Statistical analysis of cytogenetic abnormalities in human cancer cells

Variable selection for disease progression models: methods for oncogenetic trees and application to cancer and HIV

Background

Disease progression models are important for understanding the critical steps during the development of diseases. The models are imbedded in a statistical framework to deal with random variations due to biology and the sampling process when observing only a finite population. Conditional probabilities are used to describe dependencies between events that characterise the critical steps in the disease process.

Many different model classes have been proposed in the literature, from simple path models to complex Bayesian networks. A popular and easy to understand but yet flexible model class are oncogenetic trees. These have been applied to describe the accumulation of genetic aberrations in cancer and HIV data. However, the number of potentially relevant aberrations is often by far larger than the maximal number of events that can be used for reliably estimating the progression models. Still, there are only a few approaches to variable selection, which have not yet been investigated in detail.

Results

We fill this gap and propose specifically for oncogenetic trees ten variable selection methods, some of these being completely new. We compare them in an extensive simulation study and on real data from cancer and HIV. It turns out that the preselection of events by clique identification algorithms performs best. Here, events are selected if they belong to the largest or the maximum weight subgraph in which all pairs of vertices are connected.

Conclusions

The variable selection method of identifying cliques finds both the important frequent events and those related to disease pathways.

Electronic supplementary material

The online version of this article (doi:10.1186/s12859-017-1762-1) contains supplementary material, which is available to authorized users.

Methods and challenges in timing chromosomal abnormalities within cancer samples

Motivation: Tumors acquire many chromosomal amplifications, and those acquired early in the lifespan of the tumor may be not only important for tumor growth but also can be used for diagnostic purposes. Many methods infer the order of the accumulation of abnormalities based on their occurrence in a large cohort of patients. Recently, Durinck et al. (2011) and Greenman et al. (2012) developed methods to order a single tumor’s chromosomal amplifications based on the patterns of mutations accumulated within those regions. This method offers an unprecedented opportunity to assess the etiology of a single tumor sample, but has not been widely evaluated.

Results: We show that the model for timing chromosomal amplifications is limited in scope, particularly for regions with high levels of amplification. We also show that the estimation of the order of events can be sensitive for events that occur early in the progression of the tumor and that the partial maximum likelihood method of Greenman et al. (2012) can give biased estimates, particularly for moderate read coverage or normal contamination. We propose a maximum-likelihood estimation procedure that fully accounts for sequencing variability and show that it outperforms the partial maximum-likelihood estimation method. We also propose a Bayesian estimation procedure that stabilizes the estimates in certain settings. We implement these methods on a small number of ovarian tumors, and the results suggest possible differences in how the tumors acquired amplifications.

Availability and implementation: We provide implementation of these methods in an R package cancerTiming, which is available from the Comprehensive R Archive Network (CRAN) at http://CRAN.R-project.org/.

Contact:epurdom@stat.Berkeley.edu

Supplementary information:Supplementary data are available at Bioinformatics online.

-8p12-23 and +20q are predictors of subtypes and metastatic pathways in colorectal cancer: construction of tree models using comparative genomic hybridization data

A substantial body of evidence suggests the genetic heterogeneous pattern and multiple pathways in colorectal cancer initiation and progression. In this study, we construct a branching tree and multiple distance-based tree models to elucidate these genetic patterns and pathways in colorectal cancer by using a data set comprised of 244 cases of comparative genomic hybridization. We identify the six most common gains of chromosomal regions of 7p (37.0%), 7q11-32 (34.8%), 8q (48.3%), 13q (49.1%), 20p (36.1%), and 20q (50.4%), and the nine most common losses of 1p13-36 (30.9%), 4p15 (24.3%), 4q33-34 (24.3%), 8p12-23 (50.9%), 15q13-14 (23.5%), 15q24-25 (24.3%), 17p (34.8%), 18p (36.5%), and 18q (61.7%) in colorectal cancer. We classify colorectal cancer into two distinct groups: one preceding with -8p12-23, and the other with +20q. The sample-based classification tree also demonstrates that colorectal cancer can be classified into multiple subtypes marked by -8p12-23 and +20q. By comparing chromosomal abnormalities between primary and metastatic colorectal cancer, we identify five potential metastatic pathways: (-18q, -18p), (-8p12-23, -4p15, -4q33-34), (+20q, +20p), (+20q, +7p, +7q11-32), and +8q. -8p12-23 and +20q are inferred to be the two marker events of colorectal cancer metastasis. The current oncogenetic tree models may contribute to our understanding towards molecular genetics in colorectal cancer. Particularly, the metastatic pathways we describe may provide pivotal clues for metastatic candidate genes, and thus impact on the prediction and intervention of metastatic colorectal cancer.

The correlation pattern of acquired copy number changes in 164 ETV6/RUNX1-positive childhood acute lymphoblastic leukemias

The ETV6/RUNX1 fusion gene, present in 25% of B-lineage childhood acute lymphoblastic leukemia (ALL), is thought to represent an initiating event, which requires additional genetic changes for leukemia development. To identify additional genetic alterations, 24 ETV6/RUNX1-positive ALLs were analyzed using 500K single nucleotide polymorphism arrays. The results were combined with previously published data sets, allowing us to ascertain genomic copy number aberrations (CNAs) in 164 cases. In total, 45 recurrent CNAs were identified with an average number of 3.5 recurrent changes per case (range 0-13). Twenty-six percent of cases displayed a set of recurrent CNAs identical to that of other cases in the data set. The majority (74%), however, displayed a unique pattern of recurrent CNAs, indicating a large heterogeneity within this ALL subtype. As previously demonstrated, alterations targeting genes involved in B-cell development were common (present in 28% of cases). However, the combined analysis also identified alterations affecting nuclear hormone response (24%) to be a characteristic feature of ETV6/RUNX1-positive ALL. Studying the correlation pattern of the CNAs allowed us to highlight significant positive and negative correlations between specific aberrations. Furthermore, oncogenetic tree models identified ETV6, CDKN2A/B, PAX5, del(6q) and +16 as possible early events in the leukemogenic process.

