Inter-laboratory discordance of beta-2 microglobulin results: Impact on the validity of the international staging system for multiple myeloma.

Abstract

Background: Multiple myeloma prognostication is largely defined by the International Staging System (ISS)1 using serum beta-2 microglobulin (beta2M) and albumin results. Local external quality assurance data showed substantial, method-dependant variations in beta2M results. Aim(s): We determined the variability of different beta2M methods using patient samples, and whether this influences ISS prognostic scores. Method(s): beta2M stability after freeze-thawing was established first. Specimens (n=21) were then sent frozen to four laboratories using different methods/instruments for beta2M: Lab1, nephelometric (Beckman Coulter Immage); Lab2, chemiluminescent (Immulite 2000); Lab3, turbidimetric (Roche Cobas c502) and Lab 4 turbidimetric (Abbott Architect c16200). All laboratories use the same albumin method (bromocresol purple). Result(s): Lab1 produced beta2M results consistently higher than other laboratories (Figure 1): mean difference +1.55 mg/L (28.5%) (95%CI: 1.02 - 2.09, p<0.0001) compared to Lab2; +1.09 mg/L (20.1%) (95% CI: 0.78 - 1.41, p<0.0001) to Lab3; and +0.51 mg/L (9.3%) (95% CI: 0.29 - 0.73, p<0.0001) to Lab4. Compared to Lab1, ISS scores were classified one stage lower in 7, 5 and 3 patients by Lab2, Lab3 and Lab4, respectively. One patient could have been classified either stage I, II or III depending on which laboratory performed the tests. Albumin methods aligned well and had minimal impact on ISS score. Summary and Conclusion(s): Results from different beta2M methods are highly variable and harmonisation is required. This study has important implications on the validity of ISS staging for myeloma.