A phase I study of panobinostat and ruxolitinib in patients with primary myelofibrosis (PMF) and post--polycythemia vera/essential thrombocythemia myelofibrosis (post--PV/ET MF).

Ruxolitinib, a selective JAK1/JAK2 inhibitor, is the current first line therapy for myelofibrosis (MF), which reduces symptomatology and splenomegaly, but does not clearly modify disease course. Panobinostat, a histone deacetylase inhibitor, was shown to be safe and tolerable in phase I and II trials and demonstrated clinical activity in approximately a third of treated patients. Combination therapy of ruxolitinib and panobinostat showed synergistic activity in a preclinical MF model, which prompted clinical evaluation of this combination in both ruxolitinib naïve and treated MF patients. Herein, we report the results of an investigator-initiated, dose escalation, phase I trial of ruxolitinib and panobinostat in 15 patients with primary MF and post-polycythemia vera/essential thrombocythemia MF. This combination treatment proved to be safe and tolerable without dose limiting thrombocytopenia and a maximum tolerated dose of both agents in combination was not determined. The majority of patients maintained stable disease with this combination treatment and 40 % attained a clinical improvement (spleen n = 5, anemia n = 1) by modified IWG-MRT at the end of 6 cycles. This is one of the first attempts of rationally designed, JAK inhibitor-based, combination therapy studies and exemplifies the feasibility of such an approach in patients with advanced MF.

Role of minimal residual disease in the management of acute myeloid leukemia-a case-based discussion.

AML is stratified into risk-categories based on cytogenetic and molecular features that prognosticate survival and facilitate treatment algorithms, though there is still significant heterogeneity within risk groupings with regard to risk of relapse and prognosis. The ambiguity regarding prognosis is due in large part to the relatively outdated criteria used to determine response to therapy. Whereas risk assessment has evolved to adopt cytogenetic and molecular profiling, response criteria are still largely determined by bone marrow morphologic assessment and peripheral cell count recovery. Minimal residual disease refers to the detection of a persistent population of leukemic cells below the threshold for morphologic CR determination. MRD assessment represents standard of care for ALL and PML, but concerns over prognostic capability and standardization have limited its use in AML. However, recent advancements in MRD assessment and research supporting the use of MRD assessment in AML require the reconsideration and review of this clinical tool in this disease entity. This review article will first compare and contrast the major modalities used to assess MRD in AML, such as RQ-PCR and flow cytometry, as well as touching upon newer technologies such as next-generation sequencing and digital droplet PCR. The majority of the article will discuss the evidence supporting the use of MRD assessment to prognosticate disease at various time points during treatment, and review the limited number of studies that have incorporated MRD assessment into novel treatment algorithms for AML. The article concludes by discussing the current major limitations to the implementation of MRD assessment in this disease. The manuscript is bookended by a clinical vignette that highlights the need for further research and refinement of this clinical tool.

In vivo differentiation and genomic evolution in adult male germ cell tumors.

Germ cell tumors (GCTs) are the most common solid malignancy in young adult men, but the genes and genomic regions involved in their etiology are not fully defined. We report here an investigation of DNA copy number changes in GCTs using 1 Mb BAC arrays. As expected, 12p gain was the defining genomic alteration, occurring in 72/74 GCTs. Parallel expression profiling of these tumors identified potential oncogenes from gained regions (LYN and RAB25) and potential tumor suppressor genes in regions of loss (SYNPO2, TTC12, IGSF4, and EPB41L3). Notably, we observed specific genomic alterations associated with histology, including gain of 17p11.2-q21.32 and loss of 2p25.3 in embryonal carcinoma, gain of 8p23.3-12 and loss of 5p15.33-35.3, 11q23.1-25, and 13q12.11-34 in seminoma, and gain of 1q31.3-42.3, 3p, 14q11.2-32.33, and 20q and loss of 8q11.1-23.1 in yolk sac tumors (YST). Many significant genes that mapped to these regions had previously been associated with specific histologies, such as EOMES (chr3) and BMP2 (chr20) in YST and SPRY2 (chr13) and SOX17 (chr8) in seminomas. Additionally, our results suggest a model in which histologic differentiation of GCTs may drive genomic evolution.

Deregulation of FCGR2B expression by 1q21 rearrangements in follicular lymphomas.