Quantifying cancer progression with conjunctive Bayesian networks

MOTIVATION

Cancer is an evolutionary process characterized by accumulating mutations. However, the precise timing and the order of genetic alterations that drive tumor progression remain enigmatic.

RESULTS

We present a specific probabilistic graphical model for the accumulation of mutations and their interdependencies. The Bayesian network models cancer progression by an explicit unobservable accumulation process in time that is separated from the observable but error-prone detection of mutations. Model parameters are estimated by an Expectation-Maximization algorithm and the underlying interaction graph is obtained by a simulated annealing procedure. Applying this method to cytogenetic data for different cancer types, we find multiple complex oncogenetic pathways deviating substantially from simplified models, such as linear pathways or trees. We further demonstrate how the inferred progression dynamics can be used to improve genetics-based survival predictions which could support diagnostics and prognosis.

AVAILABILITY

The software package ct-cbn is available under a GPL license on the web site cbg.ethz.ch/software/ct-cbn

CONTACT

moritz.gerstung@bsse.ethz.ch.

Construction of oncogenetic tree models reveals multiple pathways of oral cancer progression

Oral cancer develops and progresses by accumulation of genetic alterations. The interrelationship between these alterations and their sequence of occurrence in oral cancers has not been thoroughly understood. In the present study, we applied oncogenetic tree models to comparative genomic hybridization (CGH) data of 97 primary oral cancers to identify pathways of progression. CGH revealed the most frequent gains on chromosomes 8q (72.4%) and 9q (41.2%) and frequent losses on 3p (49.5%) and 8p (47.5%). Both mixture and distance-based tree models suggested multiple progression pathways and identified +8q as an early event. The mixture model suggested two independent pathways namely a major pathway with -8p and a less frequent pathway with +9q. The distance-based tree identified three progression pathways, one characterized by -8p, another by -3p and the third by alterations +11q and +7p. Differences were observed in cytogenetic pathways of node-positive and node-negative oral cancers. Node-positive cancers were characterized by more non-random aberrations (n = 11) and progressed via -8p or -3p. On the other hand, node-negative cancers involved fewer non-random alterations (n = 6) and progressed along -3p. In summary, the tree models for oral cancers provided novel information about the interactions between genetic alterations and predicted their probable order of occurrence.

[Exploration of carcinogenesis based on tree models using CGH data]

Comparative genomic hybridization (CGH) can detect chromosomal deletions and amplifications of tumors, and various laboratories and public databases have accumulated a large number of CGH data, providing the opportunity to analyze the molecular mechanism of tumorigenesis in the whole genome. Tree models are generally used to study the history of biological formation and evolution in the field of bioinformatics, and evolutionary relationships between species are usually represented using phylogenetic tree. Tree models are also powerful bioinformatics tools to analyze CGH data and explore carcinogenesis. Two common tree models, the branching tree and the distanced-based tree, as well as their basic principles, methods are introduced detailedly, several technical problems in construction of tree models are discussed, and their applications in cancer research are reviewed systematically in this paper. As a generalization of single path linear model, tree models can more accurately conclude multigene, multistep, multipathway process of tumorigenesis, exploring the molecular mechanism of tumorigenesis from different angels. Apart from CGH data, tree models can be used to analyze various types of data, including high-resolution data (e.g., array-CGH data).

Construction and analysis of tree models for chromosomal classification of diffuse large B-cell lymphomas

AIM: To construct tree models for classification of diffuse large B-cell lymphomas (DLBCL) by chromosome copy numbers, to compare them with cDNA microarray classification, and to explore models of multi-gene, multi-step and multi-pathway processes of DLBCL tumorigenesis.

METHODS: Maximum-weight branching and distance-based models were constructed based on the comparative genomic hybridization (CGH) data of 123 DLBCL samples using the established methods and software of Desper et al. A maximum likelihood tree model was also used to analyze the data. By comparing with the results reported in literature, values of tree models in the classification of DLBCL were elucidated.

RESULTS: Both the branching and the distance-based trees classified DLBCL into three groups. We combined the classification methods of the two models and classified DLBCL into three categories according to their characteristics. The first group was marked by +Xq, +Xp, -17p and +13q; the second group by +3q, +18q and +18p; and the third group was marked by -6q and +6p. This chromosomal classification was consistent with cDNA classification. It indicated that -6q and +3q were two main events in the tumorigenesis of lymphoma.

CONCLUSION: Tree models of lymphoma established from CGH data can be used in the classification of DLBCL. These models can suggest multi-gene, multi-step and multi-pathway processes of tumorigenesis. Two pathways, -6q preceding +6q and +3q preceding +18q, may be important in understanding tumorigenesis of DLBCL. The pathway, -6q preceding +6q, may have a close relationship with the tumorigenesis of non-GCB DLBCL.

