Estimating Selection Bias in Previous Monoclonal Gammopathy of Undetermined Significance Research - the Importance of Screening: Results from the Population-Based Screening Study Iceland Screens, Treats or Prevents Multiple Myeloma (iStopMM)

Introduction

Monoclonal gammopathy of undetermined significance (MGUS) is a plasma cell disorder preceding multiple myeloma and related disorders, present in 4.2% of the population over the age of 50. Although usually asymptomatic, MGUS has been associated with various health-related problems, including thrombosis, infections, fractures, neuropathy, and death. Because MGUS is asymptomatic, its diagnosis is typically incidental, during clinical workup for unrelated medical issues, and therefore most individuals remain undiagnosed. Consequently, MGUS cohorts in past studies may have suffered from more comorbidities than the actual population with MGUS. This might have introduced selection bias in previous studies on MGUS, which extent has not been studied in a systematic way. Therefore, previously reported associations between MGUS and various medical issues might not be as profound as formerly observed. The aim of this study was to compare characteristics of incidentally diagnosed MGUS versus MGUS diagnosed by systematic screening, with particular focus on demographics, comorbidities, and MGUS-related factors.

Methods

The study is based on the Iceland Screens, Treats, or Prevents Multiple Myeloma (iStopMM) study. iStopMM is a population-based screening study for MGUS and a randomized controlled trial of follow-up strategies that has included 54% (n = 80,759) of the Icelandic population above 40 years of age. In total, 75,422 participants were screened for MGUS by serum protein electrophoresis (SPEP) and free light chain (FLC) assay. Information on which individuals had incidentally diagnosed MGUS (clinical MGUS) prior to participation in iStopMM were gathered from the Icelandic cancer registry and laboratory results from Landspítali University Hospital and Læknasetrið, the only laboratories in Iceland that perform SPEP. M-protein concentration, MGUS isotype and FLC ratio were obtained from the original screening samples. Comorbidity data was acquired from two high-quality national registries: Hospital Discharge Register and Register of Primary Health Care Contacts, with >95% completeness and accuracy. MGUS diagnosed by screening was further classified into MGUS with or without M-proteins (light-chain MGUS). Those with clinical MGUS were used as the reference group in all analyses. Since all individuals with clinical MGUS had M-proteins, individuals with light-chain MGUS were excluded from this study. T-test and chi-square test were used for demographic comparison; linear regression adjusting for sex and age for continuous variables, and logistic regression adjusting for sex and age for comparison of categorical variables, regarding MGUS-related factors and comorbidities.

Results

The study cohort consisted of 3,300 individuals who had MGUS with M-proteins; 224 individuals with clinical MGUS and 3,076 with screened MGUS. The clinical MGUS group was significantly older (p <0.01), more likely to live in the capital area of Iceland (p 0.02), and had a 0.14 g/dL higher mean M-protein concentration (95% confidence interval [95% CI] 0.10-0.19 g/dL, p <0.001) than those with screened MGUS. Individuals with clinical MGUS were also 1.73 times more likely to have a comorbidity (odds ratio [OR] 1.73, 95% CI 1.14-2.72, p 0.01) than those with screened MGUS, reflected in a significantly higher mean number of comorbidities (2.79 vs. 2.09, p <0.001). Finally, clinical MGUS were significantly more likely to have arrhythmias (OR 1.45, p 0.05), chronic kidney diseases (OR 2.42, p <0.001), endocrine diseases (OR 1.82, p <0.001), heart failure (OR 2.60, p <0.001), neurological diseases (OR 3.07, p <0.001) or rheumatological diseases (OR 3.25, p <0.001).

Discussion

In this large population-based study including 75,000 screened individuals, we found clinical MGUS cases to be older, more likely to live in Iceland's capital area, and have a higher M-protein concentration than those found to have MGUS while screened on the iStopMM study. Individuals with clinical MGUS also had a higher number of underlying comorbidities and were 1.5-3.3 times more likely to suffer from arrhythmias, chronic kidney diseases, endocrine disorders, heart failure, neurological diseases, and rheumatological diseases. Our findings highlight the importance of screening studies to evaluate the true epidemiological and biological implications of MGUS and suggest selection bias in prior studies.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal