Application of new acute kidney injury biomarkers in human randomized controlled trials
2016; Elsevier BV; Volume: 89; Issue: 6 Linguagem: Inglês
10.1016/j.kint.2016.02.027
ISSN1523-1755
AutoresChirag R. Parikh, Dennis G. Moledina, Steven G. Coca, Heather Thiessen‐Philbrook, Amit X. Garg,
Tópico(s)Sepsis Diagnosis and Treatment
ResumoThe use of novel biomarkers of acute kidney injury (AKI) in clinical trials may help evaluate treatments for AKI. Here we explore potential applications of biomarkers in simulated clinical trials of AKI using data from the TRIBE-AKI multicenter, prospective cohort study of patients undergoing cardiac surgery. First, in a hypothetical trial of an effective therapy at the time of acute tubular necrosis to prevent kidney injury progression, use of an indirect kidney injury marker such as creatinine compared to a new direct biomarker of kidney injury reduces the proportion of true acute tubular necrosis cases enrolled. The result is a lower observed relative risk reduction with the therapy, and lower statistical power to detect a therapy effect at a given sample size. Second, the addition of AKI biomarkers (interleukin-18 and NGAL) to clinical risk factors as eligibility criteria for trial enrollment in early AKI has the potential to increase the proportion of patients who will experience AKI progression and reduce trial cost. Third, we examine AKI biomarkers as outcome measures for the purposes of identifying therapies that warrant further testing in larger, multicenter, multi-country trials. In the hypothetical trial of lower cardiopulmonary bypass time to reduce the risk of postoperative AKI, the sample size required to detect a reduction in AKI is lower if new biomarkers are used to define AKI rather than serum creatinine. Thus, incorporation of new biomarkers of AKI has the potential to increase statistical power, decrease the sample size, and lower the cost of AKI trials. The use of novel biomarkers of acute kidney injury (AKI) in clinical trials may help evaluate treatments for AKI. Here we explore potential applications of biomarkers in simulated clinical trials of AKI using data from the TRIBE-AKI multicenter, prospective cohort study of patients undergoing cardiac surgery. First, in a hypothetical trial of an effective therapy at the time of acute tubular necrosis to prevent kidney injury progression, use of an indirect kidney injury marker such as creatinine compared to a new direct biomarker of kidney injury reduces the proportion of true acute tubular necrosis cases enrolled. The result is a lower observed relative risk reduction with the therapy, and lower statistical power to detect a therapy effect at a given sample size. Second, the addition of AKI biomarkers (interleukin-18 and NGAL) to clinical risk factors as eligibility criteria for trial enrollment in early AKI has the potential to increase the proportion of patients who will experience AKI progression and reduce trial cost. Third, we examine AKI biomarkers as outcome measures for the purposes of identifying therapies that warrant further testing in larger, multicenter, multi-country trials. In the hypothetical trial of lower cardiopulmonary bypass time to reduce the risk of postoperative AKI, the sample size required to detect a reduction in AKI is lower if new biomarkers are used to define AKI rather than serum creatinine. Thus, incorporation of new biomarkers of AKI has the potential to increase statistical power, decrease the sample size, and lower the cost of AKI trials. AKI is a research priority1Tong A. Chando S. Crowe S. et al.Research priority setting in kidney disease: a systematic review.Am J Kidney Dis. 2015; 65: 674-683Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar; it has many causes (surgery, infection, medication), affects 10% of hospitalized patients, is difficult to manage, is associated with poor outcomes, and is very costly. Although many therapies have ameliorated AKI in preclinical animal studies, few therapies have proven beneficial in human testing. There are 2 major challenges that have hampered AKI drug development. The first challenge has been the inability to adequately phenotype this syndrome. The diversity of AKI subtypes (hemodynamic, intrarenal, postrenal) and etiologies (septic, ischemic, nephrotoxic), coupled