Immune stimulation and elevated polyclonal free light chains

From Wikilite

21.1. Introduction

Figure 21.1 sFLCs in 25 patients with polyclonal hypergammaglobulinaemia (blue squares), compared with normal individuals (red crosses). (Courtesy of RA Kyle and JA Katzmann).

Figure 21.2 sFLCs in different diseases compared with blood donors. Bars show range and mean concentrations of κ (red) and λ (blue). Juv chr arthritis: juvenile chronic arthritis. Juv FMS: juvenile fibromyalgia syndrome. Patient Nos. in brackets. (Courtesy of U Hoffmann).

Diseases associated with generalised increased B-cell activation are often associated with high concentrations of polyclonal immunoglobulins and coexisting high polyclonal serum free light chains (sFLCs) (Chapter 29). This relationship was demonstrated in 25 patients with polyclonal hypergammaglobulinemia studied at the Mayo Clinic ^[1]. In some patients, concentrations of both FLCs were highly elevated, although κ/λ ratios were always within the normal range (Figure 21.1). Total immunoglobulin concentrations were as high as 54 g/L while maximum FLC concentrations were 273 mg/L for κ and 307 mg/L for λ. The correlation between immunoglobulin and sFLC levels was modest. This may have been due to impaired renal function elevating the FLC concentrations in some patients. However, data on glomerular filtration rate (GFR) were not available.

A number of recently published studies have explored the pathological associations of high polyclonal FLC concentrations in both renal (Chapter 20) and non-renal disease (below), and it has been suggested that FLC measurements could form a useful early investigation in a general health assessment ^[2].

21.2. Rheumatic diseases

Many rheumatic diseases feature polyclonal B-cell activation, high concentrations of autoimmune antibodies and polyclonal elevations of serum immunoglobulins. Excess polyclonal FLCs have been detected in the urine of these patients and indeed, their measurement may be useful for assessing disease activity. Presumably, serum analysis of FLCs in such patients would be more reliable. This is particularly applicable to patients with systemic lupus erythematosus (SLE), many of whom have high levels of urine polyclonal FLCs ^[3][4][5]. Since these patients frequently have renal impairment, sFLC concentrations may also be highly elevated.

Hoffman et al. ^[6] investigated the relationship between sFLCs and other markers of disease activity in patients with rheumatic diseases. Figure 21.2 shows the concentrations of sFLCs in the different disease groups. Patients with intercurrent illnesses were excluded from the analysis to ensure that the changes were due exclusively to the disease under study.

High FLC concentrations were found in rheumatoid arthritis, SLE, Sjögren's syndrome, vasculitis and systemic sclerosis compared with control groups in 28 patients with fibromyalgia and 19 blood donors (p <0.05) ^[6]. Furthermore, sFLC concentrations were more frequently elevated than intact immunoglobulins. In all individuals, κ/λ ratios were normal, indicating polyclonal synthesis. It was also found that sFLCs were more frequently elevated than C-reactive protein (CRP) in patients with SLE, Sjögren's syndrome and systemic sclerosis. However, the numbers of patients in some groups were insufficient for statistical analysis. As might be expected, there was a positive correlation between concentrations of sFLCs and creatinine in all patient groups.

Figure 21.3 sFLC in SLE with (A) and without (B) renal involvement. The numbers of individuals studied are shown in brackets. Bars show range and mean concentrations of κ (red) and λ (blue). (Courtesy of U Hoffmann).

Figure 21.4 Serum κ FLC concentrations (and mean values) in patients with Primary Sjögren’s Syndrome (pSS), rheumatoid arthritis (RA) and healthy controls. NR: Normal range. (Reproduced with permission from Ann Rheum Dis ^[7] and J-E Gottenburg).

Figure 21.5 sFLCs in relation to the presence of SSA and SSB antibodies in patients with Sjögren’s Syndrome. (Courtesy of X Mariette).

Systemic Lupus Erythematosus (SLE)

Hoffmann et al. ^[6] studied 45 patients with SLE and showed that sFLCs were elevated approximately 3-fold (Figure 21.2). Predictably, FLC concentrations were higher in SLE patients who had renal involvement compared with those having normal renal function (Figure 21.3).

