Zebrafish as a model to assess cancer heterogeneity,progression and relapse

Clonal evolution is the process by which genetic and epigenetic diversity is created within malignant tumor cells. This process culminates in a heterogeneous tumor, consisting of multiple subpopulations of cancer cells that often do not contain the same underlying mutations. Continuous selective pressure permits outgrowth of clones that harbor lesions that are capable of enhancing disease progression, including those that contribute to therapy resistance, metastasis and relapse. Clonal evolution and the resulting intratumoral heterogeneity pose a substantial challenge to biomarker identification, personalized cancer therapies and the discovery of underlying driver mutations in cancer. The purpose of this Review is to highlight the unique strengths of zebrafish cancer models in assessing the roles that intratumoral heterogeneity and clonal evolution play in cancer, including transgenesis, imaging technologies, high-throughput cell transplantation approaches and in vivo single-cell functional assays.KEY WORDS: Cancer stem cell, Fluorescence, Intratumoral, Single cell, Targeted therapy, Tumor     

The hypoxic microenvironment: A determinant of cancer stem cell evolution

Tumors are often viewed as unique entities with specific behaviors. However, tumors are a mixture of differentially evolved subpopulations of cells in constant Darwinian evolution, selecting the fittest clone and allowing it to outgrow the rest. As in the natural environment, the niche defines the properties the fittest clones must possess. Therefore, there can be multiple fit clones because of the various microenvironments inside a single tumor. Hypoxia is considered to be a major feature of the tumor microenvironment and is a potential contributor to the cancer stem cell (CSC) phenotype and its enhanced tumorigenicity. The acidic microenvironment around hypoxic cells is accompanied by the activation of a subset of proteases that contribute to metastasis. Because of aberrant angiogenesis and the inaccessibility of their locations, hypoxic cells are less likely to accumulate therapeutic concentrations of chemotherapeutics that can lead to therapeutic resistance. Therefore, the targeting of the hypoxic CSC niche in combination with chemotherapy may provide a promising strategy for eradicating CSCs. In this review, we examine the cancer stem cell hypothesis and its relationship to the microenvironment, specifically to hypoxia and the subsequent metabolic switch and how they shape tumor behavior.     

VDJ-Seq: Deep Sequencing Analysis of Rearranged Immunoglobulin Heavy Chain Gene to Reveal Clonal Evolution Patterns of B Cell Lymphoma

Understanding tumor clonality is critical to understanding the mechanisms involved in tumorigenesis and disease progression. In addition, understanding the clonal composition changes that occur within a tumor in response to certain micro-environment or treatments may lead to the design of more sophisticated and effective approaches to eradicate tumor cells. However, tracking tumor clonal sub-populations has been challenging due to the lack of distinguishable markers. To address this problem, a VDJ-seq protocol was created to trace the clonal evolution patterns of diffuse large B cell lymphoma (DLBCL) relapse by exploiting VDJ recombination and somatic hypermutation (SHM), two unique features of B cell lymphomas.In this protocol, Next-Generation sequencing (NGS) libraries with indexing potential were constructed from amplified rearranged immunoglobulin heavy chain (IgH) VDJ region from pairs of primary diagnosis and relapse DLBCL samples. On average more than half million VDJ sequences per sample were obtained after sequencing, which contain both VDJ rearrangement and SHM information. In addition, customized bioinformatics pipelines were developed to fully utilize sequence information for the characterization of IgH-VDJ repertoire within these samples. Furthermore, the pipeline allows the reconstruction and comparison of the clonal architecture of individual tumors, which enables the examination of the clonal heterogeneity within the diagnosis tumors and deduction of clonal evolution patterns between diagnosis and relapse tumor pairs. When applying this analysis to several diagnosis-relapse pairs, we uncovered key evidence that multiple distinctive tumor evolutionary patterns could lead to DLBCL relapse. Additionally, this approach can be expanded into other clinical aspects, such as identification of minimal residual disease, monitoring relapse progress and treatment response, and investigation of immune repertoires in non-lymphoma contexts.     

