Non-invasive cell-free DNA prenatal screening for trisomy 21 as part of primary screening strategy in twin pregnancy

OBJECTIVES

The performance of non-invasive prenatal screening using cell-free DNA testing of maternal blood in twin pregnancy is underevaluated, while serum marker-based strategies yield poor results. This study aimed to assess the performance of non-invasive prenatal screening for trisomy 21 in twin pregnancy as a first-tier test. Secondary objectives were to assess its failure rate and factors associated with failure.

METHODS

This retrospective cohort study included twin pregnancies in which non-invasive prenatal screening using cell-free DNA was performed as the primary screening strategy between May 2017 and October 2019. We used the NIPT VeriSeq® test for in-vitro diagnosis and set a fetal fraction cut-off of 4% for monochorionic pregnancies and 8% for dichorionic ones. Clinical data and pregnancy outcome were collected from physicians or midwives via a questionnaire or were retrieved directly on-site. We calculated the performance of non-invasive cell-free DNA screening for trisomy 21, analyzed its failure rate and assessed potentially associated factors.

RESULTS

Among 1885 twin pregnancies with follow-up, there were six (0.32%) confirmed cases of trisomy 21. The sensitivity of non-invasive prenatal screening for trisomy 21 was 100% (95% CI, 54.1-100%) and the false-positive rate was 0.23% (95% CI, 0.06-0.59%). The primary failure rate was 4.6%, with 4.0% being due to insufficient fetal fraction. A successful result was obtained for 65.4% of women who underwent a new blood draw, reducing the overall failure rate to 2.8%. Maternal body mass index, gestational age at screening as well as chorionicity were significantly associated with the risk of failure.

CONCLUSION

This study provides further evidence of the high performance, at an extremely low false-positive rate, of non-invasive prenatal screening in twins as part of a primary screening strategy for trisomy 21. © 2023 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

[Per partum acidosis: Interest and feasibility of cerebral Doppler during labor]

OBJECTIVES

To evaluate feasibility and interest of fetal cerebral Doppler during labor and the link with fetal pH to predict perinatal fetal asphyxia.

METHODS

Our prospective study in a university perinatal center, included patients during labor. There were no risk factors during pregnancy and patients were included after 37 weeks of pregnancy. For each patient an ultrasound with cerebral Doppler was done concomitant to a fetal scalp blood sample. We collected maternal and fetal characteristics as well as cervix dilatation, fetal heart rate analysis and fetal presentation.

RESULTS

Among 49 patients included over a period of 4 months, cerebral Doppler failed in 7 cases (11%). Majority of failure occurred at 10cm of dilatation (P=0.007, OR=14.1 [1.483; 709.1275]). Others factors like: maternal age, body mass index, parity, history of C-Section were not associated with higher rate of failure. We did not found either significant correlation between cerebral fetal Doppler and pH on fetal scalp blood sample (r=0.15) nor pH at cord blood sample (r=0.13). No threshold of cerebral Doppler is significant for fetal asphyxia prediction.

CONCLUSION

Fetal cerebral Doppler is feasible during labor with a low rate of failure but not a good exam to predict fetal acidosis and asphyxia.

Expression of the human fetal bac h19 gene in invasive cancers

We have isolated a new -ene, named BAC, which is the human equivalent of the murine H19 gene and is highly expressed in most fetal tissues and in a variety of fresh tumors. BAC was analyzed in 130 untreated invasive carcinomas of different types. The frequency of BAC-expressing cancers as well as the level of expression greatly varied among the different types of cancer and within the same type of cancer. For example, the 2.3 kb BAC transcript band was detected in 94% of breast adenocarcinomas and in only 35% of epidermoid lung carcinomas with differences of 100-fold in the level of expression between tumor specimens. The majority of tumor tissues displayed BAC expression while their normal counterpart did not with the exception of normal breast tissues which contained low but significant level of BAC transcript. It is possible that BAC expression was influenced by the presence of gene deletions in tumors. Indeed, this gene is located in chromosome 11p15, a region in which deletions have frequently been observed in human cancers. Therefore, the variable levels of expression could have a biological significance and be used as a marker of tumor progression.

