SEARCH

SEARCH BY CITATION

Keywords:

  • multiple myeloma;
  • biomarker;
  • diagnostic model;
  • magnetic beads

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. LITERATURE CITED

A diagnosis of multiple myeloma (MM) is difficult to make on the basis of any single laboratory test result. Accurate diagnosis of MM generally results from a number of costly and invasive laboratory tests and medical procedures. The aim of this work is to find a new, highly specific and sensitive method for MM diagnosis. Serum samples were tested in groups representing MM (n = 54) and non-MM (n = 108). These included a subgroup of 17 plasma cell dyscrasias, a subgroup of 17 reactive plasmacytosis, 5 B cell lymphomas, and 7 other tumors with osseus metastasis, as well as 62 healthy donors as controls. Bioinformatic calculations associated with MM were performed. The decision algorithm, with a panel of three biomarkers, correctly identified 24 of 24 (100%) MM samples and 46 of 49 (93.88%) non-MM samples in the training set. During the masked test for the discriminatory model, 26 of 30 MM patients (sensitivity, 86.67%) were precisely recognized, and all 34 normal donors were successfully classified; patients with reactive plasmacytosis were also correctly classified into the non-MM group, and 11 of the other patients were incorrectly classified as MM. The results suggested that proteomic fingerprint technology combining magnetic beads with MALDI-TOF-MS has the potential for identifying individuals with MM. The biomarker classification model was suitable for preliminary assessment of MM and could potentially serve as a useful tool for MM diagnosis and differentiation diagnosis. Anat Rec, 292:604–610, 2009. © 2009 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. LITERATURE CITED

Multiple myeloma (MM) is a clonal B-cell disorder in which malignant plasma cells (PC) accumulate in the bone marrow (BM) and result in lytic bone lesions and excessive amounts of monoclonal proteins (Hussein et al.,2002). It accounts for ∼1% of all malignant diseases and 10% of hematologic malignancies (Kyle and Rajkumar,2004). Bone pain, renal failure, susceptibility to infections, anemia, and hypercalcaemia are the most common clinical features of MM (Bataille and Harousseau,1997; Sirohi and Powles,2004). A diagnosis of multiple myeloma (MM) is difficult to make on the basis of any single laboratory test result. Accurate diagnosis generally results from a number of laboratory tests, such as bone marrow aspirate and biopsy, protein electrophoresis, immunofixation electrophoresis, quantitation of immunoglobulins in serum or urine, and image studies of the skeleton (San Miguel et al.,2006). The diagnosis of multiple myeloma is often made incidentally during routine blood tests for other conditions. For example, the existence of anemia and high serum protein levels may suggest further testing of MM. As the current laboratory tests for MM screening are generally invasive and costly, a new non-invasive, inexpensive, and convenient method is desired that can diagnose with high sensitivity and specificity.

Although monoclonal protein (M-Protein) serves as an important biomarker for diagnosis of MM, it is not specific for MM because diseases like monoclonal gammopathy of undetermined significance (MGUS) and B cell non-Hodgkin's lymphoma (NHL) can also show M-protein in serum. One approach for differentiation of monoclonal proteins in clinical settings is to identify biomarkers that are unique to MM subtypes—an approach amenable to mass spectrometry (MS)-based proteomic investigations. The discovery of biomarkers in body fluids has been advanced by the recent introduction of MS-based screening methods, such as surface-enhanced laser desorption/ionization time-of-flight (SELDI-TOF) MS and matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) MS. The first clinical investigations using SELDI-TOF MS for different cancer types (e.g., ovarian, prostate, bladder, breast, lung, brain and liver) revealed high diagnostic sensitivities and specificities (Diamandis,2002; Menon and Jacobs,2002; Petricoin et al.,2002; Guillaume et al.,2003; Howard et al.,2003; Zheng et al.,2003; Koopmann et al.,2004; Zhu et al.,2004). The limitations of SELDI-TOF analysis include high cost and difficulty in further protein identification on chips. However, affinity bead–based purification was developed to reduce costs and make proteomic procedures suitable for general MS analysis. This method uses different chemical chromatographic surfaces on an outer layer of magnetic beads to selectively purify certain subsets of proteins, allowing unbound impurities to be removed by washing with buffers. Proteins bound to the magnetic beads are then eluted, diluted, and directly analyzed by MALDI-TOF MS. Bioinformatic algorithms are used to align and integrate hundreds of mass data points from large numbers of samples. The technical performance of affinity bead purification is similar to that of ELISA, and it can be used to process many samples in parallel. This approach is sensitive and fast, features essential for clinical use. In fact, MALDI-TOF has become a powerful tool for identifying disease biomarkers by simply surveying plasma, serum, urine, or sputum samples at low costs. Baumann et al., (2005) has developed a standardized approach to proteome profiling of human serum based on magnetic bead separation and MALDI-TOF MS, and they have confirmed that magnetic bead fractionation in combination with MALDI-TOF MS is a highly sensitive and reproducible analytical platform for proteome profiling of human blood in the mass range 1000–10,000 Da. Zhang et al. (2004) successfully demonstrated the utility of manual magnetic-bead sample preparation and a low-cost bench-top MALDI-TOF MS for preliminary biomarker discovery studies in the process of constructing diagnostic models for asthma. Cheng et al. (2005) discovered that the fibrinogen α-chain fragment serves as a clinically useful biomarker for oral cancer by this technology. Analogously to the proteomic detection of various cancers, we used proteomic fingerprint technology combining magnetic beads with MALDI-TOF MS to analyze MM sera to determine whether there are distinct and reproducible protein fingerprints potentially applicable for the diagnosis of MM. We also constructed a diagnostic model of MM with the Biomarker Patterns Software using three unique biomarkers for MM in the training set and subsequently validated the accuracy of this model by use of a completely blinded test set.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. LITERATURE CITED

