Raman fiber-optical method for colon cancer detection: Cross-validation and outlier identification approach.

Endoscopy plays a major role in early recognition of cancer which is not externally accessible and therewith in increasing the survival rate. Raman spectroscopic fiber-optical approaches can help to decrease the impact on the patient, increase objectivity in tissue characterization, reduce expenses and provide a significant time advantage in endoscopy. In gastroenterology an early recognition of malign and precursor lesions is relevant. Instantaneous and precise differentiation between adenomas as precursor lesions for cancer and hyperplastic polyps on the one hand and between high and low-risk alterations on the other hand is important. Raman fiber-optical measurements of colon biopsy samples taken during colonoscopy were carried out during a clinical study, and samples of adenocarcinoma (22), tubular adenomas (141), hyperplastic polyps (79) and normal tissue (101) from 151 patients were analyzed. This allows us to focus on the bioinformatic analysis and to set stage for Raman endoscopic measurements. Since spectral differences between normal and cancerous biopsy samples are small, special care has to be taken in data analysis. Using a leave-one-patient-out cross-validation scheme, three different outlier identification methods were investigated to decrease the influence of systematic errors, like a residual risk in misplacement of the sample and spectral dilution of marker bands (esp. cancerous tissue) and therewith optimize the experimental design. Furthermore other validations methods like leave-one-sample-out and leave-one-spectrum-out cross-validation schemes were compared with leave-one-patient-out cross-validation. High-risk lesions were differentiated from low-risk lesions with a sensitivity of 79%, specificity of 74% and an accuracy of 77%, cancer and normal tissue with a sensitivity of 79%, specificity of 83% and an accuracy of 81%. Additionally applied outlier identification enabled us to improve the recognition of neoplastic biopsy samples.

Virtual staining of colon cancer tissue by label-free Raman micro-spectroscopy.

The great capability of the label-free classification of tissue via vibrational spectroscopy, like Raman or infrared imaging, is shown in numerous publications (review: Diem et al., J. Biophotonics, 2013, 6, 855-886). Herein, we present a new approach, virtual staining, that improves the Raman spectral histopathology (SHP) images of colorectal cancer tissue by combining the integrated Raman intensity image in the C-H stretching region (2800-3050 cm-1) with the pseudo-colour Raman image. This allows the display of fine structures such as the filamentous composition of muscle tissue. The morphology of the virtually stained images is in agreement with the gold standard in medical diagnosis, the haematoxylin-eosin staining. The virtual staining image also represents the whole biochemical fingerprint, and several tissue components including carcinoma were identified automatically with high sensitivity and specificity. For fast tissue classifications, a similar approach was applied on coherent anti-Stokes Raman scattering (CARS) spectral data that is faster and therefore potentially more suitable for clinical applications.

Discrepancies between sodium concentrations measured by the Kodak Ektachem 700 and by dilutional and direct ion-selective electrode analyzers.

We have identified rare (approximately 0.2% of all samples), but clinically significant, discrepancies between serum or plasma sodium concentrations measured with the Kodak Ektachem 700's direct ion-selective electrode (ISE) method and concentrations measured with two other analyzers: the Beckman Synchron CX3's dilutional ISE instrument and the Radiometer KNA2 instrument for sodium-potassium analysis by the direct ISE method. The differences do not appear to be related to any previously identified sources of discrepancy, such as variations in triglycerides, bicarbonate, total protein, albumin, or gamma-globulin, the presence of paraproteins, or interference by benzalkonium chloride from heparinized catheters. They occurred despite the use of Gen 04 reference fluid on the Ektachem. We could not identify any drug or family of drugs that the patients had taken in common and that might influence the results. Until this problem is resolved, Ektachem users should be aware of the potential for discrepancies of > 6 mmol/L in measurements of sodium concentrations.

Long-term glucose control in patients with pancreatic transplants.

OBJECTIVE: To evaluate the long-term effect on blood glucose levels of successful transplantation of part or all of an intact human pancreas in patients with insulin-dependent diabetes mellitus (IDDM). DESIGN: Cohort study. SETTING: Referral medical center. PATIENTS: Thirty-seven patients with adequate data, representative of a group of 62 patients with functioning grafts (that is, insulin-independent) at 2 years after transplantation. The 62 patients came from a total of 178 patients in the University of Minnesota series as of July 1987, for a 2-year success rate of 35% (95% Cl, 27.8% to 41.8%). These patients were compared to two diabetic control groups (18 patients with IDDM under standard insulin treatment in a university diabetes clinic and 11 patients with IDDM whose pancreas grafts had failed) and to two nondiabetic groups (14 nondiabetic patients who received immunosuppressive drugs after kidney transplantation and 196 healthy control subjects). MEASUREMENTS: Glycosylated hemoglobin was measured by the high-pressure liquid chromatography method, as total A1 (Hb A1) and the A1C subfraction (Hb A1C); results were expressed as a percentage of total hemoglobin. MAIN RESULTS: Before pancreas transplantation, the 37 patients in the study group had a mean Hb A1 of 10.8%, consistent with moderate to marked hyperglycemia and not statistically different from the levels in the diabetic control groups. All 37 patients had values above the therapeutic target range of 5.4% to 7.4%. However, at 1 and 2 years after transplantation, the mean Hb A1 value had fallen sharply to 6.7% and 6.5%, respectively, well within target range (Cl of the difference, 3.4% to 4.8%; P less than 0.001). These levels did not differ from the mean Hb A1 in the nondiabetic kidney transplant recipients but were slightly above the 6.2% value for the 196 healthy controls (Cl of the difference at 1 year, 0.2% to 0.8%). Serial values were available on 6 subjects for 5 years; these values were all well within target range. As expected, Hb A1C values were parallel to those of Hb A1. CONCLUSIONS: Pancreas transplantation, in our successful cases, lowered glycosylated hemoglobin to normal or near-normal levels that were sustained for as long as 5 years. These results compare favorably with those in our patients on standard treatment, and also with those in similar patients on intensive control reported by others. Further effort to improve transplant methods appears to be warranted.