The contribution of large genomic deletions at the CDKN2A locus to the burden of familial melanoma

Mutations in two genes encoding cell cycle regulatory proteins have been shown to cause familial cutaneous malignant melanoma (CMM). About 20% of melanoma-prone families bear a point mutation in the CDKN2A locus at 9p21, which encodes two unrelated proteins, p16^INK4a and p14^ARF. Rare mutations in CDK4 have also been linked to the disease. Although the CDKN2A gene has been shown to be the major melanoma predisposing gene, there remains a significant proportion of melanoma kindreds linked to 9p21 in which germline mutations of CDKN2A have not been identified through direct exon sequencing. The purpose of this study was to assess the contribution of large rearrangements in CDKN2A to the disease in melanoma-prone families using multiplex ligation-dependent probe amplification. We examined 214 patients from independent pedigrees with at least two CMM cases. All had been tested for CDKN2A and CDK4 point mutation, and 47 were found positive. Among the remaining 167 negative patients, one carried a novel genomic deletion of CDKN2A exon 2. Overall, genomic deletions represented 2.1% of total mutations in this series (1 of 48), confirming that they explain a very small proportion of CMM susceptibility. In addition, we excluded a new gene on 9p21, KLHL9, as being a major CMM gene.

Pharmacogenetic analysis of toxicity after 5-fluorouracil (5FU) or 5FU/oxaliplatin therapy for metastatic colorectal cancer: Preliminary results in FFCD 2000–05 trial

2508

Background: The FFCD 2000–05 randomized trial compared simplified LV5FU2 followed by FOLFOX6 (arm 1) to FOLFOX6 followed by FOLFIRI (arm 2) in the treatment of metastatic colorectal cancer. The aim was predicting the toxicity profile of oxaliplatin after the first line treatment using pharmacogenetic data. Methods: Patients (pts) with available blood samples were compared to the other pts for clinical prognostic factors (chi2 test). A logistic model was computed to test the association between polymorphisms and toxicity in each arm. An interaction test was used to assess a differential effect according to treatment (predictive effect), in order to identify a predictive effect of oxaliplatin. Grade 3–4 hematological and non-hematological toxicities (H-tox and NH-tox) at 4 months and grade 2–4 neurological at 6 months were the endpoints of the study. Thirteen genetic variants in 10 candidate genes were selected for pharmacogenetic analysis: ERCC1_04 (rs3212961), ERCC1_05 (rs11615), ERCC1_06 (rs3212948), ERCC1_24 (rs3212955), ERCC2_02 (rs1799793), ERCC2_03 (rs13181), ERCC2_06 ( rs238406 ), ERCC2_09 (rs1799787), GSTM1 (null/present), GSTT1 (null/present), TS (TSER, Ins/del6bp) and UGT1A1 (rs8175347). Genotyping was performed using Taqman probes, QMPSF and fragment analysis. Results: 327 pts (156/171) out of 410 were included (61 had no blood samples, 16 had less than 2 cycles, 3 had incomplete data on toxicity, 3 had insufficient DNA). No difference was found between included and excluded pts in the analysis for gender, age, OMS, number of metastatic organs and adjuvant chemotherapy. Pts received similar 5FU doses in both arms. Number of patients with at least one toxicity in arms 1/2 were as followed: 5/54 grade 3–4 H-tox, 28/47 grade 3–4 NH-tox, and 0/103 grade 2–4 neurological. The genotype CC of ERCC2_02 correlated with higher NH-tox at 4 months in arm 2 (p=0.0008, OR=0.31, 95%CI=[0.15–0.62] versus p=0.87, OR=0.93, CI=[0.39–2.21] in arm 1) compared to genotypes CT and TT, with borderline interaction (p=0.05). Conclusions: These preliminary results on early toxicity in first-line are in favour of an effect of ERCC2_02 on NH-tox of FOLFOX6 and a predictive effect on NH-tox of oxaliplatin.

[Table: see text]

Comprehensive analysis of CDKN2A (p16INK4A/p14ARF) and CDKN2B genes in 53 melanoma index cases considered to be at heightened risk of melanoma

OBJECTIVE

Comprehensive analysis of the 9p21 locus including the CDKN2A, ARF, and CDKN2B genes in 53 individuals from melanoma index cases considered to be at heightened risk of melanoma.

METHODS AND RESULTS

Using a combination of DNA sequencing, gene copy number by real time quantitative PCR, linkage analysis, and transcript analysis in haploid somatic cell hybrids, we found no evidence for germline alteration in either coding or non-coding domains of CDKN2A and CDKN2B. However, we identified a p14ARF exon 1beta missense germline mutation (G16D) in a melanoma-neural system tumour syndrome (CMM+NST) family and a 8474 bp germline deletion from 196 bp upstream of p14ARF exon 1beta initiation codon to 11233 bp upstream of exon 1alpha of p16(INK4A) in a family with five melanoma cases. For three out of 10 families with at least three melanoma cases, the disease gene was unlinked to the 9p21 region, while linkage analysis was not fully conclusive for seven families.

CONCLUSIONS

These data reinforce the hypothesis that ARF is a melanoma susceptibility gene and suggest that germline deletions specifically affecting p14ARF may not be solely responsible for NST susceptibility. Predisposition to CMM+NST could either be due to complete disruption of the CDKN2A locus or be the result of more complex genetic inheritance. In addition, the absence of any genetic alteration in 50 melanoma prone families or patients suggests the presence of additional tumour suppressor genes possibly in the 9p21 region, and on other chromosomes.

