Assessment of Monoclonal Gammopathies by Nephelometric Measurement of Individual Immunoglobulin / Ratios

Transcription

1 Clinical Chemistry 55: (2009) Cancer Diagnostics Assessment of Monoclonal Gammopathies by Nephelometric Measurement of Individual Immunoglobulin / Ratios Arthur R. Bradwell, 1,2* Stephen J. Harding, 2 Nicolas J. Fourrier, 2 Gregg L.F. Wallis, 2 Mark T. Drayson, 1 Hugh D. Carr-Smith, 2 and Graham P. Mead 1,2 BACKGROUND: Currently, monoclonal immunoglobulins are identified and quantified from bands on electrophoretic gels. As an alternative, clonality might be determined by measuring the separate light chain types of each Ig class to allow numerical assessment of Ig / Ig ratios, analogous to free light chain / ratios. METHODS: Using immunization, tolerization, and adsorption procedures, we prepared sheep antibodies against each of the 6 separate molecules, IgG, IgG, IgA, IgA, IgM, and IgM. Antibody targets comprised the junctional epitopes between the heavy chain and light chain domains. After purification, we assessed the antisera on a Siemens Dade-Behring BN II nephelometer for analytical quality and clinical utility. RESULTS: High-avidity, specific antibodies allowed the production of automated nephelometric immunoassays for each Ig light chain type. Laboratory comparison with serum protein electrophoresis, using dilution experiments, showed lower analytical sensitivity for monoclonal IgG detection but similar or greater sensitivity for IgA and IgM, particularly when the monoclonal bands overlaid transferrin. Results obtained from typing of monoclonal proteins into IgG, A, or M types were comparable with results obtained by immunofixation-electrophoresis methods. Initial clinical studies, in multiple myeloma patients, indicated that Ig /Ig ratios were sometimes more sensitive than immunofixation electrophoresis, provided numerical results, and correlated with changes in disease. CONCLUSIONS: Immunoassays for intact Ig / pairs are possible and should assist in the management of patients with monoclonal gammopathies American Association for Clinical Chemistry Traditionally, clinical tests for monoclonal (MC) 3 immunoglobulins have comprised serum protein electrophoresis (SPE), urine protein electrophoresis (UPE), and immunofixation electrophoresis (IFE). Although these approaches are satisfactory for measuring intact MC-Igs, they are inadequate when measuring MC serum free light chains (sflcs); to address this problem, sensitive FLC immunoassays for free and free molecules with derived / ratios have been introduced (1). Importantly, / ratios are useful because they inherently account for variable renal clearance, variations in blood volume, and tumor suppression of the alternate FLCs. Substantial clinical utility has been established for / ratios in several situations including screening for MC-Igs (2, 3); as a replacement for urine Bence Jones protein tests (1, 4); as prognostic markers for multiple myeloma (MM) (5, 6); evaluation of monoclonal gammopathy of undetermined significance (MGUS) (7); and monitoring and prognosis of other MC-gammopathies including light chain amyloidosis and B-cell chronic lymphocytic leukemia (8 10). These observations have led to the widespread measurement of sflcs and their inclusion in international guidelines for assessing MC-gammopathies (11). The clinical utility of sflc analysis results from 2 analytical factors: (1) the high sensitivity and specificity inherent in immunoassays compared with electrophoretic techniques and (2) the use of numerical / ratios to define clonality rather than the quantification of stained MC-Igs in gels. By analogy, this analytical approach might be applied to intact Ig molecules. Indeed, this concept was used when total light chain immunoassays were developed. However, such assays do not distinguish between individual Igs, and results are of no clinical value (12). In contrast, separate measure- 1 Department of Immunity and Infection, Medical School, University of Birmingham, UK; 2 The Binding Site Ltd., Birmingham, UK. * Address correspondence to this author at: Department of Immunity and Infection, Medical School, University of Birmingham, B15 2TT, UK. Fax ; a.r.bradwell@bham.ac.uk. Received February 9, 2009; accepted June 3, Previously published online at DOI: /clinchem Nonstandard abbreviations: MC, monoclonal; SPE, serum protein electrophoresis; UPE, urine protein electrophoresis; IFE, immunofixation electrophoresis; sflc, serum free light chain; MM, multiple myeloma; MGUS, monoclonal gammopathy of undetermined significance; CH1, heavy chain constant region 1; CL, light chain constant region; HLC, heavy/light chain; FcRn, neonatal Fc receptor; RID, radial immunodiffusion; EIA, enzyme immunoassay; MRC, Medical Research Council (UK). 1646

