Renal diseases and free light chains

From Wikilite

20.1. Introduction

Elevated polyclonal FLCs in serum (associated with normal κ/λ ratios) can result from increased polyclonal production, reduced renal clearance or a combination of both mechanisms. Increased production is due to proliferation of plasma cells and/or their progenitors and is frequently associated with polyclonal hypergammaglobulinaemia. This is a common finding in patients with liver diseases, connective tissue disorders, chronic infections, etc. A complete list of the relevant diseases is given in Chapter 29.

Reduced clearance of sFLCs results from impaired renal glomerular filtration rate (GFR) and is a frequent finding. This can be seen even in apparently healthy, elderly individuals who may have normal serum creatinine concentrations, but elevated polyclonal sFLCs from a slightly reduced GFR.

Frequently, acute and chronic inflammatory diseases are associated with some degree of renal impairment. The combination of increased production and reduced renal clearance leads to particularly high sFLC concentrations, classically observed in patients with systemic lupus erythematosus (SLE). Although not documented, it is likely that most patients with chronic inflammatory diseases and associated renal impairment have high concentrations of polyclonal sFLCs, but normal κ/λ ratios.

20.2. Effect of renal impairment on sFLC concentrations

Figure 20.1. Elevated sFLC concentrations in 107 patients with varying degrees of CRF. Numbers 1-8 refer to patients with monoclonal gammopathies shown in Table 20.3. Line A is at a κ/λ ratio of 0.6 and line B is the mean κ/λ ratio in patients with renal failure.

Figure 20.2. Serum κ (white) and λ (grey) FLC concentrations in CKD stages 1 – 5, in 688 patients attending a renal disease clinic plus patients on peritoneal dialysis (PD), haemodialysis (HD) and controls (Con). Data presented as box plots (1st – 3rd inter-quartile ranges, central line is median value) with whiskers (5th – 95th percentile values). Courtesy of Colin Hutchison.

The normally rapid renal clearance of sFLCs of 2-6 hours is increased to 2-3 days in complete renal failure (Chapter 3) ^[1]. Removal from the circulation then occurs through pinocytosis in all tissues but particularly by capillary endothelial cells (Chapter 10). As GFR falls, sFLC concentrations rise and may be as high as 20-30 times normal levels in end-stage renal failure (Figures 20.1 and 20.2) ^[2][3][4].

The relationship between sFLC concentrations and GFR is shown in Table 20.1. The MDRD index (Modification of Diet in Renal Disease) of GFR has a worse correlation with FLC levels than serum creatinine levels but is better than the Cockcroft-Gault formula ^[5][6]. Cystatin C concentrations have the highest correlation with FLC concentrations and arguably are the most accurate simple measure of GFR. The rising concentrations of sFLC with progressive chronic kidney disease (CKD) stage (from MDRD) are shown in Figure 20.2.

Table 20.1. Relationship between sFLC concentrations and different markers of GFR in 107 patients with CRF (correlation coefficient - R²). MDRD: Modification of Diet in Renal Disease.

Preliminary analysis of follow-up data from a prospective cohort of 1,394 patients with CKD indicated that the summation of serum κ plus λ FLC concentrations was prognostic for all-cause mortality and change in GFR (Figures 20.3 and 20.4) ^[7]. In univariate analysis, the combined FLC concentrations had greater prognostic value than the creatinine-based CKD staging system. The principal causes of death associated with high FLC concentrations were cardiovascular disease, infections and cancer.

20.3. Effect of renal impairment on serum κ/λ ratios

Figure 20.3. Relationship of patient survival with serum concentration of polyclonal FLCs (p <0.001). (Courtesy of Colin Hutchison).

Figure 20.4. Relationship of loss of renal function with baseline sFLC concentrations (Estimated GFR was based on the MDRD equation). (Courtesy of Colin Hutchison).

Figure 20.5. Increase in sFLC κ/λ ratio with increasing CKD stage plus patients on peritoneal dialysis (PD) and haemodialysis (HD) and controls (Con). Same patient data as used in Figure 20.2. Data presented as box plots (1st – 3rd inter-quartile ranges, central line is median value) with whiskers (5th – 95th percentile values). Courtesy of Colin Hutchison.

Figure 20.6. sFLCs in blood donors (red) and patients with AL amyloidosis (yellow), many of whom have renal impairment. (The λ−producing patients tend to cluster against the normal samples because of increased κ levels from reduced renal clearances. The arrows A and B show potential changes in FLC concentrations with recovery of normal renal function).

