The Quest for Better Biomarkers of Bone Turnover in CKD

• renal osteodystrophy
• bone turnover
• hyperparathyroidism

Renal osteodystrophy (ROD) is a complex heterogeneous disorder of bone that results from abnormal calcium and phosphate metabolism, decreased calcitriol synthesis, increased parathyroid hormone (PTH) levels, metabolic acidosis, and defective bone mineralization. Specifically, ROD is the bone component of CKD–Mineral and Bone Disease (CKD-MBD), a disorder of bone, mineral metabolism, and soft tissue calcifications. More than one in ten Americans has CKD¹ and CKD-MBD occurs in nearly 100% of patients with CKD. ROD results in bone loss² and fractures³ and has been linked to increased risk of vascular calcifications and cardiovascular (CV) events.⁴ For patients with CKD, compared with the general population, fractures and CV risk are >17-⁵ and >1.4-fold⁶ greater, respectively, and mortality rates after fracture and CV events are >3-⁷ and >10-fold greater,⁶ respectively. In 2010, health care–associated costs after fracture exceeded $600 million.⁷ Thus, improvements in the diagnosis and clinical management of ROD are a critical first step in the long-term goal of reducing morbidity and mortality in patients with CKD-MBD.

Diagnosis of turnover is an impediment to ROD treatment. The 2005 Kidney Disease Improving Global Outcomes (KDIGO) committee shifted the historical nomenclature of ROD type (e.g., osteitis fibrosa cystica) to a unified classification system on the basis of bone Turnover, Mineralization, and Volume (TMV), and ROD turnover is now classified as low- or high-turnover ROD. Current treatment of ROD is focused on suppressing high turnover with active vitamin D (calcitriol and analogs) and/or calcimimetics, while simultaneously avoiding the development of low-turnover ROD through excessive use of these same agents. In addition, emerging data and clinical experience suggest that ROD with bone loss or fractures may be safely managed with treatments that are used for osteoporosis (antiresorptives for high-turnover ROD; anabolics for low-turnover ROD), as long as low-turnover ROD can be identified and avoided. The primary concern in identifying and preventing the development of low-turnover ROD is that it has been associated with risk of fractures⁸ and vascular calcifications that may increase CV risk.⁴ Guidelines and clinical experience recommend that diagnosis of turnover should be obtained before starting ROD treatment, and turnover should be monitored during the course of therapy because turnover may change, thus requiring an alteration to the treatment (e.g., discontinuing calcitriol or calcimimetics for over suppression of turnover). The gold standard method to define turnover is double-labeled tetracycline transiliac crest bone biopsy with quantitative histomorphometry. However, bone biopsy is invasive, expensive, requires approximately 3-months for results, cannot be used for rapid decision making, is not easily implemented as a disease and treatment monitoring tool, and is available at only several centers worldwide. In addition, it assumes that iliac crest remodeling is representative of systemic turnover. Because these limitations render bone biopsy impractical and in the vast majority of cases impossible to use for either diagnosis or treatment monitoring, KDIGO recommends that circulating levels of PTH and bone specific alkaline phosphatase (BSAP) can be used in the clinic to diagnose and guide management of ROD.

Unfortunately, PTH and other clinically used markers of bone formation and resorption provide less than optimal discrimination of turnover type in ROD. The largest bone biopsy study to date reported that the areas under the curve (AUCs) for PTH and markers of bone formation (BSAP, procollagen type-1 N-terminal propeptide [P1NP]) are between 0.70 and 0.80, and that combining PTH with BSAP or P1NP does not improve discrimination.⁹ Therefore, although PTH and bone turnover markers discriminate turnover type, they lack sufficient accuracy to guide completely and safely our ROD treatment decisions or to be used as part of clinical trial protocols of antifracture agents in patients with CKD. Better noninvasive biomarkers of turnover in ROD are needed to overcome these important limitations and help advance patient care and the development and study of antifracture pharmacologic agents in CKD.

In this edition of the Journal of the American Society of Nephrology, Salam et al.¹⁰ report their findings on the use of bone imaging and circulating bone turnover markers as biomarkers of turnover type. In 69 patients with CKD stages 3 through 5D, they measured gold standard bone turnover by tetracycline double-labeled transiliac crest bone biopsy, obtained measures of bone structure by dual energy x-ray absorptiometry and high-resolution peripheral quantitative computed tomography (HR-pQCT), and assayed PTH and circulating markers of bone formation (BSAP, P1NP) and resorption (C-telopeptide [CTX] and tartrate resistant acid phosphatase 5b [Trap5b]). They reported that PTH and markers of bone formation and resorption all discriminated turnover. BSAP discriminated low turnover best with an AUC of 0.824, and although PTH did not discriminate low turnover it had the best discrimination of high turnover with an AUC of 0.760. They also reported that several structural features from bone imaging correlated with and discriminated turnover, most notably that greater total volumetric bone density and greater cortical bone volume/tissue volume discriminated low from non-low turnover, with AUCs of 0.811 and 0.802, respectively. Similar to other studies, combining PTH, bone turnover markers, and bone imaging measures did not improve discrimination of turnover types.

These data confirm that the quest to find better biomarkers of turnover is far from over. The diagnostic test characteristics reported by Salam et al.¹⁰ for PTH and circulating protein biomarkers of turnover are similar and consistent with those reported in previous bone biopsy series,⁹ supporting the consensus that the accuracy of these biomarkers is insufficient for ROD management at large. The use of bone structural parameters to identify bone dynamic parameters is highly controversial and doubtable because the effects of bone turnover on bone structure require about 3–6 months to be detectable on HR-pQCT imaging. Furthermore, bone biopsy findings contradict the assumption that imaging of bone structure can identify bone turnover, because we have found that lower trabecular bone volume/tissue volume and decreased cortical bone volume are present in patients with adynamic or low bone turnover, likely due to past effects of high remodeling rates.

Do these data nudge forward the paradigm of ROD treatment? On the basis of the 2017 KDIGO Guidelines, PTH and BSAP can be used in the clinic to identify turnover type because markedly high or low levels identify underlying turnover.¹¹ In cases when their levels are neither high nor low, bone biopsy is recommended if the results would affect treatment decisions (a none graded recommendation). Thus, nothing has changed. Unfortunately, almost all of our patients are affected by this complex, lethal, and often forgotten complication of CKD and one of our main impediments to developing and studying novel agents to mitigate its effects is the lack of noninvasive biomarkers that can be used for its diagnosis and treatment monitoring. Therefore, the study by Salam et al.¹⁰ provides a wake-up call to the scientific community, funding agencies, and patient advocacy and industry groups working in CKD to jump start the effort to finding better noninvasive biomarkers of ROD.

• Copyright © 2018 by the American Society of Nephrology