with the lack of accurate and rapid diagnostic tests that can distinguish between the different forms of AKI, has led to challenges in the diagnosis, classification, and prognosis of patients with the disease. In other words, it has proven difficult to enroll the "right patient" at the "right time" in AKI clinical trials. The second challenge has been the lack of validated outcome measures for AKI clinical trials. Currently, outcome measures used in early-phase clinical trials rely on disease classification systems such as AKIN, RIFLE, KDIGO, and their derivatives. These systems are primarily based on changes in the serum creatinine concentration, an indirect measure of kidney injury. Its concentration is dependent on several clinical variables, such as hydration, muscle metabolism, and medication effects.2Molitoris B.A. Urinary biomarkers: alone are they enough?.J Am Soc Nephrol. 2015; 26: 1485-1488Crossref PubMed Scopus (17) Google Scholar, 3Stevens L.A. Coresh J. Greene T. et al.Assessing kidney function — measured and estimated glomerular filtration rate.N Engl J Med. 2006; 354: 2473-2483Crossref PubMed Scopus (2300) Google Scholar Standardization generally assists with reporting of results; however, it is unclear whether a treatment's effect on AKI, as defined by these disease classification systems, will predict its effect on clinical outcomes.4Coca SG, Zabetian A, Ferket BS, et al. Evaluation of short-term changes in serum creatinine level as a meaningful end point in randomized clinical trials [e-pub ahead of print]. J Am Soc Nephrol. http://dx.doi.org/10.1681/ASN.2015060642, accessed April 14, 2016.Google Scholar A solution to both of these major challenges may lie in the development of more accurate AKI biomarkers.5Guidance for Industry and FDA Staff: Qualification Process for Drug Development Tools. Center for Drug Evaluation and Research, US Food and Drug Administration, Department of Health and Human Services; 2014.Google Scholar As in clinical practice, biomarkers are important tools in drug development. A biomarker is a measurable indicator of normal biologic processes, pathologic processes, or biologic responses to a therapeutic intervention.6Biomarkers Definitions Working GroupBiomarkers and surrogate endpoints: preferred definitions and conceptual framework.Clin Pharmacol Ther. 2001; 69: 89-95Crossref PubMed Scopus (4679) Google Scholar Biomarkers can be used in several ways to facilitate the conduct of AKI clinical trials. At enrollment they can be used as an "enrichment strategy" to (i) preferentially enroll patients with a certain type of AKI (diagnostic biomarkers), (ii) identify patients more likely to progress to a higher stage of AKI (prognostic biomarkers), and (iii) identify a subpopulation of patients most likely to respond to a particular intervention (predictive biomarkers). They can also be used as outcome measures to monitor response to therapy (pharmacodynamic biomarkers), whether it be beneficial or harmful to a patient (Figure 1). The terminology used in this article reflects the terminology currently adopted by the US Food and Drug Administration (FDA) to describe the biomarker uses within the context of drug development.5Guidance for Industry and FDA Staff: Qualification Process for Drug Development Tools. Center for Drug Evaluation and Research, US Food and Drug Administration, Department of Health and Human Services; 2014.Google Scholar As noted by the FDA, these methods of biomarker use are not mutually exclusive; for example, in diabetic kidney disease trials, urine albumin is used as a prognostic biomarker to preferentially enroll patients at higher risk of reaching the trial's end point, and as a pharmacodynamic biomarker to support proof-of-concept studies and aid in dose selection.7Jun M. Turin T.C. Woodward M. et al.Assessing the validity of surrogate outcomes for ESRD: a meta-analysis.J Am Soc Nephrol. 