Combination of multiple serum markers using an artificial neural network to improve specificity in discriminating malignant from benign pelvic masses.

A panel of four selected tumor markers, CA 125 II, CA 72-4, CA 15-3, and lipid-associated sialic acid, was analyzed collectively using an artificial neural network (ANN) approach to differentiate malignant from benign pelvic masses. A dataset of 429 patients, 192 of whom had malignant histology, was retrospectively used in the study. A prototype ANN classifier was developed using a subset of the data which included 73 patients with malignant conditions and 101 patients with benign conditions. The ANN classifier demonstrated a much improved specificity over that of the assay CA 125 II alone (87.5% vs 68.4%) while maintaining a statistically comparable sensitivity (79.0% vs 82.4%) in discriminating malignant from benign pelvic masses in an independent validation test using data from the remaining 255 patients which had been set aside and kept blind to the developers of the ANN system. A similar improvement in specificity was observed among patients under 50 years of age (82.3% vs 62.0%). The ANN system was further tested using additional serum specimens collected from 196 apparently healthy women. The ANN system had a specificity of 100.0% compared to that of 94.8% with the assay CA 125 II alone.

Understanding artificial neural networks and exploring their potential applications for the practicing urologist.

Artificial neural networks (ANNs) are complex mathematical models that are distantly based on the human neuronal structure. They are capable of modeling elaborate biologic systems without making assumptions based on statistical distributions. Preliminary work has been reported on their application in urology. The initial results have been promising, particularly as an additional tool in the detection of early prostate cancer using the ProstAsure Index, which has been the most extensively studied urologic ANN to date. We review the basic concepts behind ANNs and examine currently existing and potential future applications of this new dynamic technology both in urology and in general clinical medicine.

Evaluation of prostAsure index in the detection of prostate cancer: a preliminary report.

OBJECTIVES: Although prostate-specific antigen (PSA) has revolutionized the detection of prostate cancer, it has definite limitations with respect to its clinical sensitivity and specificity. Because a substantial number (20% to 40%) of men undergoing radical prostatectomy have a PSA level of 4.0 ng/mL or less, any new test offering diagnostic improvement must perform well in patients whose PSA level is less than or equal to 4.0 ng/mL, as well as in patients whose PSA is greater than 4.0 ng/mL. The performances of two tests, the ProstAsure index and the percent free PSA test, were evaluated in detecting cancer. METHODS: We retrospectively analyzed serum samples from 225 men who were grouped into three categories: 94 men who had a normal digital rectal examination and a serum PSA level of 4.0 ng/mL or less, 77 men who were clinically suspected of having benign prostatic hyperplasia (BPH) with a serum PSA level of 4.0 ng/mL or less, and 54 men with localized prostate cancer. The PSA assays were performed using the Hybritech and Tosoh assays and the ProstAsure index was determined by Global Health Net, Savannah, Ga. Receiver operator characteristic (ROC) curves were constructed to evaluate the performance of these two tests, and the areas under the curve were compared for significance. RESULTS: The sensitivity and specificity of detecting prostate cancer using ProstAsure were 93% and 81%, respectively. Using a cutoff value of 15%, the sensitivity and specificity of detecting cancer for percent free PSA were 80% and 74%, respectively (sensitivity increased to 93% and specificity to 59% for free PSA at 19%). In men with a total serum PSA level of 4.0 ng/mL or less, ProstAsure had a lower false-positive rate compared to free PSA level at 19% for men with or without clinical BPH as well as for men without clinical BPH using a 15% free PSA threshold. ProstAsure left fewer cancers undetected (7%) compared to free PSA at the 15% cutoff (20%). CONCLUSIONS: In this study of selected men, ROC curve analysis shows a statistically significant advantage in performance (P = 0.0023) for the ProstAsure index compared to free PSA in detecting prostate cancer.

Evaluation of a new biochemical index for the estimation of bone demineralization using artificial intelligence.