Clinical scores of SLE correlated with sFLC levels, particularly when the disease was active. In a subsequent prospective study, the clinical scores (European Consensus Lupus Activity Measurement, ECLAM ^[8]) in 8 patients were compared with a variety of laboratory parameters ^[9]. sFLC concentrations showed a strong correlation with disease activity that was not observed for CRP or erythrocyte sedimentation rate (ESR). A larger study with 75 SLE patients ^[10] also showed a strong association between total FLC concentrations and disease activity.

Primary Sjögren’s Syndrome (pSS)

Gottenberg et al. ^[7] studied 139 patients with primary Sjögren’s Syndrome (pSS). Twenty-two percent had raised sFLC levels, and mean levels were significantly higher than controls (p <0.001) (Figure 21.4), while κ/λ FLC ratios were normal in all but one patient. sFLC concentrations were significantly correlated with IgG (p <0.001), rheumatoid factor (p <0.005), β₂-microglobulin (p<0.001) and B-cell activating factor (p <0.01).

Mean sFLCs were higher in patients with autoantibodies, particularly when both anti-SSA and anti-SSB antibodies were co-occurring (Figure 21.5). Also, patients with extra-glandular involvement had higher levels than those with only glandular involvement. Interestingly, 15 patients had monoclonal FLCs, a much higher proportion than might be expected by chance. These results indicate that extra-glandular involvement in pSS is associated with intense stimulation of B-cells.

Some of these patients progress to non-Hodgkin lymphomas [particularly mucosa-associated lymphoid tissue lymphoma (MALToma) with an odds ratio of 12.9 ^[11]] although currently no biological marker is available to evaluate the individual risk for lymphoma. However, the above results show that abnormal κ/λ ratios are associated with loss of control over the proportion of heavy and light chains synthesised. [An abnormal κ/λ ratio was also recently shown to be a relevant clinical marker of malignant evolution in B-cell chronic lymphocytic leukaemia (CLL) (Chapter 18) and monoclonal gammopathy of undetermined significance (MGUS) (Chapter 19)]. Among 5 patients with pSS without MGUS who had abnormal κ/λ ratios, one had purpura and two had decreased complement C4 levels, both of which are risk factors for lymphoma.

No clonal B-cell populations could be detected in the blood of these patients, which suggests that an abnormal κ/λ ratio could be a more sensitive marker of clonality, possibly restricted initially to the site of autoimmunity. Hence, Gottenberg et al. ^[7] suggested that the predictive value of abnormal κ/λ ratios regarding the occurrence of lymphoma should be investigated in a longitudinal study of this disease.

Rheumatoid Arthritis (RA)

Gottenberg et al. ^[7] studied 50 patients with RA. 36% had raised sFLCs with mean values significantly higher than controls (p <0.001) (Figure 21.4), while sFLC κ/λ ratios were normal in all but 3 patients. sFLC concentrations were significantly correlated with IgG (p <0.04), CRP (p <0.04), and rheumatoid factor (for κ only: p <0.03), but not with anti-cyclic citrullinated peptide (CCP) antibodies. Significant correlations were observed between disease activity assessed by the Disease Activity Score 28 (DAS28) ^[12] and both κ (p=0.0004; Figure 21.6) and λ concentrations (p=0.05; data not shown). This supports the functional relationship between B-cells and disease activity. Interestingly, no correlation was observed between DAS28 and IgG, another marker of B-cell activation, but with a much longer half-life (20–25 days) than FLCs (2–6 h). The faster turnover of sFLCs might account for their observed correlation with disease activity, and suggests that they might be a good early surrogate marker for responses to treatments. Studies linking FLC concentrations to the use of drugs such as Rituximab that deplete B-cell numbers, and the development of clonal disease are underway. A further study including 710 patients with arthritis ^[13] found that polyclonal FLC (and other markers of B-cell activation) were higher in early RA than in undifferentiated arthritis. The authors concluded that B-cell activation is an early pathogenic event in the disease.

Dermatitis

A recent study of children with atopic dermatitis ^[14] reported significantly higher FLC concentrations in patients compared with controls. Levels were also higher in those with severe forms of the disease, although there were no significant associations between FLC levels and IgE or age.

21.3. Diabetes mellitus

Figure 21.6 Correlation between serum κ FLC concentrations and Disease Activity (DAS28) in patients with RA. (Reproduced with permission from Ann Rheum Dis ^[7] and J-E Gottenberg).

