Machine learning-based nonlinear regression-adjusted real-time quality control modeling: a multi-center study.

OBJECTIVES: Patient-based real-time quality control (PBRTQC), a laboratory tool for monitoring the performance of the testing process, has gained increasing attention in recent years. It has been questioned for its generalizability among analytes, instruments, laboratories, and hospitals in real-world settings. Our purpose was to build a machine learning, nonlinear regression-adjusted, patient-based real-time quality control (mNL-PBRTQC) with wide application. METHODS: Using computer simulation, artificial biases were added to patient population data of 10 measurands. An mNL-PBRTQC was created using eight hospital laboratory databases as a training set and validated by three other hospitals' independent patient datasets. Three different Patient-based models were compared on these datasets, the IFCC PBRTQC model, linear regression-adjusted real-time quality control (L-RARTQC), and the mNL-PBRTQC model. RESULTS: Our study showed that in the three independent test data sets, mNL-PBRTQC outperformed the IFCC PBRTQC and L-RARTQC for all measurands and all biases. Using platelets as an example, it was found that for 20 % bias, both positive and negative, the uncertainty of error detection for mNL-PBRTQC was smallest at the median and maximum values. CONCLUSIONS: mNL-PBRTQC is a robust machine learning framework, allowing accurate error detection, especially for analytes that demonstrate instability and for detecting small biases.

Bibenzyl-Phenylpropane Hybrids with Immunosuppressive Activities from Dendrobium Nobile Lindl.

Fifteen bibenyls and four fluorenones, including five new bibenzyl-phenylpropane hybrids, were isolated from the aerial part of Dendrobium nobile Lindl. Their structures were determined by spectroscopic methods. Bioassay on the LPS-induced proliferations of mouse splenic B lymphocytes, and Con A-induced T lymphocytes showed that compounds 1, 2, and 14 showed excellent immunosuppressive activities with IC50 values of 1.23, 1.01, and 3.87 µM, respectively, while compounds 3-4, 7, 10, 13, and 15 exhibited moderate immunosuppressive activities with IC50 values ranging from 6.89 to 14.2 µM.

A highly accurate delta check method using deep learning for detection of sample mix-up in the clinical laboratory.

OBJECTIVES: Delta check (DC) is widely used for detecting sample mix-up. Owing to the inadequate error detection and high false-positive rate, the implementation of DC in real-world settings is labor-intensive and rarely capable of absolute detection of sample mix-ups. The aim of the study was to develop a highly accurate DC method based on designed deep learning to detect sample mix-up. METHODS: A total of 22 routine hematology test items were adopted for the study. The hematology test results, collected from two hospital laboratories, were independently divided into training, validation, and test sets. By selecting six mainstream algorithms, the Deep Belief Network (DBN) was able to learn error-free and artificially (intentionally) mixed sample results. The model's analytical performance was evaluated using training and test sets. The model's clinical validity was evaluated by comparing it with three well-recognized statistical methods. RESULTS: When the accuracy of our model in the training set reached 0.931 at the 22nd epoch, the corresponding accuracy in the validation set was equal to 0.922. The loss values for the training and validation sets showed a similar (change) trend over time. The accuracy in the test set was 0.931 and the area under the receiver operating characteristic curve was 0.977. DBN demonstrated better performance than the three comparator statistical methods. The accuracy of DBN and revised weighted delta check (RwCDI) was 0.931 and 0.909, respectively. DBN performed significantly better than RCV and EDC. Of all test items, the absolute difference of DC yielded higher accuracy than the relative difference for all methods. CONCLUSIONS: The findings indicate that input of a group of hematology test items provides more comprehensive information for the accurate detection of sample mix-up by machine learning (ML) when compared with a single test item input method. The DC method based on DBN demonstrated highly effective sample mix-up identification performance in real-world clinical settings.

Evaluation of a new hemoglobin A1c analyzer for point-of-care testing.