As both serum κ and λ FLC concentrations increase with deteriorating renal function their relative amounts change slightly. While glomeruli clear monomeric κ molecules approximately 3 times faster than dimeric λ molecules (Chapter 3), pinocytosis removes both proteins at the same rate. With deteriorating renal function κ/λ ratios gradually increase and eventually equal the κ:λ production rates of approximately 1.8:1 in endstage renal failure ^[2][3][4][8].

The effect of renal impairment on κ/λ ratios is seen in several patient groups:

1. κ and λ concentrations increase with age in apparently normal individuals because of minor degrees of renal impairment (Figure 5.1). This is associated with an increase in the median κ/λ ratio from 0.49 to 0.70 in older people (Table 5.3) . This is shown as a deviation in the κ/λ ratios from 0.6 on a κ/λ plot (Figure 5.2) .

2. Patients in CRF, but without FLC monoclonal gammopathies, have higher κ/λ ratios than normal individuals (Figure 20.5).

3. Patients with AL amyloidosis caused by λ FLC monoclonal gammopathies tend to have κ/λ ratios closer to the normal range population than κ-producing patients (Figure 20.6). There is also asymmetry of κ compared with λ FLC concentrations in patients with light chain multiple myeloma (LCMM) (Figure 8.2) and monoclonal gammopathy of undetermined significance (MGUS) (Figure 19.4). Patients with these diseases frequently have some degree of renal failure, therefore κ concentrations are relatively high. Individuals with normal renal function clear κ molecules more rapidly.

The change in κ/λ ratios with increasing renal impairment is of clinical relevance when interpreting borderline results. Patients might be misclassified as having minor κ monoclonal gammopathies because of slightly increased κ/λ ratios when, in fact, they have renal impairment. Equally, patients with minor λ monoclonal gammopathies could have normal κ/λ ratios because of the relative increase in κ concentrations (Figure 20.6). A higher cut-off should be used when assessing patients with acute renal failure (Figure 13.5).

20.4. Removal of FLCs by dialysis in chronic renal failure (CRF)

Figure 20.7. Reduction of sFLCs in 40 patients with CRF undergoing four hours of haemodialysis. (Patient X had an IgAκ monoclonal gammopathy with monoclonal κ FLCs. A, B and C show axes for κ/λ ratios of 0.2, 0.6 and 2.0 respectively).

Figure 20.8. Concentrations of FLCs in the dialysate fluid of a patient with CRF, sampled every 30 minutes, during 4 hours of haemodialysis. (κ black, λ blue, κ/λ ratios red).

The pore size of haemodialysis membranes (dialysers) allows removal of small- to medium-sized protein molecules from the blood. Membranes typically have a molecular weight cut-off of 10-15kDa, and therefore the filtration efficiency for FLCs is very low. However, some dialysers are more permeable and others adsorb FLCs on to their surfaces. Figure 20.7 shows the reduction in sFLC concentrations in 40 patients with CRF over a 4-hour dialysis period using high-flux polysulphonate membranes. 60% of κ and 37% of dimeric λ molecules were removed and median κ/λ ratios changed from 1.23 to 0.7.

The quantities of FLCs in the filtrate fluid accounted for approximately 60% of the observed reductions in serum concentrations while the remainder was adsorbed onto the filtration membranes ^[9]. Multiple sampling of the dialysate fluid from a patient during the 4-hour haemodialysis period showed slowly reducing FLC concentrations as serum levels fell (Figure 20.8). Dialysate fluid κ/λ ratios decreased from 1.6 to 1.4 during the dialysis period as greater amounts of the smaller monomeric κ molecules were preferentially cleared. Some dialysers are more efficient at FLC removal because of larger pore sizes and different surface charges. However, the amounts present in CRF are hundreds of times lower than those found in LCMM. In the latter patients, “high cut-off” dialysers must be used for effective FLC removal (Chapter 13). Whether such large pore dialysers are clinically useful in preventing dialysis-associated diseases in CRF is currently being assessed.

Peritoneal dialysis may be less efficient at removing FLCs (Figure 20.1). The fluid volumes exchanged during this process are much less than in haemodialysis, which presumably accounts for the poor removal.

The clinical consequences of elevated sFLCs in renal impairment are unclear. Reports have suggested that the elevated FLCs lead to reductions in immune function and should therefore be classified as uraemic toxins ^[10]. Indeed, studies have shown that FLCs are inflammatory proteins ^[11]. It is thought that their toxicity may partly account for the relatively poor survival of many patients undergoing chronic haemodialysis.

