Liquid-based endometrial cytology: cyto-histological correlation in a population of 917 women.

OBJECTIVE: Liquid-based cytology, because of its capacity to reduce the obscuring factors and to provide thin-layer specimens, represents an opportunity to reevaluate endometrial cytology. In order to assess the utility of the liquid-based method in endometrial diagnosis, we evaluated its accuracy in comparison with histology. METHODS: Nine hundred and seventeen women scheduled for hysteroscopy were enrolled in the study. After providing informed consent, all the women proceeded sequentially to hysteroscopy, endometrial cytology and then biopsy endometrial sampling. RESULTS: Cyto-histological correlations were possible in 519 cases (57%): in 361 (39%) cases the biopsy was inadequate, in 15 (2%) the cytology was inadequate, and in 22 (2%) both were inadequate. At biopsy 25 (3%) women had adenocarcinoma, 5 (1%) had adenomatous atypical hyperplasia and 21 (2%) had simple non atypical hyperplasia. At cytology two adenocarcinomas and one adenomatous atypical hyperplasia were underrated as atypical hyperplasias and as non-atypical hyperplasia; two simple non-atypical hyperplasias were reported as negative; and eight cases were false positive (non-atypical hyperplasia at cytology, negative at biopsy). In our population, the cytology provided sufficient material more often than biopsy (P < 0.04). Sensitivity was estimated at 96%, specificity at 98%, positive predictive value at 86% and negative predictive value at 99%. CONCLUSIONS: We concluded that endometrial cytology may be an efficient diagnostic method. It could be applied to selected patients solely or in association with ultrasonography. The combination of these two noninvasive procedures may improve their diagnostic accuracy and reduce unnecessary hysteroscopies, thereby producing benefits for women and society.

International Cell Exchange, 1991.

1. During 1991, lymphocytes from 40 individuals of various races and 20 cultured B-cell lines were sent to 290 laboratories (listed in appendix) for typing. The results of the blind tests were analyzed and reported monthly. This is a summary of the year's 10 monthly exchanges. 2. Ten A-locus antigens showed 95% or greater agreement among laboratories, and 100% agreement was achieved for A2 and A3. Agreement of 95% or more was reached for 7 B-locus specificities. Both Bw54 and Bw73 showed marked improvement in detection during the last 3 years. 3. Six laboratories did not miss the assignment of any A- or B-locus antigens. The false-positive and false-negative assignment rates are given for the A- and B-loci. 4. The following variants were studied: B5-53 variants from various races, BN21, B7x40 (DT), and B7x27. 5. Aw66 assignment was found to be discrepant between the United States and European laboratories. 6. The A28 splits, Aw68 and Aw69, were studied, as were cells with different Aw68 splits. 7. Cell lines were typed by 150 laboratories, including 19 DNA laboratories. Six Class II antigens had 95% or higher detection levels, and 5 others had higher than 90% detection levels. 8. Both DNA and serology laboratories had difficulty in assigning the DRw6 splits, DRw13 and DRw14. The definition of the splits of DR3, DQw1, and DQw3 was clarified with the additional DNA typing results. 9. Amino acid sequencing confirmed a new variant of Bw53, as indicated by the assignments based on serology.

Gene mapping in the idiopathic generalized epilepsies: juvenile myoclonic epilepsy, childhood absence epilepsy, epilepsy with grand mal seizures, and early childhood myoclonic epilepsy.