2 IgG /, IgA /, and IgM / Ratios in Monoclonal Gammopathies ments of the and light chain types of IgG, IgA, and IgM might allow evaluation of individual tumor clones and give quantitative information about the immunosuppression of each nontumor Ig that is currently not available. Measurement of molecule pairs, such as IgG /IgG, would indicate clonality in the same manner as sflc / ratios. Structurally, each Ig contains unique, conformational epitopes at the junctions of the heavy chain and light chain constant regions (CH1 and CL, respectively). These regions could be targeted by specific heavy chain/light chain (HLC) antibodies to produce separate Ig assays, i.e., IgG, IgG, IgA, IgA, IgM, and IgM. The clinical utility of these assays is likely to be greatest when measurements of MC-Igs are difficult. An example is the situation in which MC-Ig bands overlie other proteins such as transferrin and are detectable only by IFE, which is nonquantitative. In addition, Ig measurements do not reliably relate to tumor production rates for 2 reasons. (1) In MM, changes in hematocrit and blood volume usually occur, causing serum concentrations of Igs to vary by 50% or more, independently of changes in tumor production (13). (2) Serum IgG concentrations are affected by variable clearance rates. Neonatal Fc receptors for IgG (FcRn) recycle IgG so that its half-life is extended from 3 days to 21 days at normal serum concentrations. At higher IgG concentrations, FcRn receptors are saturated, so IgG half-life is reduced (14, 15). Similar arguments may apply to IgA measurements (16). Consequently, monitoring patients by measuring the concentrations of MC-Igs may not reflect changing tumor production as accurately as one might suppose. This could partly explain why MC-Ig concentrations at disease presentation, in contrast to sflc / ratios, are of no prognostic value (17). These various limitations of MC-Ig quantification would be overcome by measuring HLC Ig ratios. The aims of our study were to (a) produce polyclonal antibodies against HLC epitopes that are specific for each Ig molecule; (b) define concentration ranges in normal sera and in patients with tumors; (c) compare HLC assays with SPE and IFE for MC-Ig detection in a laboratory setting; and (d) compare HLC assays with SPE and IFE as markers of disease response in patients with MM. Materials and Methods PREPARATION AND ASSESSMENT OF SPECIFIC ANTIBODIES We produced polyclonal antibodies against the conformational epitopes between the heavy chains and light chains of each Ig as follows. Sheep were immunized using individual MC-Igs of 1 class and light chain type. They were simultaneously tolerized with (1) a MC-Ig of the same Ig class but opposite light chain type, (2) MC-Igs of different classes but the same light chain type, and (3) MC-FLCs of the same light chain type. For example, we made HLC IgG antibodies by immunizing with MC-IgG and tolerizing with MC-IgG, IgA, and FLCs. We obtained antigens for immunization and tolerization by purifying MC-Igs from the sera or urine of patients with MC-gammopathies and assessed purity using silver-stained SDS-PAGE and Western blot analysis. We assessed specificity of the resulting antisera using radial immunodiffusion (RID) and nephelometry. Antigen testing panels included each MC-Ig class, subclass, and light chain type. Antisera with subclass bias were purified against immobilized MC-Igs of other subclass types. Antisera were considered specific when they reacted in a balanced manner with the appropriate panel of Ig proteins, and no reactivity occurred with other molecules including FLCs. Cross-reacting antisera were recycled through the adsorption columns until specificity was satisfactory. REFERENCE MATERIALS Reference materials were required for assay calibration. We considered polyclonal proteins preferable to the use of MC-Igs that might have unique structures. We purified individual classes of