Patients and Sample Preparation

A total of 54 patients (37 men and 17 women) with MM from the Multiple Myeloma Research Center at the Beijing Chaoyang Hospital (Beijing, China) were recruited for this study. The blood collection protocols were approved by the Institute and the informed consent was obtained. Their median age was 58 years (range from 46 to 83). All of the patients were newly diagnosed according to the Durie-Salmon staging system: stage I (n = 6), stage II (n = 22), and stage III (n = 26) MM. Out of all the patients, samples from 13 were taken again after 2 or 3 chemotherapy cycles. Blood samples from these patients were collected from serum. The 108 non-MM control serum samples were obtained from recruited healthy donors (n = 64) or from patients with other diseases. These included a subgroup of 17 plasma cell dyscrasias (monoclonal gammopathy of undetermined significance, Waldenstrom's macroglobulinemia, solitary plasmacytoma, and primary systemic amyloidosis), a subgroup of 15 with reactive plasmacytosis (rheumatic disease, chronic renal failure, polyclonal garmorphy), 5 with B cell lymphoma, and 7 other tumors with osseus metastasis. In all cases, fasting blood samples were obtained early in the morning and collected in 4 mL BD vacutainers without anticoagulants, allowed to clot at 4°C, and then, within 2 hr, centrifuged at room temperature for 5 min. The supernatant serum was transferred into other centrifuge tubes and then centrifuged for 10 min (at 4°C) at 10,000 rpm (10,000g). Each pooled serum sample was allocated into 5 tubes (100 μL/tube), frozen, and stored at −80°C for future analysis. No sample underwent more than one freeze-thaw cycle before analysis. The patients and serum samples were then divided into two groups: the “training” set and the blinded “test” set (Table 1).

Table 1. Serum samples used in training and testing sets
SamplesTraining setTesting setTotal
MM243054
Normal283462
Other disease212546
Total7389162