BACKGROUND: There has been an increasing desire for the use of point-of-care testing (POCT) by both primary care clinicians and patients. This study aimed to evaluate the performance of a new POCT analyzer for hemoglobin A1c (HbA1c) testing. METHODS: We assessed the accuracy, precision, and linearity of the POCT HbA1c analyzer (A1C EZ 2.0) with the Tosoh G8 Analyzer as comparative instrument, following the Clinical and Laboratory Standards Institute (CLSI) protocols. We evaluated sensitivity and specificity of the A1C EZ 2.0 in the clinical diagnosis of diabetes among 842 subjects from 79 communities in Beijing, China. RESULTS: Using regression analysis, the slope of the A1C EZ 2.0 vs the Tosoh G8 Analyzer was 0.9938, with an intercept of 0.0964 and a concordance correlation coefficient of 0.978. For precision, the reproducibility of CV (CVT ) were 3.7% and 2.7% at a lower (36 mmol/mol (5.4%)) and higher (107 mmol/mol (11.9%)) level of HbA1c respectively. The area under the receiver operating characteristic (ROC) curve for clinical diagnosis of diabetes was 0.911 with the HbA1c cut-off value of 44 mmol/mol (6.14%). At the HbA1c level of 48 mmol/mol (6.5%), the sensitivity and specificity were76.1% and 86.6%. CONCLUSION: The A1C EZ 2.0 has a high accuracy and precision, with a wide range of linearity, compared to a comparative laboratory instrument. It met analytical quality specifications and could be suitable for the clinical management of diabetes mellitus.

Establishment and validation of serum free light chain reference intervals in an ethnic Chinese population.

BACKGROUND: Quantification of serum free light chains (FLC) and calculation of a kappa/lambda ratio using polyclonal antisera based immunoassays provide laboratories with a sensitive alternative to urine protein electrophoresis (UPE). However, the published 0.26 - 1.65 serum FLC kappa/lambda ratio reference intervals may not be suitable for different ethnic populations (such as the Han Chinese population presented) and require validation. This is particularly important where there are significant differences in ethnicity, and hence HLA prevalence, in the population studied. METHODS: Serum FLC reference intervals were determined using 326 Han Chinese blood donor volunteers. Sensitivities and specificities for the (i) serum FLC kappa/lambda ratio reference interval and (ii) UPE analyses were determined using 68 pre-treatment, serum immunofixation (sIFE) positive multiple myeloma (MM) patient samples, 54 sera from MM patients undergoing treatment, and 56 sIFE-negative samples from patients with no plasma cell dyscrasia (PCD). RESULTS: The 100% range for the serum FLC kappa/lambda ratio in this Han Chinese population was 0.32 - 1.52. Both Han Chinese blood donors and published kappa/lambda ratio reference ranges demonstrated higher diagnostic sensitivity and specificity for PCD than was seen with UPE. Highly abnormal serum FLC kappa/lambda ratios were observed in 68% of MM patients who had a negative UPE. Furthermore, a MM screening panel of SPE plus serum FLC assays achieved 100% diagnostic sensitivity compared to 97% for a UPE plus SPE algorithm. For MM patients undergoing therapy, 70% of UPE negative samples displayed an abnormal FLC ratio. CONCLUSIONS: This study confirms the requirement to verify normal FLC reference ranges in local populations. This Han Chinese reference range is narrower than the published range. FLC testing provides a robust, sensitive, and specific alternative to classic UPE assessment.

[Models for estimating foliar Fe and Mn Concentration of Armeniaca vulgaris cv. Luntaibaixing using spectral reflectance].

Aimed at providing technology for a rapid nutrition diagnosis system of micronutrients in Armeniaca vulgaris cv. Luntaibaixing, we established an element concentration estimation model for its foliar ferrum (Fe) and manganese (Mn) concentration based on spectrum analysis. The foliar spectrum reflectance at various phenological periods of fruit development under different soil fertility conditions was measured by Unispec-SC spectrometer. By analyzing the correlation of foliar Fe, Mn concentration at various phenological periods of fruit development, the spectrum reflectance Rλ and its first-order differential f' (Rλ), we filtered out its sensitive bands. And we established an element concentration estimation model for its foliar Fe and Mn at various phenological periods of fruit development with the linear regression model. The results showed that the spectral sensitive bands of foliar Fe in fruit setting period were 873 and 874 nm, 375 and 437 nm in fruit core-hardening period, 836 and 837 nm in maturity period and 325 and 1 054 nm in post-harvest period. However, the spectral sensitive bands of Mn were 913 and 1 129 nm, 425 and 970 nm, 390 and 466 nm, 423 and 424 nm, respectively. The Fe and Mn concentration of A. vulgaris cv. Luntaibaixing leaves were the most relevant to the first-order differential f' (RD) of its spectrum reflectance, whose linear spectrum estimation model fitting degree was the highest and reached to a significant or highly significant level. It showed that the spectral sensitive bands of Fe and Mn element varied with different phenological periods of fruit development. The spectrum estimation models for its foliar Fe and Mn concentration could be established with linear model according to its first-order differential f' (Rλ).