Igs from pooled normal human sera by ammonium-sulfate precipitation, anion-exchange chromatography, and immunoaffinity chromatography. Low-level contamination with other Igs was removed by immunoaffinity adsorption and/or protein G. Fractionation of from light chain forms involved using both protein L and anti light chain immunoaffinity matrices. We assessed final purity by SDS-PAGE using silver-staining and enzyme immunoassays (EIAs). We determined subclass distribution by nephelometry (Binding Site Ltd.) and compared results with those in the CRM-470 international serum protein reference material (18). We measured concentrations of the purified proteins using the BCA reaction (Pierce Ltd.) and used the results to standardize the specific HLC assays. NEPHELOMETRIC ANALYSIS OF HLC ANTIBODIES Assay parameters were developed on a Siemens Dade- Behring BN II nephelometer. We assessed susceptibility to interference by adding known concentrations of purified materials to serum samples containing concentrations of Igs within the reference interval for each assay. Analysis was based on a paired difference protocol as proposed by the NCCLS (19). We analyzed serum samples 3 times and calculated the mean difference and 95% CIs. In addition, we measured blood donor sera by each HLC assay and compared the Clinical Chemistry 55:9 (2009) 1647

3 Fig. 1. Interference data for IgG (a) and IgA (b) HLC assays. Data points are the mean values of 3 analyses with 95% CIs shown as bars. Changes in the measured Ig concentrations upon the addition of potentially interfering substances are shown. Concentrations of added substances ( normal median serum values) were as follows: IgG 11.0 g/l ( 1.5); IgG 10.5 g/l ( 2.5); IgA 8.7 g/l ( 7); IgA 11.6 g/l ( 13); IgM 4 g/l ( 6); IgM 4 g/l ( 8); FLC 45 mg/l ( 6); FLC 51 mg/l ( 4); Chyle 282 formazin turbidity units (FTU) ( 25); hemoglobin 4.8 g/l ( 100); bilirubin 200 mg/l ( 30). summed concentrations of the Ig pairs with the respective IgG, IgA, and IgM concentrations measured by nephelometry. We assessed assay imprecision by repeated analysis of serum samples containing 3 different concentrations of Igs at low, normal, and high protein concentrations. Intraassay precision was determined by 15 analyses using a single calibration curve, and interassay precision was determined by measuring the samples once using each of 10 separate calibration curves. Linearity was determined by serial dilution of serum samples, containing known amounts of Igs, into Tris-buffered saline. HLC REFERENCE INTERVALS We measured HLC concentrations in blood donor sera by nephelometry. Sodium azide (1 g/l) was added to all samples, which were stored at 20 C until analysis. We calculated reference intervals using Microsoft Excel Analyse-It software. Briefly, data was tested for normality using the Anderson Darling test. A value of 0.05 indicated that use of parametric methods was appropriate, whereas values 0.05 indicated that nonparametric methods were needed. We calculated 95% reference intervals accordingly. LABORATORY SENSITIVITY OF HLC ASSAYS COMPARED WITH SPE AND IFE We compared the sensitivity for detection of MC-Igs by HLC ratios with SPE and IFE (Sebia Hydrasys ) using 9 individual sera containing MC-Igs (3 IgG, 4 IgA, and 2 IgM). These sera were selected on the basis of the different migration positions of the MC-Igs on SPE and included anodal and cathodal types. We added the proteins to normal sera in doubling-dilution series and assessed the sensitivity of detection by each assay. HLC ratios were considered abnormal if they were outside the appropriate 95% reference interval. For the 1648 Clinical Chemistry 55:9 (2009)

4 IgG /, IgA /, and IgM / Ratios in Monoclonal Gammopathies SPE gels, presence or absence of a band was assessed independently by 3 experienced observers. In addition, 20 routine clinical samples that were positive by SPE and IFE, and 20 that were suspicious by SPE but negative/suspicious by IFE, were assessed for clonality by HLC ratios. CLINICAL SENSITIVITY OF HLC ASSAYS COMPARED WITH SPE AND IFE To produce a preliminary assessment of the assays in a clinical setting, we tested the following sera: at clinical presentation, 18 IgG MM, 33 IgA MM, and 7 IgM (Waldenström macroglobulinemia) samples; during monitoring, 194 samples from 9 patients with IgG MM (4 IgG and 5 IgG ) and 59 samples