Deriving evolutionary tree models of the oncogenesis of endometrial adenocarcinoma

Endometrial adenocarcinoma (EAC) is the fourth leading cause of cancer death in women worldwide, but not much is known about the underlying genetic factors involved in the development of this complex disease. In the present work, we used 3 different algorithms to derive tree models of EAC oncogenesis from data on the frequencies of genomic alterations in rat chromosome 10 (RNO10). The tumor material was derived from progenies of crosses between the EAC susceptible BDII inbred rat strain and two non susceptible inbred rat strains. Data from allelic imbalance scans of RNO10 with microsatellite markers on solid tumor material and corresponding tissue cultures were used. For the analysis, RNO10 was divided into 24 segments containing a total of 59 informative microsatellite markers. The derived tree models show that genomic alterations have occurred in 11 of the 24 segments. In addition, the models provide information about the likely order of the alterations as well as their relationship with each other. Interestingly, there was a high degree of consistency among the different tree models and with the results of previous studies, which supports the reliability of the tree models. Our results may be extended into a general approach for tree modeling of whole genome alterations during oncogenesis.

A tumor progression model for hepatocellular carcinoma: bioinformatic analysis of genomic data

BACKGROUND & AIMS

It is widely recognized that genomic abnormalities underpin the development of human cancers. Aberrant patterns of chromosomal changes may represent useful information that can be used in classifying the complex traits of liver cancer cases for the genetic events involved in tumor carcinogenesis, tumor progression, and prognosis.

METHODS

Genome-wide chromosomal aberrations of 158 hepatitis B virus-associated hepatocellular carcinoma (HCC) were studied by comparative genomic hybridization (CGH). By application of a self-organizing tree algorithm, statistically significant CGH events were used to construct an evolutionary tree that could infer patient subgroups with different degrees of tumor progression. The key CGH events in the subgroups were identified. The clinical significance of the groupings and the key CGH events were examined.

RESULTS

Based on the patterns of significant chromosomal aberrations derived, 3 HCC subgroups organized in an evolutionary tree were identified. The groupings possessed information reflecting the degrees of tumor progression, including numbers of chromosomal aberrations, tumor stages, tumor sizes, and disease outcome. Gains of 1q21-23 and 8q22-24 were identified as genomic events associated with the early development of HCC. Gain of 3q22-24, however, was identified as 1 of the late genomic events found to be associated with tumor recurrence and poor overall patient survival.

CONCLUSIONS

A tumor progression model for HCC was constructed and revealed chromosomal imbalances that were significantly associated with clinical pathologic characteristics of the disease. This model explains a significant part of the variations in clinical outcome among HCC patients.

Construction of evolutionary tree models for nasopharyngeal carcinoma using comparative genomic hybridization data

Nasopharyngeal carcinoma (NPC) is a malignancy associated with remarkable racial and geographic factors. The development and progression of NPC may involve the accumulation of multiple genetic alterations over a long time. For understanding the putative order of genetic alteration in NPC tumorigenesis, we used evolutionary tree models (branching and distance-based tree models) to analyze comparative genomic hybridization (CGH) data of previously published NPC cases (n = 103). Consistent loss of 3p for both tree models was an important early event in NPC progression. Chromosome 12 gain was another important early event, and may represent a subclass different from 3p- derived subclasses of NPC. The tree models also suggested that at least two subclasses of 3p- derived NPC, one marked by 1q+, 9p-, and 13q- and the other marked by 14q-, 16q-, 9q-, and 1p-.

Subset clustering of binary sequences, with an application to genomic abnormality data

This article develops a model-based approach to clustering multivariate binary data, in which the attributes that distinguish a cluster from the rest of the population may depend on the cluster being considered. The clustering approach is based on a multivariate Dirichlet process mixture model, which allows for the estimation of the number of clusters, the cluster memberships, and the cluster-specific parameters in a unified way. Such a clustering approach has applications in the analysis of genomic abnormality data, in which the development of different types of tumors may depend on the presence of certain abnormalities at subsets of locations along the genome. Additionally, such a mixture model provides a nonparametric estimation scheme for dependent sequences of binary data.

Comparative genomic hybridization analysis of astrocytomas: prognostic and diagnostic implications

Astrocytoma is comprised of a group of common intracranial neoplasms that are classified into four grades based on the World Health Organization histological criteria and patient survival. To date, histological grade, patient age, and clinical performance, as reflected in the Karnofsky score, are the most reliable prognostic predictors. Recently, there has been a significant effort to identify additional prognostic markers using objective molecular genetic techniques. We believe that the identification of such markers will characterize new chromosomal loci important in astrocytoma progression and aid clinical diagnosis and prognosis. To this end, our laboratory used comparative genomic hybridization to identify DNA sequence copy number changes in 102 astrocytomas. Novel losses of 19p loci were detected in low-grade pilocytic astrocytomas and losses of loci on 9p, 10, and 22 along with gains on 7, 19, and 20 were detected in a significant proportion of high-grade astrocytomas. The Cox proportional hazards statistical modeling showed that the presence of +7q and -10q comparative genomic hybridization alterations significantly increased a patient's risk of dying, independent of histological grade. This investigation demonstrates the efficacy of comparative genomic hybridization for identifying tumor suppressor and oncogene loci in different astrocytic grades. The cumulative effect of these loci is an important consideration in their diagnostic and prognostic implications.