Real-time PCR-based gene dosage assay for detecting BRCA1 rearrangements in breast-ovarian cancer families

BRCA1 and BRCA2 germline mutations, mainly point mutations and other small alterations, are responsible for most hereditary cases of breast-ovarian cancer. However, the observed frequency of BRCA1 alterations is lower than that predicted by linkage analysis. Several large BRCA1 rearrangements have been identified with a variety of technical approaches in some families. We have developed a gene dosage assay based on real-time quantitative PCR and used it to extensively analyze 91 French families of breast-ovarian cancer in which no BRCA1 or BRCA2 point mutations was identified. This gene dosage method calculates the copy number of each BRCA1 exon to readily detect one, two, and three or more copies of BRCA1 target exons. In the series of 91 families at high risk of carrying BRCA1 mutations, we detected seven large rearrangements of the BRCA1 gene by using this real-time PCR approach. This simple, rapid, and semiautomated real-time quantitative polymerase chain reaction (PCR) assay is a promising alternative technique to Southern blot, bar code analysis on combed DNA, quantitative multiplex PCR of short fluorescent fragments, and cDNA length analysis for the detection of large rearrangements. Therefore, this technique should be considered as a powerful diagnostic method for breast/ovarian cancer susceptibility in clinical and research genetic surveys.

Gene expression profiles of bladder cancers: evidence for a striking effect of in vitro cell models on gene patterns

In order to assess the effect of in vitro models on the expression of key genes known to be implicated in the development or progression of cancer, we quantified by real-time quantitative PCR the expression of 28 key genes in three bladder cancer tissue specimens and in their derived cell lines, studied either as one-dimensional single cell suspensions, two-dimensional monolayers or three-dimensional spheroids. Global analysis of gene expression profiles showed that in vitro models had a dramatic impact upon gene expression. Remarkably, quantitative differences in gene expression of 2-63-fold were observed in 24 out of 28 genes among the cell models. In addition, we observed that the in vitro model which most closely mimicked in vivo mRNA phenotype varied with both the gene and the patient. These results provide evidence that mRNA expression databases based on cancer cell lines, which are studied to provide a rationale for selection of therapy on the basis of molecular characteristics of a patient's tumour, must be carefully interpreted.

Quantification of human cytomegalovirus DNA in bone marrow transplant recipients by real-time PCR

A real-time PCR assay was developed to quantify human cytomegalovirus (CMV) DNA in peripheral blood leukocytes (PBLs) of bone marrow transplantation patients. Unlike other teams, we quantified CMV and the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene using a plasmid containing both sequences as an external standard. Tenfold serial dilutions of this plasmid yielded overlapping standard curves that allowed the quantification of CMV and GAPDH gene copies in an efficient and accurate manner. Sequential blood samples (164 specimens) were collected from 16 patients. PBLs were tested by the pp65 antigenemia assay and quantitative CMV and GAPDH gene PCRs. CMV DNA was detected by PCR in 13 patients a mean of 15 days prior to the appearance of antigenemia. The administration of anti-CMV drugs led to a rapid decrease in the numbers of viral copies and positive nuclei. Real-time PCR assay results correlated with those of the CMV pp65 antigenemia assay (P < 0.00001). The TaqMan assay may be a useful tool for rapid quantification of CMV infection and for monitoring of CMV reactivation in bone marrow transplantation recipients.

Mutation and abnormal expression of the p53 gene in the viral skin carcinogenesis of epidermodysplasia verruciformis

Patients suffering from epidermodysplasia verruciformis are prone to nonmelanoma skin cancers, due to an inherited abnormal susceptibility to the oncogenic human papillomavirus type 5. Genotoxic sunlight ultraviolet B radiations are likely to be a cofactor. Lesions of two human-papillomavirus-type-5-infected epidermodysplasia verruciformis patients collected during an 8 y period were retrospectively studied for p53 mutations in exons 5 through 8 by a polymerase chain reaction single-strand conformation polymorphism technique and/or by DNA sequencing of amplified exons. Mutations were detected in 11 of 26 (42.3%) specimens, including five (62.5%) squamous cell carcinomas, three (33.3%) Bowen's carcinomas in situ, two (40%) actinic keratoses, and one (33%) benign lesion. The nine mutations characterized by sequencing were shown to be missense and to affect mutational hotspots in human cancers. Five were C-->T transitions at dicytidine sites considered as ultraviolet signature mutations. Two were transversions (C-->G and C-->A) at dicytidine sites and two were C-->T transitions at nondipyrimidine sites. A marked p53 immunoreactivity was disclosed in 72.7% of 11 invasive carcinomas, 55.6% of nine carcinomas in situ, 37.5% of eight actinic keratoses, and one of three benign lesions. This includes 81.8% of 11 specimens with a p53 mutation but also 50% of 14 specimens with no mutation detected. A dysfunction of the p53 gene is thus likely to play a part in epidermodysplasia verruciformis carcinogenesis, either due to ultraviolet-B-induced p53 mutations, as in nonmelanoma skin cancers in the general population, or involving other mutagens or mechanisms. The part played by human papillomavirus type 5 proteins expressed in epidermodysplasia verruciformis keratinocytes remains to be determined.