Idiopathic generalized epilepsies, i.e., juvenile myoclonic epilepsy (JME), childhood absence epilepsy, and epilepsy with grand mal [generalized tonic-clonic seizures (GTCS)], are the most common genetic epilepsies. Linkage studies using Bf, HLA serologic, and DNA markers by three independent investigators, one from Los Angeles and two from Berlin, have localized the JME locus to the short arm of chromosome 6 (6p). Because members of the same JME family have the same JME phenotype of childhood absence epilepsy, epilepsy with grand mal (GTCS) seizures, or early childhood myoclonic epilepsy (ECME), our observations give evidence for a single-locus etiology in 6p for JME and for at least some of the childhood absence seizures, epilepsy with grand mal (GTCS) seizures, and ECME. Studies should now address whether locus heterogeneity exists within childhood absence epilepsy, epilepsy with grand mal (GTCS) seizures, or ECME. Markers linked to JME (Bf, HLA serologic, and DNA markers in the DQ region) can be used to resolve etiologic heterogeneity. Using such markers, both linked and unlinked forms of phenotypes that are clinically indistinguishable may be detected and provide evidence for etiologic heterogeneity. Studies should also concentrate on narrowing the JME locus to 2 to 3 cm by screening families with recombinant events using RFLPs, candidate genes, and new expressed sequences on chromosome 6.

1990 cell typings in the International Cell Exchange.

1. A yearly summary of the previous year's cells typed through the International Cell Exchange allows a participating laboratory to compare its own performance with 292 currently participating exchange laboratories, inasmuch as each laboratory receives its individual antigen report. We present an annual summary for the 1990 typings of 40 cells sent for Class I and 20 cells sent for Class II antigens. 2. The mean detection percentages and the detection ranges for 21 WHO-designated (well-defined) and 21 WHO-provisional (with "w" designations, less well-defined) antigens were determined for the Class I cells typed in 1990. Seventeen WHO antigens showed 95% or greater detection levels. The remaining WHO antigens showed at least 90% agreement, with the exception of B38. More variation in detection is observed in the WHO-provisional antigens. Aw33, Bw50, Bw60, and Bw62 showed 90% or greater average detection percents. In recent years, antigens such as Bw46 and Bw70, have shown great improvement in detection. 3. The percent discrepancy rates of 8 HLA-A,B antigens typed 4 times or more in 1990 were presented as well as the total percent discrepancy rates for all Class I antigens. Comparison of 1990 figures with those of 1988 and 1989 shows a marked decrease in the total discrepancy rates. 4. The number of false negatives and false positives for the Class I antigens indicates that few laboratories have trouble typing the WHO antigens; as many as 115 laboratories had 0 misses. However, a greater number of laboratories missed the less well-defined WHO-provisional antigens: 7 laboratories had 0 misses for all the antigens and 4 laboratories had perfect records (0 false negatives and false positives) for cells typed in 1990. 5. In 1990, the cell exchange continued to study new cell variants. An A10 (A26x34) variant was detected in 4 cells and another possible variant, B7x40 (DT), was determined in 3 cells. 6. The average detection percentages and detection ranges were determined for 23 Class II antigens. Improved detection is indicated. At least 9 Class II antigens showed 90% or greater agreement level, of which at least 4 had 95% or greater mean detection. 7. Since the incorporation of DNA typing results in the B-cell line exchange reports in July 1990, the Class II antigen splits in 11 cells have been confirmed or clarified. Two Class II variants were also confirmed by the DNA laboratories. Presently, 145 serology and 7 DNA laboratories are in this exchange.(ABSTRACT TRUNCATED AT 400 WORDS)

HLA class II epitope detection by serology.

The amino acid sequence for the Class II specificities on 70 homozygous cell lines was compared with the cytotoxic reactions produced by 13,000 allogenic antisera against these lines. Many of the cytotoxic reaction patterns correlated completely with unique amino acids at a given residue. Therefore, we inferred that the antibodies were reacting directly with these amino acid-defined residues. Among 131 variable DRB1 residues, several well-correlated antisera were found to 64 (49%). Among 94 DQB1 residues, antibodies were found to 59 (63%). The location of these serologically defined epitopes on the molecule is shown in the table. We conclude that much of the serologic reactions of Class II can be explained by reactivity to the amino acid-defined epitopes. It can be shown that many of the sera that were previously considered multispecific are directed against these epitopes. These serologically defined epitopes are clearly immunogenic and are probably important in transplant matching.

HLA epitopes detected by serology.