Sample Pretreatments and Proteomic Analysis

In the proteomic profiling analysis, the serum samples from the diseased and control groups were randomized, and the investigator was blinded to their identity. Serum samples were pretreated with weak cation exchange (WCX) magnetic beads (SED™) (Beijing SED Science & Technology, Inc.). Ten microlitres of serum samples were denatured by addition of 20 μL of U9 buffer (9 mol/L urea, 20 g/L CHAPS, 50 mmol/L Tris-HCl, pH 9.0) and incubated for 30 min at 4°C. Denatured serum samples were diluted with 370 μL binding buffer (50 mmol/L sodium acetate, 1 mL/L Triton X-100, pH 4.0). At the same time, 50 μL of WCX magnetic beads were placed in a PCR-tube and the tube was placed in a magnet separator for 1 min, after which the supernatant was discarded carefully by using a pipette. The magnetic beads were then washed twice with 100 μL binding buffer. Then, 100 μL of the diluted serum sample was added to the activated magnetic beads, mixed, and incubated for 1 h at 4°C, after which the beads were washed twice with 100 μL binding buffer. Following binding and washing, the bound proteins were eluted from the magnetic beads using 10 μL of 0.5% trifluoroacetic acid. Then, 5 μL of the eluted sample was diluted 1:2 fold in 5 μL of SPA (saturated solution of sinapinic acid in 50% acetonitrile with 0.5% trifluoroacetic acid). Two microliters of the resulting mixture was aspirated and spotted onto the gold-coated ProteinChip arrays. After air-drying for approximately 5 min at room temperature, protein crystals on the chip were scanned with the ProteinChip (Model PBS IIc) reader (Ciphergen) to determine the masses and intensities of all peaks over the range m/z 1,000 to 50,000. The reader was set up as follows: mass range (1,000 to 50,000 Daltons), optimized mass range (2,000 to 15,000 Daltons), laser intensity (210), and sensitivity (9). Mixtures of peptide/protein calibrators [Angiotensin I (human; m/z 1,296.5), Dynorphin A (209–225, porcine; m/z 2,147.5), ACTH (1–24, human; m/z 2,933.5), Beta-endorphin (61–69, human; m/z 3,465), Insulin (bovine; m/z 5,733), Ubiquitin (bovine; m/z 8,564)] were added to calibrate the PBS IIc reader.

Bioinformatics and Biostatistics

Peak detection was performed with Biomarker Wizard software 3.1 (Ciphergen). The m/z ratios between 2,000 and 20,000 were selected for analysis because this range contained the majority of the resolved protein and peptides. The m/z range between 0 and 2,000 was eliminated from analysis to avoid interference from adducts, artifacts of the energy-absorbing molecules, and other possible chemical contaminants. Peak detection involved baseline subtraction, mass normalization using a common calibrant peak (m/z 6,635.09) in human serum, and normalization to the total ion current intensity with a minimum m/z of 2,000, using an external normalization coefficient of 0.2 (normalization factor for individual spectrum = 0.2/average ion current for each spectrum) for spectra obtained at different times or locations. Biomarker Wizard software 3.1 was used to compile spectra and detect peaks that were consistently present across a minimum of 10% of the spectra with a signal-to-noise ratio of ≥2.0. Selected peaks were clustered using a second-pass peak selection within a 0.3% mass window. Sample statistics were performed on spectra from each of the fractions separately (MM vs. non-MM). Peak intensities were considered statistically significantly different at P-values below 0.05.

Data Processing with Biomarker Patterns Software

Biomarker Patterns Software 5.0 (Ciphergen) was used for spectrum post-processing and generation of a proteomic fingerprint. The workflow of data analysis was the following:

1. Peak detection and alignment.

2. Selection of peaks with the highest discriminatory power.

3. Data analysis using a decision tree algorithm.

A random sampling (from samples of MM patients, patients with other diseases, and healthy donors) with two strata (MM and non-MM) was used to separate the entire data set into training and test data sets. The training data set consisted of proteomic fingerprint spectra from 24 MM and 49 non-MM serum samples. The validity and accuracy of the classification algorithm were then challenged with a blinded test data set consisting of 30 MM and 59 non-MM samples. Construction of the decision tree classification algorithm was performed as described previously (Adam et al.,2002) with modifications based on the Biomarker Patterns Software 5.0. Classification trees were split into two branches or nodes, using one rule at a time. We set the target variable level at 2 and the minimum value at 0, and the decision was made based on the presence or absence and the intensity of one peak, using the Gini or Twoing method, favoring even splits from 0.00 to 2.00 and varied by 0.2 each time, with V-fold cross-validation from 6 to 12 changed by 2 for the growth of 88 trees. The lowest cost tree (Relative cost: 0.247; Method = 0; Advanced = 10,1; Testing = 12) was selected for the final test.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. LITERATURE CITED

Protein Fingerprint Analysis

Proteomic data from the samples of the training set (consisting of 24 MM patients and 49 controls) were analyzed with Biomarker Wizard software 3.1. Fifty-six discriminating m/z peaks were found between the MM and non-MM groups (Table 2). The decision tree classification algorithm was constructed using 4088 peaks [56 peaks × (24 + 49) spectra] of statistical significance. The classification algorithm used three peaks with the following m/z: 8,131 (marker 1), 11,660 (marker 2), and 22,752 (marker 3), and generated four terminal nodes (Fig. 1). These discriminatory peaks efficiently split MM specimens into terminal nodes 1 and 2, and non-MM samples into terminal nodes 3 and 4. Mass peaks showed the average mean intensity ratio of MM vs. non-MM > 3 and a P value close to 0 (Table 3). Representative spectra of two MM specimens aligned with that of two healthy controls and two patients with diseases that may have to be differentiated with MM when diagnosed are shown in Fig. 2. The three fingerprints required for pattern recognition in the classifier (Fig. 1). This decision algorithm (serving as the MM diagnostic model) correctly classified 24 of 24 (100%) of the MM samples and 46 of 49 (93.88%) of the non-MM controls in the training set.