from 5 with IgA MM (4 IgA and 1 IgA ). ETHICS PERMISSION Sera from blood donors were obtained with approval of the South Birmingham Health District Ethics Committee. MM sera were obtained under permission for analyzing samples from the UK Medical Research Council (MRC) Myeloma trials. Results SPECIFICITY OF ANTISERA Initial immunization and tolerization procedures resulted in strongly reactive antisera against each HLC molecule in addition to cross-reacting antibodies to other Igs by RID assays. The latter were successfully removed using affinity absorption, and the final purified reagents were produced using positive affinity chromatography against the target HLC MC-Ig. Three antisera batches were made for IgG, IgG, IgA, and IgA, and 1 batch each for IgM and IgM. REFERENCE MATERIALS All 6 polyclonal Ig protein preparations were produced, i.e., IgG, IgG, IgA, IgA, IgM, and IgM. Each was 99% pure by silver-stained SDS-PAGE, and light chain specific as indicated by EIA, dot-blot analysis, and Western blotting. Contamination with other Igs and light chain types was 0.5% by EIA. Nonreducing SDS-PAGE indicated that both IgA preparations were composed of polymeric (approximately 10%) and monomeric (approximately 90%) forms, IgG preparations were monomeric, and IgM preparations were pentameric. All were comparable with normal plasma as shown by Western blotting. Except for IgG4, the percentage compositions of Ig subclasses in the purified IgG and IgA preparations were similar to CRM-470 proportions (in brackets): IgG, IgG1:55.9, IgG2:40.3, IgG3:3.1, IgG4:0.7; IgG, IgG1:60.2, IgG2: Table 1. HLC concentrations and HLC Ig /Ig ratios in blood donor sera and concentrations of tumor and non-tumor HLC concentrations and HLC Ig /Ig ratios in patients with multiple myeloma (IgG and IgA) and Waldenström macroglobulinemia (IgM). 32.1, IgG3:7.2, IgG4:0.45; IgA, IgA1:80, IgA2:20; IgA, IgA1:84.8, IgA2:15.2 (CRM-470, IgG1:54, IgG2: 36, IgG3:6.1, IgG4:4.0. IgA1:79.1, IgA2:20.9). IgM and IgM preparations were difficult to purify from n Median (range) Blood donor sera HLC immunoglobulins, g/l IgG ( ) a IgA ( ) a IgM ( ) a IgG ( ) a IgA ( ) a IgM ( ) a HLC ratios IgG /IgG ( ) a IgA /IgA ( ) a IgM /IgM ( ) a Tumor sera Tumor HLC type, g/l IgG ( ) IgG ( ) IgA ( ) IgA ( ) IgM 4 34 ( ) IgM ( ) Nontumor HLC type, g/l IgG ( ) IgG ( ) IgA ( ) IgA ( ) IgM ( ) IgM ( ) HLC ratios IgG /IgG 9 52 ( ) IgG /IgG ( ) IgA /IgA ( ) IgA /IgA ( ) IgM /IgM ( ) IgM /IgM ( ) a 95% range. Clinical Chemistry 55:9 (2009) 1649

5 Fig. 2. Blood donor sera showing summed concentrations of HLC immunoglobulin pairs compared with nephelometric measurements of total IgG (n 109), IgA (n 191), and IgM (n 118). Data points are the individual results from the normal range samples shown in Table 1. each other because of polymerization. This delay prevented their use in the study. As an interim measure, we assumed a 60/40 mass ratio of IgM /IgM in normal human serum. ASSAY RESULTS BY NEPHELOMETRY The results of the assay development showed that the HLC antisera produced typical nephelometric assays. Antisera volumes of 40 L were mixed with buffer volumes of 140 L and 10 L of supplementary reagents that included polyethylene glycol The 6 HLC assays were linear over the sections of the calibration curves used. Assay times were between 12 and 18 min with 6 curve points. Sample dilutions were 1/20 to 1/100 with detection limits of 84 mg/l (IgG ), 42 mg/l (IgG ), 21 mg/l (IgA ), 16 mg/l (IgA ), 10 mg/l (IgM ), and 6 mg/l (IgM ), allowing measurement of all patient samples. The limits of detection were determined by the lowest calibration point at the minimum sample dilution and were confirmed to be 2SD from the mean of saline blanks run in triplicate. For IgA, intraassay imprecision was between 2.4% and 3.8% and interassay imprecision between 1.5% and 3.5%. For IgG, intraassay imprecision was between 2.2% and 7.4%. There was insufficient material for full imprecision assessments of the IgG and IgM HLC assays, but these studies are planned. Interference data is shown in Fig. 1 for IgG and IgA HLC assays. The IgG HLC assays showed some interference upon addition of monoclonal IgM or IgM at 6- or 8-fold normal 1650 Clinical Chemistry 55:9 (2009)