We report here the molecular cloning and characterization of a t(1;14)(q21;q32) in a follicular lymphoma (FL) with an unusual BCL2 aberration. Fluorescence in situ hybridization (FISH) and Southern blot analysis of tumor cells identified the translocation breakpoint within the 5' switch region of IGHG (Sgamma). We cloned the chimeric breakpoint region approximately 1.5 kbp downstream from the HindIII site of 5'Sgamma2 on chromosome 14q32 and identified a 360-bp novel segment with homology to the CpG island clone 11h8. Two BAC clones containing this sequence were isolated and mapped to 1q21 by FISH. BAC 342/P13 contained sequences homologous to Fcgamma receptors 2A, 3A, 2B, 3B, and a heat shock protein gene HSP70B. The translocation brought the Sgamma2 region of a productive IGH allele 20 approximately 30 kbp upstream of FCGR2B. As a result of the translocation, the b2 isoform of FCGR2B was overexpressed in the tumor. Screening of a panel of 76 B-cell lymphomas with 1q21-23 cytogenetic aberrations by Southern blot analysis using breakpoint probes identified an additional FL with a t(14;18)(q32;q21) and a breakpoint in the FCGR2B region. These results suggest that FCGR2B may be deregulated by 1q21 aberration in BCL2 rearranged FLs and possibly play a role in their progression.

ID gene expression varies with lineage during differentiation of pluripotential male germ cell tumor cell lines.

Human male germ cell tumors (GCTs) comprise an excellent model system for understanding the molecular events controlling cellular differentiation and lineage decision. Pluripotential embryonal carcinoma cell lines derived from GCTs can be induced to undergo terminal differentiation along specific lineages dependent upon the differentiating agent. We report here that one such cell line, NTera2/clone D1 (NT2/D1), previously shown to undergo differentiation along a neuronal lineage by all-trans-retinoic acid (RA), can be induced along a distinct non-neuronal lineage by the mammalian morphogens, bone morphogenetic proteins-2 and -4 (BMP-2 and -4). Very little is known regarding the molecular events that govern such human lineage decisions. In this study, the role of the ID (inhibitor of differentiation and DNA-binding) family of genes that act as inhibitors of the function of helix-loop-helix (HLH) transcriptional activators involved in lineage commitment was investigated using two pluripotential GCT cell lines as a model system. In the differetiation programs studied, Id1 was noted to decline, an event often associated with the decrease in proliferative rate occurring during differentiation. However, differences in the expression of ID2 and ID3 family members were detected between the programs. Notably, an increase in Id3 during RA-induced differentiation of NT2/D1 cells was observed, while Id2 levels increased during BMP-2 and -4 treatment of NT2/D1 cells and during the induction of an endodermal-like differentiation program in the cell line, 27X-1. The pluripotential male GCT cell lines comprise a unique system in which the roles of specific genes such as the ID family of genes in human cell differentiation and lineage decision can be studied.

Expression of ID genes in differentiated elements of human male germ cell tumors.

The ID genes are members of a family of genes that encode helix-loop-helix (HLH)-containing proteins. The Id proteins, unlike other HLH proteins, lack an adjacent DNA binding domain and hence act as dominant negative regulators of HLH transcription factors that have been implicated in control of cellular differentiation. Although the role of Id genes in murine development has been documented, their roles in human embryogenesis remain unknown. In this study, human male germ cell tumors (GCTs) were used as a model for examining the expression of the ID genes in various histologies that are reflective of different temporal phases of human development. In seminomas, little or no expression of IDI, ID2, and ID3 was detected, consistent with the uncommitted germ cell-like phenotype of this tumor histology. Likewise, GCTs with histologies reflective of extraembryonic and embryonic patterns of differentiation exhibited patterns of expression of the three ID genes often similar to those noted during murine development. It was also evident, as revealed by ID expression patterns, that despite the overall aberrant spatial differentiation patterns displayed by these tumors, some tissue-tissue interactions reminiscent of those observed during normal embryogenesis are retained. Thus, adult male GCTs offer a unique system in which the role of genes such as the IDs can be studied in human embryogenesis.

Genetic analysis of the APAF1 gene in male germ cell tumors.