Epigenetically regulated clonal heritability of CTA expression profiles in human melanoma

The intratumoral heterogeneity of cancer testis antigens (CTA) expression, which is driven by promoter methylation status, may hamper the effectiveness of CTA‐directed vaccination of melanoma patients. Thus, we investigated whether the intratumoral heterogeneity of CTA expression is inherited at cellular level, or evolves throughout cellular replication, leading to a phenotypically unstable tumor cell population with reduced immunogenicity and/or able to escape immune control. Utilizing a previously characterized ex vivo clonal model of intratumoral heterogeneity of CTA expression in melanoma, Mel 313 MAGE‐A3‐low clone 5 (clone 5^M3‐low) and MAGE‐A3‐high clone 14 (clone 14^M3‐high) were sub‐cloned and analyzed for CTA profile. Molecular assays demonstrated that levels of MAGE‐A3 expression were highly conserved among generated sub‐clones, as compared to parental clones. A similar behavior was identified for an extensive panel of other CTA investigated. Inherited levels of MAGE‐A3 expression correlated with the extent of promoter methylation among clone 5^M3‐low and clone 14^M3‐high sub‐clones analyzed. Treatment of clone 5^M3‐low with a DNA hypomethylating agent (DHA) resulted in an up‐regulated expression of MAGE‐A3, which was inherited at single cell level, being still detectable at day 60 in its sub‐clones. Bisulfite sequencing demonstrated that also MAGE‐A3 promoter methylation status was inherited among sub‐clones generated from DHA‐treated clone 5^M3‐low and strictly correlated with MAGE‐A3 expression levels in investigated sub‐clones. Similar results were obtained for additional CTA studied. Altogether our findings demonstrate that constitutive and DHA‐modified CTA profiles of melanoma cells are clonally inherited throughout cellular replications, thus providing relevant insights to improve the effectiveness of CTA‐based immunotherapy. J. Cell. Physiol. 223: 352–358, 2010. © 2010 Wiley‐Liss, Inc.     

Hormone resistance and neuroendocrine differentiation due to accumulation of genetic lesions during clonal evolution of prostate cancer

Progression of malignant tumors is largely due to clonal evolution of the primary tumor, clones acquiring different sets of molecular genetic lesions. Lesions can confer a selective advantage in proliferation rate or metastasis on the tumor cell population, especially if developing resistance to anticancer therapy. Prostate cancer (PCa) provides an illustrative example of clinically significant clonal evolution. The review considers the genetic alterations that occur in primary PCa and the mechanism whereby hormone-refractory PCa develops on hormone therapy, including mutations and alternative splicing of the androgen receptor gene (AR) and intratumoral androgen synthesis. Certain molecular genetic lesions determine resistance to new generation inhibitors (AR mutations that block the antagonist effect or allow other hormones to activate the receptor) or lead to neuroendocrine differentiation (repression of the AR signaling pathway, TP53 mutations, and amplification of the AURKA or MYCN oncogene). Multistep therapy based on the data about somatic mutations associated with progression and metastasis of the primary tumor can be expected to significantly improve the survival of patients with advanced PCa in the nearest future.     

scDPN for High-throughput Single-cell CNV Detection to Uncover Clonal Evolution During HCC Recurrence

Single-cell genomics provides substantial resources for dissecting cellular heterogeneity and cancer evolution. Unfortunately, classical DNA amplification-based methods have low throughput and introduce coverage bias during sample preamplification. We developed a single-cell DNA library preparation method without preamplification in nanolitre scale (scDPN) to address these issues. The method achieved a throughput of up to 1800 cells per run for copy number variation (CNV) detection. Also, our approach demonstrated a lower level of amplification bias and noise than the multiple displacement amplification (MDA) method and showed high sensitivity and accuracy for cell line and tumor tissue evaluation. We used this approach to profile the tumor clones in paired primary and relapsed tumor samples of hepato-cellular carcinoma (HCC). We identified three clonal subpopulations with a multitude of aneuploid alterations across the genome. Furthermore, we observed that a minor clone of the primary tumor containing additional alterations in chro-mosomes 1q, 10q, and 14q developed into the dominant clone in the recurrent tumor, indicating clonal selection during recurrence in HCC. Overall, this approach provides a comprehensive and scalable solution to understand genome hetero-geneity and evolution.     

