Biomarker research holds the potential for developing new, more accurate and powerful IVDs.
Evidence has been accumulating that several novel biomarkers will become powerful clinical tools for detecting treatment-induced acute kidney injury (AKI) and diagnosing impaired kidney function.1 In 2004, FDA launched the Critical Path Initiative (CPI) with the purpose of improving drug development.2 In order to support CPI, the Critical Path Institute, an independent non-profit institute, was developed in 2005 to facilitate collaboration among FDA scientists, academia, and industry for the sake of public good. In 2006, the Critical Path Institute announced the Predictive Safety Testing Consortium (PSTC), a public-private partnership with the specific aim of bringing together pharmaceutical companies to share and eventually improve the current safety testing strategies for drugs in development.
|Photo courtesy of Pacific Biomarkers Inc.|
CPI inspired Pacific Biomarkers Inc. (Seattle) to develop a biomarker validation strategy that would identify novel biomarkers useful for assessing the drug-induced toxicity of drugs in development.3 Following detailed analytical validations, the biomarkers qualified in this program would be used in clinical trials with the goal of becoming accurate and powerful IVDs that would either replace or augment less sensitive tests. This article will summarize the work that Pacific Biomarkers has initiated, describe some individual biomarkers in the program, and outline their utility as potential IVDs.
Aims and Approaches
The current costs associated with drug development from discovery to final approval have been estimated to be $1 billion or more.4 Drug-induced organ toxicity accounts for 30% of all drugs that fail prior to reaching the market.5 The sooner such toxicity is discovered, the sooner further development costs can be curtailed, enabling drug developers to focus on other safer drugs. Early detection of organ toxicity would not only reduce the cost of drug development but also avert injuries to patients involved in clinical trials.
Thus, an appropriate acute organ injury biomarker panel has the potential to detect organ-specific toxic drug effects that currently are being missed in clinical trials. The early detection of organ injury in clinical trials could reduce drug development timelines by a number of years. Such detection might also reduce overall healthcare costs if organ injury could be diagnosed early enough to avoid severe and costly complications in patients using the drug after it is approved.
The key strategic goal of Pacific Biomarkers’ program is to provide pharmaceutical and biotech companies with services for testing robust novel biomarkers that have undergone thorough analytical validation and clinical qualification to diagnose early organ injury. This kind of work is now being conducted in response to recommendations by FDA for improving drug development as outlined in CPI, which currently is moving forward under the direction of PSTC and the Health and Environment Sciences Institute.2
CPI’s aim is to characterize not only the analytical performance of such novel biomarkers for detecting organ injury but also their clinical performance through collaborations among various pharmaceutical companies. The anticipated qualification of organ injury biomarkers will ultimately demand the development of stringent standardization procedures, which currently are nonexistent. As this article will discuss in the following sections, the analytical performance of some AKI biomarkers has already been characterized while the validation of others is underway.
The best approach to accomplishing the goals of the CPI program is to validate each biomarker using rigorous protocols for singlicate assays that generate the most robust data possible. During this process, modifying the assay methods or validating new ones may be necessary. Solid analytical performance is necessary to ensure that the data generated from the studies evaluating the novel biomarkers provides the best possible basis for choosing appropriate biomarkers for further development. Once a panel of 5-7 biomarkers with strong clinical performance has been identified, investigations on multiplex procedures for these biomarkers can be initiated.
Unfortunately, for many biomarkers, analytical performance deteriorates significantly during multiplexing. Nevertheless, testing multiple biomarkers at the same time using a multiplexed design is crucial for some physiological conditions. The obvious advantage is a comprehensive analysis of one particular pathophysiological pathway. However, this approach may reduce the overall analytical performance of the examined biomarkers, thereby reducing the data’s utility.
|Table I. Renal safety biomarker panel in urine.|
Therefore, considerable efforts should be given toward establishing sound performance for the individual biomarkers before attempts can be made to perform them on a multiplex platform. The preliminary work will provide assay performance targets to be used for subsequent multiplex assay evaluation. Hence, identifying methods and platforms with solid analytical and diagnostic performance in the early phases is vital before studies evaluating biomarker clinical utility are initiated (see Table I).