A multi-model fusion algorithm as a real-time quality control tool for small shift detection.

BACKGROUND: Patient-based real-time quality control (PBRTQC), a complement to traditional QC, may eliminate matrix effect from QC materials, realize real-time monitoring as well as cut costs. However, the accuracy of PBRTQC has not been satisfactory as physicians expect till now. Our aim is to set up a artificial intelligence-based QC for small error detection in real laboratory settings. Taking tPSA as our unique research subject, data extraction, data stimulation, data partition, model construction and evaluation were designed. METHODS: 84241 deidentified results for tPSA were extracted from Laboratory Information System of Aviation General Hospital. The data set was accumulated by way of data simulation. Independent training and test datasets were separated. After three classification models (RF, SVM and DNN) in ML constructed and weighted by information entropy, a multi-model fusion algorithm was generated. Performance of the fusion model was evaluated by comparing with optimal PBRTQC. RESULTS: For 4 PBRTQC methods, MovSO showed overall better performance for 0.2 µg/L bias and optimal MNPed was equal to 200. For the fusion model, MNPeds were less than 12 for all biases, and ACC surpassed MovSO nearly 100 times. Except for 0.01 µg/L bias, ACC was more than 0.9 for the rest of biases. FPR was apparently lower than MovSO, only 0.2% and 0.1%. CONCLUSION: The fusion model shows outstanding performance and reduces incorrect and omitting error detection, adaptable for the real settings.

Advancement and prospects of tumor gene therapy.

Gene therapy is one of the most attractive fields in tumor therapy. In past decades, significant progress has been achieved. Various approaches, such as viral and non-viral vectors and physical methods, have been developed to make gene delivery safer and more efficient. Several therapeutic strategies have evolved, including gene-based (tumor suppressor genes, suicide genes, antiangiogenic genes, cytokine and oxidative stress-based genes) and RNA-based (antisense oligonucleotides and RNA interference) approaches. In addition, immune response-based strategies (dendritic cell- and T cell-based therapy) are also under investigation in tumor gene therapy. This review highlights the progress and recent developments in gene delivery systems, therapeutic strategies, and possible clinical directions for gene therapy.

Rapid diagnosis of the alpha-thalassemia-1 Southeast Asian type deletion using a single tube real-time SYBR-polymerase chain reaction combined with dissociation curve analysis.

Dear Sir, A single tube polymerase chain reaction (PCR) with three primers and SYBR GREEN1 combined with dissociation curve analysis was set up that can clearly differentiate between Hb Bart's hydrops fetalis, normal subjects and - -(SEA) heterozygotes. This method seems to be simpler than that using a two-tube real-time SYBR-PCR with two different primer sets followed by analyses of DeltaC(T) and C(T) ratio.

Facile Synthesis of Monodispersed Co Nanoparticles on Titanium Carbides for Hydrolysis of Ammonia Borane at Mild Temperature.

Herein, we successfully immobilized non-noble Co nanoparticles on titanium carbides (MXene) for the hydrolysis of ammonia borane by a simple co-reduction route. The synthesized Co nanoparticles with the size of 3.2 nm were monodispersed on MXene surface. The Co NPs/MXene exhibited excellent catalytic performance for the hydrolysis of ammonia borane with TOF value of 39.9 molH2 mol-1cat min-1 at 323 K. The enhanced catalytic property was mainly due to the ultrafine nanoparticles formed on MXene surface. More importantly, the catalytic activity for hydrolysis of ammonia borane did not significantly decrease for up to 6 recycling tests, indicating the remarkable reusability of Co NPs/MXene. Furthermore, this study opens up a new strategy for the preparation of non-noble metallic nanocatalysts with high performance for hydrolysis reactions in practical hydrogen storage systems, thereby fast-tracking the application of ammonia borane in fuel cells.

Monitoring novel modified hemoglobin using mass spectrometry contributes to accurate blood glucose management of the Han Chinese population.