There is evidence that many of the so-called multispecific cytotoxic antisera react to the amino acid-defined epitopes of the HLA-A and B loci molecules. This evidence is based on the finding that a group of sera produced allelic reactions which corresponded to amino acid substitutions. Antisera for 29 amino acid variants were identified at 67 variable residues on the A locus and for 17 of the 51 variable sites on the B locus. The results were validated by showing that the reactions produced by testing cells from 111 individuals on an epitope typing tray fit the Hardy Weinberg equilibrium. This means that the alleles defined by the antisera behaved as alleles in the random population. Tests of homozygous cells on the epitope typing trays produced completely mutually exclusive reaction patterns as expected on the basis of the amino acid substitutions. Two examples are cited in which immunization across a single HLA "antigen" difference existed between donor and recipient, resulting in a "multispecific" antiserum. The specificities contained in the serum exactly fit an amino acid difference between the donor and recipient.

An association between acute retinal necrosis syndrome and HLA-DQw7 and phenotype Bw62, DR4.

Human leukocyte antigen (HLA) typing was performed on 27 white patients with acute retinal necrosis syndrome. Antigens for the HLA-A, -B, -C, -DR and -DQ loci were determined, and frequencies were compared with racially matched controls. There was a statistically significant increase in the frequency of HLA-DQw7 (11 of 20 [55%] of patients vs 294 of 1546 [19%] of controls, P = .0004, relative risk 5.20) that remained significant at the P = .05 level when the P value was multiplied by the number of antigens tested. The HLA phenotype Bw62, DR4 is also more frequent than in normal control populations (4 of 25 [16%] of patients vs 26 of 1023 [2.6%] of controls, relative risk 7.49). These results support an association between the acute retinal necrosis syndrome and certain HLA specificities and suggest a possible immunogenetic predisposition to the syndrome in some patients.

Mapping the gene for juvenile myoclonic epilepsy.

The practice of epileptology at a molecular level, where gene products are identified by gene mapping, will soon be possible for a growing number of epilepsies. Juvenile myoclonic epilepsy (JME) is the first of such epilepsies to be mapped to a chromosome, namely chromosome 6p21.3. Family studies of 68 JME probands from California revealed 50% of all families reported seizures in first- or second-degree relatives. Twelve percent of all family members other than the proband had epileptic seizures. Eighty percent of symptomatic siblings and 6% of asymptomatic siblings had diffuse 4- to 6-Hz multispike-wave complexes. Twelve percent of asymptomatic parents had diffuse, nonspecific slow waves mixed with spikes or sharp waves. JME is tightly linked to the Bf-HLA loci in chromosome 6. No matter what mode of inheritance is assumed, linkage to the clinical manifestations of JME and its associated EEG traits is indicated by lod scores over 3.0, as long as "EEG affected" but clinically asymptomatic family members are counted as affected during LIPED analysis. Studies are now being done to further localize the JME site. At the same time, further linkage studies should decide if JME is heterogeneous within itself and whether the same JME site in 6p21.3 underlies absence and grand mal epilepsies.

Fifteen-year overview of the International Cell Exchange.

1. Through the International Cell Exchange, 4 cells were shipped for blind typing to tissue typing laboratories on a monthly basis for the past 15 years. As many as 288 laboratories worldwide currently participate in this exercise. 2. The frequency of detection of a total of 85 HLA-A,B,C specificities by the laboratories has been determined. Each specificity has been classified as to the reliability of identification. 3. Lymphocytes from 27 donors were sent for retesting in as long as a 13-year interval. This provided an accurate picture of the advances in percent detection by the laboratories. 4. At least 5 variants were found by shipping cells to the laboratories: variants of A9, B21, B5, Bw52, and B40. 5. B cell lines were sent to investigate the HLA-DR antigens. At least 11 specificities had a concordance rate higher than 80% and 5 had a rate of more than 95%. 6. HLA-A,B antibodies sent for antibody characterization were classified by a new mass program for the likely epitope against which the sera were directed. As many as 28 different epitopes were detected by the sera sent in the serum exchange.

HLA serologic epitopes.

A total of 25,000 antisera from parous women was evaluated for antibody specificity toward HLA epitopes in the A,B, and C loci. It was determined that most of the reactivity patterns against a large panel could be accounted for by the amino acid substitutions at the known residues. It has been surprising that each of the variable amino acids at each residue against which antisera have been found belonged on either the A, B, or C locus.

Overview and epitope matching.

1. Kidney graft survival rates have stabilized over the past 4 years, suggesting that gains achieved with CsA have plateaued. The overall 1-year graft survival is 77% for first cadaver donor transplants, 90% for parental donors, and 93% for HLA-identical sibling donors. Patient survival for all categories is now over 95%. 2. The UNOS 6-antigen match program has resulted in outstanding graft survivals. Of 88 kidneys which were transplanted into first graft recipients, the 6-antigen match kidney had a 1-year graft survival of 91% compared to 74% for the contralateral kidney transplanted locally (p less than 0.008). 3. In highly sensitized patients with more than 80% PRA the shipped 6-antigen matched kidney had a 91% 1-year graft survival rate compared to 72% survival in comparable control patients from the registry (p less than 0.005). In patients with less than 80% PRA, 6-antigen matched kidneys had 90% 1-year graft survival compared to 80% in controls (p less than 0.00001). 4. The spread of 1-year graft survival rates at 68 centers that performed more than 100 transplants was 63-94%. The cumulative graft survival of 6-antigen matches performed at 129 different centers was 90%. Thus, the strong center effect was neutralized by the use of matched transplants. 5. In contrast to the 1% of patients who would receive O-B,DR mismatched transplants on a random basis, if kidneys were shared in the national pool, 74% could receive such a transplant. We therefore propose that a keep one-share one policy be adopted for all kidneys harvested in the United States. If no 0-B,DR mismatched patient is available, both kidneys will be kept by the harvesting center. Since 63% of kidneys are currently being shared with other centers for various reasons, the 75% sharing suggested by the new system should not impose a hardship on transplant centers. The UNOS payback agreement will be replaced by this agreement by which shipping will be done only to achieve excellent matches. 6. In order to achieve a better method of excluding the worst mismatches, an attempt was made to develop a new method of mismatching using amino acid sequences of the HLA specificities. Donor and recipient types can be converted to amino acid sequences and the mismatching done on the basis of amino acids of mismatch at each residue or position. For each residue, specific combinations of amino acid substitutions were examined individually to determine their effect on graft survival. From these studies, a list of "immunogenic" amino acids was prepared, and graft survival was then computed for increasing numbers of amino acids of mismatch.(ABSTRACT TRUNCATED AT 400 WORDS)

Significance of the HLA molecular structure to transplantation.

1. Residues in the first, second, and third domains of HLA-A,B,C, and the first domain of DR beta, DQ alpha, and DQ beta molecule have been assigned to unique A,B,C or DR specificities from the known amino acid sequences. Antibodies were noted to correlate with most of these variable amino acid residues. 2. We therefore conclude that most of the variable residues in the first domain function as immunogens against which the antibodies had been elicited. 3. If the variants at these positions are immunogens, then it follows that matching for transplantation should be done by considering mismatched amino acids rather than the private specificities, as performed today. 4. Molecular matching cannot be performed immediately for the HLA-A,B,C specificities since many specificities are not yet sequenced. For the DR and DQ specificities, the sequences are established, but antisera identifying the subtypes of DR and DQ are only now becoming available. Once patients are typed for the 23 DR beta alleles, 8 DQ alpha alleles and 13 DQ beta alleles, matching should be feasible. 5. Molecular matching combines both public and private specificity matching since it assumes that both types of antigens are distinct. A single extended DR mismatch may involve as many as 76 amino acid residues of mismatch. 6. True cross-reactivity of the HLA molecules can eventually be established through structural studies. Most of the previously described "cross-reactivities" are likely to be shared determinants.

HLA antigens and other genetic markers in the Mapuche Indians of Argentina.