6 IgG /, IgA /, and IgM / Ratios in Monoclonal Gammopathies Fig. 3. HLC concentrations in patients with IgG (A), IgM (B), and IgA (C) monoclonal proteins at disease presentation compared with the blood donor samples (showing lines at the 95% range for HLC ratios). (D), SPE of normal human serum (NHS) and 5 MM samples shown in 3C, in which SPE measurements were unreliable. (E) Total IgA measurements plus HLC concentrations and ratios for the 6 samples in D. *Samples with increases in the nontumor HLC concentrations. concentrations, respectively. The single IgM and IgM batches precluded statistically valid interference studies. Limited evaluations showed similar interference when adding the same amounts of IgG, IgG, IgA, and IgA proteins. NORMAL SERA Results from testing blood donor sera for the 6 HLC assays are shown in Table 1 (top) as median values and 95% ranges. Figs. 2 and 3 show the individual sample results together with 95% ranges for the HLC ratios. SUMMED CONCENTRATIONS OF HLC ASSAYS COMPARED WITH IgG, IgA, AND IgM CONCENTRATIONS Results from summing the concentrations of separate HLC immunoglobulin pairs in blood donor sera compared with individual IgG, IgA and IgM measurements are shown in Fig. 2. Comparison of the summed concentrations of the IgM HLC assays with total IgM appeared less satisfactory, possibly due to lower specificity. LABORATORY SENSITIVITY OF HLC ASSAYS COMPARED WITH SPE AND IFE Two sets of results were generated in comparing assay sensitivities. First, HLC assay sensitivity was evaluated for detection of MC-Igs by dilution into normal sera. For the 3 MC-IgGs, the limits of detection of SPE were 0.48, 0.85, and 0.43 g/l compared with HLC results of 0.95, 1.7, and 1.7 g/l, respectively. For the 4 MC-IgA proteins, SPE results were 0.38, 0.54, 0.99, and 1.88 g/l compared with HLC results of 0.38, 1.09, 0.5, and 0.47 g/l. The 2 MC-IgMs produced results of 0.71 and 0.74 g/l for SPE compared with 1.4 and 0.74 g/l for HLC assays. Thus, SPE was more sensitive for IgG samples but similar for IgA and IgM. One of the MC-IgAs overlaid transferrin, resulting in SPE being 4-fold less sensitive than the HLC ratio (for other examples, see Clinical Chemistry 55:9 (2009) 1651

7 Table 2. SPE samples (9 of 40) that were discordant between IFE and HLC ratios. Sample SPE IFE HLC ratios a pig, polyclonal Ig. sflc ratios ( :, mg/l) 1 Positive Minor IgG Normal Normal 2 Positive Minor IgG Normal FLC Positive 2 FLC bands Normal FLC Positive IgM IgM IgM 0.6 Normal 5 Positive Minor IgA IgG 3.9 Normal 6 Suspicious (pig) a Normal/suspicious IgA 3.5 Normal (114:70) 7 Suspicious (pig) Normal/suspicious IgA 3.5 Normal (100:69) 8 Suspicious Normal/suspicious IgA 2.9 Normal (22:16) 9 Suspicious Normal/polyclonal IgG 4.6 Normal (23:28) Fig. 3D and E). It should be noted that the MC-Igs were diluted into normal sera, so there was no immunosuppression of nontumor Igs. Therefore, results are not comparable with true clinical samples. Second, sensitivity was evaluated by comparison of HLC and IFE assays. Of the 40 SPE positive or suspicious samples that were tested by IFE, 31 showed concordance between IFE and HLC for typing the MC-Igs. Nine samples showed differences between IFE and HLC typing (Table 2, with serum FLC results also included). Two minor IgG bands by IFE (samples 1 and 2) were normal by HLC assays, whereas 4 abnormal HLC samples (samples 6 9) were normal/suspicious by IFE. Three samples produced alternative assessments of clonality. Sample 3 was correctly negative by HLC assays because it contained only MC-FLCs. Sample 4 was biclonal by IFE but had a dominant IgM by HLC assays. Sample 5 had a minor IgA by IFE but IgG by HLC analysis. Subsequent IgG subclass testing showed IgG1 33%, IgG2 56%, IgG3 1.7%, and IgG4 10.2%, indicative of IgG subclass dysregulation. Samples 6 and 7 had increased polyclonal FLCs and hypergammaglobulinemia. CLINICAL SENSITIVITY OF HLC ASSAYS COMPARED WITH SPE AND IFE Diagnosis. HLC concentrations and ratios (medians and 100% ranges) in patients with MM and Waldenström macroglobulinemia at clinical presentation are shown in Table 1. All tumor sera had abnormal HLC ratios, and many had immunosuppression of the nontumor HLC although 2 IgAs and 1 IgM had increases in the nontumor HLCs (Fig. 3). Five IgA samples that were difficult to quantify by SPE are shown in Fig. 3D and are compared with HLC results in Fig. 3C and 3E. Immunosuppression of the nontumor HLC concentrations produced abnormal HLC ratios that