2015; 26: 2289-2302Crossref PubMed Scopus (37) Google Scholar Novel biomarkers used in isolation may not perform as well as a composite biomarker score in combination with existing clinical information and traditional tests.8Bihorac A. Chawla L.S. Shaw A.D. et al.Validation of cell-cycle arrest biomarkers for acute kidney injury using clinical adjudication.Am J Respir Crit Care Med. 2014; 189: 932-939Crossref PubMed Scopus (342) Google Scholar, 9Kashani K. Al-Khafaji A. Ardiles T. et al.Discovery and validation of cell cycle arrest biomarkers in human acute kidney injury.Crit Care. 2013; 17: R25Crossref PubMed Scopus (794) Google Scholar To examine the potential utility of new AKI biomarkers in drug development trials, we analyzed data from the Translational Research Investigating Biomarker Endpoints in AKI (TRIBE-AKI) cohort, a multicenter, prospective cohort of patients undergoing cardiac surgery. We evaluated biomarkers in various roles (diagnostic, prognostic, predictive, and pharmacodynamic) in hypothetical and simulated AKI trials, and demonstrate their impact on detectable relative risk, required sample size, and trial cost. Current AKI diagnostic criteria rely on the serum creatinine concentration, which is a filtration marker and indirectly measures kidney injury. Use of serum creatinine in prior clinical trials may have led to the enrollment of heterogeneous patient populations, where only a subpopulation had an AKI subtype likely to respond to the intervention being tested.10Waikar S.S. Betensky R.A. Emerson S.C. et al.Imperfect gold standards for kidney injury biomarker evaluation.J Am Soc Nephrol. 2012; 23: 13-21Crossref PubMed Scopus (205) Google Scholar In this regard, new diagnostic AKI biomarkers, which are released upon injury to specific renal tubular segments, may more accurately reflect a diagnosis of acute tubular necrosis (ATN). Interleukin 18 (IL-18), released from the proximal tubule upon injury, has been shown to differentiate ATN from other forms of kidney disease.11Parikh C.R. Jani A. Melnikov V.Y. et al.Urinary interleukin-18 is a marker of human acute tubular necrosis.Am J Kidney Dis. 2004; 43: 405-414Abstract Full Text Full Text PDF PubMed Scopus (428) Google Scholar Biomarkers such as neutrophil gelatinase–associated lipocalin (NGAL), kidney injury molecule-1 (KIM-1), and liver-type fatty acid–binding protein (L-FABP), as well as traditional biomarkers such as urine microalbumin and fractional excretion of sodium, also help diagnose ATN.12Belcher J.M. Sanyal A.J. Peixoto A.J. et al.Kidney biomarkers and differential diagnosis of patients with cirrhosis and acute kidney injury.Hepatology. 2014; 60: 622-632Crossref PubMed Scopus (206) Google Scholar Traditional and novel biomarker combinations, including filtration and injury biomarkers, could be used in clinical trials to enroll patients with a form of AKI most likely to respond to an intervention. Selection of these patients for enrollment would be expected to reduce the sample size needed to detect a therapeutic effect if it in truth exists. For example, a hypothetical drug that treats ATN may not be effective in treating other conditions that raise the serum creatinine concentration. Such conditions include hemodynamic causes such as volume depletion, hepatorenal and cardiorenal syndromes, glomerulonephritides, acute interstitial nephritis, and urinary tract obstruction. If a clinical trial uses elevation of serum creatinine as the enrollment criterion, it will invariably lead to enrollment of patients with and without ATN. Table 1 models the results of a trial of a hypothetical therapy for the early treatment of AKI that reduces the risk of AKI progression by 30% in patients with true ATN, but has no effect in patients without ATN. Depending on the non-ATN patient proportion in the enrolled trial sample, this could result in a substantial loss of statistical power (Figure 2a) and require a higher sample size to reliably detect the true treatment effect if it exists (Figure 2b).Table 1Example of early treatment trial in acute kidney injuryaTreatment administered at the time of clinical diagnosis of AKI by serum creatinine.AKI etiologyDichotomous outcomeObserved relative riskIntervention event rateControl event rate100% ATN + 0% non-ATN14%20%0.790% ATN + 10% non-ATN13.1%18.5%0.7180% ATN + 20% non-ATN12.2%17%0.7270% ATN + 30% non-ATN11.3%15.5%0.7360% ATN + 40% non-ATN10.4%14%0.7450% ATN + 50% non-ATN9.5%12.5%0.760% ATN + 100% non-ATN5%5%1.0Impact of changing the proportion of enrolled patients with true acute tubular necrosis (ATN) in a trial where a therapy reduces the relative risk (RR) of progressive kidney injury by 30% in patients