Genetic architecture of creatinine metabolism: disentangling muscle mass and kidney function

Chronic kidney disease (CKD) affects over 850 million individuals worldwide (Jager KJ et al. Nephrol Dial Transplant 2019;34:1803-5). Patients with CKD are at increased risk of cardiovascular disease and premature mortality. The glomerular filtration rate (GFR) represents the most widely used measure of kidney function, and is used to classify patients according to disease severity. Although very precise GFR measurement is highly complex, it can be estimated using serum creatinine and some basic parameters like age and sex, using equations such as the eGFRCKD-EPI formula (Levey A et al. Ann Int Med 2009;150:604-12). Creatinine is continuously secreted by muscle cells; therefore, changes in body composition affect serum creatinine and thereby the creatinine-based eGFR (eGFRcrea). Consequently, eGFR is less reliable in persons with very low or high muscle mass. Over recent years, several large, international genome-wide association study (GWAS) efforts have successfully identified large numbers of genes related to creatinine-based eGFRcrea and eGFRcrea-based CKD diagnosis (Figure 1, Wuttke M et al. Nat Genet 2019;51:957-72). Yet, these eGFRcrea-based findings may at least partly be driven by variability in muscle mass.
An alternative approach to the assessment of kidney function is to use creatinine clearance, based on serum creatinine and 24-hour urinary creatinine excretion (see 5.5 for formula). Creatinine clearance has the advantage that it is not dependent of muscle mass; it has, however, the disadvantage that its widespread use is limited by the more complex additional requirement of 24-hour urine collections, which are not routinely collected in most centers (and cohort studies). In Lifelines, however, these data are available, allowing for the first large-scale GWAS of creatinine clearance, and comparison with eGFRcrea GWAS results (Figure 2). We will generate a new eGFRcrea dataset in UGLI and add this to existing publicly available data involving >1 million individuals (Wuttke M et al. Nat Genet 2019).
Creatinine is produced by skeletal muscle cells. Since creatinine is produced by muscle cells at a steady rate, and is similarly excreted by the kidneys, the 24-hour urinary creatinine excretion is a reliable measure of muscle mass. Thus, a creatinine excretion GWAS can provide important information on the genetic architecture of muscle mass. Prior studies have used different markers of muscle mass such as appendicular lean mass (Pei et al. Commun Biol 2020;3:608, Korostishevski M et al. Eur J Hum Genet 2016;24:277-8) or handgrip strength (Crosslan H et al. Am J Physiol Regul Integr Comp Physiol 2020;319:R184-R194). Yet, creatinine excretion is considered the most reliable marker of muscle mass. Muscle mass, in turn, is considered one of the main indicators of health, and most chronic diseases are characterized by loss of muscle mass. Several studies, including (unpublished) data from the Lifelines cohort, indicate that muscle mass as reflected by 24-hour urinary creatinine excretion, is strongly and inversely associated with a higher risk of premature mortality. The next question is whether interventions that increase muscle mass may also improve clinical outcomes. Decoding its genetic architecture may provide new targets for intervention to address this important question. 
In this project we propose to perform GWAS of creatinine clearance (representing kidney function) and creatinine excretion (representing muscle mass). In addition, we aim to compare the outcomes of these two analyses with existing (and newly generated) eGFRcrea GWAS data (Figure 2). This approach will also allow us to pinpoint which eGFRcrea-related loci reported in large-scale GWAS analysis involving >1 million individuals are actually driven by muscle mass, and therefore do not reflect kidney function.