The use of artificial neural networks (ANN) for the identification of a positive correlation between the QuiOs quotient, a single-valued indicator of bone mineral density, and multisite dual energy x-ray absorptiometry (DEXA) measurements is described. The measurements were obtained using the Hologic QDRR-2000 x-ray bone densitometer in a multicenter clinical trial including 374 female patients. The QuiOs quotient estimates bone mineral density and determines the severity of bone density loss. This quotient is calculated by using a proprietary software program to perform a multivariate analysis of the results of testing of serum levels of calcium, phosphate, total alkaline phosphate, two alkaline phosphatase isoenzymes (liver and intestine), estradiol, and progesterone. The results show that given the four DEXA measurements obtained from the bone densitometer, the trained neural network can predict whether the corresponding QuiOs score of the same patient will be below the cutoff score of 0.690. The findings in this study indicate a positive correlation between DEXA measurements and the QuiOs quotient obtained from two different sources.

Comparative study of a biochemical index for the estimation of bone demineralization and dual photon absorptiometry.

Multivariate classification methods were used to create an early detection technique for determining bone density. This biochemical index (QuiOs) is clinically useful as a potential adjunct in identifying the presence of biochemical deficiencies known to cause osteopenia and the devastating effects of osteoporosis. The test uses the following serum concentrations of a predetermined set of blood constituents: calcium, phosphorus, alkaline phosphatase (ALP), two alkaline phosphatase isoenzymes (liver and intestine), estradiol, and progesterone. Using the results of these six biochemical and hormonal tests, a correlation equation was developed that demonstrates a nonlinear relationship between QuiOs and Ward's triangle of DPA. A sensitivity of 86% and a specificity of 80% was demonstrated for this biochemical index against DPA in this clinical trial.

Immunological responses to a murine mammary adenocarcinoma. Outgrowth of pulmonary occult metastases induced by immunosuppressive perturbations.

Intravenously (i.v.) or subcutaneously (s.c.) injected T1699 tumor cell subpopulations (Ts, and TLI-1) were destroyed more rapidly in BALB/c nu/nu athymic mice than in syngeneic DBA/2J mice. The number of artificial pulmonary metastases of TLI-1 tumors was significantly lower, and the growth rate of s.c. Ts, and TLI-1 tumors was correspondingly slower in the nude mice. No macroscopic outgrowth of pulmonary metastases of Ts tumors was detected in either group of s.c. tumor-bearers, even though the s.c. tumors progressed and killed the hosts. Unexpectedly, however, s.c. implantation in normal DBA/2J mice of lung homogenates not only from DBA/2J but also from s.c. Ts tumor-bearing BALB/c nu/nu mice produced tumors, suggesting that significant numbers of clonogenic Ts cells were present in the lungs of animals without spontaneous outgrowth of pulmonary metastases. Perturbations of s.c. Ts tumor-bearing DBA/2J or BALB/c nu/nu mice with immunosuppressive drugs or with anti-mouse thymocyte serum, respectively, induced rapid outgrowth of pulmonary metastases.

Mitogenic augmentation of T cells from immunodepressed cancer patients by a murine interleukin-1 (IL-1).

An interleukin-1 (IL-1) lymphocyte-activating factor, with an approximate molecular weight of 13,000-16,000 was isolated from the culture supernatants of a murine macrophage-like cell line, P388D1. Contrary to the general belief that murine mediator is incapable of stimulating human lymphocyte mitogenic responses, leukoagglutinin (LA)-induced mitogenesis of peripheral blood lymphocyte (PBL) were significantly affected by murine IL-1. On one hand, when PBL were obtained from either normal healthy individuals or from cancer patients who still possessed normal levels of general immunocompetence, their mitogenic responses were not augmented any further by the murine mediator. Instead, slight but significant suppression were noted in most cases. On the other hand, the LA-induced responses of PBL from 5 immunodepressed cancer patients were markedly augmented by murine IL-1.

Antigenic cross-reactivity between human leukemic cells and dinitrophenylated normal leukocytes.