allowed identification of clonality in patients who had normal concentrations of the tumor HLCs. Monitoring. Of the 9 IgG MM patients studied, 4 did not achieve complete response by IFE or by HLC ratios. In 3 of 5 patients achieving complete response by IFE and HLC, subsequent relapse was indicated earlier by HLC ratios. Of the 5 IgA patients, all 26 samples that were positive by IFE had abnormal HLC / ratios. In 2 of 5 patients, abnormal HLC ratios suggested residual disease when IFE was negative. For 1 patient, the MC-IgA was obscured by another protein in the SPE gel but could be monitored by HLC ratios. In a second patient, abnormal HLC ratios indicated a slow relapse more than a year before IFE became positive. One IgG MM patient was studied in detail during 2 remissions and relapses and illustrates several features of HLC assays. Between days 400 and 600, SPE/ IFE tests were negative, but gradually increasing HLC ratios suggested early progression of disease, an observation supported by the subsequent relapse (Fig. 4). (1) At some time points, HLC ratios were more sensitive than SPE and IFE for assessing residual disease, thereby, allowing earlier detection of relapse. (2) HLC / ratios had larger changes during monitoring than SPE. (3) HLC / ratios provided information about the tumor cell kill vs nontumor plasma cell kill during the second course of chemotherapy. IgG measurements indicated a tumor response but HLC ratios were stable, indicating no selective tumor cell killing. This was in contrast to the first course of chemotherapy that had huge selective tumor cell killing. Indeed, the patient terminally relapsed after the second chemotherapy course, suggesting that the HLC measurements provided the correct analysis of poor tumor response Clinical Chemistry 55:9 (2009)

8 IgG /, IgA /, and IgM / Ratios in Monoclonal Gammopathies Fig. 4. Immunoglobulin measurements during monitoring of a patient with IgG multiple myeloma showing IgG /IgG ratios, serum protein electrophoresis, monoclonal IgG by scanning densitometry, and total IgG by nephelometry. C, cyclophosphamide; CR, complete response; NR, normal range; V, vincristine; A, adriamycin; M, melphalan; P, prednisolone. Discussion Target epitopes for the HLC antisera comprise only a small part of the total Ig surface area, so development of specific polyclonal antisera was challenging. Results of specificity testing and comparisons with total immunoglobulin concentrations indicated that some crossreacting antibodies remained and improvements are warranted. An alternative approach would be to develop MC-antibodies. Structural variability in the target epitopes, however, might make the production of universal reagents difficult, as is apparent for FLCs. Furthermore, it is unlikely that MC-antibodies would provide good precipitating reagents for nephelometric and turbidimetric assays. Use of the HLC antisera in nephelometry was relatively straightforward. Once the correct specificity had been established, the high avidity of the antibodies enabled the production of reproducible assays. HLC epitopes are situated on both arms and both sides of Ig molecules, so they readily cross-linked with the antibodies. Initial results showed narrow reference intervals for HLC Ig pairs, but further studies will be required that include elderly individuals and hospital populations. A few of the sera studied, which were normal by IFE, showed high HLC / ratios and high sflc / ratios (Table 2). Increased sflc / ratios (indicating a relative excess of production) have previously been observed in association with increased polyclonal Igs (3). The HLC results support the observation of relative increases in the production of to molecules. This may be linked to an excess of MC-Igs seen in patients with non-hodgkin lymphoma and chronic lymphocytic leukemia (10, 20). Laboratory comparison of SPE with the HLC assays provided a guide to their clinical use. SPE has a limit of quantification of 0.5 to 2 g/l depending on the position of MC-Ig bands in relation to other proteins and how narrowly they focus in the gels (1). Hence, Clinical Chemistry 55:9 (2009) 1653