Construction of tree models for pathogenesis of nasopharyngeal carcinoma

Pathogenesis of nasopharyngeal carcinoma (NPC) is a multistep and multipathway process that cannot be fully explained by a fixed linear progression model. We used distance-based and branching-tree methods to construct more general tree-like models for NPC carcinogenesis from 170 comparative genomic hybridization (CGH) samples previously published in five smaller studies. Imbalances were classified into "overlap regions," each containing the most commonly gained or lost band on each chromosome arm as well as adjacent bands that were gained or lost almost as often. The chromosome abnormalities associated with NPC were -3p26-13 (48.9%), -11q22-25 (38.1%), -16q12-24 (38.1%), -14q24-32 (32.4%), -13q21-32 (22.3%), -9p23-21(21.6%), +12p12 (46%), +12q13-15 (43.9%), +1q22-32 (33.1%), +3q13.1-26.2 (30.2%), and +8q22.1-24.2 (27.3%). NPC can be classified into two groups, one marked by +12p12 and +8q22.1-24.2 and the other by -3p26-13, -11q22-25, -14q24-32, and +1q22-32. The tree models predicted -3p26-13 and +12p12 as early events and +8q22.1-24.2 as a late event. The predictions for -3p26-13 and +8q22.1-24.2 were consistent with previous studies. The prediction for +12p12 is being reported for the first time. Many known NPC-related genes on chromosomal regions of these tree models are discussed, some of which may merit additional study. The potential applications of tree models are also discussed.

Genetic differences detected by comparative genomic hybridization in head and neck squamous cell carcinomas from different tumor sites: construction of oncogenetic trees for tumor progression

For a better understanding of genetic alterations in head and neck squamous cell carcinoma (HNSCC), we applied comparative genomic hybridization (CGH) in the analysis of 75 HNSCCs, comprised of 18 pharyngeal squamous cell carcinomas (PSCCs), 23 laryngeal squamous cell carcinomas (LSCCs), and 34 oral squamous cell carcinomas (OSCCs). The three subgroups of HNSCC showed significant differences in genetic alteration patterns. Overall, PSCC and LSCC had more copy number aberrations (CNAs) per tumor than did OSCC. Apparent differing patterns of high-level amplification were also observed. The smallest recurrent chromosomal regions of high-level amplification (> or = 15% of cases) were 7q22, 8q24.1, and 11q12-13 in PSCC and 3q26.1-29 in OSCC. According to single frequency and combined frequencies of CNAs, we concluded that the most important chromosomal events for progression of head and neck cancer were +3q, +5p, +8q, and -3p for all subgroups of HNSCC; additionally, +7q, +17q, -9p, and -13q for PSCC; +7p, +9q, +11q12-13, +14q, and +17q for LSCC; and +1p and +11q12-13 for OSCC. To identify further important genetic alterations and the relationships among the alterations, we constructed oncogenetic tree models for tumor progression of HNSCC from CGH data using branching and distance-based tree models. The tree models predicted that: (1) +3q21-29 was the most important early chromosomal event, and -3p, which occurred after +3q21-29, was also an important chromosomal event for all subsites of HNSCC; (2) +8q is the second most important early chromosomal event; (3) there may be at least three subgroups of HNSCC: one characterized by -3p, -9p, +7p, and -13q; another by +5p, +9qter, and +17p; and the other by +8q and +18p. These results suggest that different chromosomal aberrations may play a role in the initiation and/or progression of different subgroups of HNSCC.

Estimating an oncogenetic tree when false negatives and positives are present

Human solid tumors are believed to be caused by a sequence of genetic abnormalities arising in the tumor cells. The understanding of these sequences is extremely important for improving cancer treatment. Models for the occurrence of the abnormalities include linear structure and a recently proposed tree-based structure. In this paper we extend the pure oncogenetic tree model by introducing false positive and false negative observations. We state conditions sufficient for the reconstruction of the generating tree. As an example we analyze a comparative genomic hybridization data set and show that addition of the error model significantly improves the ability of the model to describe the data.

Graph models of oncogenesis with an application to melanoma

We describe several analytical techniques for use in developing genetic models of oncogenesis including: methods for the selection of important genetic events, construction of graph models (including distance-based trees, branching trees, contingency trees and directed acyclic graph models) from these events and methods for interpretation of the resulting models. The models can be used to make predictions about: which genetic events tend to occur early, which events tend to occur together and the likely order of events. Unlike simple path models of oncogenesis, our models allow dependencies to exist between specific genetic changes and allow for multiple, divergent paths in tumor progression. A variety of genetic events can be used with the graph models including chromosome breaks, losses or gains of large DNA regions, small mutations and changes in methylation. As an application of the techniques, we use a recently published cytogenetic analysis of 206 melanoma cases [Nelson et al. (2000), Cancer Genet. Cytogenet.122, 101-109] to derive graph models for chromosome breaks in melanoma. Among our predictions are: (1) breaks in 6q1 and 1q1 are early events, with 6q1 preferentially occurring first and increasing the probability of a break in 1q1 and (2) breaks in the two sets [1p1, 1p2, 9q1] and [1q1, 7p2, 9p2] tend to occur together. This study illustrates that the application of graph models to genetic data from tumor sets provide new information on the interrelationships among genetic changes during tumor progression.