Individual Markers and Their Uses
Different consortia and key opinion leaders have identified approximately 30 biomarkers for AKI.6 The biomarkers with the greatest current level of analytical and clinical evidence are being targeted and investigated in Pacific Biomarkers’ present program. They fall into three categories as summarized here.
The first group includes biomarkers of specific renal tissue injury. For example, kidney injury molecule 1 (KIM-1) is a type 1 transmembrane protein that is not detectable in normal kidney tissue but is expressed at high levels in human and rodent kidneys with dedifferentiated proximal tubule epithelial cells after ischemic or toxic injury.7 Neutrophil gelatinase-associated lipocalin (NGAL), another biomarker of specific renal injury, is a ubiquitous 25 kDa protein generally expressed in low concentrations but which increases greatly in the presence of epithelial injury and inflammation.8 Also, a biomarker specific for proximal tubular damage is n-acetyl glucosaminadase (NAG).9 Increased interleukin 18 (IL-18), another marker in this group, has been associated with a variety of causes of renal injury from endotoxemia to cisplatin toxicity, while yet another, clusterin, is thought to reflect renal ischemia-reperfusion injury. All of these biomarkers are presently available for use in clinical trials.10
The second group of biomarkers, which includes cystatin C, albumin, total protein levels, and β2-microglobulin, are good global kidney markers assessing glomerular function.11-14 Most of these markers have been known for some time, and thus a great amount of clinical evidence has accumulated to establish their clinical utility. However, a drawback of these biomarkers is their lack of specificity to localized regions of the kidney. All four of these markers are also currently available to support clinical trials.
The third group includes emerging or more exploratory AKI biomarkers. Thus, α-gluthathione S-transferase (α-GST) is believed to be specific for proximal tubule damage, as is retinol binding protein-4 (RBP-4), trefoil factor 3 (TTF3), osteoactivin, and calbindin.15-17 Meanwhile, liver-type fatty acid binding protein (L-FABP) is a promising biomarker for tubular injury, while ϒ-glutamyl transferase (ϒ-GT) and pi-glutathione S-transferase (Π-GST) may indicate tubular epithelium injury and distal tubular epithelium injury, respectively.18-20 Finally, type IV collagen levels could be useful as an indicator of glomerular injury.21 To date, α-GST and RBP-4 are ready to support new clinical research, while osteoactivin will be available soon. The rest of these biomarkers must still be considered emerging biomarkers with not enough evidence to support clinical trials.
Potential Uses as IVDs
In the near future, a handful of biomarkers that can be used as clinical IVDs will likely be identified by different consortia and receive FDA approval. Such approval must be preceded by additional clinical validation and strong analytical data suggesting that these biomarkers are fit for IVD use. All the biomarkers highlighted above do have potential utility as IVDs. However, only a handful will most likely pass FDA’s biomarker validation process.
In addition to detecting AKI, the biomarkers will eventually need to fulfill at least the following three major clinical roles: general diagnostic utility for physicians to identify patients with potential AKI; diagnosis of AKI in patients in intensive care units with septic shock or following major surgeries (e.g., coronary artery bypass graft surgery or kidney transplant); and as tools to follow and evaluate kidney function in, among others, diabetic patients who are prone to kidney disease.
Existing serum biomarkers that detect kidney toxicity include creatinine and blood urea nitrogen. However, both have limitations since substantial injury is required before an increase in these biomarkers is observed. While measuring the glomerular filtration rate is sometimes used to assess renal function, it is tedious, can require injection of chemicals, and in the end may still not be a reliable marker for detecting early kidney injury. Thus, the biomarkers available for AKI all lack sensitivity and specificity for early detection of impaired kidney function.
Optimal new IVDs for AKI will have to satisfy the following multiple criteria: identify kidney injury before serum creatinine levels increase; reflect the degree of toxicity; be specific to a localized site in the kidney; track the progression of injury and recovery; predict outcomes; act as surrogate endpoints useful for clinical interventional studies; and be readily detectable in available body fluids.6,22
FDA’s IVD database lists about 300 products involving the detection of creatinine levels that the agency approved between 1976 and 2010.23 The number of products available for creatinine detection highlights the environment in which the new kidney biomarker studies are being conducted. A blend of various methods and platforms is available to healthcare professionals for measuring creatinine, the traditional kidney function marker. Similar technologies will most likely be applied to measuring new AKI biomarkers and will hopefully result in the same reliable and reproducible results.
However, before getting to that point, such novel biomarkers will have to go through the thorough clinical validation and qualification approval procedure as recommended by FDA.24 This process begins with a submission of biomarker data for review by the Interdisciplinary Pharmacogenomic Review Group (IRPG) at FDA. Consisting of various experts from different FDA centers, the IPRG team reviews the biomarker qualification data together with other biomarker data submitted through the voluntary data submission process. This evaluation allows FDA, together with the applicant, to design a biomarker qualification study that would produce the data that would justify either the acceptance or rejection of the suggested biomarker. While this process is long and tedious, each one of the novel organ safety biomarkers needs to go through it before a consensus can be reached on which biomarkers are qualified for IVD use.
The IVD industry has already shown enormous interest in developing these novel biomarkers for diagnostic purposes. An interesting example of what the future of IVDs for AKI might be is the measurement of urine NGAL. Bioporto Diagnostics (Gentofte, Denmark) is developing methods for the early diagnosis of AKI in critical illness, nephrotoxicity, and kidney transplantation using NGAL. Bioporto has an established ELISA method for NGAL and is introducing an automated (research use only) assay that may be set up on an open channel on a variety of random-access chemistry analyzers. The Bioporto assay would allow NGAL analyses in urine, plasma, and serum samples. Bioporto has reported a mean of 63 ng/mL (range 37–106 ng/mL, n=80) in EDTA plasma and 5.3 ng/mL (range 0.7–9.8 ng/mL, n=7) in urine of healthy donors. Patients in intensive care units exhibit NGAL concentrations in urine ranging from 110 ng/mL to 40,000 ng/mL (n=11) and from 66 ng/mL to 922 ng/mL in serum (n=11).
Another major player that is developing a diagnostic NGAL assay is Abbott Laboratories (Abbott Park, IL), which has developed a urine NGAL test for it random access autoanalyzer, the Architect. This test is being used in certain European countries after it received the CE Mark of certification last year. Less than a year ago, Abbott submitted the data that it has generated with its NGAL assay to FDA for approval as the first novel diagnostic biomarker for detecting AKI in the United States. Future biomarker testing using IVD devices featuring speed and accuracy comparable to the Architect but which incorporate other specific biomarkers is now being developed as the result of the AKI biomarker program discussed in this article.
Pacific Biomarkers’ current strategy proposes that once the utility of a new biomarker is established using an immunoassay with well-documented high-quality performance characteristics, it will be necessary to transfer the biomarker to fast, high-throughput automated systems to accommodate its use in clinical diagnostic laboratories serving hospitals and physicians for the diagnosis and monitoring of acute kidney injury.
1. B Borrell, “Biomarkers for Kidney Damage Should Speed Drug Development,” in Nature News [online] May 10, 2010 [cited 4 November 2010]; available from Internet: www.nature.com/news/2010/100510/full/news.2010.232.html.
2. “Critical Path Initiative,” U.S. Food and Drug Administration Website [cited 4 November 2010]; available from Internet: www.fda.gov/ScienceResearch/SpecialTopics/CriticalPathInitiative/default....
3. “AKI Biomarker Plan,” Pacific Biomarkers Inc. Website [cited 4 November 2010]; available from Internet: www.pacbio.com/publications/white-papers-AKI.php.
4. CP Adams and VV Brantner VV, “Spending on New Drug Development,” Health Economics, Policy, and Law 19, no. 2 (2010): 130-141.
5. F Ge and QY He, “Genomic and Proteomic Approaches for Predicting Toxicity and Adverse Drug Reactions,” Expert Opinion on Drug Metabolism & Toxicology 5, no. 1 (2009): 29-37.
6. JV Bonventre, et al., “Next-Generation Biomarkers for Detecting Kidney Toxicity,” Nature Biotechnology 28, no. 5 (2010): 436-440.
7. WK Han, et al., “Human Kidney Injury Molecule-1 is a Tissue and Urinary Tumor Marker of Renal Cell Carcinoma,” Journal of the American Society of Nephrology 16, no. 4 (2005): 1126-1134.