c-erbB-2 (HER-2/neu) gene amplification is a better indicator of poor prognosis than protein over-expression in operable breast-cancer patients

Our aim was to compare the prognostic value of c-erbB-2 gene amplification analyzed by Southern blot with that of protein (p185) over-expression measured by immunohistochemistry in 172 patients with operable breast cancer (BC). Amplification and p185 over-expression were found in 31 (18%) and 51 (30%) BCs, respectively. All but 1 of the tumors showed both amplification and over-expression, while 21 (12%) tumors displayed over-expression without amplification. The risk of death associated with c-erbB-2 gene amplification and p185 over-expression was evaluated by multivariate analysis, taking into account tumor size, histoprognostic grade, hormone receptors and axillary node status. During a mean follow-up of 9.5 (+/-2) years, node involvement (p < 0.001), c-erbB-2 gene amplification (p = 0.02) and negative hormone receptors (p = 0.02) were found to be independent prognostic indicators of the risk of death. Over-expression of p185 with no amplification was not correlated with this risk. When the risk of death associated with c-erbB-2 amplification was studied according to chemo- and hormone therapy, no significant difference was observed between subgroups of subjects. Amplification was also associated (p = 0.02) with the risk of multifocal distant metastases (i.e., metastases detected concomitantly in at least 2 sites) and, thus, with BC aggressiveness. These data show the importance of c-erbB-2 gene amplification in predicting the long-term outcome of patients and in selecting eligible patients for c-erbB-2-targeted therapies.

[A functional gene map is required to adapt therapy of metastatic neuroblastoma]

Neuroblastoma is a very common solid tumor which arises in childhood and shows an extreme heterogeneity at the clinical, histological and genetic levels. Besides age and stage, N-myc amplification and 1p deletion are prognostic factors of the disease: in Europe, these genetic markers are used to conduct therapy. In France, N-myc amplification is a factor of bad prognosis which leads, in all forms of the disease including localised forms and metastatic forms of children aged of less than 1 year, to a myeloablative treatment with autologous hematopoietic stem cells transplantation. By contrast, N-myc amplification has no impact on the survival of children aged of more than 1 year with a poor prognosis (30% overall survival, 5 years) but this genetic abnormality is taken into account to treat primary tumor of these patients. In an attempt to find out prognostic factors of these aggressive forms of the disease, various pathways (apoptosis, differentiation angiogenesis, detoxication, immune response) have been recently surveyed, but studies have been carried out on a limited number of genes. Moreover, experimental models of human metastatic neuroblastoma have been obtained in which variations of genes transcript levels involved in these pathways, are observed. The current break-through of cDNA microarrays allows to develop a dynamic transcriptomic scanning of these models as well as of tumors and bone marrows from patients upon conventional chemotherapy. This technology will enable: i) to define molecular entities of the metastatic disease; ii) to apply adapted treatment; iii) to develop new therapeutic strategies.

Genomic and allelic expression status of the p73 gene in human neuroblastoma

BACKGROUND AND PROCEDURE

The p53 gene homologue, p73, is located on the 1p36-3 locus, which is frequently deleted in human neuroblastoma (NB). A survey of 61 NB showed that among 33% of informative cases, p73 loss of heterozygosity (LOH) occurred in 7 of 20 (35%).

RESULTS

LOH pattern of vicinal markers suggested that the p73 gene could not be considered as the candidate NB suppressor gene. Moreover, comparative measurements of allelic expression in tumors and corresponding patient lymphocytes indicate that pure biallelism is much more frequent in lymphocytes than in tumors (71% vs 30%, P= 0.05), which suggests that disequilibrated allelic expression is associated with NB disease.

CONCLUSION

Therefore, in the p73 LOH NBs, the p73 gene could be altered in the maintained allele not by mutations [Ishimiya et al.: Med Pediatr Oncol, this issue], but rather by an abnormal transcription.