9 cathodal IgG bands are more easily detected than anodal IgA bands that may also contain polymeric elements. For HLC assays, on a simple arithmetic basis, low-concentration MC-IgGs are less likely to produce abnormal ratios than IgA or IgM. For example, 1 g MC-IgG is less abnormal as a percentage of 15 g normal polyclonal IgG than1gmc-igm is in 2 g normal IgM. However, both are detected equally by SPE. Hence, HLC assays are likely to be more sensitive when assessing MC-IgA and MC-IgM. This was apparent when MC-Igs were diluted into normal serum and the sensitivity of detection assessed. However, there is no contribution of immunosuppression in the HLC ratios when diluting samples into normal sera compared with actual patient samples. In addition, IgA and IgM producing tumors are relatively larger than IgG-producing tumors at similar serum protein concentrations because the former Igs have shorter half-lives (5 6 days for IgA and IgM vs 21 days for IgG). Because large tumors produce greater bone marrow suppression of normal immunoglobulins, greater disturbance of IgA and IgM HLC ratios may occur compared with IgGproducing tumors with similar Ig concentrations. Analytical assessment of MC-Igs using SPE is unreliable in the presence of overlapping proteins. Because this frequently applies to MC-IgAs, such proteins might be better quantified by HLC reagents. HLC immunoassays also provide numerical results compared with visual assessments of IFE bands that contain high proportions of the precipitating second antibody. HLC assays should, therefore, allow better assessment of changes in MC-Igs when they are detectable only by IFE. Occasionally, HLC assays were insensitive. One diagnostic serum sample was abnormal by SPE/IFE but normal by HLC and FLC tests (Table 2). Published data indicate that samples with minor MC-IgGs but normal FLCs are usually MGUS with little clinical consequence (7). By analogy, low-concentration MC proteins by SPE that have normal HLC and FLC results are probably in a low-risk clinical category. The clinical relevance of immunosuppression for the nontumor HLCs remains to be determined. However, these are the first assays that provide a measure of immunosuppression in the separate IgG, IgA, and IgM niches of the bone marrow and lymph node compartments. Thus 2 patients with MC-IgA and 1 patient with MC-IgM had increases in the nontumor HLC concentrations, but the clinical relevance remains to be determined. Measurements of HLC ratios might prove helpful when monitoring patients. In Fig. 4, HLC ratios appeared more sensitive than SPE and IFE for assessing early progression of disease and appeared to provide information about the treatment-related tumor cell kill vs nontumor plasma cell kill. HLC ratios may be reliable measurements during monitoring because they are not affected by changes in blood volume, hematocrit, and variable metabolism (via FcRn receptors for IgG) that affect assays for total Igs. Such results could help in decisions on the effectiveness of treatments. Studies on the prognostic role of HLC ratios in MM and other MC-diseases will also be of interest. In conclusion, immunoassays have been developed that can determine individual immunoglobulin / ratios. These may be clinically useful when MC-Igs are of low concentration and/or are hidden by other protein bands. Furthermore, such measurements include information on the immunosuppression of individual immunoglobulins, which could prove of interest when assessing disease progression and responses to therapy. Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article. Authors Disclosures of Potential Conflicts of Interest: Upon manuscript submission, all authors completed the Disclosures of Potential Conflict of Interest form. Potential conflicts of interest: Employment or Leadership: A.R. Bradwell, The Binding Site Ltd.; S.J. Harding, The Binding Site Ltd.; N.J. Fourrier, The Binding Site Ltd.; G.L.F. Wallis, The Binding Site Ltd.; H.D. Carr-Smith, The Binding Site Ltd.; G.P. Mead, The Binding Site Ltd. Consultant or Advisory Role: None declared. Stock Ownership: A.R. Bradwell, The Binding Site Ltd. Honoraria: None declared. Research Funding: None declared. Expert Testimony: None declared. Other: (Financial Disclosure) A.R. Bradwell is a director and major shareholder of The Binding Site Ltd. H.D. Carr-Smith and G.P. Mead are directors, while S.J. Harding, N.J. Fourrier, and G.L.F. Wallis are employees. Role of Sponsor: The funding organizations played no role in the design of study, choice of enrolled patients, review and interpretation of data, or preparation or approval of manuscript. Acknowledgments: We wish to acknowledge the technical contributions to the production and assessment of the HLC assays by the following employees of The Binding Site Ltd: Josie Hobbs, Richard Hampton, Simon Blackmore, Jenny Harris, and Paul J. Showell. 