BACKGROUND: The goal of this study was to detect novel modified forms of hemoglobin using mass spectrometry (MS) and to investigate the effect of modified hemoglobin on HbA1c and fasting plasma glucose (FPG). METHODS: This study was conducted on 1200 subjects aged >25 years. Hemoglobin from the above-mentioned subjects was detected using direct-infusion electrospray ionization-MS, and HbA1c and FPG were measured according to the manufacturer's instructions. Regression analysis was performed to estimate the correlations and interactions among HbA1c, FPG, and modified hemoglobin. RESULTS: Multiple modified forms (α1, α2, α3, ß1, ß2, and ß3) of hemoglobin were observed using MS. Statistical analyses indicated that modified hemoglobin was significantly correlated with FPG (p ≤ .01). The association of FPG with α1% (p = .021) and ß3% (p < .001) values was independent of HbA1c% and other modified forms of hemoglobin. Interaction analyses implied two significant interaction effects of HbA1c% with gender (ß = -0.184, p = .007) and α3% (ß = -0.104, p < .001) on FPG. The relationship between HbA1c% and FPG was stronger in males than in females, and a decreased level of α3% also affected the association of HbA1c% and FPG. CONCLUSIONS: This MS-based method is an effective tool for monitoring glycated forms of hemoglobin than traditional approaches. For the Han Chinese population, multiple-glycated hemoglobin affects the association of FPG with HbA1c%, and the correlation between FPG and HbA1c% in females is different from that in males. These data suggest that the HbA1c criteria for the diagnosis and monitoring of diabetes should be established according to genders and modified types of hemoglobin.

Detection of alpha-thalassemia in China by using multiplex ligation-dependent probe amplification.

The multiplex ligation-dependent probe amplification (MLPA) method was used to analyze 118 DNA samples from 90 alpha-thalassemia (alpha-thal) patients and 28 normal persons from Southern China, where the main causes of alpha-thal are three large deletions (-alpha3.7, -alpha4.2, and --SEA) and two point mutations in the alpha-globin gene cluster on chromosome 16. The results, detected by the P140B HBA kit, were in complete concordance with the results detected by multiplex polmymerase chain reaction (m-PCR) and real-time PCR. The advantages and limitations of the techniques are discussed. We concluded that MLPA was a rapid and reliable method to determine the cause of both deletional and nondeletional alpha-thal in China.

[Molecular characterization of vancomycin-resistant enterococci isolates].

OBJECTIVE: To investigate the prevalence and genotype of vancomycin-resistant enterococci (VRE)faecium and faecalis isolates. METHODS: Thirty-eight non duplicate vancomycin -resistant enterococci isolates were collected from 2003 to 2007 in Beijing Chaoyang hospital. The prevalence of VRE was analyzed using WHONET software. These strains were processed by brain heart infusion agar screening in the presence of vancomycin (6 mg/L), and were analyzed for genotypic characteristics using multiplex PCR. The homology of the isolates was determined by pulsed-field gel electrophoresis (PFGE). RESULTS: Most VRE strains were isolated from the ICU patients (34.2% , 13/38). The vancomycin resistance rate (vancomycin MIC> or =16 mg/L) was 2.6% in 2003, 1.8% in 2004, 0.7% in 2005, 6. 8% in 2006, and 3.3% in Jan - Mar 2007. Fifteen vancomycin-resistant Enterococci faecalis isolates were identified as vanB genotype by PCR and sequencing. (the vancomycin MIC ranged from 16 to 48 mg/L and the teicoplanin MIC ranged from 0.38 to 0.5 mg/L). The vanA gene was confirmed by PCR and sequencing in twenty-three vancomycin-resistant Enterococcus faecium (vancomycin MIC > or =256 kg/L, teicoplanin MIC = 2-32 mg/L). Twelve vancomycin-resistant Enterococcus faecium isolates showed incongruence between phenotype and genotype for glycopeptides resistance (vanA genotype and vanB phenotype). The PFGE analysis revealed 10 different PFGE types. Fourteen vancomycin-resistant Enterococci faecalis isolates belonged to the same clone. 21 vancomycin-resistant Enterococci faecium isolates belonged to 9 PFGE types. The four most prevalent clones was A, D, E and F. CONCLUSION: The VRE strains isolated from Beijing Chaoyang Hospital belong to vanA and vanB genotypes. Vancomycin -resistant Enterococci faecalis is monoclonal, and vancomycin-resistant Enterococci faecium is polyclonal. Some of the vancomycin-resistant Enterococcus faecium isolates show incongruence between phenotype and genotype.

Evaluation of the efficacy and safety of elemene in treating malignant pleural effusion caused by tumors: A PRISMA guided meta-analysis.