Transfer of Ku86 RNA antisense decreases the radioresistance of human fibroblasts

Ku86 has been shown to be involved in DNA double-strand break (DSB) repair and radiosensitivity in rodents, but its role in human cells is still under investigation. The purpose of this study was to evaluate the radiosensitivity and DSB repair after transfection of a Ku86-antisense in a human fibroblast cell line. Simian virus 40-transformed MRC5V1 human fibroblasts were transfected with a vector (pcDNA3) containing a Ku86-antisense cDNA. The main endpoints were Ku86 protein level, Ku DNA end-binding and DNA protein kinase activity, clonogenic survival, and DSB repair kinetics. After transfection of the Ku86-antisense, decreased Ku86 protein expression, Ku DNA end-binding activity, and DNA protein kinase activity were observed in the uncloned cellular population. The fibroblasts transfected with the Ku86-antisense showed also a radiosensitive phenotype, with a surviving fraction at 2 Gy of 0.29 compared with 0.75 for the control and 20% of unrepaired DSB observed at 24 hours after irradiation compared with 0% for the control. Several clones were also isolated with a decreased level of Ku86 protein, a surviving fraction at 2 Gy between 0.05 and 0.40, and 10-20% of unrepaired DSB at 24 hours. This study is the first to show the implication of Ku86 in DSB repair and in the radiosensitivity of human cells. This investigation strongly suggests that Ku86 could constitute an appealing target for combining gene therapy and radiation therapy.

[Oncogenic factors of metastatic dissemination in neuroblastoma]

Disseminated neuroblastoma frequently show a very poor prognosis. N-myc gene amplification, 1p deletion and lack of CD44 gene expression, are all genetic factors associated with the disease's dissemination. Human neuroblastoma xenografts in nude mice has permitted to characterize, in disseminated neuroblasts, oncogenes overexpression, inactivation of tumor suppressor genes as well as detoxifying genes activation which contributes to increase cellular resistance to chemotherapy. These genetic abnormalities permit to propose a nosology of this very aggressive pediatric solid tumor. Hopefully, this genetic classification could be of great value for new therapeutic approaches.

[High incidence of p53 mutations in primary and metastatic head and neck tumors. Frequent protein overexpression in normal epithelium]

Mutation of the p53 tumor suppressor gene is the most commonly observed gene alteration in human cancers. In order to identify new prognostic factors and tumor aggressiveness in squamous cell head and neck carcinomas, we analyzed 50 node metastases and 28 primary tumors including 13 matched specimens for p53 alterations. Mutations were found in 54 (69%) tumors, 76% of which were missense, 9% were nonsense and 15% were microdeletions or microinsertions. Twenty-five mutations were transitions mostly G-->A (40%) and 20 were transversions mostly G-->T (25%) thus confirming the role of tobacco carcinogens in the induction of these mutations. For eight patients mutations were observed in matched primary tumors and metastases, indicating clonal dissemination of tumor cells in most of these carcinomas. Furthermore the incidence of mutations was not different in primary tumors and node metastases indicating that this gene alteration was not related to the metastatic dissemination. No correlation was found between mutation and clinical parameters, the 8-year survival rates were not different (log rank test: P = 0.49) in patients with and without mutation. There was a good correlation between p53 mutation and protein overexpression (Fisher's exact test: P < 10(-4). Interestingly, immunostaining was also observed in basal cells from normal mucosa and in early lesions adjacent to the primary tumor in 11/15 specimens irrespective of the presence of mutation in the corresponding tumors. p53 protein overexpression may therefore constitute a biomarker for early stages of carcinogenesis of the head and neck epithelium.

High incidence of loss of heterozygosity and abnormal imprinting of H19 and IGF2 genes in invasive cervical carcinomas. Uncoupling of H19 and IGF2 expression and biallelic hypomethylation of H19

The few imprinted genes characterized so far include the insulin-like growth factor-2 gene (IGF2) coding for a foetal growth factor and the H19 gene whose normal function is unknown but which is likely to act as an RNA with an antitumour effect. IGF2 is expressed by the paternal allele and H19 by the maternal allele. This reciprocal expression is quite interesting because both H19 and IGF2 genes are located close to each other on chromosome 11p15.5 in a region subject to loss of heterozygosity (LOH). Moreover, loss of imprinting (LOI) or biallelic expression has been proposed as an epigenetic mechanism for tumorigenesis in a variety of human cancers including Wilms' tumour. In this study we report the LOH, LOI and methylation status of H19 and IGF2 genes in 29 invasive cervical carcinomas of different clinical stages. Fourteen (48%) and 13 (45%) tumours were heterozygous for H19 and IGF2 respectively. LOH for H19 and IGF2 genes were found in 2 of 14 (14%) and 3 of 13 (23%) informative tumours, respectively. LOI of H19 and IGF2 was detected in 2 of 12 (17%) and 5 of 10 (50%) tumours with no LOH, respectively. More interestingly, monoallelic expression of the otherwise silent H19 allele (allele switch) was observed in 2 of 12 (17%) tumours and biallelic expression of IGF2 was detected in one specimen of normal cervix adjacent to the tumour. The expressing H19 allele, and to a lower degree also the silent allele, were hypomethylated in tumours suggesting that demethylation of both H19 alleles may be associated with an early step of imprinting alteration. In cervical cancer H19 and IGF2 expressions could be independently regulated. In conclusion, our data suggest that H19 and IGF2 genes, via deletions and/or abnormal imprinting, could play a crucial role in a large proportion (58%) of cervical cancers where they may be associated with disease progression.

High incidence of p53 alterations (mutation, deletion, overexpression) in head and neck primary tumors and metastases; absence of correlation with clinical outcome. Frequent protein overexpression in normal epithelium and in early non-invasive lesions

We have analysed 78 head and neck carcinomas (50 node metastases and 28 primary tumors including 13 matched specimens) in 65 patients for p53 alterations. Mutations were found in 54 (69%) tumors. Of the 53 mutations within exons, 40 (76%) were missense, five (9%) nonsense and eight (15%) microdeletions or microinsertions. Twenty-five (47%) mutations were transitions mostly G-->A (40%) and 20 (38%) were transversions, mostly G-->T (25%), thus confirming the role of tobacco carcinogens in the induction of these mutations. The incidence of mutations was not different in primary tumors (68%) and node metastases (70%) indicating that this gene alteration was not related to the metastatic dissemination. For eight patients, mutations were observed in matched primary tumors and metastases, indicating clonal dissemination of tumor cells in most of these carcinomas. There was a good correlation between mutations and protein overexpression (Fisher's exact test P < 10(-4). Immunostaining was also observed in basal cells from normal epithelium and in early lesions adjacent to the primary tumor in 11/15 (73%) specimens irrespective of the presence of mutation in the corresponding tumors. These data confirm that p53 overexpression is an early event in the multistep process of epithelial cell carcinogenesis. Loss of heterozygosity for the TP53 locus was detected in 54% of tumors but no association was found with mutation (Fisher's exact test P = 0.14). No mdm-2 amplification was detected in any tumors. No correlation was found between mutation and clinical parameters, the 5-year survival rates were not different (log rank test P = 0.39) in patients with and without mutation. In conclusion, we have shown that p53 gene mutations and deletions and protein overexpression are frequent in the most aggressive head and neck carcinomas but are not associated with disease progression. The presence of protein in normal mucosa and in non-invasive lesions may constitute a biomarker for early stages of carcinogenesis.

The p53 and mdm-2 genes in human testicular germ-cell tumors

Mutations in the p53 gene are common in many cancers. They have been documented to occur in about 55% of all cancers of 51 different cell and tissue types. These mutations are accompanied by overexpression of the p53 protein in the nucleus of the cell, and this protein has lost its tumor suppressor function. In this study, 25 testicular germ-cell (TGC) tumors were tested for p53 mutations and the level of p53 protein expression. While 67% of the tumors overproduced the p53 protein in the nucleus of 10-60% of their cells, in all cases the DNA sequence of exons 4-9 of the p53 gene was wild type. In this tumor type, there was apparently no selection pressure for p53 mutations. The mdm-2 gene residues on chromosome 12 (12q13-q14), a chromosome often altered in TGC tumors. mdm-2 gene amplification (2.5- to 10-fold) was detected in three (12%) of these TGC tumors. These three tumors, and eight additional TGC tumors, overexpressed mdm-2 mRNA. There was a good correlation between overexpression of p53 protein and overexpression of mdm-2 mRNA (P = 0.01). This may well result from the fact that the level of mdm-2 mRNA is regulated by the p53 level. These studies demonstrate that TGC tumors fail to be selected for p53 mutations but nonetheless frequently expressed high levels of wild-type p53 protein in the cell nucleus. Perhaps this produces the excellent response to radiation and chemotherapy of these tumors, which generally have a good prognosis. Wild-type p53 may mediate apoptosis in these cells in response to the DNA damage.

Characterization of a new p53-mutated and hpv-negative human squamous-cell cervical carcinoma-derived cell-line

A new cell line, designated IGR/Cut40, has been established from a stage II squamous cell carcinoma of the uterine cervix. These cells, which have a tetraploid DNA content and typical epithelial features, displayed a high proliferation rate and a powerful tumorigenic potential in immunodeficient mice. No human papillomavirus (HPV) DNA was detected using PCR and consensus primers. Sequencing of p53 cDNA revealed a mutation CGC(Arg)-->CAC(His) at codon 175 of the gene encoding for an abundant nuclear protein. IGR/Cut40 cell line should permit a better understanding of the HPV-infection-unrelated tumorigenesis of the uterine cervix.