Distance-based reconstruction of tree models for oncogenesis

Comparative genomic hybridization (CGH) is a laboratory method to measure gains and losses in the copy number of chromosomal regions in tumor cells. It is hypothesized that certain DNA gains and losses are related to cancer progression and that the patterns of these changes are relevant to the clinical consequences of the cancer. It is therefore of interest to develop models which predict the occurrence of these events, as well as techniques for learning such models from CGH data. We continue our study of the mathematical foundations for inferring a model of tumor progression from a CGH data set that we started in Desper et al. (1999). In that paper, we proposed a class of probabilistic tree models and showed that an algorithm based on maximum-weight branching in a graph correctly infers the topology of the tree, under plausible assumptions. In this paper, we extend that work in the direction of the so-called distance-based trees, in which events are leaves of the tree, in the style of models common in phylogenetics. Then we show how to reconstruct the distance-based trees using tree-fitting algorithms developed by researchers in phylogenetics. The main advantages of the distance-based models are that 1) they represent information about co-occurrences of all pairs of events, instead of just some pairs, 2) they allow quantitative predictions about which events occur early in tumor progression, and 3) they bring into play the extensive methodology and software developed in the context of phylogenetics. We illustrate the distance-based tree method and how it complements the branching tree method, with a CGH data set for renal cancer.

Somatic deletions in hereditary breast cancers implicate 13q21 as a putative novel breast cancer susceptibility locus

A significant proportion of familial breast cancers cannot be explained by mutations in the BRCA1 or BRCA2 genes. We applied a strategy to identify predisposition loci for breast cancer by using mathematical models to identify early somatic genetic deletions in tumor tissues followed by targeted linkage analysis. Comparative genomic hybridization was used to study 61 breast tumors from 37 breast cancer families with no identified BRCA1 or BRCA2 mutations. Branching and phylogenetic tree models predicted that loss of 13q was one of the earliest genetic events in hereditary cancers. In a Swedish family with five breast cancer cases, all analyzed tumors showed distinct 13q deletions, with the minimal region of loss at 13q21-q22. Genotyping revealed segregation of a shared 13q21 germ-line haplotype in the family. Targeted linkage analysis was carried out in a set of 77 Finnish, Icelandic, and Swedish breast cancer families with no detected BRCA1 and BRCA2 mutations. A maximum parametric two-point logarithm of odds score of 2.76 was obtained for a marker at 13q21 (D13S1308, theta = 0.10). The multipoint logarithm of odds score under heterogeneity was 3.46. The results were further evaluated by simulation to assess the probability of obtaining significant evidence in favor of linkage by chance as well as to take into account the possible influence of the BRCA2 locus, located at a recombination fraction of 0.25 from the new locus. The simulation substantiated the evidence of linkage at D13S1308 (P < 0.0017). The results warrant studies of this putative breast cancer predisposition locus in other populations.

Inferring tree models for oncogenesis from comparative genome hybridization data

Comparative genome hybridization (CGH) is a laboratory method to measure gains and losses of chromosomal regions in tumor cells. It is believed that DNA gains and losses in tumor cells do not occur entirely at random, but partly through some flow of causality. Models that relate tumor progression to the occurrence of DNA gains and losses could be very useful in hunting cancer genes and in cancer diagnosis. We lay some mathematical foundations for inferring a model of tumor progression from a CGH data set. We consider a class of tree models that are more general than a path model that has been developed for colorectal cancer. We derive a tree model inference algorithm based on the idea of a maximum-weight branching in a graph, and we show that under plausible assumptions our algorithm infers the correct tree. We have implemented our methods in software, and we illustrate with a CGH data set for renal cancer.

Lack of homozygously inactivated p73 in single-copy MYCN primary neuroblastomas and neuroblastoma cell lines

We examined 18 neuroblastoma cell lines and 32 primary single-copy MYCN tumor specimens to determine whether mutations of p73, a novel p53-related gene located in chromosome band 1p36.33, contribute to the genesis or progression of childhood neuroblastoma. By fluorescence in situ hybridization, 16 of the 18 cell lines, but only 3 of the 32 primary tumors, had evidence of a deleted p73 allele. Sequence analysis of the p73 coding region in the mRNAs expressed by these cell lines and tumors did not reveal inactivating mutations, suggesting that p73 is not homozygously inactivated in neuroblastoma. However, several novel splice forms of p73 mRNAs were identified, including one without exon 11 that predominated in multiple MYCN-amplified cell lines. Its encoded p73 protein differed from other splice forms in that the C-terminus was derived from an alternative reading frame. Further study of the functional properties of the protein encoded by this splice form of p73 will be needed to determine whether it contributes to the pathogenesis of childhood neuroblastoma with MYCN gene amplification.

Clinical implications of cytogenetic abnormalities in melanoma

Unlike leukemia, in which specific reciprocal translocations are frequently observed, melanomas involve complex recurring chromosome anomalies. Analysis of the constituted genome of melanoma patients should identify cancer susceptibility genes and at-risk individuals in families with a history of melanoma. The first of these genes to be cloned is the cell cycle regulatory protein inhibitor--the p16 gene-- and a second gene locus for melanoma predisposition has been linked to the chromosome 1p36 band region. Detection of the most common somatic genetic alterations in melanoma enhances our understanding of molecular mechanisms of melanoma development and may lead to genetic markers in melanoma. Some alterations may be used to identify interesting subpopulations. Others may be of prognostic value when they are considered in tandem with clinical data.

