Serum immunoglobulin free light-chain measurement in primary amyloidosis: prognostic value and correlations with clinical features

Primary systemic or light-chain amyloidosis (AL) is characterized by multiorgan deposition of amyloid fibrils derived from immunoglobulin free light chains (FLCs), either κ or λ.^1-5  The introduction of a nephelometric FLC assay (Freelite) has enabled quantification of circulating FLCs.^6-12  FLC assay, used along with serum and urine protein electrophoresis and immunofixation, significantly improves the detection of monoclonal proteins in AL.¹³  The FLC assay by measuring the amyloid precursor protein provides us a unique opportunity to study disease biology. We undertook this study in a large cohort of patients with long follow-up, to better define the impact of the FLC measurements on clinical characteristics and survival.

The current study included patients with biopsy proven AL seen at Mayo Clinic between 1980 and 2006, who had FLC measurements within 90 days of diagnosis performed as part of clinical evaluation or subsequently on stored serum. Proof of a clonal plasma cell process, either by presence of monoclonal protein (on serum or urine protein electrophoresis or immunofixation or serum FLC assay) or presence of clonal marrow plasma cells, was required. Of the 1938 patients seen during this period, 730 (38%) satisfied the criteria. The study was conducted with approval from Mayo Clinic Institutional Review Board.

Major organ (cardiac, hepatic, or renal) involvement was defined as previously described. Renal, cardiac or hepatic involvement required a positive biopsy of the respective organ or 24-hour urine protein excretion > 0.5 g/d, an interventricular septal thickness > 12 mm, or an alkaline phosphatase > 1.5× normal, respectively. We used decreased serum carotene as a marker for intestinal involvement and resultant malabsorption.

Serum FLC quantitation was carried out as previously described using the Freelite FLC assay (The Binding Site Limited). The clonal light-chain is considered the “involved” FLC (iFLC) and the other is referred to as the “uninvolved” FLC (uFLC), with the numerical difference between the 2 denoted by dFLC.

The χ² and Fisher exact tests were used to compare differences between nominal variables and the Mann-Whitney U test or Kruskal-Wallis test for continuous variables. Kaplan-Meier analysis was used for analyzing overall survival (OS), and survival curves were compared using the log-rank test.¹⁴  Curves were generated with all patients surviving beyond 10 years censored at that time. Multivariate analysis was performed using the Cox Proportional Hazards model.¹⁵ 

The median age was 63.3 years (range, 32-90 years) with 463 (63%) males; and the estimated median follow-up from diagnosis was 58.4 months with 212 patients (29%) alive at the time of analysis. The baseline laboratory and clinical features are described in Table 1. The κ/λ FLC ratio was abnormal (< 0.26 or > 1.65) in 644 patients (88%), consistent with previous reports comparing the FLC assay to electrophoretic tests in serum and urine.^7,9-11  Based on immunofixation, marrow immunohistochemistry or FLC assay, the clonal light-chain was determined to be λ in 528 (72.3%) patients, unlike in myeloma where κ is more often (60%) the clonal light chain.^9,16  Also in contrast to myeloma, only 366 (51.3%) of the 714 patients with immunofixation results, had a detectable heavy chain. The median iFLC and dFLC was higher for κ-AL (31.4 and 29.4 mg/dL, respectively) compared with 19.4 and 18.2 mg/dL, respectively, for the λ-AL. This is in contrast with myeloma where the median involved κ and λ concentrations were 37.1 and 71.3 mg/dL, respectively in one large study.¹⁶  This is likely a reflection of the higher tumor burden and the higher prevalence of renal insufficiency in the κ-AL patients (Table 1).

Our study highlights interesting associations between the clinical features and the type and levels of FLC as shown in Table 1. Overall, cardiac, renal, and liver involvement was seen in 65%, 55%, and 20% of patients, respectively. Patients with κ-AL were more likely to have liver and GI tract involvement. While the proportion of patients with nephrotic range proteinuria was higher among λ-AL patients, the proportion with serum creatinine > 1.5 mg/dL was higher among κ-AL patients. No relationship was found between frequency of cardiac involvement and type of light chain.