Rapid Phenotypic and Genomic Change in Response to Therapeutic Pressure in Prostate Cancer Inferred by High Content Analysis of Single Circulating Tumor Cells

Timely characterization of a cancer''s evolution is required to predict treatment efficacy and to detect resistance early. High content analysis of single Circulating Tumor Cells (CTCs) enables sequential characterization of genotypic, morphometric and protein expression alterations in real time over the course of cancer treatment. This concept was investigated in a patient with castrate-resistant prostate cancer progressing through both chemotherapy and targeted therapy. In this case study, we integrate across four timepoints 41 genome-wide copy number variation (CNV) profiles plus morphometric parameters and androgen receptor (AR) protein levels. Remarkably, little change was observed in response to standard chemotherapy, evidenced by the fact that a unique clone (A), exhibiting highly rearranged CNV profiles and AR+ phenotype was found circulating before and after treatment. However, clinical response and subsequent progression after targeted therapy was associated with the drastic depletion of clone A, followed by the sequential emergence of two distinct CTC sub-populations that differed in both AR genotype and expression phenotype. While AR- cells with flat or pseudo-diploid CNV profiles (clone B) were identified at the time of response, a new tumor lineage of AR+ cells (clone C) with CNV altered profiles was detected during relapse. We showed that clone C, despite phylogenetically related to clone A, possessed a unique set of somatic CNV alterations, including MYC amplification, an event linked to hormone escape. Interesting, we showed that both clones acquired AR gene amplification by deploying different evolutionary paths. Overall, these data demonstrate the timeframe of tumor evolution in response to therapy and provide a framework for the multi-scale analysis of fluid biopsies to quantify and monitor disease evolution in individual patients.     

Addressing heterogeneity of individual blood cancers: the need for single cell analysis

Cancer heterogeneity is a significant factor in response to treatment and escape leading to relapse. Within an individual cancer, especially blood cancers, there exists multiple subclones as well as distinct clonal expansions unrelated to the clinically detected, dominant clone. Over time, multiple subclones and clones undergo emergence, expansion, and extinction. Although sometimes this intra-clonal and inter-clonal heterogeneity can be detected and/or quantified in tests that measure aggregate populations of cells, frequently, such heterogeneity can only be detected using single cell analysis to determine its frequency and to detect minor clones that may subsequently emerge to become drug resistant and dominant. Most genetic/genomic tests look at the pooled tumor population as a whole rather than at its individual cellular components. Yet, minor clones and cancer stem cells are unlikely to be detected against the background of expanded major clones. Because selective pressures are likely to govern much of what is seen clinically, single cell analysis allows identification of otherwise cryptic compartments of the malignancy that may ultimately mediate progression and relapse. Single cell analysis can track intra- or inter-clonal heterogeneity and provide useful clinical information, often before changes in the disease are detectable in the clinic. To a very limited extent, single cell analysis has already found roles in clinical care. Because inter- and intra-clonal heterogeneity likely occurs more frequently than can be currently appreciated on a clinical level, future use of single cell analysis is likely to have profound clinical utility.     

Solving the puzzle of metastasis: the evolution of cell migration in neoplasms

Background

Metastasis represents one of the most clinically important transitions in neoplastic progression. The evolution of metastasis is a puzzle because a metastatic clone is at a disadvantage in competition for space and resources with non-metastatic clones in the primary tumor. Metastatic clones waste some of their reproductive potential on emigrating cells with little chance of establishing metastases. We suggest that resource heterogeneity within primary tumors selects for cell migration, and that cell emigration is a by-product of that selection.

Methods and Findings

We developed an agent-based model to simulate the evolution of neoplastic cell migration. We simulated the essential dynamics of neoangiogenesis and blood vessel occlusion that lead to resource heterogeneity in neoplasms. We observed the probability and speed of cell migration that evolves with changes in parameters that control the degree of spatial and temporal resource heterogeneity. Across a broad range of realistic parameter values, increasing degrees of spatial and temporal heterogeneity select for the evolution of increased cell migration and emigration.

Conclusions

We showed that variability in resources within a neoplasm (e.g. oxygen and nutrients provided by angiogenesis) is sufficient to select for cells with high motility. These cells are also more likely to emigrate from the tumor, which is the first step in metastasis and the key to the puzzle of metastasis. Thus, we have identified a novel potential solution to the puzzle of metastasis.     