thumbnail image

Figure 1. Diagram of the classification tree in the training set. The numbers in the rectangles are the mass values, followed by the peak intensity values. For example, the mass value under the rectangle is 8,131 Da and the intensity is <5.88. N in the rectangles is the sum of MM and non-MM. The number on the left side of the pie charts represents the MM samples and the number on the right side of the pie charts represents the non-MM samples.

Download figure to PowerPoint

thumbnail image

Figure 2. A representative spectra. Alignment of representative MM (labeled as MM1 and MM2) and non-MM controls [patients with disease that may have to be differentiated from MM when diagnosed (labeled as D1 and D2) and healthy controls (labeled as N1 and N2)] spectra with the mass range (boxed) for the three biomarkers with m/z of 8,131, 11,660 and 22,752.

Download figure to PowerPoint

Table 2. Fifty-six discriminating m/z peaks between MM group and the control group
m/zPm/zPm/zPm/zP
  1. m/z means mass-to-charge ratio. P was generated by peak comparison between MM and non-MM group. Peaks labeled by ▴ were selected as biomarkers for MM diagnostic model.

8,1311.0 × 10−82,7995.6 × 10−52,1733.0 × 10−42,0860.001
8,9183.7 × 10−72,2635.9 × 10−52,1533.0 × 10−42,6430.001
8,5915.3 × 10−76,8488.0 × 10−513,7333.0 × 10−47,6410.002
22,7527.7 × 10−72,4638.4 × 10−56,8324.0 × 10−411,6600.002
23,4078.1 × 10−72,4228.8 × 10−523,5895.0 × 10−411,3910.002
8,5552.0 × 10−611,7511.0 × 10−46,6226.0 × 10−43,2950.002
2,2142.0 × 10−52,0481.0 × 10−42,2406.0 × 10−47,8420.002
8,6872.2 × 10−59,2741.0 × 10−42,4867.0 × 10−42,1060.002
24,0702.2 × 10−54,4701.0 × 10−49,4747.0 × 10−47,5580.003
7,7572.4 × 10−52,2312.0 × 10−47,9168.0 × 10−44,2830.005
2,1283.2 × 10−52,3722.0 × 10−44,6438.0 × 10−46,4290.005
2,9553.4 × 10−52,1952.0 × 10−414,0089.0 × 10−43,8840.006
2,6863.6 × 10−52,8892.0 × 10−44,0649.0 × 10−43,4510.006
2,0154.2 × 10−52,0392.0 × 10−44,1710.0017,8140.007
Table 3. Biomarker statistics for MM vs non-MM spectra and decision tree classification
m/zPMMNon-MM
MeanSDMeanSD
8,1311.0 × 1.0−82.231.4812.799.06
22,7527.7 × 10−70.300.451.050.79
11,6600.0022.002.532.541.17

Masked Analysis of the MM Diagnostic Model

Analysis of spectra from the completely blinded test set (30 MM; 34 normal controls; 25 disease controls) accurately identified 26 of 30 (86.67%) MM specimens and accurately classified 48 of 59 (81.36%) of the controls as non-MM (Table 4). By categorizing 24 into terminal node 1 and 2 into terminal node 2; for the non-MM controls, all of the 34 normal people were successfully identified as non-MM cases; among them, 32 were categorized into terminal node 4 and the other two into terminal node 3; for the disease controls, 14 of 25 patients were correctly identified as non-MM, and the other 11 who were incorrectly recognized as MM patients included 3 of 5 MGUS patients, 2 of 3 Waldenstrom's macroglobulinemia patients, both of the Solitary plasmacytoma patients, 2 of 5 the B cell lymphoma patients, and 2 of 5 other tumor with osseus metastasis patients.