A total of 107 Mapuche Indians living in western Argentina were studied with respect to 16 genetic systems. For HLA, there were a few differences in relation to previous studies; and considering the averages observed in 15 other South American tribes, Mapuche Indians showed low values for A2, A9 and C3, but high ones for A28 and B16. This is the first report of the presence (in low frequencies, 1-6%) of alleles C2, C6 and C7, as well as of DR antigens (most frequent alleles DR4 and DR2) in South American Indians. Some peculiar reactions shown by products of locus B suggest the presence of antigens that are characteristic of the Mapuche. As for the other systems, the frequencies of R1 (Rh) and PGM1(1) were lower but those for r (Rh), GLO1 and Hp1 were higher than the averages obtained considering previous studies of this ethnic group. Other salient findings were the variability observed in the PGM2 and C3 systems, and the low prevalence of Bfs.

Antigenos HLA donantes de sangre portadores cronicos asintomaticos del antigeno de superfice (HBsAg) de la hepatitis viral tipo B. / HLA antigens in blood donors asymptomatic chronic carriers of hepatitis type B surface antigen (HBsAg)

La relacion entre los antigenos de histocompatibilidad del sistema HLA y la incidencia de positividad para el antigeno de superficie del virus de la hepatitis tipo B (HBsAg) ha sido estudiada por diversos investigadores en 3 grupos principales de individuos: a) pacientes con hepatitis cronica activa; b) pacientes con insuficiencia renal cronica, y c) portadores cronicos asintomaticos del HBsAg. Nuestro estudio se efectuo en 37 portadores cronicos asintomaticos del HBsAg, todos ellos donantes de sangre, en su mayoria hombres, en los cuales la presencia de HBsAg fue confirmado por radioinmunoensayo durante un ano como minino.Se utilizaron como grupo control normal, 257 personas de ambos sexos, blancos, sin relacion de parentesco. Se calculo el riesgo relativo que posee un individuo de ser portador de HBsAg segun la presencia de un determinado antigeno de histocompatibilidad. Se estudiaron 13 antigenos del locus A, 15 del locus B y 4 del locus C. Se encontro un aumento estadisticamente significativo en la frecuencia de los antigenos del locus B, el B 15 y el Bw21. Si se multiplica el valor p obtenido por el numero de antigenos estudiados (32), para obtener el valor corregido, solo la presencia aumentada del alelo Bw21 conserva su significado estadistico (p corrigido = 0,03). Aumentos en las frecuencias de los antigenos B15 y Bw21 han sido senalados tambien por otros autores

Antigenos HLA donantes de sangre portadores cronicos asintomaticos del antigeno de superfice (HBsAg) de la hepatitis viral tipo B. / HLA antigens in blood donors asymptomatic chronic carriers of hepatitis type B surface antigen (HBsAg)

La relacion entre los antigenos de histocompatibilidad del sistema HLA y la incidencia de positividad para el antigeno de superficie del virus de la hepatitis tipo B (HBsAg) ha sido estudiada por diversos investigadores en 3 grupos principales de individuos: a) pacientes con hepatitis cronica activa; b) pacientes con insuficiencia renal cronica, y c) portadores cronicos asintomaticos del HBsAg. Nuestro estudio se efectuo en 37 portadores cronicos asintomaticos del HBsAg, todos ellos donantes de sangre, en su mayoria hombres, en los cuales la presencia de HBsAg fue confirmado por radioinmunoensayo durante un ano como minino.Se utilizaron como grupo control normal, 257 personas de ambos sexos, blancos, sin relacion de parentesco. Se calculo el riesgo relativo que posee un individuo de ser portador de HBsAg segun la presencia de un determinado antigeno de histocompatibilidad. Se estudiaron 13 antigenos del locus A, 15 del locus B y 4 del locus C. Se encontro un aumento estadisticamente significativo en la frecuencia de los antigenos del locus B, el B 15 y el Bw21. Si se multiplica el valor p obtenido por el numero de antigenos estudiados (32), para obtener el valor corregido, solo la presencia aumentada del alelo Bw21 conserva su significado estadistico (p corrigido = 0,03). Aumentos en las frecuencias de los antigenos B15 y Bw21 han sido senalados tambien por otros autores