with ATN, but has no effect on outcomes in the absence of ATN. We assume that 95% of non-ATN cases will respond to standard of care, and the acute kidney injury (AKI) will not progress. With these assumptions, the table shows the effect on observed RR with increasing proportion of non-ATN cases. Examples of calculation for 60% ATN + 40% non-ATN: Intervention rate: 60% (14%) + 40% (5%) = 10.4%. Control rate: 60% (20%) + 40% (5%) = 14%. RR = 10.4% / 14% = 0.74.a Treatment administered at the time of clinical diagnosis of AKI by serum creatinine. Open table in a new tab Impact of changing the proportion of enrolled patients with true acute tubular necrosis (ATN) in a trial where a therapy reduces the relative risk (RR) of progressive kidney injury by 30% in patients with ATN, but has no effect on outcomes in the absence of ATN. We assume that 95% of non-ATN cases will respond to standard of care, and the acute kidney injury (AKI) will not progress. With these assumptions, the table shows the effect on observed RR with increasing proportion of non-ATN cases. Examples of calculation for 60% ATN + 40% non-ATN: Intervention rate: 60% (14%) + 40% (5%) = 10.4%. Control rate: 60% (20%) + 40% (5%) = 14%. RR = 10.4% / 14% = 0.74. AKI prognostic biomarkers can help identify a patient subpopulation at high risk of developing severe AKI or other poor clinical outcomes. Enrolling patients at greater risk of reaching these study end points could lead to a substantial reduction in the required sample size and duration of follow-up for trials (recognizing there is additional effort to find patients who are eligible for trial participation).13Draft Guidance for Industry: Enrichment Strategies for Clinical Trials to Support Approval of Human Drugs and Biological Products. Center for Drug Evaluation and Research / Center for Biologics Evaluation and Research / Center for Devices and Radiological Health, US Food and Drug Administration; 2012.Google Scholar In other renal settings such as diabetic kidney disease and polycystic kidney disease, prognostic biomarkers (albuminuria, reduced glomerular filtration rate [GFR], total kidney volume) have been used in drug development as inclusion criteria to enrich clinical trials.14Fried L.F. Duckworth W. Zhang J.H. et al.Design of combination angiotensin receptor blocker and angiotensin-converting enzyme inhibitor for treatment of diabetic nephropathy (VA NEPHRON-D).Clin J Am Soc Nephrol. 2009; 4: 361-368Crossref PubMed Scopus (109) Google Scholar, 15Fried L.F. Emanuele N. Zhang J.H. et al.Combined angiotensin inhibition for the treatment of diabetic nephropathy.N Engl J Med. 2013; 369: 1892-1903Crossref PubMed Scopus (802) Google Scholar, 16de Zeeuw D. Akizawa T. Audhya P. et al.Bardoxolone methyl in type 2 diabetes and stage 4 chronic kidney disease.N Engl J Med. 2013; 369: 2492-2503Crossref PubMed Scopus (701) Google Scholar, 17Torres V.E. Abebe K.Z. Chapman A.B. et al.Angiotensin blockade in late autosomal dominant polycystic kidney disease.N Engl J Med. 2014; 371: 2267-2276Crossref PubMed Scopus (187) Google Scholar A number of observational studies have evaluated prognostic biomarkers (novel, traditional, risk models) of AKI in various clinical settings to identify high-risk patients; these studies indicate that measuring biomarkers at or before the time of AKI can improve our ability to detect patients at high risk for progressing to more severe AKI or a longer duration of AKI.8Bihorac A. Chawla L.S. Shaw A.D. et al.Validation of cell-cycle arrest biomarkers for acute kidney injury using clinical adjudication.Am J Respir Crit Care Med. 2014; 189: 932-939Crossref PubMed Scopus (342) Google Scholar, 9Kashani K. Al-Khafaji A. Ardiles T. et al.Discovery and validation of cell cycle arrest biomarkers in human acute kidney injury.Crit Care. 2013; 17: R25Crossref PubMed Scopus (794) Google Scholar, 18Parikh C.R. Devarajan P. Zappitelli M. et al.Postoperative biomarkers predict acute kidney injury and poor outcomes after pediatric cardiac surgery.J Am Soc Nephrol. 2011; 22: 1737-1747Crossref PubMed Scopus (294) Google Scholar, 19Parikh C.R. Coca S.G. Thiessen-Philbrook H. et al.Postoperative biomarkers predict acute kidney injury and poor outcomes after adult cardiac surgery.J Am Soc Nephrol. 