Evaluation of the Impact of Renal Failure on Correlation and Concordance between 2 Free Light Chain Assays

Moreau C, Autier B, Cavey T, Rouger E, Norwood J, Bendavid C, et al. Evaluation of the Impact of Renal Failure on Correlation and Concordance between 2 Free Light Chain Assays. Clin Lymphoma Myeloma Leuk 2016.

Freelite and N Latex FLC analysis was performed on 1215 serum samples from patients with or without monoclonal gammopathy (MG+ or MG-).

• There is no definition of how MG is defined in the methods section.
• 876/1215 samples were from patients with MG (368 patients).
• 818/1215 samples had renal function data available (455 patients).

Freelite and N Latex FLC results were compared for all 1215 samples.

• There was a good correlation for κ and λ assays (Pearson’s correlation: κ: R=0.952, λ: R=0.930).
• However, the assays are not equivalent (Pearson’s κ: r²=0.906, λ: r²=0.866).
• Passing-Bablok regression for the κ and λ assays demonstrated that the bias between the two assays was neither constant nor-proportional.
  • N Latex FLC results tended to be higher at low FLC concentrations.
  • Freelite results tended to be higher at high FLC concentrations.

The 818 samples were grouped according to their renal function (eGFR, ≥60, 31-60, 15-30, <15 mL/min/1.73m²):

• A good correlation was observed between Freelite and N Latex FLC κ and λ results irrespective of renal function (κ: R=0.926 – 0.974, λ: R=0.874 – 0.919).

The clinical interpretation of κ/λ FLC ratio for Freelite and N Latex FLC was compared.

• 12.5% (102/818) of samples demonstrated discordant interpretations.

Patients were grouped according to their renal function and MG status (+/-).

• Agreement increased as renal function decreased.
  • Concordance: eGFR≥60: 79.2%; 31-60: 88.8%; 15-30: 89.0%; <15: 92.4%.
  • Kappa coefficient: eGFR≥60: 0.63; 31-60: 0.80; 15-30: 0.81; <15: 0.88.

The FLC results of 41 patients with at least six consecutive samples were analysed

• Note: The κ/λ FLC ratio is plotted on the monitoring graphs in the manuscript but dFLC/iFLC is recommended in the guidelines
• 19/41 had a normal κ/λ FLC ratio for both assays during follow-up
• 22/41 had an abnormal κ/λ FLC ratio during follow-up
  • 11/22 patients showed parallel results with both assays
  • 11/22 had discrepant result 
    • 4/11 were significant
      • 3/4 patients had a partial response (PR) with N Latex FLC but either no response or progressive disease (NR/PD) with Freelite
      • 1/4 patients had a PR with Freelite and NR/PD with N Latex FLC

 ‘We confirmed that quantitative discrepancies remain between the 2 assays and the methods are not interchangeable in the follow-up of MG.’

‘Interpretation must take into account clinical data and the same assay must be used for patient follow-up.’

View Publication