To assess the relationship between FLC burden and clinical features, patients were divided into a high (> 19.6 mg/dL) and low (<= 19.6 mg/dL) FLC group, using the median. Patients with high dFLC had more frequent and severe cardiac involvement with lower ejection fraction, and higher levels of cardiac biomarkers troponin T and NT-ProBNP, consistent with previous reports (Table 1).⁹  Similarly, higher dFLC was associated with more severe GI and renal involvement.

Previous studies have demonstrated prognostic value for FLC in different plasma cell disorders including MGUS, myeloma, and amyloidosis.^17,18  In AL, high baseline FLC level was associated with poor outcome in patients undergoing stem cell transplant and a reduction in FLC was associated with improved outcome.^19-21  One study did not find a prognostic value for baseline FLC levels, but was limited by small patient numbers.¹¹  In the current study, the median overall survival (OS) for the 86 patients with a normal κ/λ FLC-ratio (no clonal excess of light chain) was 63.6 months compared with 16.2 months for the remaining 644 patients (P < .001; Figure 1A). In terms of the FLC burden, the median OS for patients with high dFLC was 10.1 months compared with 38.2 months for those with low dFLC (P < .001; Figure 1B). Given the significantly different median value for the dFLC between κ-AL and λ-AL patients, we repeated the analyses using the respective medians for determining the high and low groups (29.4 mg/dL for κ-AL patients, 18.2 mg/dL for λ-AL patients). The results were similar; the median OS among patients with a high dFLC was 10.9 months compared with 37.1 months (P < .001) (Figure 1C). Given that treatments for AL have changed in recent years, we repeated the analysis using the more recent group of patients (1998-2006). The results were similar with a median OS for the high dFLC group of 11 months compared with 66 months for the rest (P < .001). In a multivariate model including NT-ProBNP, troponin T, number of organs involved, ventricular septal thickness, ejection fraction (EF), circulating plasma cells, and serum uric acid level, dFLC was an independent predictor of survival. The impact of elevated free light chains on survival likely represents the increased availability of precursor light chain for amyloid fibril formation. Interestingly, this may also explain the poor prognosis seen with t(11;14) in AL as translocations involving the heavy chain locus are associated with higher FLC levels.^22,23 

We also examined if the type of light chain influenced outcome and found no relationship; median OS was 18.4 months for κ-AL patients compared with 19 months for λ-AL patients (P = .2). However, the fact that outcomes were similar despite lower levels of iFLC and dFLC for λ-AL patients suggest that λ-light chains might be more “amyloidogenic.” Interestingly patients without an identifiable heavy chain had an inferior survival, 12.6 compared with 29.3 months for those with a heavy chain identified (P = .02). It is important to note that patients without a heavy chain also had higher dFLC (25.5 vs 15.3 mg/dL; P < .001), which likely impacted the extent of organ deposition and outcome. However, in a multivariate model including dFLC and the presence/absence of heavy chain, both were independently prognostic for survival. Unbound light chains may be inherently more prone to undergo misfolding into an amyloid configuration and might explain this finding.

In conclusion, the results of this study provide several valuable observations. The association between type of light chain and organ involvement provides unique insights and can potentially improve our understanding of biology. Finally, it provides an assessment of the prognostic value of FLC measurements in AL paving the way for its incorporation in future risk stratification models.

This work was supported in part by Hematologic Malignancies Program, Paul Calabresi K12 grant (S.K.) and grants CA62242, CA93842, and CA10080 from the National Cancer Institute, National Institutes of Health, and the Department of Health and Human Services.

National Institutes of Health

Contribution: S.K. designed the study, analyzed the data, and wrote the paper; R.J.C. performed the FLC assays; and A.D., J.A.K., D.R.L., C.L.C., S.R.Z., M.Q.L., S.R.H., F.K.B., N.L., M.R.-A., R.A.K., S.V.R., and M.A.G. were involved in the writing and reviewing of the manuscript.

Conflict-of-interest disclosure: A.D. has received honorarium from Binding Site. The remaining authors declare no competing financial interests.

Correspondence: Shaji Kumar, MD, Division of Hematology, Mayo Clinic, 200 First St SW, Rochester, MN 55905; e-mail: kumar.shaji@mayo.edu.