Table 4. The prediction results of the diagnostic model for MM in the masked analysis
GroupCasesCorrect-classedAccurate %
MM302686.67
Control594881.36

Proteomic Analysis of MM Sera in Terms of Monoclonal Immunoglobulin Type and Disease Stage

All of the MM patients were divided into two groups according to their monoclonal immunoglobulin light chain, namely, κ and λ type. Serum samples from MM patients were applied to MALDI-TOF MS analysis and several, discriminating peaks were found between the κ and λ type MM groups. Compared to the λ type, peaks at 5,190 Da, 6,191 Da, 6,308 Da, 6,364 Da, 7,098 Da, and 11,858 Da increased, and those at 22,752 Da, 22,964 Da and 23,188 Da decreased in the κ type group. Three peaks with an m/z greater than 24 kDa (24,374, 24,553, and 24,664) also increased in the κ type group, but did not increase as much as the other increased peaks. The peak at 6,191 Da may serve as a biomarker (Fig. 3) to discriminate between the two types of MM. The mean peak height is 8.218 in κ type MM and 2.365 in λ type.

thumbnail image

Figure 3. The marker 6,191 Da in κ and λ type MM.

Download figure to PowerPoint

When comparing all types of MM (IgG, IgA, IgD, κ light chain only, and λ light chain only MM), the peaks at 6,364 Da and 6,449 Da were most significantly decreased in IgD MM (P = 0.002, P = 0.001) (Fig. 4).

thumbnail image

Figure 4. 6,364 Da in IgD MM and other MM patients.

Download figure to PowerPoint

For newly diagnosed MM patients, mass spectra in different disease stages were compared. There were no significantly different peaks (P < 0.01) among the different stages I, II, and III. In the 13 follow-up samples, there were no obvious changes in the peaks that corresponded to their therapy responses.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. LITERATURE CITED

Multiple myeloma (MM) is the second most common hematologic malignancy affecting at least 32,000 new patients per year worldwide and accounting for approximately 1% of all malignant diseases and 10% of hematologic malignancies. Conventional diagnosis in MM is based on a series of laboratory tests and medical procedures, such as examination of bone marrow, bone marrow biopsy, protein electrophoresis, immunofixation electrophoresis, quantitation of immunoglobulins in serum or urine, and image studies of the skeleton. Patients with monoclonal gammopathy of undetermined significance (MGUS), amyloidosis, other lymphoproliferative disorders with paraproteinemia, pain in the lumbar spine, and renal insufficiency are not rare occurrences among the population, and it is necessary to distinguish MM from these diseases (The International Myeloma Working Group,2003).

Despite intensive research of MM, its diagnosis still remains an imperfect process; therefore, novel diagnostic approaches are needed. As development of new proteomic technology continues, a new method with MALDI-TOF MS and proteomic pattern recognition software has become more and more reliable for construction of disease diagnostic models. The method is both sensitive and fast, features essential for clinical use, and it serves as a powerful tool for identifying disease biomarkers by simply surveying plasma, serum, urine, or sputum samples at low cost. This method, based on biomarkers that are unique to the targeted diseases, is amenable to mass spectrometry (MS)-based proteomic investigations.

We used a well-developed standardized approach (Baumann et al.,2005) to proteome profiling of human serum based on magnetic bead separation and MALDI-TOF MS; this approach, consisting of magnetic bead fractionation in combination with MALDI-TOF MS, was confirmed to be a highly sensitive and reproducible analytical platform for proteome profiling of human blood. Several studies have constructed disease diagnostic models or discovered clinically useful biomarkers by means of this approach. (Zhang et al.,2004; Cheng et al.,2005). In addition, because magnetic beads provide a large adsorbable surface, they have much more potential for adsorption of peptides and proteins from serum than ProteinChip. Combining the advantages of magnetic beads and MALDI-TOF MS can help in discovering more low-abundant proteins in sera.

In our study, 56 discriminating m/z peaks were found between the MM and non-MM groups, and the classification algorithm used a panel of three peaks with m/z of 8,131 (marker 1), 11,660 (marker 2), and 22,752 (marker 3) to identify MM and non-MM samples. Because MM is a multifactorial bone disease, a combination of multiple markers should be more reliable and powerful in diagnosis with high specificity and sensitivity.