Dinitrofluorobenzene-modified normal human leukocytes (DNP-LK) have been shown to evoke the production of antibodies, which following absorption with erythrocytes and unmodified leukocytes (UN-LK) had residual leukoagglutinating activity against human leukemic cells (HLC). In the present study we observed that rabbit antisera prepared against cell membrane extracts of DNP-modified granulocytes (DNP-GM) or lymphocytes (DNP-LM) were cytotoxic to HLC. By passive hemagglutination, these antisera were specifically reactive with DNP-GM or DNP-LM conjugated to sheep erythrocytes (SRBC) but not with DNP-modified bovine serum albumin (DNP-BSA) or bovine gamma globulin (DNP-BGG) nor with UN-LK, suggesting that the antisera were devoid of anti-DNP antibodies. Rabbit anti-DNP-BSA or anti-DNP-BGG failed to react with DNP-LK, and anti-DNP antibodies in these sera could not be absorbed by DNP-LK, suggesting that DNP groups either were not expressed on the surface of DNP-LK or were not detectable by these methods. These data, together with our recent finding that dinitrophenylation alters the expression of histocompatibility antigens, suggest that neoantigens cross-reactive with HLC antigens are induced by DNP modification of membrane antigens of normal leukocytes, and that the antibodies produced against these DNP-modified cells are directed mainly against the modified protein and not against the DNP moiety per se.

Autoimmune neutropenia in early infancy: a review.

Fourteen infants with autoimmune neutropenia reported in the literature have been reviewed. Autoantibodies directed against their own neutrophils were demonstrated in the sera of these infants by agglutination, complement-dependent cytotoxicity, and immunofluorescence techniques. These antibodies were highly specific and were directed against antigens present on neutrophils. Among the currently known neutrophil antigens (NA1, NA2, NB1, NC1, NE1, ND1, 9A), antibodies reacting with either NA1 or NA2 have been identified frequently in the sera of infants with autoimmune neutropenia. Good correlation was demonstrated between the presence or absence of autoantibodies and the episodes of neutropenia in many cases. Antibodies from the patients also reacted with neutrophils from their parents and from normal unrelated volunteers when they shared the neutrophil-specific antigen against which the antibody was directed. Antibodies demonstrable by complement-dependent cytotoxicity appeared to detect different antigens which may also cause autoimmune neutropenia. Infants with this disorder were healthy at birth and for a few months afterwards, then became chronically ill with such symptoms as intermittent fever, diarrhea, and infections. Their hemoglobin levels, lymphocyte and platelet counts, and other immunological studies were normal except for severe to moderate neutropenia.

Role of neutrophil antigen NA1 in an infant with autoimmune neutropenia.

A nontransfused 14-month-old female infant was investigated for persistent neutropenia of eight months' duration, with absolute neutrophil counts ranging from 410 to 935 cu mm. The patient's sera reacted with neutrophils from her own peripheral blood, from normal donors, and from her mother, all these having the neutrophil antigen NA 1, but not with neutrophils from NA 1-negative donors, including the father. The autoantibody was detectable by capillary agglutination and by indirect immunofluorescence techniques but not by complement-dependent cytotoxicity. No antibody was found in the mother's serum. Studies on three occasions showed good correlation between the appearance of circulating autoantibody and the peripheral neutrophil counts. Our observations, together with previously published reports, suggest a possible relationship of NA 1 antigen and the disease susceptibility of NA 1-positive infants to autoimmune neutropenia.

An improved method for rapid layering of Ficoll-Hypaque double density gradients suitable for granulocyte separation.

A simple technique is described to layer Ficoll-Hypaque double density gradients suitable for rapid isolation of human granulocytes using disposable plastic syringes and 20-gauge needles. The syringes without plungers were kept at 45 degrees angle initially at the top of the test tube, the tip of the needle touching the side of the test tube. While layering the gradients, the syringes were progressively adjusted at 3 different positions which enhanced the smooth flow by forming a continuous stream from the tip of the needle to the top of the lower gradient resulting in excellent interfaces without causing turbulence or inadvertent mixing. This method is useful particularly when unknown sera are screened against granulocytes or lymphocytes from a large number of normal donors.

Quantitative assay for lectin-induced cytoagglutination by means of an electronic particle-counting technique.