Evolutionary dynamics of two related malignant plasma cell lines

Cancer is the consequence of sequential acquisition of mutations within somatic cells. Mutations alter the relative reproductive fitness of cells, enabling the population to evolve in time as a consequence of selection. Cancer therapy itself can select for or against specific subclones. Given the large population of tumor cells, subclones inevitably emerge and their fate will depend on the evolutionary dynamics that define the interactions between such clones. Using a combination of in vitro studies and mathematical modeling, we describe the dynamic behavior of two cell lines isolated from the same patient at different time points of disease progression and show how the two clones relate to one another. We provide evidence that the two clones coexisted at the time of initial presentation. The dominant clone presented with biopsy-proven cardiac AL amyloidosis. Initial therapy selected for the second clone that expanded leading to a change in the diagnosis to multiple myeloma. The evolutionary dynamics relating the two cell lines are discussed and a hypothesis is generated in regard to the mechanism of one of the phenotypic characteristics that is shared by these two cell lines.Key words: multiple myeloma, amyloidosis, chemotherapy, clonal evolution, selection     

Evolution of Fitness in Experimental Populations of Vesicular Stomatitis Virus

The evolution of fitness in experimental clonal populations of vesicular stomatitis virus (VSV) has been compared under different genetic (fitness of initial clone) and demographic (population dynamics) regimes. In spite of the high genetic heterogeneity among replicates within experiments, there is a clear effect of population dynamics on the evolution of fitness. Those populations that went through strong periodic bottlenecks showed a decreased fitness in competition experiments with wild type. Conversely, mutant populations that were transferred under the dynamics of continuous population expansions increased their fitness when compared with the same wild type. The magnitude of the observed effect depended on the fitness of the original viral clone. Thus, high fitness clones showed a larger reduction in fitness than low fitness clones under dynamics with included periodic bottleneck. In contrast, the gain in fitness was larger the lower the initial fitness of the viral clone. The quantitative genetic analysis of the trait ``fitness' in the resulting populations shows that genetic variation for the trait is positively correlated with the magnitude of the change in the same trait. The results are interpreted in terms of the operation of MULLER's ratchet and genetic drift as opposed to the appearance of beneficial mutations.     

Report on the Eleventh International Workshop on the Identification of Transcribed Sequences 2001. November 9-11, 2001. Washington, DC, USA

Glioblastoma multiforme (GBM) is characterized by intratumoral heterogeneity as to both histomorphology and genetic changes, displaying a wide variety of numerical chromosome aberrations the most common of which are monosomy 10 and trisomy 7. Moreover, GBM in vitro are known to have variable karyotypes within a given tumor cell culture leading to rapid karyotype evolution through a high incidence of secondary numerical chromosome aberrations. The aim of our study was to investigate to what extent this mitotic instability of glioblastoma cells is also present in vivo. We assessed the spatial distribution patterns of numerical chromosome aberrations in vivo in a series of 24 GBM using two-color in situ hybridization for chromosomes 7/10, 8/17, and 12/18 on consecutive 6-microm paraffin-embedded tissue slides. The chromosome aberration patterns were compared with the histomorphology of the investigated tumor assessed from a consecutive HE-stained section, and with the in vitro karyotype of cell cultures established from the tumors. All investigated chromosomes showed mitotic instability, i.e., numerical aberrations within significant amounts of tumor cells in a scattered distribution through the tumor tissue. As to chromosomes 10 and 17, only monosomy occurred, as to chromosome 7 only trisomy/polysomy, apparently as a result of selection in favor of the respective aberration. Conversely, chromosomes 8, 12, and 18 displayed scattered patterns of monosomy as well as trisomy within a given tumor reflecting a high mitotic error rate without selective effects. The karyotypes of the tumor cell cultures showed less variability of numerical aberrations apparently due to clonal adaptation to in vitro conditions. We conclude that glioblastoma cells in vivo are characterized by an extensive tendency to mitotic errors. The resulting clonal diversity of chromosomally aberrant cells may be an important biological constituent of the well-known ability of glioblastomas to preserve viable tumor cell clones under adaptive stress in vivo, in clinical terms to rapidly recur after antitumoral therapy including radio- or chemotherapy.     