Cytogenetic and molecular analyses have shown that the chromosome band 12q22 is recurrently deleted in male germ cell tumors (GCTs), indicating the presence of a candidate tumor suppressor gene (TSG) in this region. To identify the TSG, we mapped the APAF1 gene, a proapoptotic mammalian homologue of ced-4, to chromosomal band 12q22, that suggested that this might be the candidate deleted gene in GCTs. We further localized the gene between the polymorphic markers D12S1671 and D12S1082 at 12q22 to determine the role of APAF1 in the pathogenesis of GCT, and we characterized its normal genomic structure and analyzed its alterations in GCTs. The APAF1 gene comprises 27 exons, with the coding region spanning 26. The region containing APAF1 was found to be deleted in GCT by fluorescence in situ hybridization analysis, but without evidence of coding sequence alterations. RT-PCR and Western blot analysis showed APAF1 gene expression at detectable levels in all GCT cell lines analyzed. An aberrant-sized APAF1 protein was seen in one cell line. This and 2 other cell lines carrying APAF1 deletions also exhibited defects in dATP-mediated caspase-3 activation. Caspase-3 activity was effectively restored by addition of recombinant caspase-9 and APAF1 proteins, and to a lesser extent by caspase-9 alone, but not by APAF1 alone. These data do not support a TSG role for APAF1, but defects in other components of the apoptotic pathway that may be related to 12q22 deletion cannot be ruled out. Genes Chromosomes Cancer 28:258-268, 2000.

Genetics and biology of adult human male germ cell tumors.

Adult human male germ cell tumors (GCTs) provide a unique opportunity to study the generation of a transformed pluripotential cell from a totipotential GC in lineage differentiation and on the path to gametogenesis. The pluripotentiality of the tumor cells manifests as histological differentiation into GC-like undifferentiated (SE), primitive zygotic (EC), embryonal-like somatically differentiated (TE), and extra-embryonally differentiated (CC, YST) phenotypes. The tumors and cell lines derived from them comprise exceptional model systems for the molecular analysis of human embryonal cell fate and lineage differentiation. The majority of GCTs show exquisite sensitivity to cisplatin-based treatment and have served as models for the development of chemotherapy for solid tumors. Until recently, the molecular mechanisms of GC transformation, GCT differentiation, or GCT chemotherapy sensitivity and resistance were understood poorly. Very recent studies of GCTs have suggested that: (a) overexpression of cyclin D2 is a very early, possibly the oncogenic, event in GC tumorigenesis; (b) differentiation in GCTs may be governed by several possibly interacting pathways, such as loss of regulators of GC totipotentiality and of embryonic development, and genomic imprinting; and (c) chemotherapy sensitivity and resistance may be rooted in part in a p53-dependent apoptotic pathway. In this review, these new data are discussed in the context of GC and GCT biology, and several novel testable genetic models are proposed.

MUC1 is activated in a B-cell lymphoma by the t(1;14)(q21;q32) translocation and is rearranged and amplified in B-cell lymphoma subsets.

The band 1q21 is among the most common sites affected by chromosomal translocations in lymphoid, myeloid, epithelial, and sarcomatous lesions. In non-Hodgkin's lymphoma (NHL), translocations and duplications affecting this chromosomal site are frequently, but not exclusively, seen in association with primary abnormalities such as the t(14;18)(q32;q21) and t(8;14)(q24;q32) translocations, suggesting a role for 1q21 rearrangements in tumor progression. We report here the characterization and cloning of breakpoints in a case of extranodal ascitic B-cell lymphoma with a t(1;14)(q21;q32) translocation. The breakpoints on the der(1) and der(14) chromosomes were mapped by fluorescence in situ hybridization and Southern blot analysis and cloned using an IGHG (Cgamma) probe. The translocation linked the IGHG4 switch (Sgamma4) sequences of the productively rearranged allele to chromosome 1 sequences downstream of MUC1, leaving the MUC1 transcriptional unit intact. MUC1 was markedly overexpressed in the tumor at the mRNA and protein levels relative to lymphoma cell lines lacking a 1q21 rearrangement. Presumably, MUC1 transcription is aberrantly regulated by the IGHA (Calpha) 3' enhancer element retained on the same chromosome. Screening of a panel of B-cell lymphomas by Southern blot analysis identified a subset with a 3' MUC1 breakpoint and another with low-level amplification of MUC1. MUC-1 mucin has previously been shown to be frequently overexpressed in human epithelial cancers and to be associated with tumor progression and poor clinical outcome. Thus, MUC1 activation by chromosomal translocation, rearrangement, and amplification, identified here for the first time in NHL, is consistent with its suggested role in tumorigenesis. (Blood. 2000;95:2666-2671)

Chromosomal amplification is associated with cisplatin resistance of human male germ cell tumors.

