Cerebrospinal fluid and free light chains

From Wikilite

During inflammation of the central nervous system there is usually synthesis of intrathecal immunoglobulins. Since the blood-brain barrier largely prevents their escape into blood, they gradually accumulate in the CSF. They are then detectable as oligoclonal bands on electrophoretic gels or can be quantitated by protein assays.

When determining clinical relevance, abnormal immunoglobulins need to be assessed in the context of small amounts of serum immunoglobulins that may have diffused into the CSF from the blood. In particular, if the patient’s serum contains monoclonal immunoglobulins produced in the bone marrow, some will cross the blood-brain barrier making interpretation of intrathecal production difficult. Similarly, if there is inflammation of the meninges, serum proteins will enter the CSF more readily and not only be of intrathecal origin. Hence, CSF measurements are always made in relation to serum proteins. When assessing immunoglobulin production, CSF/blood immunoglobulin concentration ratios are compared with CSF/blood albumin ratios because albumin is never synthesised in the brain. Similarly, CSF IgG oligoclonal bands are assessed on electrophoretic gels run alongside corresponding serum samples. However, commonly used techniques such as iso-electric focussing are only qualitative, rather time consuming and interpretation may be difficult ^[1].

Consequently, alternative markers of intrathecal inflammation have been studied, particularly CSF FLCs ^{[2][3][4][5][6]}. These are produced alongside intact immunoglobulins and accumulate in the CSF. Moreover, they do not enter CSF from the blood in significant amounts because of their large size and low serum concentrations. The detection methods for CSF FLCs have included isoelectric focusing, quantitation by enzyme/radio-immunoassays and nephelometry, but results have not been of general interest. With the development of simple FLC nephelometric assays, there has been renewed interest in their measurement in CSF.

Figure 22.1. Box plot of κ FLC concentration (mg/L) in CSF from (1) normal individuals, (2) patients with multiple sclerosis and (3) other neurological diseases. NR: upper limit of normal range for κ in the CSF was 0.5mg/L. (Courtesy of KJ Lackner).

Figure 22.2. Quotients of κ FLC concentrations in CSF and serum plotted against the respective albumin quotients. Samples are from patients in Figure 22.1. (Courtesy of KJ Lackner).

Using these new FLC assays, Fischer et al. ^[7] studied CSF/serum pairs from 95 patients who had been investigated for intrathecal immunoglobulin synthesis. Of these, 24 were negative for oligoclonal immunoglobulin synthesis and 71 were positive, comprising 49 with multiple sclerosis and 22 with other neurological diseases (Figure 22.1). The median κ concentrations in the patients with neurological diseases were high. (Curiously, the λ FLC concentrations were unreliable). A κ FLC index was then constructed: κFLC index = κ FLC in CSF/κ FLC in serum divided by albumin in CSF/albumin in serum.

When samples with increased albumin leakage were excluded, there was no overlap between normal and disease groups (cut-off level of κ concentrations: 0.5mg/L). This indicated that determination of κ FLC concentrations in CSF provided information similar to that of oligoclonal band measurements, providing samples with raised albumin levels were excluded. Furthermore, when using the κFLC index (Figure 22.2) only two normal samples were misclassified. They concluded that CSF κ FLC measurements may be a useful diagnostic procedure for detecting and potentially monitoring intrathecal immunoglobulin synthesis.

Others have published similar findings. Desplat-Jego et al. ^[8] showed that of 89 patients studied, the κFLC index was more sensitive but less specific than the comparable IgG CSF index or CSF oligoclonal bands. 2 patients were only positive by the κFLC index, so it was a useful complementary test for the diagnosis of multiple sclerosis. It was considered that because the CSF FLC assay was easy, rapid and automated, it could be included in CSF studies in patients with suspected intrathecal inflammation. Further studies by Lewis et al. ^[9] with 27 patients, Presslauer et al. ^[10] with 367 patients and Arneth et al. ^[11] with 110 patients supported these findings.

Additional support for the clinical value of CSF κ FLC measurements has come from its relationship with disability outcome. Rinker et al. ^[12] reviewed the clinical records of 57 patients with multiple sclerosis over a 15 year median follow-up period. Raised levels of CSF κ FLCs (by radioimmunoassay) predicted progression to the need for ambulatory assistance during the study with a specificity of 87.5% and a positive predictive value of 89%.

Monoclonal FLC production in the brain has been assessed in patients with lymphomatous meningitis (LM). This is a serious complication of malignant lymphoma and can be difficult to diagnose. Hildebrandt et al. ^[13] studied 17 patients, 5 of whom were positive for LM by CSF cytology, immunocytochemistry and imaging procedures. All patients with LM had elevated CSF κ/λ ratios as did 4 of the control population of patients with cerebral infections. There is some doubt about the sensitivity of the λ FLC assay for CSF samples but a larger study is underway

References