8. KM Schmidt-Ott, et al., “Neutrophil Gelatinase-Associated Lipocalin-Mediated Iron Traffic in Kidney Epithelia,” Current Opinion in Nephrology and Hypertension 15 (2006): 442-449.
9. K Damman, et al., “Tubular Damage in Chronic Systolic Heart Failure is Associated with Reduced Survival Independent of Glomerular Filtration Rate,” Heart 96, no. 16 (2010): 1297-1302.
10. JA Leslie and KK Meldrum, “The Role of Interleukin-18 in Renal Injury,” Canadian Medical Association Journal 178, no. 2 (2008): 145-148.
11. VS Vaidya, et al., “Urinary Biomarkers for Sensitive and Specific Detection of AKI in Humans,” Clinical and Translational Science 1, no. 3 (2008): 200-208.
12. CJ Wiedermann, W Wiedermann, and M Joannidis, “Hypoalbuminemia and AKI: a Meta-Analysis of Observational Clinical Studies,” Intensive Care Medicine 36, no. 10 (2010): 1657-1665.
13. J Barratt and P Topham, “Urine Proteomics: the Present and Future of Measuring Urinary Protein Components in Disease,” Canadian Medical Association Journal 177, no. 4 (2007): 361-368.
14. IK Al-Taee, et al., “The Clinical Significance of β2-Microglobulin in End-Stage Renal Disease,” Saudi Journal of Kidney Diseases and Transplantation 14 (2003): 492-496.
15. BK Van Kreel, MA Janssen, and G Kootstra, “Functional Relationship of Alpha-Glutathione S-Transferase and Glutathione S-Transferase Activity in Machine-Preserved Non-Heart-Beating Donor Kidneys,” Transplant International 15, no. 11 (2002): 546-549.
16. Y Yu, H Jin, and D Holder, “Urinary Biomarkers Trefoil Factor 3 and Albumin Enable Early Detection of Kidney Tubular Injury,” Nature Biotechnology 28, no. 5 (2010): 470-477.
17. S Sourial, M Marcusson-Ståhl, and K Cederbrant, “Meso Scale Discovery and Luminex Comparative Analysis of Calbindin D28K,” Journal of Biomedicine and Biotechnology Epub (2009): 187426.
18. MA Ferguson, et al., “Urinary Liver-Type Fatty Acid-Binding Protein Predicts Adverse Outcomes in AKI,” Kidney International 77, no. 8 (2010): 708-714.
19. G Targher, et al., “Relationship Between Serum Gamma-Glutamyltransferase and Chronic Kidney Disease in the United States Adult Population: Findings from the National Health and Nutrition Examination Survey 2001-2006,” Nutrition, Metabolism, and Cardiovascular Diseases 20, no. 8 (2010): 583-590.
20. JL Koyner and PT Murray, “Alpha Glutathione S-Transferase and Pi Glutathione S-Transferase are Novel Biomarkers of AKI Following Adult Cardiac Surgery,” World Congress of Nephrology, Milan 2009 [cited 5 November 2010]; available from Internet: www.argutusmed.com/Acute_Kidney_Injury/Default.469.html.
21. Y Furumatsu, et al., “Urinary Type IV Collagen in Nondiabetic Kidney Disease,” Nephron Clinical Practice 117 (2011): c160-c166.
22. CSC Bouman, LG Forni, and M Joannidis, “Biomarkers and AKI: Dining with the Fisher King?” Intensive Care Medicine 36, no. 3 (2009): 381-384.
23. “In Vitro Diagnostics,” U.S. Food and Drug Administrations Website [cited 9 November 2010]; available from Internet: www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfivd/index.cfm.
24. F Goodsaid and F Frueh, “Biomarker Qualification Pilot Process at the U.S. Food and Drug Administration,” AAPS Journal 9, no. 1 (2007): E105-E108.
Timothy H. Carlson, PhD, is director of laboratory services at Pacific Biomarkers Inc. (Seattle). He can be reached at email@example.com.
Amar A. Sethi, MD, PhD, is vice president of research and development at Pacific Biomarkers Inc. (Seattle). He can be reached at firstname.lastname@example.org.