Estimating GFR from a Panel of Endogenous Filtration Markers (Panel eGFR)

描述（由申请人提供）：评估肾功能是医学实践不可或缺的一部分。然而，即使是基于血清肌酐和胱抑素 C 的最准确的 GFR 估计值在选定人群中也会存在偏差，而且总体上并不精确。测量的 GFR 是估计 GFR 降低的唯一验证性测试，但不实用，因此在大多数临床实践或研究环境中不进行。 我们的长期目标是开发与测量的 GFR 一样准确的 GFR 估计值，需要更少的人口统计或临床变量，并且只需一个血液样本即可分析一组内源性滤过标记物，临床实验室可以自动报告这些标记物，用作临床实践和研究中的验证性测试。我们认为过去的一个关键缺陷 改进 GFR 估计的尝试是寻找单一理想的滤过标记。我们的目标是评估新型内源性滤过标记物，并确定由 4-7 个标记物（2-4 个新型标记物和 2-3 个成熟标记物）组成的“GFR 组合”，与 GFR 估计方程一起使用来报告“eGFR 组合”。基于统计概念并经我们的初步数据证实，我们的中心假设是，即使每个新标记物并不比肌酐或半胱氨酸蛋白酶抑制剂 C 更准确，面板 eGFR 也可以比当前 GFR 估计更准确。在面板中包含多个标记物的基本原理是减少每个标记物的不相关非 GFR 决定因素的偏差，减少包含人口或临床变量的需要，从而提高每个附加标记物的精度。预期结果是 GFR 面板和 GFR 估计方程，可用于报告面板 eGFR，接近测量 GFR 的准确性。 我们的研究团队慢性肾脏病流行病学合作组织 (CKD-EPI) 在生物标志物评估、GFR 估计和 CKD 流行病学方面拥有丰富的经验。在当前的提案中，我们将使用 9 项研究中 5390 名受试者的样本来检查 3 种已确定的（尿素、肌酐和胱抑素 C）和 4 种新型过滤标记物 [3 种低分子量血清蛋白：¿-微量蛋白 (BTP)、¿-2 微球蛋白 (B2M) 和肿瘤相关胰蛋白酶抑制剂 (TATI)；和 1 种代谢物：2-(¿-吡喃甘露糖基)-L-色氨酸（MPT，也称为 Tryp-Man），或另一种代谢物]。我们的具体目标是 (1) 评估新型内源性滤过标记物，以将其纳入具有完善滤过标记物的组（GFR 组）中。 (2) 开发 GFR 估计方程，与多个滤过标记一起使用来报告面板 eGFR。我们有足够的能力在研究中以及汇总数据集中的子组中检验我们的假设。 拟议的工作具有高度创新性，因为它增强了使用单一标记物估计 GFR 的传统策略，并且将能够开发出与使用单一血液样本测量的 GFR 一样准确的 GFR 估计值。拟议的工作意义重大，因为它将促进 GFR 估计方程的开发，以确认临床实践和研究中估计 GFR 的下降。

DESCRIPTION (provided by applicant): Assessing kidney function is an integral part of the practice of medicine. However, even the most accurate GFR estimates based on serum creatinine and cystatin C are biased in selected populations and imprecise in all. Measured GFR is the only confirmatory test for decreased estimated GFR, but is not practical and consequently, is not performed in most clinical practice or research settings. Our long-term goal is to develop GFR estimates that are as accurate as measured GFR, requiring fewer demographic or clinical variables and only a single blood sample to assay a panel of endogenous filtration markers, which can be reported automatically by clinical laboratories for use as a confirmatory test in clinical practice and research. We think that a critical flaw in past attempts to improve GFR estimation is the search for a single ideal filtration marker. Our objective is to evaluate novel endogenous filtration markers and to identify a "GFR panel" consisting of 4-7 markers (2-4 novel and 2-3 well-established markers) for use with GFR estimating equations to report a "panel eGFR". Our central hypothesis, based on statistical concepts and confirmed by our preliminary data, is that the panel eGFR can be substantially more accurate than current GFR estimates even if each novel marker is not more accurate than creatinine or cystatin C. The rationale for including multiple markers in a panel is to diminish bias from non-correlated non-GFR determinants of each marker, reduce the need for inclusion of demographic or clinical variables, thereby increasing precision with each additional marker. The expected outcome is a GFR panel and GFR estimating equations that can be used for reporting panel eGFR that approaches the accuracy of measured GFR. Our research team, the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI), has extensive experience in evaluation of biomarkers, GFR estimation, and CKD epidemiology. In the current proposal, we will use specimens from 5390 subjects in 9 studies to examine 3 established (urea, creatinine and cystatin C) and 4 novel filtration markers [3 low molecular weight serum proteins: ¿-trace protein (BTP), ¿-2 microglobulin (B2M), and tumor associated trypsin inhibitor (TATI); and 1 metabolite: 2-(¿-mannopyranosyl)-L-tryptophan (MPT, also known as Tryp-Man), or another metabolite]. Our specific aims are (1) to evaluate novel endogenous filtration markers for inclusion in a panel with well-established filtration markers (GFR panel). (2) To develop GFR estimating equations for use with multiple filtration markers to report a panel eGFR. We have adequate power to test our hypotheses within studies, and within subgroups in the pooled dataset. The proposed work is highly innovative because it augments the traditional strategy of estimating GFR using a single marker, and will enable development of GFR estimates that are as accurate as measured GFR using a single blood sample. The proposed work is significant because it will facilitate development of GFR estimating equations for confirmation of decreased estimated GFR in clinical practice and research.