Figure 21.7 sFLC concentrations in 745 patients with early diabetes mellitus (circles) compared with the normal population (crosses) and showing patients with MGUS (diamonds) (1.9%).

Two early studies identified a relationship between urine FLC concentrations and rapidly progressive diabetes mellitus. Thus, 20 years ago, it was noted that the urinary excretion of κFLCs (and the κFLC/albumin excretion ratio) was significantly higher in type 1 diabetes mellitus patients than in patients with nondiabetic proteinuria, and that diabetic patients with proliferative retinopathy had higher urine κFLC excretion than those without retinopathy (λFLC assays were not available) ^[15]. Subsequently, the same authors suggested that elevated FLC/albumin excretion ratios were an early indication of diabetic nephropathy and they directly implicated a renal cause of the FLC leakage rather than excess production ^[16]. This was supported by their finding of normal sFLC concentrations. However, in the absence of sensitive serum assays this interpretation was, perhaps, premature.

A recent study by our group analysed FLC levels in both serum and urine of Type 2 diabetic patients to determine if they were an early marker of diabetic kidney disease ^[17]. It was clear that diabetic patients had significantly raised serum and urine concentrations of polyclonal FLCs (Figure 21.7) before overt renal impairment developed (p <0.001). κ concentrations were higher than λ concentrations and 1.9% of patients had MGUS (confirmed by immunofixation electrophoresis (IFE)). A good correlation existed between sFLC concentrations and various markers of GFR including serum creatinine, cystatin-C [κ: R=0.55 (p<0.01); λ: R=0.56 (p<0.01)] and estimated GFR (Figure 21.8 shown for κ only). South-Asian diabetic patients had higher sFLCs than Caucasian diabetic patients. This finding was independent of renal function and suggestive of underlying inflammation.

Urinary FLC concentrations were raised in diabetic patients (p <0.001). Sixty-eight percent of patients with normal urinary albumin/creatinine ratios (ACRs) had abnormal urinary FLC/creatinine ratios. Urine FLC concentrations correlated with urinary ACR [κ: R=0.32 (p<0.01); λ: R=0.25 (p<0.01)]. However, some patients had normal GFR (estimated using the Modification of Diet in Renal Disease (MDRD) study equation) with high concentrations of sFLCs indicating increased production, and suggestive of generalised inflammation/vasculopathy. Perhaps retinopathy and nephropathy are the most readily observed clinical signs of a generalised inflammatory process that is apparent from raised FLC production.

Since polyclonal FLCs are potentially nephrotoxic, increased concentrations may contribute to progressive nephropathy. It has also been suggested that monoclonal FLCs may play a role in some patients’ renal disease ^[18]. Indeed, mesangial monoclonal FLC deposits observed in renal biopsies of patients with renal impairment are sometimes similar in appearance to those found in diabetic glomerulosclerosis ^[19]. Furthermore, FLC MGUS is observed in patients with renal impairment (Chapter 20), so it may be an additional risk factor for progressive nephropathy.

Thus, Type 2 diabetic patients have significantly raised concentrations of serum and urinary polyclonal FLCs before overt renal disease occurs, and measurement of polyclonal FLCs could possibly provide a useful tool in early diagnosis of diabetic kidney disease.

21.4. Infectious diseases with elevated polyclonal serum free light chains (sFLCs)

Figure 21.8 κ sFLCs in 745 patients with Type 2 diabetes correlated with the estimated GFR - MDRD (p <0.001).

Figure 21.9 sFLCs in patients with pneumonia compared with other diseases. The slight increase in κ (black) compared with λ concentrations (blue) may be related to renal impairment. (Reproduced with permission from Zeitschrift für Rheumatologie ^[6] and U Hoffman).

An early report by Sölling ^[20] documented modest increases in sFLCs in a small number of patients with tuberculosis and chronic bronchitis. Elevations in sFLCs were also found in patients with active sarcoidosis, at approximately twice the concentration found in normal individuals. A later study by Hoffman et al. demonstrated elevated levels of polyclonal sFLCs in patients with acute pneumonia ^[6] (Figure 21.9).