Improved method for identifying alpha 1-antitrypsin Pi M subtypes by isoelectric focusing in agarose.

We describe an improved method for the classification of alpha 1-antitrypsin variants by isoelectric focusing in agarose. Identification of the three Pi M subtypes can now be made by using a narrow-range carrier ampholyte (pH 4.2-4.9) and pretreating serum with dithioerythritol-iodoacetic acid to enhance band resolution. Phenotype results for two groups of Pi M homo- and heterozygotes are compared to illustrate the improved accuracy of the new method.

Response of serine antiproteases to growth hormone therapy in growth hormone deficient children.

Growth hormone regulates the hepatic mRNA levels of alpha 1-antitrypsin and two contrapsin-like mRNAs in the rat. To determine whether growth hormone regulates similar serine protease inhibitors in humans, we measured serum alpha 1-antitrypsin, alpha 1-antichymotrypsin, and antithrombin III by radioimmunodiffusion in 16 growth hormone deficient children before and after growth therapy. Of the 19 determinations made, 17/19 showed an increase in alpha 1-antitrypsin after administration of growth hormone, 198.6 +/- 39.1 mg/dl before growth hormone and 239.4 +/- 44 mg/dl after growth hormone (p = 0.005). Specificity of the response for alpha 1-antitrypsin was indicated by the fact that neither alpha 1-antichymotrypsin or antithrombin III values changed after growth hormone (p = 0.6 and 0.5, respectively). These data are compatible with the hypothesis that growth hormone regulates serine protease inhibitors in humans and suggests that investigation of other members of the serpin gene family might prove fruitful in defining additional growth hormone target genes.

Contribution of monocyte-macrophage system to serum alpha 1-antitrypsin.

The major serum antiprotease is alpha 1-antitrypsin (A1AT). Deficiency of A1AT can result in infantile cirrhosis and premature emphysema, both of which have a high degree of morbidity and significant mortality. Although synthesized primarily by the liver, A1AT has been histochemically localized in monocytes and macrophages in vitro and has been shown to be produced in tissue culture of monocyte-macrophage origin. This study was planned to quantitatively and qualitatively assess the in vivo monocyte-macrophage system contribution to serum A1AT. We used bone marrow transplantation (BMT) as an experimental method because there is commanding evidence that after engraftment, the monocyte-macrophage system of the recipient is replaced by that of donor origin. Protease inhibitor (Pi) typing was done on 150 potential BMT recipients and on their potential donors before transplantation. From these initial recipients, 92 eventually underwent transplantation, and 11 recipient-donor pairs, in which each donor's Pi type contained a band not in the recipient's Pi type, were chosen for the study. Six recipients survived beyond 100 days after BMT, and in these cases the donor contained either an S or an M2 band in his or her Pi type not present in the recipient. Using a silver stain method on diluted serum of known M1M2 and MS types, we were able to detect a 2% dilution of the S band and a 25% dilution of the M2 band. When the same method was applied to gels used in typing recipient Pi after BMT, we were unable to detect any contribution to serum A1AT by the donor monocyte-macrophage system.

Immunonephelometry of apolipoprotein B with a centrifugal analyzer.

We developed an automated immunonephelometric assay for the measurement of apolipoprotein B (apo B) with a light-scattering microcentrifugal analyzer. Pretreating specimens with a dilute solution of Tween 20 or triglyceride lipase decreased the nephelometric response of apo B. Polyethylene glycol is included in the reaction mixture, and the reaction is complete within 4 min. The method is precise (CV = 6.5%, mean = 0.68 g/L) and the standard curve is linear to an apo B concentration of 2.8 g/L. Lipemia does not interfere with the method if grossly lipemic specimens are centrifuged to remove chylomicrons.

Fatal liver disease associated with alpha 1-antitrypsin deficiency PiM1/PiMduarte.