OBJECTIVE: Elemene is widely used to treat malignant pleural effusion in China. This meta-analysis aimed to evaluate the efficacy and safety of elemene in treating malignant pleural effusion. METHODS: Electronic databases including Pubmed, the Cochrane Library, Embase and Chinese biomedical literature database were searched until March 2017. Clinical controlled trials (CCTs) assessing the efficacy and safety of elemene in the treatment of malignant pleural effusion were included. The quality of the included studies was evaluated using the quality evaluation criteria of the Cochrane Handbook version 5.1.0. RESULTS: A total of 46 CCTs were included, with 2992 patients. Results of meta-analysis showed that elemene significantly improved the overall response rate (ORR) in controlling malignant pleural effusion (risk ratio [RR] = 1.16; 95% CI: 1.08-1.23; P < .05). Subgroup results showed that the ORR of elemene in the treatment of lung cancer patients with malignant pleural effusion (RR = 1.20, 95% CI: 1.07-1.34; P < .05) was higher than that of other cancers (RR = 1.14, 95% CI: 1.05-1.23; P < .05). Meanwhile, elemene did not significantly increase the incidences of chest pain and fever (P > .05). CONCLUSION: Elemene is suggested to have the ability of improving the treatment outcome of malignant pleural effusion with acceptable safety.

Improvement in the Quality of HbA1c Determination by Using Commutable Specimens With IFCC-Assigned Values.

BACKGROUND: To assess the usefulness of commutable secondary reference materials with International Federation of Clinical Chemistry (IFCC)-assigned values, for improving the quality of hemoglobin A1c (HbA 1c ) determination. METHODS: We recalibrated and evaluated 3 point-of-care-test (POCT) devices via 3 different method-specific central laboratory analyzers that were calibrated using commutable specimens with IFCC-assigned values. The staff members who performed POCT were also evaluated before and after training. RESULTS: HbA 1c levels measured with POCT devices showed significantly lesser bias after external mathematical calibration. The interlaboratory CVs for HbA 1c measurements decreased from 12% to 4% after training of POCT-device-operating personnel in the NycoCard group. The CVs in the DCA Vantage and Afinion groups also improved after training. CONCLUSION: Calibration of laboratory devices by specimens with IFCC-assigned values and by external mathematical calibration could improve the accuracy of POCT HbA 1c measurements. Also, standardized training could improve precision in POCT HbA 1c measurements, especially for semiautomated HbA 1c POCT devices.

Proteomic analysis for finding serum pathogenic factors and potential biomarkers in multiple myeloma.

BACKGROUND: Multiple myeloma (MM) is a malignant tumor, which takes the second place in malignant blood disease. The clinical symptoms are complicated that make more difficult to diagnose and therapy. Lots of researches focus on the proteins about MM in order to solve those problems. We used proteomic methods to find potential biomarkers in MM patients. METHODS: We applied the peptide ligand library beads (PLLBs) to deplete high abundance proteins in serum for finding potential pathogenic factors and biomarkers of MM. Using 1D-Gel-liquid chromatography-tandem mass spectrometry (LC-MS/MS), we identified 789 and 849 unique serum proteins in MM patients and in healthy controls, respectively. RESULTS: Twenty-two proteins were found differentially expressed between the two groups including serum amyloid A protein, vitamin D-binding protein isoform-1 precursor, plasma kallikrein, and apolipoprotein A-I. Changes of integrin alpha-11 and isoform-1 of multimerin-1 were validated with Western blotting. The linkage of the differentially expressed proteins and the pathogenesis pathways of MM were discussed. CONCLUSIONS: PLLB combined with 1D-gel-LC-MS/MS analysis is an efficient method to identify differentially expressed proteins in serum from patients with MM.

[Plasmid-mediated cephalosporinase among extended-spectrum-beta-lactamase-producing Escherichia coli and Klebsiella pneumoniae].