Cytogenetics of 158 patients with regional or disseminated melanoma. Subset analysis of near-diploid and simple karyotypes

We report on the cytogenetic analyses of 158 cases of metastatic malignant melanoma, comprised of 63 cases with regional disease (RD) and 95 cases with distant (metastatic) disease (DD). Clonal structural abnormalities were identified in 126 (80%) cases and were significantly increased ( < 0.01 after adjusting for multiple comparisons) on chromosomes (in order of frequency of involvement) 1, 6, 7, 11, 9, and 3. Clustering of breakpoints occurred at 1p36, 1p22-q21, 6p11-q21, 9p, 11q23-qter, 13p (especially for cases with DD), and 19q13. The most common clonal numerical abnormalities, in a subset of 49 near-diploid cases were -10, -22, -9, +7, -19, and -Y. Analysis of chromosome segment gains and losses (CSRP) showed frequent loss of chromosomes 6 and 10, followed by equal rates of involvement of chromosomes 1, 7, and 9. Whole or segmental losses of chromosome 9 (especially 9p) correlate well with recent molecular genetic studies identifying putative suppressor genes, and are also likely important genetic abnormalities. However, based on the frequency of abnormalities in this large series of metastatic melanomas, it is likely that structural abnormalities of 1 and 6, and 10 are important in the pathogenesis of sporadic advanced melanoma.

Alterations in the PITSLRE protein kinase gene complex on chromosome 1p36 in childhood neuroblastoma

p58cdc2L1, a protein kinase implicated in apoptotic signaling, is one of eight separate kinases encoded by three tandemly duplicated and linked genes, which we have termed PITSLRE A, B and C. One allele of this complex on chromosome 1 was either deleted or translocated in each of 18 neuroblastoma cell lines with cytogenetically apparent 1p alterations. A protein encoded by this locus, PITSLRE gamma 1, was absent in three of the lines and a smaller, apparently truncated, PITSLRE polypeptide was found in another line. These findings identify a novel gene complex on chromosome 1 that encodes a protein kinase subfamily. We suggest that the PITSLRE locus may harbour one or more tumour suppressor genes affected by chromosome 1p36 modifications in neuroblastoma.

Clonal chromosome abnormalities in 54 cases of ovarian carcinoma

As a prelude to assessing the relationship of chromosome alterations to clinical outcome in ovarian carcinoma, we report on the cytogenetic analysis on short-term cultures from 54 patients. All patients had histopathologically confirmed malignancy, with the majority of cases demonstrating serous ovarian adenocarcinomas. Structural alterations were evident in 52 cases, whereas numeric changes were identified in 13 cases. The most notable numeric abnormalities were loss of the X-chromosome (9/13 total cases) and +7 (3/9 diploid cases). Structural alterations most frequently involved chromosomes 1, 3, 6, 7, 11, and 12. Chromosomal breakpoints were shown to cluster in several chromosomal banding regions, including 1p36, 1p11-q21, 3p23-p10, 7p (especially 7p22), 11p, 11q, 12p13-q12, and 12q24. The frequency of structural alterations involving the following chromosome arms was found to be significantly increased: 1p (p < 0.01), 7p (p < 0.01), 11p (p < 0.01), 11q (p < 0.05), and 12p (p < 0.05). An analysis of the net gain or loss of chromosome segments was also performed, with the most consistent tendency observed being over-representation of 1q and chromosome 7, deletion of 1p, and loss of the X chromosome.

Clonal chromosome abnormalities in human breast carcinomas. I. Twenty-eight cases with primary disease

Cytogenetic analysis was performed on a selected series of short-term cultures of primary breast carcinomas from 28 patients. All patients had histopathologically confirmed malignancies, with the majority (25/28 cases) demonstrating infiltrating ductal carcinoma. All 28 cases evidenced clonal chromosome abnormalities, with 10/28 displaying only numeric aberrations, whereas 18/28 displayed clonal structural alterations. In near-diploid tumors the most common numeric changes were -17 and -19. However, trisomy 7 was the only numeric change in two near-diploid tumors. Structural chromosome alterations were primarily isochromosomes, apparent terminal deletions, and unbalanced non-reciprocal translocations. Chromosomes I (10/18-56%) and 6 (8/18-44%) were most frequently altered in this series. Breakpoints of clonal structural abnormalities were shown to cluster to several chromosome segments, including 1p22-q11, 3p11, 6p11-13, 7p11-q11, 8p11-q11, and 19q13. Analysis of the gain or loss of specific chromosome segments revealed that the most consistent tendency was over-representation of 1q, 3q, and 6p.

Clonal chromosome abnormalities in human breast carcinomas. II. Thirty-four cases with metastatic disease

Cytogenetic analysis was performed on a selected series of short-term cultures from 34 patients with documented metastatic breast carcinoma. The majority of tumor cells were hyperdiploid, with clonal structural alterations observed in 94% of patients (32/34). The most common numeric changes were -2, -15, and -18. Chromosome 20 was the most frequently over-represented (in near-3n tumors only). Clonal structural chromosome alterations included isochromosomes, terminal deletions, and, most frequently, unbalanced non-reciprocal translocations. Chromosomes most often involved in structural rearrangements included 1, 7, 11, and 6 (accounting for 24.7%, 10.3%, 9.1%, and 7.0% of breakpoints, respectively). When the breakpoints of clonal structural abnormalities were analyzed, they were shown to cluster to several chromosome segments, including 1p11-q21, 7pter, 11p12-q12, and 6q11-21. An analysis of the net gain or loss of specific chromosome segments was also performed, with the most consistent tendency observed being the over-representation of 1q, 6p, 7, and 11. The most frequent losses included 1p, 6q, 7, and 11q.

