morphologic and phenotypic features. Spike-in mCTC recoveries were between 90 and 91%. More than 80% of single-digit spike-in mCTCs were identified and even a single cell in 7.5 mL could be found. Analysis of single SKBR3 mCTCs identified presence of a known TP53 mutation by both PCR and whole exome sequencing, and confirmed the reported karyotype of this cell line. Patient sample CTC counts matched or exceeded CellSearch CTC counts in a small feasibility cohort. CONCLUSION: The AccuCyte--CyteFinder system is a comprehensive and sensitive platform for identification and characterization of CTCs that has been applied to the assessment of CTCs in cancer patient samples as well as the isolation of single cells for genomic analysis. It thus enables accurate non-invasive monitoring of CTCs and evolving cancer biology for personalized, molecularly-guided cancer treatment.

Acquisition of mitochondrial dysregulation and resistance to mitochondrial-mediated apoptosis after genotoxic insult in normal human fibroblasts: a possible model for early stage carcinogenesis.

Acquisition of death-resistance is critical in the evolution of neoplasia. Our aim was to model the early stages of carcinogenesis by examining intracellular alterations in cells that have acquired apoptosis-resistance after exposure to a complex genotoxin. We previously generated sub-populations of BJ-hTERT human diploid fibroblasts, which have acquired death-resistance following exposure to hexavalent chromium [Cr(VI)], a broad-spectrum genotoxicant. Long-term exposure to certain forms of Cr(VI) is associated with respiratory carcinogenesis. Here, we report on the death-sensitivity of subclonal populations derived from clonogenic survivors of BJ-hTERT cells treated with 5 µM Cr(VI) (DR1, DR2), or selected by dilution-based cloning without treatment (CC1). Following Cr(VI) treatment, CC1 cells downregulated expression of the anti-apoptotic protein Bcl-2 and exhibited extensive expression of cleaved caspase 3. In contrast, the DR cells exhibited no cleaved caspase 3 expression and maintained expression of Bcl-2 following recovery from 24 h Cr(VI) exposure. The DR cells also exhibited attenuated mitochondrial-membrane depolarization and mitochondrial retention of cytochrome c and SMAC/DIABLO following Cr(VI) exposure. The DR cells exhibited less basal mtDNA damage, as compared to CC1 cells, which correlates with intrinsic (non-induced) death-resistance. Notably, there was no difference in p53 protein expression before or after treatment among all cell lines. Taken together, our data suggest the presence of more resilient mitochondria in death-resistant cells, and that death-resistance can be acquired in normal human cells early after genotoxin exposure. We postulate that resistance to mitochondrial-mediated cell death and mitochondrial dysregulation may be an initial phenotypic alteration observed in early stage carcinogenesis.

Ultrastructural localization of extracellular matrix proteins of the lymph node cortex: evidence supporting the reticular network as a pathway for lymphocyte migration.

BACKGROUND: The lymph node (LN) is a crossroads of blood and lymphatic vessels allowing circulating lymphocytes to efficiently recognize foreign molecules displayed on antigen presenting cells. Increasing evidence indicates that after crossing high endothelial venules, lymphocytes migrate within the node along the reticular network (RN), a scaffold of fibers enwrapped by fibroblastic reticular cells (FRC). Light microscopy has shown that the RN contains specific extracellular matrix (ECM) proteins, which are putative molecular "footholds" for migration, and are known ligands for lymphocyte integrin adhesion receptors. RESULTS: To investigate whether ECM proteins of the RN are present on the outer surface of the FRC and are thus accessible to migrating lymphocytes, ultrastructural immunohistochemical staining of cynomolgus monkey LN was performed using antibodies to human ECM proteins that were successfully employed at the light microscopic level. The fibrillar collagens I and III were observed primarily within the reticular network fibers themselves. In contrast, the matrix proteins laminin, fibronectin, collagen IV, and tenascin were observed within the reticular fibers and also on the outer membrane surface of the FRC. CONCLUSIONS: These findings suggest a molecular basis for how the RN functions as a pathway for lymphocyte migration within the lymph node.

Upregulation of Poly (ADP-Ribose) Polymerase-1 (PARP1) in Triple-Negative Breast Cancer and Other Primary Human Tumor Types.

Poly (ADP-ribose) polymerase-1 (PARP1) is a key facilitator of DNA repair and is implicated in pathways of tumorigenesis. PARP inhibitors have gained recent attention as rationally designed therapeutics for the treatment of several malignancies, particularly those associated with dysfunctional DNA repair pathways, including triple-negative breast cancer (TNBC). We investigated the PARP1 gene expression profile in surgical samples from more than 8,000 primary malignant and normal human tissues. PARP1 expression was found to be significantly increased in several malignant tissues, including those isolated from patients with breast, uterine, lung, ovarian, and skin cancers, and non-Hodgkin's lymphoma. Within breast infiltrating ductal carcinoma (IDC) samples tested, mean PARP1 expression was significantly higher relative to normal breast tissue, with over 30% of IDC samples demonstrating upregulation of PARP1, compared with 2.9% of normal tissues. Because of known DNA repair defects, including BRCA1 dysfunction, associated with TNBC, exploration of PARP1 expression in breast cancers related to expression of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) led to the observation that negative expression of any of the 3 receptors was associated with upregulation of PARP1 expression, compared with receptor-positive tissues. To validate these observations, an independent set of breast adenocarcinomas was evaluated and demonstrated >2-fold upregulation of PARP1 in approximately 70% of primary breast adenocarcinomas, including TNBC, compared with syngeneic nonmalignant breast tissues. Immunohistochemistry (IHC) showed that upregulation of the PARP1 gene was consistent with increased protein expression in TNBC. These analyses suggest a potential biological role for PARP1 in several distinct malignancies, including TNBC. Further investigation of PARP1 as a biomarker for the therapeutic activity of PARP inhibitor-based therapy is warranted.