A method employing an electronic particle-counting technique was used to quantify lectin-induced agglutination of human granulocytes and lymphocytes with either concanavalin A or wheat germ agglutinin. The number and mean volume of single cells and aggregates in the presence of increasing concentrations of lectin were computed from 95% confidence intervals. Agglutination depended on both the number of free cells and the number and size of the cell clusters. Changes in these two variables were mutually independent of one another, and both were simultaneously determined. An index of agglutination that takes the effect of these two variables into account was defined as (formula: see text) VA equals mean volume of cell aggregates, NS equals number of single cells, NA equals number of cell aggregates, VS equals mean volume of single cells, rb equals r at a given lectin concentration, and ra equals r in the absence of lectin. For any combination of lectin and cell type, the agglutination curve, as described by zeta, consisted of two components: a) a flat region in which zeta remained constant with increasing lectin concentrations and b) a region in which zeta increased linearly as a function of the logarithm of lectin concentration. The shapes of these curves offered two parameters for quantitative comparison of agglutinability: 1)threshold concentration, defined as the minimum concentration of lectin (microgram/ml) required to bring about a measurable rise in zeta and 2) the concentration gradient, defined as the change in zeta for an increase of one log unit in the concentration of the lectin in the range beyond the threshold concentration. This method offers a high degree of quantification and provides reliable information that can be meaningfully correlated with cell surface characteristics.

Relationship between concanavalin A-induced agglutinability of murine leukemia cells and their propensity to form heterotypic aggregates with syngeneic lymphoid cells.

The agglutinabilities of the murine leukemia cell lines L1210, P388, and C1498 were determined in the presence of concanavalin A (Con A) by a quantitative cytoagglutination assay. The propensity of these cells to form heterotypic aggregates with normal syngeneic spleen cells, those obtained from mice carrying the respective leukemias in ascitic form, and syngeneic lung cells also were determined. Con A caused agglutination of all five types of leukemia cells and the resulting agglutination patterns had certain characteristics. Threshold concentrations of Con A, below which no significant cytoagglutination occurred, were very low. A steady increase in zeta, a previously defined index of agglutination that simultaneously takes into account the number of free cells and the number and size of the aggregates, was observed with increasing concentrations of Con A until a plateau was reached at 25-50 microgram/ml and this extended over a wide range of lectin concentrations (50-1,000 micrograms/ml). Self-aggregation of leukemia cells was not observed, and their propensity to form heterotypic aggregates with syngeneic lung cells was negligible. However, all leukemia cell lines formed measurable aggregates with spleen cells from both normal and leukemia-bearing mice; these aggregates usually reached a maximum plateau between 30 and 35 minutes of incubation and remained constant thereafter. Aggregation of leukemia cells with spleen cells from leukemic mice always was greater than that with spleen cells from normal mice. Con A agglutinability of leukemia cells was correlated with their propensity to form heterotypic aggregates, which suggests that Con A agglutinability of leukemia cells was correlated with their propensity to form heterotypic aggregates, which suggests that Con A receptor carbohydrate moieties may be involved in the intercellular adhesion leading to heterotypic aggregate formation.

Differences in mitogenic responses of murine T cells to two distinct phytohemagglutinin (PHA) subcomponents.

Two distinct mitogenic subcomponents of phytohemagglutinin (PHA)--leucoagglutinin (LA) and "purified" PHA--apparently stimulated different subpopulations of murine T cells. In the DBA/2J strain, the mitogenic responses of splenic lymphocytes to LA reached maximal levels after 24 to 36 hr exposure and almost completely disappeared by 48 hr, whereas maximal responses to PHA were maintained after 48 hr incubation. The levels of LA-responding T cells were highest in DBA/2J spleens at 5 weeks of age but markedly declined by 9 weeks of age, whereas thymic levels of LA-responding T cells reached a maximum at 9 weeks of age and remained maximal past 15 weeks of age. PHA-responding cells, in contrast, reached maximal levels in both the spleen and thymus of DBA/2J mice at 9 weeks of age. In C57BL/6J mice, splenic and thymic lymphocytes responded similarly to both components, except that the response of splenic lymphocytes to PHA reached a maximum after shorter incubation time and declined sooner than their response to LA. The mitogenic responses of C57BL/6J thymocytes to both components were already at their peak by 5 weeks of age and almost totally disappeared by 9 weeks, whereas the responses of splenic lymphocytes were maximal at 9-15 weeks of age. The responses of DBA/2J splenocytes to LA was significantly augmented by PHA, but LA markedly suppressed the proliferative responses to PHA.

Binding of fluoresceinated lectins to normal and dinitrofluorobenzene treated human leucocytes.

Using fluoresceinated lectins we have shown the receptor distribution on normal human granulocytes and lymphocytes following tagging with 1-fluoro- 2, 4-dinitrobenzene (DNFB). DNP-tagged cells exhibited strong, smooth membrane staining and produced smaller patches dispersed uniformly over the entire cell surface.