Chemotherapy resistance of tumors is an important biological and clinical problem. Studies from many tumor types have indicated that resistance may be based on multiple genetic pathways. Human male germa cell tumors (GCTs) are an especially good model system to study the genetic basis of tumor sensitivity and resistance to chemotherapy. GCTs are exquisitely sensitive to treatment with DNA-damaging drugs such as cisplatin, rarely exhibit TP53 gene mutations, express normal p53 protein, and undergo p53-mediated apoptosis upon drug treatment. A small proportion of tumors (20-30% of metastatic lesions) escape the apoptotic response and result in treatment resistance. We have recently shown (J. Houldsworth, et al., Oncogene, 16: 2345-2359, 1998) that in a subset of such tumors, resistance is linked to TP53 gene mutations. In a further search for genetic mechanisms underlying resistance, we subjected a panel of 17 tumors from relapse-free patients (sensitive) and 17 chemotherapy-resistant tumors to comparative genomic hybridization analysis to identify possible amplified regions (implying amplified/overexpressed genes) associated with resistance. With the exception of 12p11.2-12, high level amplification was not detected in any of the sensitive tumors. We have identified eight amplified regions (1q31-32, 2p23-24, 7q21, 7q31, 9q22, 9q32-34, 15q23-24, and 20q11.2-12) in five resistant tumors, which suggests that chromosomal and, hence, gene amplification may comprise a pathway to drug resistance. Identification of amplified/overexpressed genes at these sites may elucidate new genetic pathways of chemotherapy resistance in GCTs and possibly also in other tumors.

Chromosomal and gene amplification in diffuse large B-cell lymphoma.

Chromosomal translocations leading to deregulation of specific oncogenes characterize approximately 50% of cases of diffuse large B-cell lymphomas (DLBL). To characterize additional genetic features that may be of value in delineating the clinical characteristics of DLBL, we studied a panel of 96 cases at diagnosis consecutively ascertained at the Memorial Sloan-Kettering Cancer Center (MSKCC) for incidence of gene amplification, a genetic abnormality previously shown to be associated with tumor progression and clinical outcome. A subset of 20 cases was subjected to comparative genomic hybridization (CGH) analysis, which identified nine sites of chromosomal amplification (1q21-23, 2p12-16, 8q24, 9q34, 12q12-14, 13q32, 16p12, 18q21-22, and 22q12). Candidate amplified genes mapped to these sites were selected for further analysis based on their known roles in lymphoid cell and lymphoma development, and/or history of amplification in tumors. Probes for six genes, which fulfilled these criteria, REL (2p12-16), MYC (8q24), BCL2 (18q21), GLI, CDK4, and MDM2 (12q13-14), were used in a quantitative Southern blotting analysis of the 96 DLBL DNAs. Each of these genes was amplified (four or more copies) with incidence ranging from 11% to 23%. This analysis is consistent with our previous finding that REL amplification is associated with extranodal presentation. In addition, BCL2 rearrangement and/or REL, MYC, BCL2, GLI, CDK4, and MDM2 amplification was associated with advanced stage disease. These data show, for the first time, that amplification of chromosomal regions and genes is a frequent phenomenon in DLBL and demonstrates their potential significance in lymphomagenesis.

Human male germ cell tumor resistance to cisplatin is linked to TP53 gene mutation.