Cytogenetics and clinical correlations in breast cancer

We describe an ongoing study examining the relevance of cytogenetic changes in primary and metastatic breast cancer. Tumors samples from breast cancer patients are karyotype using G-banding techniques, and all chromosomal findings, including the presence of structural and numeric clonal abnormalities, are documented in a dynamic patient data base. Information on essential clinical characteristics is collected, and patients are followed longitudinally for disease recurrence, progression, and survival. Statistical analyses will examine potential correlations between specific abnormalities and clinical features of the disease, and survival differences between patients will be examined as a function of karyotypic differences. We anticipate that this research will provide insights on the natural history of breast cancer and help direct the search for the underlying molecular mechanisms of tumor genesis and progression.

Nonrandom structural and numerical chromosome changes in non-small-cell lung cancer

Cytogenetic studies were performed on 27 tumor cell lines (most of which were derived from metastatic lesions) and four fresh malignant pleural and pericardial effusions from 30 patients with non-small-cell lung cancer (non-SCLC). Many clonal structural (deletions and nonreciprocal translocations) and numerical abnormalities were found in each specimen. Statistical analysis revealed these changes were nonrandomly distributed among the chromosomes. A statistically significant number of chromosomal breakpoints were seen in regions 1q1, 1q3, 3p1, 3p2, 3q1, 3q2, 7q1, 13p1, 14p1, 15p1, and 17q1 when the regions were compared to the total haploid complement. In addition, when a given region was compared to other regions within the same chromosome, statistically significant numbers of breakpoints were noted for regions 1q3, 5q1, 7q1, 13p1, 14p1, 15p1, 16q2, 17q1, and 21p1. Specific chromosome bands showing the most frequent involvement in structural abnormalities were (in descending order) 3p14.2, 3q21, 19q13, 11p15, 1q11, 7q11, 1q21, 3p23, and 3p21. The breakpoints indicate areas to look for new dominant oncogenes activated by translocations, while the areas of deletions and loss of material by nonreciprocal translocations highlight areas to search for recessive oncogenes. These cytogenetic studies represent strong evidence that multiple genetic lesions are associated with the development of metastatic lung cancer, and provide a roadmap to search for new genes involved in the pathogenesis of lung cancer.

Numerical chromosome changes in 165 malignant tumors. Evidence for a nonrandom distribution of normal chromosomes

The numbers of normal copies of each of the chromosomes in representative karyotypes from 165 malignant tumors of the bladder, breast, cervix, colorectum, and testis studied in this laboratory or described in the literature were assessed to determine whether particular chromosomes were over- or underrepresented. For each chromosome, the mean number of copies was expressed as a percentage of the number expected on the basis of the total number of chromosomes in the karyotypes. The most highly represented autosomes in the tumors as a whole were, in descending order of frequency, numbers 7, 20, 12, 19, 21, and 3, while those most underrepresented were numbers 10, 1, 4, 5, 14, 17, 11, and 18. In tumors of males, the Y tended to be underrepresented. The X was highly represented in the testicular tumors (there were usually two or more copies) and in colorectal tumors of males, but not in the other tumor categories studied. For the tumors as a whole, statistically significant differences could be demonstrated between pairs of autosomes that were at opposite ends of the frequency range. Differences between tumors at the different sites studied were not demonstrable. It is suggested that the determination of the number of normal copies of chromosomes, i.e., whether there are more or fewer than expected, may usefully complement observations on structural changes by reflecting the presence of oncogenes and tumor-suppressor genes, respectively. It may also point to chromosomes that are involved in significant genic changes in which cytogenetic observations on structural changes are equivocal.

Loss of heterozygosity for the short arm of chromosome 1 in human neuroblastomas: correlation with N-myc amplification

Partial monosomy of the short arm of chromosome 1 is the most consistent cytogenetic abnormality found in human neuroblastomas, but its overall frequency and significance are unclear. Using a panel of chromosome-1-specific DNA probes that identify restriction fragment length polymorphisms, we demonstrate that 13 of 47 human neuroblastomas (28%) have somatic loss of heterozygosity (LOH) at one or more loci on the distal short arm of chromosome 1. the chromosomal region that shows LOH most consistently is between 1p36.1 and 1p36.3; loss of a gene or genes in this region may be critical for the development or progression of neuroblastomas. The region of LOH in human neuroblastoma may resemble that described for pheochromocytoma, medullary thyroid carcinoma, and melanoma, which are also tumors of neural-crest origin. Although LOH for distal chromosome 1p can occur in early stages of neuroblastoma, the loss usually occurs in tumors of advanced clinical stages. LOH for the short arm of chromosome 1 correlates significantly with N-myc amplification, suggesting that these two genetic events are related. Indeed, these two lesions appear to characterize a genetically distinct subset of particularly aggressive neuroblastomas.