OBJECTIVE: To investigate the prevalence and genotype of plasmid-mediated cephalosporinase (AmpC) beta-lactamase in extended-spectrum-beta-lactamase-producing (ESBL) Escherichia coli and Klebsiella pneumoniae. METHODS: 24 strains of cefoxitin-resistant ESBL-producing E. coli and 8 strains of K. pneumoniae were collected from January to December 2001 at Beijing Chaoyang Hospital. Analytical isoelectric focusing electrophoresis was used to measure the pI of the beta-lactamase. Conjugation experiment was used to study the transfer of cefoxitin resistance. The homology of the isolates was determined by pulsed field gel electrophoresis (PFGE). Plasmid-mediated AmpC enzyme genes were amplified and sequenced by using multiplex PCR. RESULTS: The prevalence of ESBL-producing E. coli and K. pneumoniae were 16.8% (49/292) and 16.5% (35/212), respectively. The prevalence of AmpC enzyme among ESBL-producing E. coli and K. pneumoniae isolates were 2.0% (1/49) and 17.1% (6/35), respectively. These 7 isolates produced DHA-1 AmpC enzyme. One strain of K. pneumoniae could transfer cefoxitin resistance to the recipient. Among 7strains, 5 strains produced CTX-M-3 and 2 produced SHV-12 ESBL. These 7 strains also produced TEM-1 broad-spectrum enzymes. These strains harbored 2 - 5 plasmids and one of them were 33 - 36 kb. PFGE showed these strains came from a variety of clones. CONCLUSIONS: In this hospital, 7 strains of the ESBL-positive E. coli and K. pneumoniae produced both DHA-1AmpC enzyme and CTX-M-3/SHV-12 ESBL. These 7 strains were from different clones.

Point-of-care testing of HbA1C is traceable to IFCC reference method by external calibration.

BACKGROUND: POCT for HbA1C is widely used in China. However, the lack of traceability of POCT leads to poor comparability of patient results. METHODS: The first step was the evaluation of the precision of NycoCard and DCA by using two-level patient specimens. The second step was the calibration of the central laboratory instrument G8 HBA1C Variant with samples whose values had been assigned by an IFCC reference method. The third step was the assignment of values to 50 fresh whole blood patient specimens by this calibrated G8. The fourth step was to use these 50 fresh whole blood patient specimens to calibrate and to revise the POCT instruments. The fifth step was to confirm whether these 50 specimens were required through mathematical calculations. RESULTS: The low and high CVs at levels were 3.61% and 1.85% for NycoCard but 1.71% and 2.85% for DCA. The linear equation of NycoCard to calibrated G8 and that of DCA to calibrated G8 were Y=0.8530X+0.6409 and Y=0.8995X+0.3891, respectively, and the correlation coefficient for every POCT instrument was greater than 0.985. By external calibration of POCT instruments, the mean deviation detected by NycoCard was reduced from -4.0±3.4mmol/mol to 0.5±3.9mmol/mol, and that by DCA went down to 0.2±3.3mmol/mol. The minimum specimen size for the external calibration of POCT instrument was 10. CONCLUSION: POCT measurement traceability can be established by external calibration. Using an external calibration mode improves the comparability of POCT patient results.

Bcl-2/IgH expression in minimal bone marrow infiltration by follicular lymphoma cells.

The purpose of this study was to investigate the roles of bcl-2 chromosomal translocation and Bcl-2 protein expression in follicular lymphoma (FL) minimal bone marrow (BM) infiltration. We identified the same bcl-2/IgH fusion gene in paraffin-embedded lymph node (LN) samples and BM samples using immunohistochemistry (IHC), immunocytochemistry (ICC), cytologic morphology and fluorescence in situ hybridization (FISH). The presence of the Bcl-2/IgH fusion gene in the BM samples and paraffin-embedded LN samples from 56 patients with follicular lymphomas was detected using FISH. The Bcl-2 protein levels in BM and paraffin-embedded tissues were quantified using ICC and IHC, respectively. Approximately 78.6% (44/56) of the paraffin­embedded LN tissue sections that underwent FISH analysis had a bcl-2/IgH translocation. The primary lesion was also positive for the bcl-2/IgH fusion gene, as were the BM minimal infiltrates. The bcl-2/IgH rearrangement occurred in 88.6% (39/44) of the BM specimens. The bcl-2/IgH recombination rate in stage III/IV cancers was significantly different to that observed in stage I/II cancers (p=0.041). In 59% (23/39) of the cases with t(14;18), Bcl-2 was found to be present as assessed by ICC. Positive Bcl-2 ICC staining and the t(14;18) translocation (as detected using FISH) were positively correlated (p=0.028). We then applied the FISH method to slides that had previously been morphologically evaluated using Wright-Giemsa staining; any slides with at least one abnormal cell were subjected to FISH analysis following staining. The assessment of bcl-2/IgH translocation status may contribute to the better detection of minimal BM infiltration by FL cells. Utilizing FISH and cytologic morphology techniques allows for earlier and more accessible BM examination.