Advanced Intestinal Cancers often Maintain a Multi-Ancestral Architecture

A widely accepted paradigm in the field of cancer biology is that solid tumors are uni-ancestral being derived from a single founder and its descendants. However, data have been steadily accruing that indicate early tumors in mice and humans can have a multi-ancestral origin in which an initiated primogenitor facilitates the transformation of neighboring co-genitors. We developed a new mouse model that permits the determination of clonal architecture of intestinal tumors in vivo and ex vivo, have validated this model, and then used it to assess the clonal architecture of adenomas, intramucosal carcinomas, and invasive adenocarcinomas of the intestine. The percentage of multi-ancestral tumors did not significantly change as tumors progressed from adenomas with low-grade dysplasia [40/65 (62%)], to adenomas with high-grade dysplasia [21/37 (57%)], to intramucosal carcinomas [10/23 (43%]), to invasive adenocarcinomas [13/19 (68%)], indicating that the clone arising from the primogenitor continues to coexist with clones arising from co-genitors. Moreover, neoplastic cells from distinct clones within a multi-ancestral adenocarcinoma have even been observed to simultaneously invade into the underlying musculature [2/15 (13%)]. Thus, intratumoral heterogeneity arising early in tumor formation persists throughout tumorigenesis.     

Mechanism of relapse in pediatric acute lymphoblastic leukemia

Relapse following initial chemotherapy remains a barrier to survival in approximately 20% of children suffering from acute lymphoblastic leukemia (ALL). Recently, to investigate the mechanism of relapse, we analysed clonal populations in 27 pairs of matched diagnosis and relapse ALL samples using PCR-based detection of multiple antigen receptor gene rearrangements. These clonal markers revealed the emergence of apparently new populations at relapse in 13 patients. In those cases where the new ‘relapse clone’ could be detected in the diagnosis population, there was a close correlation between length of first remission and quantity of the relapse clone in the diagnosis sample. A shorter length of time to first relapse correlated with a higher quantity of the relapsing clone at diagnosis. This observation, together with demonstrated differential chemosensitivity between sub-clones at diagnosis, indicates that relapse in ALL patients may commonly involve selection of a minor intrinsically resistant sub-clone that is undetectable by routine PCR-based methods. From a clinical perspective, relapse prediction may be improved with strategies to detect minor potentially resistant sub-clones early during treatment, hence allowing intensification of therapy. Together with the availability of relevant in vivo experimental models and powerful technology for detailed analysis of patient specimens, this new information will help shape future experimentation towards targeted therapy for high-risk ALL.     

Expression of enzymes and oncogene induced after radiotherapy and/or chemotherapy in patients with brain tumors.

The relationship between the degree of the expression of Cu/Zn SOD, GST-pi and bcl-2 in the initial and recurrent tumor tissue after radiotherapy and/or chemotherapy and the cellular heterogeneity obtained from DNA content by image cytometry was investigated. Subjects were 7 patients who had glial tumors which were surgically removed at onset and removed a second time at recurrence. Radiotherapy and chemotherapy were also administered after initial resection. Immunoreactivity for copper/zinc super oxide dismutase (Cu/Zn SOD), GST (glutathione-S-transferase)-pi, and bcl-2 were evaluated from routinely prepared tissue blocks. Tumors were classified into two groups by cytometric analysis of DNA ploidy in the G2M cell cycle phase. One tumor group consisted of single clonal cells in both the initial and recurrent tumors and the other group consisted of tumors with polyclonal cells in the initial and recurrent tumor. In this study, one patient (case 3) with single clonal cell glioblastoma at recurrence did not show high Cu/Zn SOD activity after radiotherapy and chemotherapy but showed a short survival time after recurrence. In three patients (cases 1, 2, 3) with single clonal-cell glioblastoma, the recurrent tumor cells showed high GST-pi immunoreactivity and survival time was short after recurrence. Tumor cells in two patients (cases 5, 7) with single clonal cell anaplastic glioma at recurrence, showed high GST-pi immunoreactivity and had a short survival time after recurrence. In three single clonal glioblastomas (cases 1, 2, 3), the recurrent tumor showed the increased bcl-2 immunoreactivity and showed a short survival time after recurrence. In two patients (case 5, 7) with single clonal cell anaplastic glioma at recurrence, tumor cells showed a high bcl-2 immunoreactivity and these patients showed a short survival time after recurrence. Although the number of subjects is very small, our study shows that the immunoreactivity of bcl-2 and GST-pi in malignant gliomas may be very important factors in radio- and chemosensitivity, and shows that GST-pi is induced by radiation and anti-cancer drugs.     