Male germ cell tumors (GCTs) are uniquely sensitive to cisplatin-based chemotherapy, with more than 90% of newly diagnosed cases cured. The underlying cause for resistance to treatment in 20-30% of metastatic lesions remains to be identified. Unlike other solid tumors, no mutations in the TP53 gene have been identified to date in random panels of GCT specimens, which could account for the exquisite sensitivity of these tumors to genotoxic insult. However, in a panel of resistant GCTs that did either not respond to cisplatin-based chemotherapy or subsequently relapsed and resulted in the death of the patient, we have now identified a subset of tumors to contain TP53 mutations within exons 6-9. A cell line derived from one of these tumors (228A) displayed the same TP53 mutation as the tumor specimen, expressed only mutant TP53 mRNA, and exhibited a relative resistance to cisplatin in vitro in comparison to a cell line (218A) derived from a responsive tumor with wild-type TP53. The resistant cell line displayed a much reduced apoptotic cell death and did not exhibit an induction of transcription of the p53-responsive genes WAF1 and MDM2 following cisplatin treatment, compared to that observed in the sensitive cell line. The levels of bax, an agonist of apoptosis, were found to be reduced in the resistant cell line. The simplest explanation for the resistance of this subset of GCTs that are resistant to cisplatin-based chemotherapy, is the inability of the cells to mount an apoptotic response following exposure due to a functionally inactivating mutation in the TP53 gene.

The cytogenetic theory of the pathogenesis of human adult male germ cell tumors. Review article.

Human male germ cell tumors (GCTs) represent a biological paradox because, in order to develop into a pluripotential tumor, a germ cell destined to a path of limited or no proliferation must acquire the potential for unlimited proliferation. In addition, it must acquire the ability to elicit embryonal differentiation patterns without the reciprocal inputs from fertilization and the imprinting-associated genomic changes which are a part of normal embryonal development. Although much speculated about, the genetic mechanisms underlying these properties of male GCTs remain enigmatic. Recent cytogenetic and molecular genetic analyses of these tumors are providing new insights and new testable hypotheses. Based on our recent work, we propose two such hypotheses. One relates to the mechanism of germ cell transformation and germ cell tumor development. We suggest that the invariable 12p amplification noted as early as in carcinoma in situ/intratubular germ cell neoplasia (CIS/ITGCN) lesions leads to deregulated overexpression of cyclin D2, a cell cycle G1/S checkpoint regulator with oncogeneic potential. Such overexpression reinitiates the cell cycle. We visualize this happening during the pachytene stage of meiosis through aberrant recombinational events which lead to 12p amplification. The other hypothesis relates to the origin of primary extragonadal GCTs. By comparing cytogenetic changes in primary mediastinal versus gonadal lesions, we propose that, in contrast to long-standing speculation that primary extra-gonadal tumors arise from embryonally misplaced primordial germ cells, these lesions arise from migration of transformed gonadal germ cells.

Aberrant expression of cyclin D2 is an early event in human male germ cell tumorigenesis.

Human male germ cell tumors (GCTs) arise in the spermatocytic lineage, and subsets display embryonal-like differentiation. Virtually all GCTs exhibit multiple copies of the short arm of chromosome 12, even in carcinoma in situ/intratubular germ cell neoplasia, the earliest recognizable neoplastic lesion of germ cells. Among the candidate amplified genes mapped to 12p, expression of the cyclin D2 gene (CCND2) was deregulated in a panel of GCT cell lines, with the relative level of steady-state mRNA and protein inversely correlated with the pattern of differentiation characteristic of the cell line. GCT cell lines with a more differentiated phenotype, as indicated by an immunophenotypic analysis, displayed lower cyclin D2 expression with a concurrent increase in expression of the cell cycle inhibitor p21. In the GCT cell lines in which cyclin D2 was highly expressed, cyclin D2 was in complex with its expected catalytic partners (Cdk4 and Cdk6). Whereas no detectable cyclin D2 expression was evident in normal human germ cells, cyclin D2 was expressed in the abnormal germ cells of all carcinoma in situ/intratubular germ cell neoplasia lesions studied. In GCT specimens that displayed no evidence of differentiation (seminoma) or primitive differentiation (embryonal carcinoma), cyclin D2 expression was detected. However, in tumor specimens with certain patterns of differentiation (teratoma and yolk sac tumor), expression was down- or up-regulated depending on the pattern. Our data suggest that aberrant cyclin D2 expression is an early event in germ cell tumorigenesis.

REL proto-oncogene is frequently amplified in extranodal diffuse large cell lymphoma.