2011; 22: 1748-1757Crossref PubMed Scopus (185) Google Scholar, 20Mishra J. Dent C. Tarabishi R. et al.Neutrophil gelatinase-associated lipocalin (NGAL) as a biomarker for acute renal injury after cardiac surgery.Lancet. 2005; 365: 1231-1238Abstract Full Text Full Text PDF PubMed Scopus (1946) Google Scholar Therefore, these prognostic biomarkers may enrich AKI trials. An example of this approach and its impact on the trial cost is presented in Table 2.Table 2Cost reduction by using biomarkers as enrollment criteria to predict clinical acute kidney injuryEnrollment criteriaSample size detailsTrial screeningTotal trial cost (screening + trial)aExpressed in millions.Clinical risk factorsbTRIBE-AKI cohort (high risk for AKI based on clinical risk factors; see Materials and Methods for details).Known insultcCardiopulmonary bypass time >120 minutes.Kidney injury biomarkerAKI event rate control armRequired sample size (RR 0.7)$500/patient screening cost$10,000/patient trial costdCost estimated assuming lower cost of $10,000 per patient, which may be more appropriate for a short-duration AKI therapeutic trial.$45,000/patient trial costeCost estimated based on $45,000 per patient. This cost is estimated based on recently reported pharmaceutical-sponsored trials in cardiology and endocrinology (from Roy21).Screen failure rateScreening cost (in dollars)Total cost (in millions)% Cost reductionfPercentage reduction in cost is calculated using trial with clinical risk factors as reference.Total cost (in millions)% Cost reductionfPercentage reduction in cost is calculated using trial with clinical risk factors as reference.✓——4.3%69020%$3451$69.0Reference$314.0Reference✓✓—6.9%421163%$5691$47.834%$195.238%✓—pNGAL > 200 ng/ml6.1%481056%$5466$53.626%$221.929%✓✓pNGAL > 200 ng/ml9.1%315578%$7171$38.747%$149.253%✓—uIL-18 > 60 pg/ml10.4%272080%$6800$34.053%$129.259%✓✓uIL-18 > 60 pg/ml12.2%229488%$9559$32.555%$112.864%AKI, acute kidney injury; AKI event, doubling of serum creatinine or requiring dialysis; pNGAL, plasma neutrophil gelatinase–associated lipocalin; RR, relative risk; uIL-18, urinary interleukin-18.a Expressed in millions.b TRIBE-AKI cohort (high risk for AKI based on clinical risk factors; see Materials and Methods for details).c Cardiopulmonary bypass time >120 minutes.d Cost estimated assuming lower cost of $10,000 per patient, which may be more appropriate for a short-duration AKI therapeutic trial.e Cost estimated based on $45,000 per patient. This cost is estimated based on recently reported pharmaceutical-sponsored trials in cardiology and endocrinology (from Roy21Roy A. Stifling new cures: the true cost of lengthy clinical drug trials. Project FDA report, Manhattan Institute For Policy Research, 2012.Google Scholar).f Percentage reduction in cost is calculated using trial with clinical risk factors as reference. Open table in a new tab AKI, acute kidney injury; AKI event, doubling of serum creatinine or requiring dialysis; pNGAL, plasma neutrophil gelatinase–associated lipocalin; RR, relative risk; uIL-18, urinary interleukin-18. To explore the effects of an enrichment strategy for enrolling patients at high risk for AKI soon after surgery, in the TRIBE-AKI cohort we utilized a combination of a known renal insult (cardiopulmonary bypass [CPB] time >120 minutes) and novel biomarkers (plasma NGAL and urine IL-18) obtained within 6 hours after the surgery (but before development of clinical AKI). We calculated the rate of developing severe AKI with various enrichment strategies, and calculated the sample size needed to detect a relative risk of 0.7 of a given intervention with 80% power and 5% alpha. We used the following assumptions: The cost of screening was $500 based on the cost of blood and urine testing and research staff effort in the extra screening step. The cost per patient enrolled in a trial was variable, and depended on the intervention cost, follow-up duration, and follow-up laboratory investigation cost. A recent cost analysis of phase 3 trials in obesity, cardiovascular disease, and diabetes showed that $45,000 to $50,000 was spent per patient enrolled.21Roy A. Stifling new cures: the true cost of lengthy clinical drug trials. Project FDA report, Manhattan Institute For Policy Research, 2012.Google Scholar We also did the cost analysis for a less