In the masked test, 26 of 30 MM patients were successfully recognized, and all 34 healthy people were correctly classified into the non-MM group; however, this test was not able to satisfactorily differentiate MM patients from other plasma cell dyscrasias and some other malignant diseases with osseus metastasis. We may enlarge our case groups in our further studies to identify more markers that discriminate between MM and other diseases in order to avoid misdiagnoses. In our routine work, the differential diagnosis between MM and MGUS was still a challenge (Ocqueteau et al.,1998). In some cases, there is no single parameter that allows such a distinction between both entities in all cases. In a relatively significant proportion of cases, the patient's follow-up was the only discriminating factor. According to patient follow-up data, a subset of individuals was not correctly categorized.

In our diagnostic model, the three peaks with different m/z values may be biomarkers unique for MM or for some other disease. Dan et al. (2006) identified an 11.5 kDa positive protein marker derived from serum amyloid A1; its theoretical mass is 11.68 kDa, and the mass of our marker 2 (11.66 kDa) is very similar to that of serum amyloid A1. Tolson et al. (2004) found that the marker is related to renal cancer and detected multiple variants of SAA-1. It is a positive marker in severe acute respiratory syndrome (SARS) (Ren et al.,2004) and indicates relapse in nasopharyngeal cancer (Cho et al.,2004). Serum amyloid A is an acute phase protein that is associated with circulating high-density lipoproteins and that modulates lipid trafficking and immune responses. It is also the precursor protein in reactive amyloidosis (Dan et al.,2006). This information will help us in further investigations.

There are also some peaks related to different MM types (heavy chain type and light chain types); these will help us investigate type-related markers of MM. As there were no significant differences in the peaks from spectra of samples in different disease stages and there were no significant changes during therapy, the discriminatory power of the markers is most likely reflected by the malignant nature of the tumor rather than its progression.

The number of specimens analyzed in this study limited the validity of the results to some degree. Further independent validation and some protein-peptide interaction studies using cross-linker combined with MALDI-TOF for understanding protein functions are needed (Sinz et al.,2005). In addition, validation data sets should be from different sources than those of the original training data set. This is one way to ensure that the performance of the selected biomarkers is not influenced by systematic biases between the disease and control specimens.

In conclusion, we have shown that using proteomics approaches, such as magnetic beads and MALDI-TOF MS in combination with bioinformatics tools, could facilitate the discovery of new biomarkers and provide a rapid and mass-accurate mode of analysis for the detection of multiple disease-related proteins simultaneously, reproducibly, and in high-throughput. Using the panel of three selected biomarkers, we could achieve high sensitivity and specificity for the detection of multiple myeloma.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. LITERATURE CITED

The authors thank IMS Health Management, Ltd. to provide some analysis for healthy controls.