Comparative genomic hybridization (CGH) analysis of DNA extracted from a diffuse lymphoma with a large cell component (DLLC) that displayed double minute chromosomes upon conventional karyotypic analysis indicated overt amplification of DNA sequences derived from the 2p13-15 region. Southern blot analysis of this tumor DNA with a cDNA probe for the proto-oncogene REL, previously mapped to 2p14-15, indicated a greater than 35-fold amplification of REL. To determine the incidence of REL amplification and possible clinical or histologic association with DLLC, a panel of 111 tumor DNAs from DLLC specimens was screened for REL amplification by Southern blot analysis. A copy number of > or = 4 was noted in 26 cases (23%). Southern blot analysis of these 26 tumor DNAs with a cDNA probe for TGFA, mapped to 2p13, indicated lack of coamplification except in one case. Another member of the Rel/NF-kappa B family of transcriptional activators, RELA/p65 mapped to 11q13, was amplified in five cases as determined by Southern blot analysis using a cDNA probe. Nineteen of the 26 DLLC (73%) with REL amplification were primary extranodal lymphomas. As a group, the tumors with REL amplification demonstrated an increased frequency of chromosomal aberrations previously associated with tumor progression, suggesting an oncogenic effect of amplified REL in B-lymphoid cells that already contained a transforming genetic lesion. Thus, REL amplification is a frequent event in DLLC, and probably constitutes a progression-associated marker of primary extranodal lymphomas. This study shows the usefulness of the CGH technique in identifying chromosomal regions overrepresented in tumors that can point to amplified genes and may be correlated with clinical features of the disease.

Comparative genomic hybridization: an overview.

Comparative genomic hybridization (CGH) is a newly described molecular-cytogenetic assay that globally assays for chromosomal gains and losses in a genomic complement. In this assay, normal human metaphase chromosomes are competitively hybridized with two differentially labeled genomic DNAs (test and reference), which upon fluorescence microscopy, reveal the chromosomal locations of copy number changes in DNA sequences between the two complements. Application of CGH to DNAs extracted from fresh frozen specimens and cell lines of various tumor types has revealed a number of recurring chromosomal gains and losses that were undetected by traditional cytogenetic analysis. Few previously known sites were found to be in higher copy number, or lost by CGH, while many novel amplified regions were identified. These regions warrant further molecular genetic studies aimed at isolating the perturbed genes. Since CGH can also be performed on DNA extracted from formalin-fixed paraffin-embedded archived tumor specimens with few modifications, gains and losses of genetic material can be determined for specimens that would otherwise be unanalyzable. Prospective and retrospective application of CGH to tumor specimens would permit correlative studies to be performed, possibly identifying diagnostic and prognostic indicators of disease. CGH may also have a future role in detection and identification of chromosomal abnormalities in prenatal diagnosis and in dysmorphic anomalies.

Frequent allelic deletions and loss of expression characterize the DCC gene in male germ cell tumors.

The DCC tumor suppressor gene has been shown to be frequently deleted or its expression reduced or absent in colorectal, gastro-intestinal, pancreatic, prostatic, and breast carcinomas, and glioblastomas. By allelotype analysis using the DCC-flanking polymorphic marker D18S5 we have previously shown that allelic deletions at 18q21 occur in 40% of male germ cell tumors (Murty et al., 1994). In order to further understand the role of DCC gene in germ cell tumorigenesis, we evaluated deletions by loss of heterozygosity (LOH) and mRNA expression by RT-PCR in tumor tissues and cell lines. Analysis of 61 paired normal-tumor DNAs using the probes D18S5, JOSH 4.4 (a polymorphism within the DCC locus) and a (CA)n polymorphism in an intron of DCC revealed that 45% of GCTs had allelic deletions. In addition, two homozygous deletions were found in the DCC gene among 91 (61 used in the LOH analysis and an additional 30) tumor DNAs when screened with the cDNA probes (pDCC 1.65, pDCC 1.9 and pDCC 1.0). By RT-PCR analysis of four normal testes, nine GCT cell lines and 14 tumor tissues, DCC gene expression was detected in all four normal testes, while four (45%) GCT cell lines and one (7%) tumor specimen showed lack of expression. In addition, DCC expression was highly reduced in three (21%) tumor tissues. The high frequency of LOH at 18q21 was characteristic of seminomas as well as all subsets of non-seminomas in primary as well as metastatic states. Frequent allelic loss in all histologic subsets, homozygous deletions, and loss of expression of DCC suggest that suppression of this gene's function is an early event in GCT development.