expensive intervention by assuming $10,000 cost per patient, which may be realistic for a therapy in AKI that does not require many years of follow-up. By using a combination of known insult (CPB time >120 minutes) and injury markers (IL-18 or NGAL), we demonstrate that trial cost can be decreased by 29% to 64% (Table 2). The main drawback of using a highly selective strategy, as evident from these tables, is the high screen failure rate and the need to screen a large number of AKI patients (which would require additional clinical sites and investigators). Second, since this approach uses biomarkers obtained around the time of injury (cardiac surgery), it will not be applicable to therapies that need to be administered before the injury. Predictive biomarkers identify the patients most likely to respond to a particular treatment based on the candidate drug's mechanism of action. They are a powerful enrichment strategy and reduce trial size and cost by identifying a subgroup of patients who have higher likelihood of responding to intervention and thus demonstrate a larger effect size.5Guidance for Industry and FDA Staff: Qualification Process for Drug Development Tools. Center for Drug Evaluation and Research, US Food and Drug Administration, Department of Health and Human Services; 2014.Google Scholar Predictive biomarkers have been used with success in other diseases to enhance clinical trials and provide personalized treatments. A classic example of this strategy is the drug Herceptin (trastuzumab). In clinical trials, it was found that HER-2–overexpressing breast cancer patients had 5-month survival with Herceptin as compared to 2-month survival in patients overall.22Piccart-Gebhart M.J. Procter M. Leyland-Jones B. et al.Trastuzumab after adjuvant chemotherapy in HER2-positive breast cancer.N Engl J Med. 2005; 353: 1659-1672Crossref PubMed Scopus (4216) Google Scholar, 23Accurso F.J. Rowe S.M. Clancy J.P. et al.Effect of VX-770 in persons with cystic fibrosis and the G551D-CFTR mutation.N Engl J Med. 2010; 363: 1991-2003Crossref PubMed Scopus (638) Google Scholar Similarly, using a patient's molecular signature in AKI as an inclusion criterion could aid in the evaluation of novel therapeutics and personalized AKI treatments. For example, a drug acting on the apoptosis pathway via the NLRP-3 inflammasome complex would be more effective in patients in whom this AKI pathway is active. A biomarker, such as urine IL-18, that is released upon activation of the NLRP-3 inflammasome pathway could be a predictive biomarker to select patients for enrollment in a trial of such a drug. While both prognostic and predictive biomarkers can be used as enrichment strategy, there is an important distinction in their application. Prognostic biomarkers identify a subgroup of patients at higher risk of an event (e.g., need for dialysis, progression to a higher stage of AKI, mortality) and enrich the cohort enrolled in a trial for the desired event. Prognostic biomarkers can improve the absolute risk reduction achieved by an intervention by increasing the overall event rate but do not change the relative risk reduction. The biomarker used in prognostication may not be causally related to AKI progression, or be a target for therapy. Predictive biomarkers, on the other hand, identify a subgroup of patients who are most likely to respond to a particular intervention and in theory improve both the observed relative and absolute risk reduction with a therapy. A predictive biomarker is often a therapeutic target or causally related to the outcome. Pharmacodynamic biomarkers assess whether a biological response, favorable or unfavorable, has occurred in a patient receiving a therapeutic intervention. As end points, they assist in detecting efficacy or toxicity of the intervention and quantify the extent of this response, thereby also allowing selection of optimal dose of the intervention. Pharmacodynamic biomarkers indicating efficacy are valuable in early proof-of-concept trials. Biomarkers that respond rapidly to treatment could facilitate early decision making and shorten often lengthy proof-of-concept studies. Biomarkers examining clinical treatment effectiveness (i.e., improved survival or reduced disease progression) are being tested and validated in various forms of chronic kidney disease. For example, a reduction in proteinuria in patients with diabetic kidney disease is associated with a decreased progression from chronic kidney disease to end-stage renal disease.24Atkins R.C. Briganti E.M. Lewis J.B. et al.Proteinuria reduction and progression to renal failure in patients with type 2 diabetes mellitus and overt nephropathy.Am J Kidney Dis. 2005; 45: 281-287Abstract Full Text Full Text PDF PubMed Scopus (312) Google Scholar, 25de Zeeuw D. Remuzzi G. Parving H.