There is an additional concern about the toxicity of the high FLC load on the kidneys in patients during the progression to end-stage renal failure. As the nephron mass declines with increasing renal impairment, the remaining nephrons, which are hyperfiltrating, are exposed to increasing levels of potentially toxic FLCs. This may contribute to an accelerating decline in renal function ^[12].

20.5. Monoclonal FLCs in CRF

Figure 20.9. sFLC concentrations in 595 patients with CKD (blue squares) showing patients with MGUS (black squares) and the normal population (red diamonds).

Figure 20.10. sFLC concentrations in 465 renal transplant recipients (circles) showing patients with MGUS (red diamonds) and the normal population (crosses).

It is of interest that 8 of 107 (7%) patients in CRF represented in Figure 20.1 and Table 20.3 had monoclonal gammopathies. None was associated with clinically apparent plasma cell dyscrasias and, indeed, only one had been identified prior to the study. Three of the patients had FLC-only MGUS; five other patients had intact immunoglobulin monoclonal proteins identified by serum immunofixation electrophoresis (IFE) of which 2 had associated abnormal FLC ratios.

Hutchison et al. ^[13][2] assessed the prevalence of MGUS in patients with CKD and renal transplant recipients using sFLC analysis and IFE for intact monoclonal immunoglobulins. Of 595 patients studied in the CKD group, 10.6% had MGUS, 3 times more than an aged-matched population (Figure 20.9). In the transplant patients, 8.3% had MGUS (Figure 20.10).

The reason for the high MGUS prevalence is unclear. However, monoclonal FLCs may contribute to an increased rate of renal deterioration in patients with or without underlying renal damage. Dingli et al. ^[14] identified a relationship between focal and segmental glomerulosclerosis and plasma cell disorders in a study of 13 patients. It was considered likely that some of the pathological damage was due to monoclonal FLCs. Studies of renal biopsies in patients with chronic renal disease have shown that AL-amyloidosis is under-recognised and approximately 3% of biopsies have monoclonal light chains deposited in renal tissues without systemic features ^[15][16].

Many of these patients with light chain monoclonal diseases require chemotherapy, but monitoring treatment from repeated biopsies is impractical. sFLC analysis should allow these patients to be identifed early and monitored for the appropriate chemotherapy in order to eliminate their clonal disease ^[17]. It may be that these patients with renal failure should not be transplanted until their clonal plasma cell disorder is properly treated, as the donated kidney may not survive long ^[18][19].

As regards monoclonal FLCs in the renal transplant recipients, MGUS may result from the effects of immunomodulatory drugs. The time course of MGUS development and its clinical and pathological consequences are unknown. Further studies are needed.

Table 20.3. Analysis of 8 samples with monoclonal gammopathies (5 with monoclonal light chains only) in 107 patients with chronic renal failure (from Figure 20.1).

Clinical Case History No 10

Clinical case history No 10. Influence of renal function on sFLC concentrations.
A 55-year-old woman presented with uncomplicated IgGκ ΜΜ and was given three courses of vincristine, adriamycin (doxorubicin), dexamethasone (VAD) followed by high-dose melphalan and a peripheral blood stem cell transplant (PBSCT). She was well immediately following the transplant but then developed a degree of renal impairment. Careful attendance to fluid and electrolyte balance restored normal renal function. While she was in hospital, her sFLCs were monitored daily as part of an on-going investigation into their role as markers of treatment responsiveness (Figure 20.11 and Chapter 13). The effect of high-dose melphalan (HDM) was to reduce both κ and λ FLCs to below normal concentrations with a favourable relative reduction in the tumour FLC levels. Subsequently, from day 15, as bone marrow engraftment took place, both FLCs increased and the κ/λ ratio normalised. FLCs then increased above normal with a rising κ/λ ratio indicating some renal impairment. This observation was supported by an increase in serum creatinine concentrations. After day 22, FLC concentrations normalised and the κ/λ ratio fell, alongside serum creatinine levels. In this patient, the combination of serial sFLC and serum creatinine measurements allowed assessment of renal dysfunction and bone marrow recovery. Changes in the κ/λ ratio may need to be corrected for renal impairment if there is confusion in interpreting the results. In addition, since patients with plasma cell dyscrasias are frequently elderly and are treated with nephrotoxic drugs, repeated measurements of sFLCs might assist in assessing renal function.
	Figure 20.11. Changes in FLCs and creatinine while monitoring a patient with MM after HDM and PBSCT (Courtesy of G Pratt).

References