| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
Department of Clinical Chemistry and Molecular Diagnostics, Philipps University of Marburg, Marburg, Germany;
aaddress correspondence to this author at: Department of Clinical Chemistry and Molecular Diagnostics, Philipps University of Marburg, 35033 Marburg, Germany; e-mail herzumi{at}med.uni-marburg.de
Quantitative measurements of plasma and urinary paraprotein concentrations play a major role in the monitoring of patients with multiple myeloma. The concentrations are routinely estimated from the size of the M-spike on protein electrophoresis (PEL) or by automated immunologic assays for IgG, IgA, IgM, IgE, or IgD. In the case of light chain myeloma and intact immunoglobulin myeloma with predominant light chain production, light chain concentrations could, until recently, be measured only by the size of the urinary light chain M-spike on PEL or by the measurement of the total (free and bound) light chain concentrations.
A latex-enhanced assay (Freelite; The Binding Site, Ltd.) measuring free light chains (FLCs) in serum and urine has recently become available for the BNII (Dade Behring) analyzer. The Myeloma Management Guidelines (1) recommend the Freelite test for serial monitoring of the FLCs in serum, but periodic 24-h urine collection is still required for Bence Jones proteinuria (BJP) and total urinary protein (TUP) quantification. Depending on the glomerular and tubular function, serum and urine FLC concentrations may not change to the same degree (2), so that monitoring of serum FLC alone is questionable for revealing the actual degree of disease in patients with BJP and tubular dysfunction.
We evaluated the analytical performance of the immunochemical test for serum and urine with the BNII analyzer. The test uses antibodies that specifically recognize an epitope of the common region of 
and 
light chains that is "hidden" when the light chains are attached to the immunoglobulin heavy chain (3).
Intra- (within) and interassay (day-to-day total) imprecision (CV) was determined with control material and with patient samples containing high and low concentrations of polyclonal or monoclonal FLCs (Table 1
). The high CV observed for the serum sample with a high monoclonal FLC concentration may reflect the variable degree of polymerization, which is common at high FLC concentrations. This phenomenon has been described previously, most notably for FLCs (4). The linearity of urine samples from patients with BJP, evaluated as the correlation coefficient of the linear regression line of the measured vs expected values in serial linear dilutions, was good for FLC (9 dilutions; range, 118–865 mg/L; slope, 0.885; intercept, 81 mg/L; r = 0.90) and FLC (17 dilutions; range, 17–14 900 mg/L; r = 0.97). Linearity of TUP measured simultaneously by the benzethonium chloride method (Roche Diagnostics) was very good for the sample with FLC (9 dilutions; range, 0.03–0.58 g/L; slope, 1.0145; intercept, –0.06 g/L; r = 0.995) and FLC (17 dilutions; range, 0–3.01 g/L; slope, 0.945; intercept, –0.01 g/L; r = 0.96). The linearity of serum samples was also determined for both FLCs: (9 dilutions; range, 88–984 mg/L; slope, 1.0754; intercept, 151.4 mg/L; r = 0.6984); (8 dilutions; range, 384-6600 mg/L; slope, 1.022; intercept, –50.29 mg/L; r = 0.9912). Because we observed extremely high concentrations of FLCs, much higher than the TUP, in some urine samples with BJP, we studied the reliability of the Freelite test in urine.
|
We measured FLCs and TUP in 105 urine samples (87 patients) on which immunofixation electrophoresis (IFE; SEBIA) had been performed. We measured TUP by the benzethonium chloride and biuret method with the Hitachi 917 analyzer and by the modified biuret and pyrocatechol violet dry-chemistry method with the Vitros 250 analyzer (Johnson & Johnson). Urine IFE showed 63 samples with monoclonal bands; 20 were FLC, 15 were FLC, 21 were intact immunoglobulins plus (n = 10) or (n = 11) FLCs, and 7 were intact immunoglobulins without FLC.
The FLC concentration ranges were 1–4800 mg/L for and 1–14 200 mg/L for . The lowest FLC concentration with an associated monoclonal band by IFE was 4 mg/L. Considering a / ratio outside the interval 1:2.71–1:0.25 (3) to be abnormal, we identified the FLC type of the BJP, as shown by the monoclonal bands in the urine IFE, with a sensitivity of 87% and a specificity of 53%.
We determined the imprecision (CV) of the TUP methods, using the same urine sample with BJP. The CVs were 7.2% (0.06 g/L) for the benzethonium chloride, 12% (0.56 g/L) for the biuret, 4.2% (0.48 g/L) for the modified biuret, and 6.7% (0.05 g/L) for the pyrocatechol violet method. The total urinary FLC concentration exceeded the benzethonium chloride TUP in 54 of 105 cases, the biuret TUP in 26 of 105 cases, the modified biuret TUP in 17 of 105 cases, and the pyrocatechol violet TUP in 46 of 105 cases, with maximum differences of 11, 9.2, 7.8, and 14 g/L, respectively.
To assess recovery of FLC by TUP methods, we measured FLC and TUP in a normal urine sample without bands by IFE ( = 17.60 mg/L; = 7.44 mg/L) to which we had added purified and light chains. The solutions of purified material were provided and quantified by radial immunodiffusion by The Binding Site. The final concentrations of and FLCs added were 1240 and 930 mg/L, respectively. The linearity for the urine sample with added FLCs, evaluated as the linear regression line of the measured vs expected values in serial linear dilutions, was good for both (5 dilutions; range, 816–9530 mg/L; R = 0.945) and (6 dilutions; range, 834–1600 mg/L; R = 0.9441). TUP measurements of the samples showed good linearity for all methods. Recovery of the purified FLCs, however, differed among the 4 TUP methods. For , the TUP results were 0.12, 3.9, 2.39, and 0.39 g/L for the benzethonium chloride, biuret, modified biuret, and pyrocatechol violet, respectively, and for , the TUP results were 0.61, 3.2, 2.65, and 0.57 g/L, respectively.
Previous authors have emphasized the difficulty of measuring clones of FLCs, as their structures are heterogeneous and can be modified through pH, polymerization, and oligomerization (5)(6)(7)(8). Both of the routinely used methods for monitoring BJP, urine PEL and TUP, are unspecific for FLCs and have several drawbacks. Urine PEL is time-consuming and insensitive, requires previous concentration, and is subject to interference from other small urinary proteins in a tubular proteinuria pattern, which frequently occurs in such patients (9)(10)(11). The TUP methods show variable, partial recovery of the FLCs (12)(13). The Freelite assay provides specific quantification of BJP and has acceptable analytical performance. Because the diagnostic performance is poor, monoclonality needs to be confirmed by IFE (14).
We conclude that monitoring of renal involvement and BJP in patients with FLC myeloma can be improved by measuring both TUP and FLC in urine. Monitoring of the TUP concentration should be performed with the same assay.
Acknowledgments
We thank I. Pietrek and R. Heinz for excellent technical support and The Binding Site, Ltd., for the purified light chains.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