-H. et al.Proteinuria, a target for renoprotection in patients with type 2 diabetic nephropathy: lessons from RENAAL.Kidney Int. 2004; 65: 2309-2320Abstract Full Text Full Text PDF PubMed Scopus (805) Google Scholar Furthermore, efficacy biomarkers can be incorporated in adaptive trial designs, where change in biomarkers on follow-up samples can be used to modify sample size or inclusion criteria during the course of the trial, further streamlining the clinical trial process. Pharmacodynamic biomarkers of efficacy may have potential of being accepted as surrogate outcomes if the effect of a specific treatment on biomarkers correlates strongly and consistently with meaningful clinical outcomes. While cross-sectional and short-term prospective studies have identified a number of biomarkers associated with AKI, there is a lack of evidence from longitudinal and interventional studies that validate the utility of any biomarker for monitoring disease activity or clinical response. Pharmacodynamic biomarkers that monitor toxicity or harm hold potential to serve as safety end points in clinical trials, particularly in phase 1 and 2 trials for dose selection. In fact, such application to monitor safety in drug development is quite prevalent. For example, serum creatinine is routinely monitored in the drug development process of nonrenal drugs to detect kidney toxicity during dose selection. However, serum creatinine lacks sensitivity and specificity for detecting true kidney injury and is not an ideal biomarker for safety.2Molitoris B.A. Urinary biomarkers: alone are they enough?.J Am Soc Nephrol. 2015; 26: 1485-1488Crossref PubMed Scopus (17) Google Scholar Biomarkers with higher sensitivity for kidney injury may be preferable as better markers of kidney damage. A panel of new urine biomarkers recently received FDA qualification for monitoring renal toxicity in preclinical studies.26Review of Qualification Data for Biomarkers of Nephrotoxicity Submitted by the Predictive Safety Testing Consortium. Center for Drug Evaluation and Research, US Food and Drug Administration; 2009.Google Scholar To demonstrate the potential application and benefit of a pharmacodynamic biomarker in AKI clinical trials, we use TRIBE-AKI data and simulate trials of 2 interventions, which in the literature have been shown to reduce the risk of AKI. In the first example, we use CPB time as an intervention to show that a smaller sample size would be needed with biomarkers to demonstrate the beneficial effect of shorter CPB time on kidney injury (Table 3, Supplementary Table S1 online). In the second example, we show that the beneficial effect of statins on AKI is evident if biomarkers are used as outcomes, whereas creatinine-based outcomes are not statistically significant (Figure 3).Table 3Sample size for hypothetical cardiopulmonary bypass time trial using various biomarkers as pharmacodynamic end pointsEvent rate in control arm (CPB time > 120 min)Minimal detectable relative riskRequired sample sizeReclassification % in control arm. AKI by biomarker versus clinical AKIbReclassification percentage was calculated in the control arm as the percentage of individuals with improved reclassification when defining AKI by elevated biomarker levels compared to clinical AKI defined by elevation in serum creatinine. Improvements in reclassification occurred in the following 2 circumstances: (i) there was no clinical AKI by serum creatinine, but AKI by elevated biomarker levels, or (ii) there was clinical AKI but no AKI by biomarker elevation (see shaded cells in Supplementary Table S2).AKI > 50% or dialysisAKI > 100% or dialysisClinical AKI defined by serum creatinineAKI > 50% or dialysis26%0.5320……AKI > 100% or dialysis7%0.571945……AKIN stage 1 or higher49%0.59204……AK
Referência(s)