Abnormal serum concentrations of proteins in Parkinson's disease.

Blood serum was used to identify protein biomarkers for diagnosis of Parkinson's disease (PD) using analytically validated quantitative 2D-gel electrophoresis, and single variable and multivariate statistics. Using banked samples from a first medical center, we identified 57 specific protein spot biomarkers with disease-specific abnormal levels in serum of patients with PD, Alzheimer's disease, amyotrophic lateral sclerosis and similar neurodegenerative conditions (337 samples), when compared to age-matched normal controls (132 samples). To further assess their clinical usefulness in Parkinson's disease, we obtained prospective newly drawn blood serum samples from a second (56 PD, 30 controls) and third (6 PD, 48 controls) medical center. The protein concentrations of the 57 biomarkers were assessed by 2D-gel electrophoresis. Stepwise linear discriminant analysis selected a combination of 21 of the 57 as optimal to distinguish PD patients from controls. When applied to the samples from the second site, the 21 proteins had sensitivity of 93.3% (52 of 56 PD correctly classified), specificity of 92.9% (28 of 30 controls correctly classified); 15 of 15 patients with mild, 28 of 30 with moderate to severe symptoms, and all of the 6 PD patients from the third site were correctly classified. Eleven of the 21 proteins showed statistically significant abnormal concentrations in patients with mild symptoms, and 14 in patients with moderate-severe symptoms. The protein identities reflect the heterogeneity of Parkinson's disease, and thus may provide the capability of monitoring the blood for a diverse range of PD pathophysiological mechanisms: cellular degeneration, oxidative stress, inflammation, and transport.

A strategy for predicting the chemosensitivity of human cancers and its application to drug discovery.

The U.S. National Cancer Institute has used a panel of 60 diverse human cancer cell lines (the NCI-60) to screen >100,000 chemical compounds for anticancer activity. However, not all important cancer types are included in the panel, nor are drug responses of the panel predictive of clinical efficacy in patients. We asked, therefore, whether it would be possible to extrapolate from that rich database (or analogous ones from other drug screens) to predict activity in cell types not included or, for that matter, clinical responses in patients with tumors. We address that challenge by developing and applying an algorithm we term "coexpression extrapolation" (COXEN). COXEN uses expression microarray data as a Rosetta Stone for translating from drug activities in the NCI-60 to drug activities in any other cell panel or set of clinical tumors. Here, we show that COXEN can accurately predict drug sensitivity of bladder cancer cell lines and clinical responses of breast cancer patients treated with commonly used chemotherapeutic drugs. Furthermore, we used COXEN for in silico screening of 45,545 compounds and identify an agent with activity against human bladder cancer.

MicroRNA expression profiles for the NCI-60 cancer cell panel.

Advances in the understanding of cancer cell biology and response to drug treatment have benefited from new molecular technologies and methods for integrating information from multiple sources. The NCI-60, a panel of 60 diverse human cancer cell lines, has been used by the National Cancer Institute to screen >100,000 chemical compounds and natural product extracts for anticancer activity. The NCI-60 has also been profiled for mRNA and protein expression, mutational status, chromosomal aberrations, and DNA copy number, generating an unparalleled public resource for integrated chemogenomic studies. Recently, microRNAs have been shown to target particular sets of mRNAs, thereby preventing translation or accelerating mRNA turnover. To complement the existing NCI-60 data sets, we have measured expression levels of microRNAs in the NCI-60 and incorporated the resulting data into the CellMiner program package for integrative analysis. Cell line groupings based on microRNA expression were generally consistent with tissue type and with cell line clustering based on mRNA expression. However, mRNA expression seemed to be somewhat more informative for discriminating among tissue types than was microRNA expression. In addition, we found that there does not seem to be a significant correlation between microRNA expression patterns and those of known target transcripts. Comparison of microRNA expression patterns and compound potency patterns showed significant correlations, suggesting that microRNAs may play a role in chemoresistance. Combined with gene expression and other biological data using multivariate analysis, microRNA expression profiles may provide a critical link for understanding mechanisms involved in chemosensitivity and chemoresistance.

Transcript and protein expression profiles of the NCI-60 cancer cell panel: an integromic microarray study.

To evaluate the utility of transcript profiling for prediction of protein expression levels, we compared profiles across the NCI-60 cancer cell panel, which represents nine tissues of origin. For that analysis, we present here two new NCI-60 transcript profile data sets (A based on Affymetrix HG-U95 and HG-U133A chips; Affymetrix, Santa Clara, CA) and one new protein profile data set (based on reverse-phase protein lysate arrays). The data sets are available online at http://discover.nci.nih.gov in the CellMiner program package. Using the new transcript data in combination with our previously published cDNA array and Affymetrix HU6800 data sets, we first developed a "consensus set" of transcript profiles based on the four different microarray platforms. Using that set, we found that 65% of the genes showed statistically significant transcript-protein correlation, and the correlations were generally higher than those reported previously for panels of mammalian cells. Using the predictive analysis of microarray nearest shrunken centroid algorithm for functional prediction of tissue of origin, we then found that (a) the consensus mRNA set did better than did data from any of the individual mRNA platforms and (b) the protein data seemed to do somewhat better (P = 0.027) on a gene-for-gene basis in this particular study than did the consensus mRNA data, but both did well. Analysis based on the Gene Ontology showed protein levels of structure-related genes to be well predicted by mRNA levels (mean r = 0.71). Because the transcript-based technologies are more mature and are currently able to assess larger numbers of genes at one time, they continue to be useful, even when the ultimate aim is information about proteins.