Allelic loss and somatic differentiation in human male germ cell tumors.

The complex but poorly understood human male germ cell tumors offer unusual opportunities for the genetic analysis of malignant transformation and embryonal differentiation in a pluripotential stem cell lineage. Histologically, these tumors are divided into two major subgroups, seminomas which are characterized by inability to express embryonal differentiation, and non-seminomas which are characterized by ability to express embryonal as well as extra-embryonal patterns of differentiation. To understand the role of genetic factors in the development of these tumors and the regulation of differentiation expressed by them, we carried out a detailed allelotype analysis by the loss of heterozygosity assay. This analysis revealed frequent deletions in known tumor suppressor genes (RB1, DCC, NME), a number of previously described sites of candidate tumor suppressor genes (3p, 9p, 9q, 10q, 11p, 11q and 17p), as well as several novel sites (2p, 3q, 5p, 12q, 18p and 20p). Our results also showed that well differentiated teratomas exhibit a significantly higher level of allelic loss compared to the less differentiated embryonal carcinomas. In addition, certain loci and genes exhibited frequent non-random deletion in teratomas (D3S32, D3S42, D5S12, D10S25, D11S12, RB1, TP53, NME1, NME2, D17S4, D18S6 and D20S6) and embryonal carcinomas (IFNB, D9S27). Among these loci, the NME genes were notable for a high degree of genetic loss (> 70%) in teratomas. These results suggested that nonrandom loss or inactivation of certain genes may be associated with tumor development and loss or inactivation of other genes may be associated with somatic differentiation.

Gene amplification in non-Hodgkin's lymphoma.

Among 426 consecutively ascertained and karyotypically abnormal non-Hodgkin's lymphoma (NHL) tumours, cytological evidence for gene amplification in the form of homogeneously staining regions (HSRs) was encountered in nine cases of large cell diffuse lymphoma (LC-DL). The mean age of patients with HSRs was 62.9 years and four died within a year of diagnosis. To identify candidate gene(s) amplified in these tumours, we performed a Southern blot analysis of tumour DNA using probes for 23 known protooncogenes and the multidrug resistance gene, PGY1. Besides a two-fold amplification of the BCL2 gene in two cases, no evidence for overt amplification of any of the genes assayed was found. To confirm DNA amplification in these specimens we performed the DNA in-gel renaturation assay. Evidence for presence of amplified DNA fragments was obtained in four of seven specimens. These results suggest amplification of a novel gene(s). To our knowledge, this is the first formal study of gene amplification in a large consecutively ascertained series of fresh lymphoma biopsies.

Molecular cytogenetic analysis of i(12p)-negative human male germ cell tumors.

The i(12p) chromosome has been shown to characterize more than 80% of male germ cell tumors (GCTs) and is an important diagnostic marker. Although recent cytogenetic analyses of GCTs have defined nonrandom chromosome abnormalities in these tumors, no attempt has so far been made to compare i(12p)-positive and -negative tumors in terms of their cytogenetic, histologic, and clinical features. During a 5-year period, we have ascertained 202 GCTs, of which 117 had clonally abnormal karyotypes. Among the latter, 91 had one or more copies of i(12p), whereas 26 lacked an i(12p). We report here the karyotypic analysis of these 26 i(12p)-negative GCTs. In this group, nonrandom sites of chromosomal rearrangements included 12p13 (9/26) and 1p11-q11 (5/26). Comparison of the cytogenetic features of i(12p)-negative tumors with i(12p)-positive tumors revealed the only significant difference to be rearrangements affecting 12p13 in the former (35%) as compared to their absence in the latter (3%). Hybridization of metaphase preparations of 9 i(12p)-negative tumors with a chromosome 12 painting probe and with a microdissected 12p painting probe revealed extra copies of chromosome 12 segments incorporated into marker chromosomes whose composition could not otherwise be resolved by banding analysis; all were shown to be derived from 12p. These data demonstrate that both i(12p)-negative and -positive groups are characterized by an increased copy number of 12p, which is consistent with a lack of significant clinical or biological difference between them. An increased 12p copy number thus is a specific aberration of significance to the development of germ cell tumors.