Gene amplification in human neuroblastomas: basic mechanisms and clinical implications

Extrachromosomal double minutes (DM) and homogeneously staining regions (HSR) of metaphase chromosomes have been reported frequently in human neuroblastomas and tumor derived cell lines. We and other investigators have determined that virtually all DM- and HSR-bearing neuroblastoma cell lines have multiple copies of the oncogene N-myc. Circumstantial evidence suggests that the extrachromosomal DM may be the level at which amplification takes place, with subsequent linear integration into HSR. Indeed, DM may represent circular molecules. The global organization and fine structure of the amplified sequences remains to be determined, but the unit of amplification is probably between 2 X 10(5) and 2 X 10(6) base pairs (bp). In some cases, the process of amplification may involve somatic recombination with distant DNA segments. We have determined that N-myc amplification is a common finding in primary neuroblastomas from untreated patients. We identified N-myc amplification, ranging from 3- to 300-fold, in 34 of 89 cases (38%). N-myc amplification is found almost exclusively in tumors from patients with advanced stages of disease (stages III and IV, p less than 0.01). In addition, the presence of amplification is highly correlated with rapid tumor progression (p less than 0.001). Thus, amplification of the N-myc oncogene is a new intrastage prognostic factor, and N-myc amplification appears to play an important role in determining disease progression.

Marker chromosomes in direct preparations of human large bowel tumors

A survey of 257 marker chromosomes in 48 primary human large bowel adenocarcinomas showed that 44% were markers with recognized patterns. Chromosomes #1, #3, #5, #8, #9, #13, and #17 were involved most frequently. Markers related to chromosomes #7 and X were not seen in any recognizable form. The unidentified chromosomes were classified as markers with abnormal banding regions. In correlating tumor location and stage of invasion with markers, there were fewer markers in tumors from the right side. However, there was little difference in the number of markers seen in the left-sided tumors, irrespective of histopathologic stage, suggesting that function and microenvironmental conditions between various parts of the colon may be related to these differences. The most striking observation is that 21% of all tumors analyzed were without any obvious markers.

24-36-Chromosome clones in human malignancies. Cytogenetic interrelationships, clinical significance and patient age

The literature has been reviewed for cases of malignant disease showing cellular clones with 24-36 chromosomes. Such cases are characterized by aggressive disease. Together the clones with 24-36 chromosomes compose a remarkably consistent non-random cytogenetic pattern, which demonstrates that different chromosomes are of different value for cellular survival and clonal propagation. Considering that the tissues of origin are very different (various solid tumours and different leukaemias), the close mutual cytogenetic relationship between the clones indicates that the cytogenetic pattern is tissue non-specific. Karyotypic non-specificity of cells with very different phenotypes is an apparent contradiction, which raises important questions concerning the relation between karyotype and phenotype.

Nonrandom chromosomal changes in untreated retinoblastomas

The karyotypic patterns of 15 retinoblastomas were examined. Five tumors were found to have two distinct stem lines and, therefore, the chromosomal patterns of 20 tumor cell lines are reported. Three nonrandom chromosomal changes, namely, a loss of a chromosome #13, the presence of an i(6p), or a trisomy of 1q were observed. The potential importance of these chromosomal changes in tumor development is discussed, particularly the loss of a chromosome #13 or the gain of an i(6p). At least one of the three chromosomal changes was found in 75% of the tumor lines analyzed.

Cytogenetic features of acute nonlymphoblastic leukemia in 73 children and adolescents

We examined the leukemia cells of 81 consecutively admitted children and adolescents with acute nonlymphoblastic leukemia (ANLL) to determine the frequency and specificity of chromosomal abnormalities. Karyotypes were obtained for 73 (90%) of the 81 children, and 36 (49%) were abnormal. The modal karyotypes for the cases were tightly clustered in the diploid range; only 5 (7%) were hypodiploid, with 45 chromosomes each, and only 2 (3%) had greater than 50 chromosomes. Specific chromosomal abnormalities in the abnormal karyotypes were compared to morphologic subgroups of ANLL. An 8;21 translocation was found in 6 of 9 cases with M2 morphology but was also found in 1 case with M1 morphology. One of 4 with M3 (progranulocytic) morphology had a 15;17 translocation, and another had a 17q deletion. A structural abnormality in 11q was found in 3 of 7 patients with M5 (monoblastic) morphology, 2 of whom had a 9;11 translocation. The only case of M6 had a 22q-or Philadelphia chromosome in addition to other abnormalities. Statistical analysis of 27 abnormal karyotypes showed preferential structural rearrangement of 8q and 21q. We conclude that, in children as well as adults, specific structural abnormalities are correlated with certain morphologic subgroups of ANLL. However, other chromosomal changes associated with prior mutagenic exposure of adult ANLL were uncommon in children, which may suggest a difference in pathogenesis.

Karyotypic progression in human tumors

Karyotypic progression may be viewed in at least two ways. One approach seeks evidence for increasing and progressive deviation from the normal chromosome pattern in tumors. The clearest examples, found in some leukemias, are those in which successive karyotypic changes are superimposed on an already aberrant cell population. Evidence of chromosomal progression within solid tumors is far less frequent, possibly because the tumors themselves are at a relatively late stage in their evolution. An alternative approach, therefore, attempts to correlate the extent of karyotypic deviation with other aspects of tumor progression. Recent data, based on classical cytogenetic analyses and flow cytometry, are presented to determine relationships between karyotype and specific origin and morphology of tumors. The predominant theme which emerges, not surprisingly, is that the more deviant chromosome patterns are associated with other measures of increased biologic malignancy. What is surprising is the degree to which these properties are expressed in primary tumors and the relative lack of evidence for further karyotypic evolution with recurrence or metastasis. Examples of genetic instability, evolution through polyploidy, gene amplification, and selection for specific chromosomal rearrangement are found in populations of premalignant and malignant human cells. There is increasing recognition of the importance of tumor-specific chromosome aberrations in the stepwise progression from the normal to the fully neoplastic cell.