A 53-yr-old man with a rare form of partial alpha 1-antitrypsin deficiency, PiM1/PiMduarte, died of endstage cirrhosis. Typical cytoplasmic alpha 1-antitrypsin globules were present in the hepatocyte cytoplasm. Initial protease inhibitor phenotyping on the patient was reported as normal PiM1 in more than one laboratory. This case emphasizes the diagnostic importance of alpha 1-antitrypsin and illustrates the point that protease inhibitor phenotyping without family genotyping may be misleading in heterozygous patients with liver disease.

Nephelometric assay of apolipoprotein A-I with a centrifugal analyzer.

We developed an automated immunonephelometric assay for quantification of human apolipoprotein A-I (apo A-I) with a fluorescence light-scattering microcentrifugal analyzer. The presence of polyethylene glycol and Tween 20 in the reaction mixture ensures maximum exposure of the antigenic sites of the apoprotein so that immune complex formation occurs more rapidly (reaction is complete within 2 min) and to a greater extent. Lipemia and hemolysis do not interfere with the measurement of apo A-I. The method requires only 10 microL of specimen and is fast and easy to perform. Results vary linearly with apo A-I concentrations to 2.5 g/L. Assay precision (CV) was 3.1% for a specimen with an apo A-I concentration of 1.45 g/L, and the lower limit of detection was 0.15 g/L. Values for a candidate Reference Material agree well with those reported in an international survey (Clin Chem 1985;30:223-8).

Artifactual elevation of serum creatinine level due to fasting.

Abnormal serum creatinine (1.6 mg/dL) and creatinine clearance (33 mL/min) levels found in a 50-year-old woman during fasting were corrected with refeeding. Five healthy subjects who fasted for 96 hours demonstrated an increase in their mean serum creatinine level from 1.0 +/- 0.08 to 1.7 +/- 0.11 mg/dL as determined by Jaffé's method. This increase was probably an artifact caused by the rise in the serum acetoacetate level during fasting. The serum creatinine level determined by an enzymatic method and serum urea nitrogen level did not change substantially during the fast. We conclude that fasting may cause an artifactual increase in the serum creatinine level determined by Jaffé's method, the method used by most clinical laboratories.

Enzyme inactivation in serum before determination of total bile acids.

Endogenous NADH-generating enzymes must be inactivated before total serum bile acids can be measured accurately by the direct enzymic method. To do this, we pretreat the sera with NaOH, in a final concentration of 0.1 mol/L. Consequently, lactate dehydrogenase activity at least as high as 30 000 U/L is destroyed, obviating blank determinations. Values for bile acid in serum, so obtained, agree with values obtained after pretreatment with heat, an alkali-methanol solution, or sodium pyruvate, but our pretreatment has the advantages of ease, speed, economy, and negligible blank values.

Effect of protein on the determination of total bile acids in serum.

We measured total serum bile acids on a fluorescence-light-scattering micro centrifugal analyzer by the direct enzymatic method with 3 alpha-hydroxysteroid dehydrogenase (EC 1.1.1.50) and with resazurin as a fluorogenic electron acceptor. We found that serum protein has an inhibitory effect on the measurement of bile acids, but this effect was eliminated by adding bovine serum albumin to the reaction mixture in a final protein concentration (12.2 g/L) that was high compared with that contributed by a normal serum specimen. The assay is a sensitive method that reaches equilibrium in 5 min. The method is microscale (5 microL of sample, 150 microL of working reagent), is easy to perform, and is accurate (analytical recovery = 104.1%) and precise (CV = 11.1 and 5.7% on specimens with bile acid concentrations of 7.6 and 35.4 mumol/L, respectively). Normal values are 1-12 and less than 9 mumol/L on nonfasting and fasting individuals, respectively. Pure 3 alpha-hydroxysteroid dehydrogenase must be used: we found several enzyme preparations that gave falsely high values for bile acid.

Improved simultaneous gas-chromatographic analysis for homovanillic acid and vanillylmandelic acid in urine.

We describe an improved gas-chromatographic method for the simultaneous quantitation of the catecholamine metabolites, homovanillic acid (3-methoxy-4-hydroxyphenylacetic acid) and vanillylmandelic acid (3-methoxy-4-hydroxymandelic acid). Our improvements in the method of Muskiet et al. (Clin. Chem. 23: 863, 1977) include a shorter program time and a longer silylation interval. Recovery and precision data obtained by this improved technique are similar to those of Muskiet et al. Vanillylmandelic acid results (y) were compared with those by the method of Pisano et al. (Clin. Chim. Acta 7: 285, 1962). The relation is expressed by the equation y = 0.52 + 1.05x (Sy . x = 2.33 mg/24 h and r = 0.997). Results for homovanillic acid (y) were compared with those by the method of Knight and Haymond (Clin. Chem. 23: 2007, 1977); the equation was y = 0.84 + 0.90x (Sy . x = 2.04 and r = 0.97). Retention times are also reported for several phenolic acids and other related compounds found in urine.