Expression of anionic glutathione S transferase (GST pi) gene in carcinomas of the uterine cervix and in normal cervices

The aim of the present study was to analyse in invasive carcinomas of the uterine cervix, the anionic glutathione S transferase (GST pi) gene, possibly implicated in the drug resistance of human cancers. Total RNA preparations obtained from invasive cervical cancers (106 specimens), carcinomas in situ (CIS) (three specimens) and normal cervical epitheliums (24 specimens) were analysed by Northern and slot blot hybridisation. A 0.7 kb GST pi transcript band was detected in all the cervical specimens. GST pi mRNA levels were lower in normal cervix (mean: 0.7 +/- 0.1 arbitrary units) than in invasive carcinomas (mean: 2.5 +/- 1.5 units) (Student test P less than 10(-4)). However no significant difference was observed between invasive cancers of advanced stages (III and IV) and those of early stages (I and II). The presence of human papillomavirus in cancers and in normal cervices did not influence significantly the GST pi mRNA level. Neither amplification nor gross rearrangement of GST pi gene could be observed after Southern blot analysis of genomic DNA. In conclusion, our data indicate that the presence of high levels of GST pi transcripts in invasive cancers may be a consequence of the multiple biochemical changes which accompany cervical carcinogenesis.

Prognostic value of c-myc proto-oncogene overexpression in early invasive carcinoma of the cervix

The prognostic effect of c-myc oncogene overexpression was assessed in a multivariate analysis of 93 patients with invasive carcinoma of the cervix, stage Ib, IIa, and IIb proximal. The treatment was based on the association of brachytherapy-colpohysterectomy and lymphadenectomy. Analysis of c-myc gene expression was done using Northern and slot blot hybridization techniques. Overexpression of c-myc (ie, levels at least three times the mean observed in normal tissues) was present in 33% of the tumors. The proportion of carcinomas with c-myc overexpression significantly increased with the size of the primary tumor (P = .04). No relationship was found between c-myc overexpression and the other clinical and histologic parameters, including the nodal status. The relative risk of relapse (overall, pelvic failure, distant metastases) was analyzed in a Cox's proportional hazards model. Three factors were significantly related to the risk of overall relapse when the multivariate analysis was performed, namely, the tumor size, the nodal status, and c-myc expression. A combination of c-myc expression and the nodal status provided a very accurate indication of the risk of relapse. Indeed, patients with negative nodes had a 3-year disease-free survival rate of 93% (95% confidence interval [Cl], 79% to 98%) when c-myc was expressed at a normal level, whereas this rate was only 51% (95% Cl, 26% to 63%) when c-myc was overexpressed (log-rank test, P = .02). In addition, in the subgroup of patients with positive nodes, this rate was 44% (95% Cl, 25% to 77%) and 15% (95% Cl, 4% to 49%) when c-myc gene was expressed at normal level, or overexpressed, respectively. Finally, c-myc gene overexpression was, in the multivariate analysis, the first factor selected by the model regarding the risk of distant metastases.

Analysis of the c-Ha-ras-1 gene for deletion, mutation, amplification and expression in lymph node metastases of human head and neck carcinomas

The c-Ha-ras gene was analysed by Southern blot hybridisation in 67 specimens of lymph node metastases and in 25 specimens of primary tumours obtained from 85 untreated patients with head and neck squamous cell carcinoma. The loss of one c-Ha-ras allele was observed in 10/46 (22%) tumours from heterozygous patients for this locus. Different genes, located as the c-Ha-ras gene on the short arm of chromosome 11, were also found to be deleted suggesting that the deletion of other genes could play a role in aggressiveness of head and neck carcinomas. Using polymerase chain reaction, mutation at codon 12 was detected in only 2/54 (3.8%) tumours but no mutation involving codon 61 was found. Neither gene amplification nor gene rearrangement could be observed. Total RNA was prepared from 79 of these tumour specimens and analysed by Northern and slot blot hybridisation. A 1.2 kb c-Ha-ras transcript band was detected in all the RNA preparations. Relatively high c-Ha-ras transcript levels were found in 18% of lymph node metastases and in 21% of primary tumours, indicating no significant differences between these cancers. Moreover, the c-Ha-ras mRNA levels were not significantly greater in the primary tumours than in the normal mucosae in 10/12 cases for which both tissues were analysed. These data indicate that c-Ha-ras gene does not seem to be strongly involved in head and neck carcinomas at that advanced stage of the disease, as this was previously reported for earlier clinical stages.