1. Bradwell AR, Carr-Smith HD, Mead GP, Tang LX, Showell PJ, Drayson MT, Drew R. Highly sensitive automated immunoassay for immunoglobulin References free light chains in serum and urine. Clin Chem 2001;47: Bakshi NA, Guilbranson R, Garstka D, Bradwell AR, Keren DF. Serum free light chain (FLC) measurement can aid capillary zone electrophoresis (CZE) in detecting subtle FLC M-proteins. Am J 1654 Clinical Chemistry 55:9 (2009)

10 IgG /, IgA /, and IgM / Ratios in Monoclonal Gammopathies Clin Path 2005;124: Hill PG, Forsyth JM, Rai B, Mayne S. Serum free light chains: an alternative to the urine Bence Jones proteins screening test for monoclonal gammopathies. Clin Chem 2006;52: Katzmann JA, Dispenzieri A, Kyle RA, Snyder MR, Plevak MF, Larson DR, et al. Elimination of the need for urine studies in the screening algorithm for monoclonal gammopathies by using serum immunofixation and free light chain assays. Mayo Clin Proc 2006;81: Kyrtsonis M-C, Vassilakopoulos TP, Kafasi N, Sachanas S, Tzenou T, Papadgiannis A, et al. Prognostic value of serum free light chain ratio at diagnosis in multiple myeloma. Br J Haematol 2007;137: Snozek CLH, Katzmann JA, Kyle RA, Dispenzieri A, Larson DR, Clark RJ, et al. Prognostic value of the serum free light chain ratio in patients with newly diagnosed myeloma: proposed incorporation into the International Staging System. Leukemia 2008;22: Rajkumar SV, Kyle RA, Therneau TM, Melton LJ III, Bradwell AR Clark RJ, et al. Serum free light chain ratio is an independent risk factor for progression in monoclonal gammopathy of undetermined significance. Blood 2005;106: Lachmann HJ, Gallimore R, Gillmore JD, Carr- Smith HD, Bradwell AR, Pepys MB, Hawkins PN. Outcome in systemic AL amyloidosis in relation to changes in concentration of circulating free immunoglobulin light chains following chemotherapy. Br J Haematol 2003;122: Dispenzieri A, Lacy MQ, Katzmann JA, Rajkumar SV, Abraham RS, Hayman SR, et al. Absolute values of immunoglobulin free light chains are prognostic in patients with primary systemic amyloidosis undergoing peripheral blood stem cell transplantation. Blood 2006;107: Pratt G, Harding S, Holder R, Fegan C, Pepper C, Oscier D, et al. Abnormal serum free light chain ratios are associated with poor survival and may reflect biological subgroups in patients with chronic lymphocytic leukaemia. Br J Haem 2009; 144: Dispenzieri A, Kyle R, Merlini G, Miguel JS, Ludwig H, Hajek R, et al. International Myeloma Working Group guidelines for serum-free light chain analysis in multiple myeloma and related disorders. Leukemia 2009;23: Kyle RA. Sequence of testing for monoclonal gammopathies: serum and urine assays. Arch Pathol Lab Med 1999;123: Alexanian R. Blood volume in monoclonal gammopathy. Blood 1977;49: Kim J, Hayton W, Robinson J, Anderson C. Kinetics of FcRn-mediated recycling of IgG and albumin in humans: pathophysiology and therapeutic implications using a simplified mechanism-based model. Clin Immunol 2007;122: Akilesh S, Christianson G, Roopenian D, Shaw A. Neonatal FcR expression in bone marrow-derived cells functions to protect serum IgG from catabolism. J Immunol 2007;179: Wines BD, Hogarth PM. IgA receptors in health and disease. Tissue Antigens 2006;68: Durie B, Jacobson J, Barlogie B, Crowley J. Magnitude of response with myeloma frontline therapy does not predict outcome: importance of time to progression in Southwest Oncology Group chemotherapy trials. J Clin Oncol 2004; 22: Schauer U, Stemberg F, Rieger CHL, Borte M, Schubert S, Riedel F, et al. IgG subclass concentrations in certified reference material 470 and reference values for children and adults determined with The Binding Site reagents. Clin Chem 2003;49: NCCLS. Interference testing in clinical chemistry; proposed guideline. NCCLS Publication EP7-P. Villanova (PA): NCCLS; Martin W, Abraham R, Tait S, Clark RJ, Bone N, Geyer SM, et al. Serum-free light chain: a new biomarker for patients with B-cell non-hodgkin lymphoma and chronic lymphocytic leukaemia. Translat Res 2007;149: Clinical Chemistry 55:9 (2009) 1655