Clonal variation and aging of diploid fibroblasts : Cinematographic studies of cell pedigrees

Time-lapse cinematographic (TLC) analysis of clones of human diploid fibroblasts indicate heterogeneity in clonal division behaviour. Variations are noted in interdivision time, clone size and generations per clone. Correlation coefficients for interdivision times of sister pairs are high in young clones and generally low in aged clones. A consistent division pattern at all population doubling levels is one of low average interdivision time for early and late generations of a clone and high average interdivision time for the middle range of generations of a clone. The clonal division patterns observed experimentally have been duplicated in computer simulated pedigrees. The computer model is based on an oscillating system which allows for flux of regulator substances. The critical concentrations of regulator substances determine the clonal division pattern for a given progenitor cell.     

Phenomenon of evolution of clonal chromosomal abnormalities in childhood acute myeloid leukemia

An analysis of chromosomal abnormalities in bone-marrow cells was performed in 116 children with diagnoses of acute myeloid leukemia (AML). The frequency of the evolution of clonal chromosomal abnormalities in AML constituted 42.3%. Quantitative abnormalities of chromosomes 8, 9, and 21, as well as the secondary structural abnormalities in the chromosomal regions 12p12, 9p22, 9q22, 9q34, 11q14–23, and 6q2, were the most abundant. Quantitative abnormalities were registered in 26.7% cases. The basic mechanism of evolution of the leukemic clone contained trisomy, deletions, and monosomy. The frequency of evolution was seven times higher in the age group of up to 2 years and twofold higher in the age group of up to 5 years. The high frequency of evolution at t(15;17)(q22;q22) was established, while its absence was revealed at inv(16)(p13q22). Patients with clonal evolution were characterized by the increased frequency of relapses and earlier death before reaching remission, which might be explained by the severe initial state of those patients. The conception of abnormalities in the evolution of the clone was proposed to occur at certain stages as follows: (1) appearance of balanced rearrangements; (2) trisomy occurrence; (3) loss of chromosomal material. The occurrence of an unbalanced genome during evolution possesses advantages in the clonal proliferate activity and may be related to its response to chemotherapy. An identity in abnormal chromosomal structure was revealed as a result of the comparison of karyotypes during diagnostics and during relapse, which could be evidence of the initial induction of some types of evolution of chromosomal abnormalities in leukemic cells in AML children by the chemical agents.     

Intratumoral heterogeneity: Clonal cooperation in epithelial-to-mesenchymal transition and metastasis

Although phenotypic intratumoral heterogeneity was first described many decades ago, the advent of next-generation sequencing has provided conclusive evidence that in addition to phenotypic diversity, significant genotypic diversity exists within tumors. Tumor heterogeneity likely arises both from clonal expansions, as well as from differentiation hierarchies existent in the tumor, such as that established by cancer stem cells (CSCs) and non-CSCs. These differentiation hierarchies may arise due to genetic mutations, epigenetic alterations, or microenvironmental influences. An additional differentiation hierarchy within epithelial tumors may arise when only a few tumor cells trans-differentiate into mesenchymal-like cells, a process known as epithelial-to-mesenchymal transition (EMT). Again, this process can be influenced by both genetic and non-genetic factors. In this review we discuss the evidence for clonal interaction and cooperation for tumor maintenance and progression, particularly with respect to EMT, and further address the far-reaching effects that tumor heterogeneity may have on cancer therapy.     

Tumorigenic,invasive, karyotypic,and immunocytochemical characteristics of clonal cell lines derived from a spontaneous canine anaplastic astrocytoma

Summary Tumor cells from a spontaneously arising canine astrocytoma were isolated and cloned. Three clonally derived cell lines (DL3580 clone 1, DL3580 clone 2, and DL3580 clone 3) were developed and found to express glial fibrillary acidic protein (GFAP) as well as epidermal growth factor receptor (EGFR/c-erbB₁). The cell lines were tumorigenic as subcutaneous xenografts or as intracranial implants in athymic mice, or both. Both the monolayer astrocytoma cells and the xenograft tumor cells from clone 2 were aneuploid, with a modal number of 84 chromosomes per metaphase; clones 1 and 3 were also aneuploid with modal numbers of 82 and 75/79, respectively. The histology of both the initial spontaneously occurring tumor in the dog and the intracranial astrocytoma in athymic mice demonstrated features of diffuse infiltration into normal brain. These newly developed canine glioma cell lines are karyotypically stable for 1 yr in culture and carry the same marker chromosomes as the parental lines. These glioma cell lines may serve as models for investigating mechanisms of glioma invasion into brain. Additionally, clonal cell lines with divergent properties isolated from the same tumor may assist in studies of the molecular basis of astrocytoma progression and heterogeneity.