Chronic viral infection is a significant cause of elevated polyclonal immunoglobulins and sFLCs. Terrier et al.^[21] studied 59 patients with chronic hepatitis C virus infections (HCV) and mixed cryoglobulinaemia (MC) at different stages of evolution to NHL. The MC comprised type II cryoglobulins with immune complexes of monoclonal IgM directed against polyclonal IgG. 17 patients had no MC, 7 had asymptomatic MC, and 35 had MC vasculitis, 9 of whom had B-cell NHL.

The results showed elevated sFLCs in nearly 50% of patients. Furthermore, mean polyclonal sFLC concentrations and the frequency of abnormal sFLC κ/λ ratios progressively increased with worsening disease category (p <0.001 and p=0.002) (Figure 21.10), increasing cryoglobulin concentrations (p <0.0001 and p=0.0016) and the severity of the B-cell disorder (p=0.045 and p=0.0012). Among patients with an abnormal sFLC ratio at baseline, FLC ratios correlated with the virological response to HCV treatment (Figure 21.11). The authors concluded that in HCV-infected patients, abnormal sFLC ratios were very interesting markers, and were consistently associated with the presence of MC vasculitis and/or B cell NHL. After anti-viral therapy, the sFLC ratio could be used as a surrogate marker for the control of the HCV-related lymphoproliferation.

Elevated concentrations of sFLCs have been demonstrated in HIV positive individuals ^[22], and higher levels were associated with significantly increased risk of progression to non-Hodgkin lymphoma. Neither abnormal FLC ratio nor immunoglobulin concentration produced a similar association. The authors speculated that FLC measurement may have a clinical utility for assessing risks of NHL in HIV patients and/or may have more general applications as a marker of polyclonal B-cell activation.

21.5. Lymphoma

Figure 21.10 Abnormal sFLC ratios are progressively more frequent with worsening HCV disease category. (MC: mixed cryoglobulinaemia. B-NHL: B cell non-Hodgkin lymphoma). (Reproduced with permission from Ann Rheum Dis ^[21] and B Terrier).

Figure 21.11 Abnormal sFLCs ratios in HCV infections may normalise with successful anti-viral treatment. (EOF: end of follow-up). (Reproduced with permission from Ann Rheum Dis ^[21] and B Terrier).

The poor prognosis associated with high FLC concentrations in diffuse large B-cell lymphoma has already been described (Chapter 18). The normal FLC ratio present in many of these patients indicates that the elevations were due to polyclonal B-cell activation and were not tumour-derived. This is certainly the case for Hodgkin’s lymphoma where the tumour cells are incapable of producing immunoglobulin proteins. De Filippi and colleagues reported elevated FLC concentrations in 47% of 119 patients with Hodgkin’s lymphoma. The FLC concentration predicted event-free survival in patients with early-stage but not late-stage disease (a situation also seen in CLL, Chapter 18). In a separate case series of 3 patients with refractory Hodgkin’s lymphoma ^[23], it was noted that sFLC concentrations decreased after lenalidomide treatment and rose transiently during disease “flares” suggesting an association with the disease process.

21.6. Free light chains as bioactive molecules in inflammatory diseases

Since FLCs are part of the antigen binding site of intact immunoglobulin molecules, they are bioactive. This has been observed for both polyclonal and monoclonal FLCs and has recently been reviewed (Figure 21.12) ^[24][25].

For instance, biological activities of FLCs have been shown in patients with immediate hypersensitivity-like responses and contact sensitivity dermatitis. Since FLCs can activate mast cells (which contain a range of biologically active molecules), their potential for causing or contributing to inflammatory and other diseases such as asthma is high. FLCs are possibly one of the components in the active inflammatory processes that are apparent in chronic renal failure. Furthermore, different monoclonal FLCs have been shown to bind specific target molecules that can stimulate antiangiogenic activity or have proteolytic potential. The complementarity-determining regions of FLCs have sufficient variability and flexibility to mimic almost any biological molecule.

In this context, removal of FLCs using “high cut-off” dialysers may be helpful in reducing inflammation in chronic kidney disease (Chapter 13). It has also been suggested that their inflammatory actions might be usefully blocked by novel FLC binding peptides ^[25]. To help resolve these issues, studies on purified polyclonal FLCs from patients with different diseases are required.

Figure 21.12 Biological actions of sFLCs include enzymatic acitivity, binding to tissue substrates and mast cells. (Reproduced with permission from Trends Pharmacol Sci ^[25] and M Thio).

References