LITERATURE CITED

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. LITERATURE CITED
  • Adam BL, Qu Y, Davis JW, Ward MD, Clements MA, Cazares LH, Semmes OJ, Schellhammer PF, Yasui Y, Feng Z, Wright GL Jr. 2002. Serum protein fingerprinting coupled with a pattern-matching algorithm distinguishes prostate cancer from benign prostate hyperplasia and healthy men. Cancer Res 62: 36093614.
  • Bataille R, Harousseau JL. 1997. Multiple myeloma. N Engl J Med 336: 16571664.
  • Baumann S, Ceglarek U, Fiedler GM, Lembcke J, Leichtle A, Thiery J. 2005. Standardized approach to proteome profiling of human serum based on magnetic bead separation and matrix-assisted laser desorption/Ionization time-of-flight mass spectrometry. Clin Chem 51: 973980.
  • Cheng AJ, Chen LC, Chien KY, Chen YJ, Chang JTC, Wang HM, Liao CT, Chen IH. 2005. Oral cancer plasma tumor marker identified with bead-based affinity-fractionated proteomic technology. Clin Chem 51: 22362244.
  • Cho WC, Yip TT, Yip C, Yip V, Thulasiraman V, Ngan RK, Yip TT, Lau WH, Au JS, Law SC, Cheng WW, Ma VW, Lim CK. 2004. Identification of serum amyloid A protein as a potentially useful biomarker to monitor relapse of nasopharyngeal cancer by serum proteomic profiling. Clin Cancer Res 10: 4352.
  • Dan A, Delmiro FR, Marios CP, Sergio AR, Mark H, Alison L, Edward TJS, Achim S, Richard P, Charlotte FJR, Sanjeev K. 2006. Identification of diagnostic markers for tuberculosis by proteomic fingerprinting of serum. Lancet 368: 10121021.
  • Diamandis EP. 2002. Proteomic patterns in serum and identification of ovarian cancer. Lancet 360: 170.
  • Guillaume E, Zimmermann C, Burkhard PR, Hochstrasser DF, Sanchez JC. 2003. A potential cerebrospinal fluid and plasmatic marker for the diagnosis of Creutzfeldt-Jakob disease. Proteomics 3: 14951499.
  • Howard BA, Wang MZ, Campa MJ, Corro C, Fitzgerald MC, Patz EF Jr. 2003. Identification and validation of a potential lung cancer serum biomarker detected by matrix-assisted laser desorption/ionization-time of flight spectra analysis. Proteomics 3: 17201724.
  • Hussein MA, Juturi JV, Lieberman I. Multiple myeloma: present and future. 2002. Curr Opin Oncol 14: 3135.
  • Koopmann J, Zhang Z, White N, Rosenzweig J, Fedarko N, Jagannath S, Canto MI, Yeo CJ, Chan DW, Goggins M. 2004. Serum diagnosis of pancreatic adenocarcinoma using surface-enhanced laser desorption and ionization mass spectrometry. Clin Cancer Res 10: 860868.
  • Kyle RA, Rajkumar SV. 2004. Multiple myeloma. N Engl J Med 351: 18601873.
  • Menon U, Jacobs I. 2002. Screening for ovarian cancer. Best Pract Res Clin Obstet Gynaecol 16: 469482.
  • Ocqueteau M, Orfao A, Almeida J, Blade' J, Gonzalez M, Ramon GS, Consuelo LB, Maria JM, Hernandez J, Escribano L, Caballero D, Rozman M, San Miguel JF. 1998. Immunophenotypic characterization of plasma cells from monoclonal gammopathy of undetermined significance (MGUS) patients. Implications for the differential diagnosis between MGUS and multiple myeloma. Am J Pathol 152: 16551664.
  • Petricoin EF, Ardekani AM, Hitt BA, Levine PJ, Fusaro VA, Steinberg SM, Mills GB, Simone C, Fishman DA, Kohn EC, Liotta LA. 2002. Use of proteomic patterns in serum to identify ovarian cancer. Lancet 359: 572577.
  • Ren Y, He QY, Fan J, Jones B, Zhou Y, Xie Y, Cheung CY, Wu A, Chiu JF, Peiris JS, Tam PK. 2004. The use of proteomics in the discovery of serum biomarkers from patients with severe acute respiratory syndrome. Proteomics 4: 34773484.
  • San Miguel JF, Gutiérrez NC, Mateo G, Orfao A. 2006. Conventional diagnostics in multiple myeloma. Eur J Cancer 42: 15101519.
  • Sinz A, Kalkhof S, Ihling C. 2005. Mapping protein interfaces by a trifunctional cross-linker combined with MALDI-TOF and ESI-FTICR mass spectrometry. J Am Soc Mass Spectrom 16: 19211931.
  • Sirohi B, Powles R. 2004. Multiple myeloma, seminar. Lancet 363: 875887.
  • The International Myeloma Working Group. 2003. Criteria for the classification of monoclonal gammopathies, multiple myeloma and related disorders: a report of the International Myeloma Working Group. Br J Haematol 121: 749757.
  • Tolson J, Bogumil R, Brunst E, Beck H, Elsner R, Humeny A, Kratzin H, Deeg M, Kuczyk M Mueller G A, Mueller CA, Flad T. 2004. Serum protein profiling by SELDI mass spectrometry: detection of multiple variants of serum amyloid alpha in renal cancer patients. Lab Invest 84: 845856.
  • Zhang XY, Leung SM, Morris CR, Shigenaga MK. 2004. Evaluation of a novel, integrated approach using functionalized magnetic beads, bench-top MALDI-TOF-MS with prestructured sample supports, and pattern recognition software for profiling potential biomarkers in human plasma. J Biomol Tech 15: 167175.
  • Zheng PP, Luider TM, Pieters R, Avezaat CJ, van den Bent MJ, Sillevis Smitt PA, Kros JM. 2003. Identification of tumor-related proteins by proteomic analysis of cerebrospinal fluid from patients with primary brain tumors. J Neuropathol Exp Neurol 62: 855862.
  • Zhu XD, Zhang WH, Li CL, Xu Y, Liang WJ, Tien P. 2004. New serum biomarkers for detection of HBV-induced liver cirrhosis using SELDI protein chip technology. World J Gastroenterol 10: 23272329.