Non-invasive diagnosis of acute rejection in renal transplant patients using mass spectrometry of urine samples - a multicentre diagnostic phase III trial
Final Report Abstract
This prospective, single-arm, multicentre, phase III diagnostic study was conducted to demonstrate that a pre-specified urinary peptide marker set (index test) can replace the indicated biopsy (reference test) in detecting acute T cell-mediated rejection (TCMR) of renal allografts (i.e. in-place validation of the index test). The primary objective was to determine the accuracy of the index test in terms of sensitivity and specificity in comparison to the indicated biopsy as the gold standard. A major deviation from the study protocol was the inclusion of archived patient samples (n=306) in order to achieve the planned sample size (n=600), resulting in a final overall sample size of 635 patients. The dataset consisted of 82 patients (13%) with acute TCMR grade I-III and 547 cases (87%) classified as borderline rejection or non-rejection, according to the Banff classification. Applying the pre-defined peptide marker set and thresholds, the sensitivity of the index test was 0.66 (95% CI 0.56-0.76) and the specificity 0.47 (95% CI 0.43-0.51) to predict acute TCMR. The area under the curve (AUC) was 0.60, the positive predictive value 0.16 (95% CI 0.12-0.20), and the negative predictive value 0.90 (95% CI 0.87-0.94). Thus, performance of the index test was not sufficient according to the pre-specified thresholds of 83% for sensitivity and 70% for specificity. Severity of acute TCMR was not reflected by the peptide marker set. Borderline cases of acute TCMR were not specifically detected by the peptide marker set. A number of confounding factors were identified which appeared to lead to false positive results of the index test in patients without rejection, particularly higher age and male gender, higher donor age and female donor gender, other graft injury than rejection, inflammation/infection, and therapy with heparin and calcium supplements. Statin therapy was associated with lower scores of the index test in all patients, suggesting systematic effects towards false negative diagnostic results. Performance of the peptide marker set was examined in relation to a secondary, independent histomorphological re-assessment of the biopsies. The inter-observer agreement (Krippendorff’s alpha) for diagnosing acute TCMR was 0.49 (95% CI 0.32-0.62). Application of the peptide marker set to the re-assessed biopsy diagnoses resulted in a slightly improved performance of index test, with an AUC of 0.63, a sensitivity of 0.73 (95% CI 0.58-0.88) and a specificity of 0.45 (95%-CI 0.40-0.50). Outcome of the patients after the index biopsy and index test was used in a secondary analysis to examine if missed diagnosis of acute TCMR in the index biopsy might have affected the performance of the index test. A total number of 47 patients were identified without TCMR in the index biopsy but with an acute rejection episode within 2 months after the index biopsy. Including these cases into the group of patients with acute TCMR in the index biopsy, no improvement in the diagnostic performance of the index test was observed. Based on the analysis of potential confounders of the index test and considering the different settings in which the pre-specified peptide marker set had been established and in which this test was applied, an alternative peptide marker set was established, using an independent sample from a European trial (NCT02832661) that more closely resembled the setting of this study. Application of the newly established marker set showed a small increase in test performance. Compared to the results with the pre-specified marker set, the AUC increased to 0.70, whereas sensitivity decreased from 0.66 to 0.63 (95% CI 0.59-0.68) and specificity increased from 0.47 to 0.62 (95% CI 0.51-0.72). Conclusion: The pre-specified urine peptide marker set was not sufficiently sensitive and specific to predict acute T cell-mediated rejection.
Publications
-
Non-invasive diagnosis of acute rejection in renal transplant patients using mass spectrometry of urine samples - a multicentre phase 3 diagnostic accuracy study. BMC Nephrology 2015, 16:153
Zapf A, Gwinner W, Karch A, Metzger J, Haller H, Koch A
-
Establishing Biomarkers in Transplant Medicine: A Critical Review of Current Approaches. Transplantation 2016,100:2024-38
Anglicheau D, Naesens M, Essig M, Gwinner W, Marquet P
-
Proteomics for rejection diagnosis in renal transplant patients: Where are we now? World J Transplant 2016, 6:28-41
Gwinner W, Metzger J, Husi H, Marx D