Somatic deletions and mutations of c-Ha-ras gene in human cervical cancers

The c-Ha-ras-1 locus was analysed in cervical cancers and shown to exhibit the loss of one allele in 36% of heterozygous tumours and a mutation at codon 12 in 24% of tumours at advanced stages. 40% of tumours with mutation contained also a deletion. A c-myc gene activation was found in 100% and 70% of tumours containing mutation and deletion respectively. This suggests that the two proto-oncogenes cooperate for the progression of cervical cancers. Furthermore as more than 90% of these tumours contained also human papillomavirus sequences, our data strongly suggest that multiple genetic events are involved in the genesis and progression of most cervical cancers.

C-myc proto-oncogene expression and prognosis in early carcinoma of the uterine cervix

Expression of the c-myc gene was studied by northern blot and slot blot hybridisation in 72 specimens of stage I or II squamous cell carcinoma of the uterine cervix. In 25 of the 72 tumours c-myc proto-oncogene was overexpressed (ie, at levels 4-20 times higher than in normal tissues). Patients whose tumours showed c-myc overexpression had an eight-fold greater incidence of early relapse than the other patients (p = 0.001). The 18-month relapse-free survival rates were, respectively, 49% and 90% for these two groups of patients.

A type-II DNA topoisomerase and a catenating protein from the transplantable VX2 carcinoma

It has recently been suggested that topoisomerases could be important targets for several DNA intercalating drugs used in cancer therapy. This prompted us to purify and characterize a type II topoisomerase in a highly tumorigenic transplantable rabbit tumor isolated from a skin carcinoma associated with cottontail rabbit papillomavirus. We have found that the decatenating activity present in tumor cells was 40-100 times higher than that in the rabbit liver, while no activity could be found in skin extracts. The type II topoisomerases purified from tumor and liver cells consist of two subunits with molecular masses of about 160 kDa. The conditions of the reactions of relaxation, unknotting and decatenation catalyzed by these topoisomerases II were found to be similar to those observed with enzymes of other eukaryotic cells. In the course of the purification of the VX2 enzyme, we isolated and characterized a protein of about 30 kDa in whose presence the topoisomerase II was able to catenate very efficiently supercoiled DNA molecules. This protein has the same electrophoretic mobility as an H1-2 histone, and cross-reacts with an anti-H1 antiserum. The VX2 topoisomerase II as well as the VX2 tumor should constitute useful models for assays of antitumoral drugs.

A type I DNA topoisomerase from Trypanosoma cruzi

A type I DNA topoisomerase has been isolated from the nuclei of the flagellate Trypanosoma cruzi, using poly(ethylene glycol) fractionation and chromatography on hydroxyapatite and on phosphocellulose. The relaxation activity was ATP-independent, enhanced by Mg2+ and spermidine. The enzyme removed supercoils from negative and positive superhelical DNAs. Topoisomerase activity was associated with a polypeptide of Mr about 65000 as shown by glycerol gradient centrifugation and by electrophoresis on sodium dodecyl sulfate/polyacrylamide gels.

The kinetoplast of DNA of Trypanosoma gambiense: comparison with the kDNA of Trypanosoma equiperdum

The molecular components of the kinetoplast DNA (kDNA) network of Trypanosoma gambiense have been studied and compared with those of the very closely related species T. equiperdum, previously studied in detail. The kDNA of T. gambiense contains about 80 maxicircles of 20 kilobase pairs and 4000 minicircles of 1 kilobase pairs. The restriction cleavage sites of 7 restriction endonucleases have been mapped on the T. gambiense maxicircle. The majority of these sites were also found in T. equiperdum maxicircles; however their relative positions which are different do not allow us to conclude to relatedness of maps. Maxicircles of the Cairn or of the rolling circle type have been observed and thought to be replicative intermediates. Experiments on renaturation kinetics and hybridization after blotting transfer, show that T. gambiense and T. equiperdum maxicircles have base sequences in common. The T. gambiense minicircles are heterogeneous in base sequence, in contrast to the T. equiperdum minicircles which are homogeneous. The minicircles of the two species have also common base sequences.

Complete nucleotide sequence of minicircle kinetoplast DNA from Trypanosoma equiperdum

The kinetoplast DNA of Trypanosoma equiperdum is composed of about 3000 supercoiled minicircles of 1000 base pairs and about 50 supercoiled maxicircles of 23,000 base pairs topologically interlocked so as to form a compact network. Minicircles of T. equiperdum, which are homogeneous in base sequence, were purified by equilibrium CsCl centrifugation and used as starting material for DNA sequence analysis. One minicircle is composed of 1012 base pairs and has an adenine.thymine base pair content of 72.8%. The termination codons are uniformly distributed along the molecule and restrict the coding potentiality of the molecule to oligopeptides of about 20 amino acids. The molecule contains three dyad symmetries and a sequence of 12 nucleotides is repeated six times. We also noted the presence of a region of about 130 base pairs that is almost perfectly homologous with that of the minicircles from the closely related species T. brucei.