Operational checks for real-time urinary biomarker measurements for assessment of the quality of deceased donor kidneys Authors: Mitali Barik, Serena DSouza, Wassim Obeid, David Hu, Sophia Xu, Mona Doshi, Sumit Mohan, Elyssa Gansner, Gary Marklin, Nick Olden, Wrenn Woodard, Walter Herczyk, Heather Thiessen Philbrook, Chirag R Parikh Published: 2026 ABSTRACT AbstractLateral flow devices (LFDs) were integrated into the clinical organ procurement workflow of two organ procurement organizations (OPOs) to obtain measurements of urinary biomarkers uromodulin (UMOD), osteopontin (OPN), and chitinase-3-like protein 1 (YKL-40) in deceased donors. We report on the feasibility of expanding the procurement workflow with this point-of-care testing, as well as methods of monitoring study performance and sample integrity through quality checks (QCs). Data QCs were implemented at OPO sample collection, transport/storage, processing, sample shipment back to the data coordinating center, and data transcription. From April 2022 to April 2025, 62 laboratory staff at two OPOs measured repair biomarkers of 958 samples from 879 unique donors. 4% of samples (n = 35) were excluded due to sample QCs, such as being left unfrozen beyond acceptable protocol ranges. These exclusions resulted in 923 samples from 851 unique donors [720 donors after brain death (DBD), 131 donors after circulatory death (DCD)]. Median processing time of LFD biomarker measurements across OPOs was 46 min (IQR: 40–55) and 5.1% cases required LFD remeasurement due to invalid readings. Developed QCs were used to identify samples suitable for research purposes and have been summarized into a checklist for use in multidisciplinary sciences.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-026-48155-5.Keywords: Lateral flow devices, Sample integrity, Urine, Organ procurement organization, Donors, Transplantation, Deceased donors, Quality checksSubject terms: Biological techniques, Biomarkers, Health care, Medical research Introduction Although the number of deceased donor kidney transplants (DDKTs) in the United States is increasing, rates of organ non-utilization and delayed graft function post-transplantation have also increased1,2. Organ procurement organizations (OPOs) play an important role in the organ donation process by coordinating the procurement, assessment, transport and transplantation of organs. The assessment of the quality of the procured organ is vital as it determines if the organ needs biopsy, pumping, should be used locally and ultimately influences if the organ will be utilized or eventually discarded. Our research team has identified three urinary repair biomarkers [uromodulin (UMOD), osteopontin (OPN), and chitinase-3-like protein 1 (YKL-40)] associated with decreased graft failure in DDKT recipients3,4. These biomarkers may have the potential to improve the assessment of kidney quality for renal transplantation5,6. Translation of these biomarkers to clinical practice requires point-of-care measurement that can be readily incorporated into OPO workflows. Existing immuno-assay platforms such as enzyme-linked immunosorbent assay (ELISA) or Meso Scale Discovery (MSD) assays are unsuitable as these techniques typically require specialized training, costly laboratory equipment, and up to four hours of wait time for measurement readout. We addressed these limitations by developing lateral flow devices7 (LFDs) specific to our biomarkers of interest [UMOD, OPN, YKL-40] in collaboration with Global Access Diagnostics (Boston, MA). In contrast to ELISA or MSD assays, LFDs are well-suited for integration into clinical settings, requiring minimal training or laboratory equipment and completing measurements within 20 min. To assess their utility, we collaborated with OPOs who agreed to integrate the LFD into their workflow. The objective of this study was to (a) integrate LFD measurement in the OPO workflow and (b) maintain sample integrity for biomarker measurement at the OPO and future measurements. Results The OPO workflow with specific QC in our protocol is outlined in Fig. 1, augmented by a broader QC checklist in Figure S3. From April 2022 to April 2025, 62 trained staff across the two OPOs measured repair biomarkers for 958 samples from 879 unique donors (Fig. 2). Enrollment trends at OPO #1 remained relatively consistent. Following adjudication by the data coordinating center (DCC) on any discrepancies with times monitored and data quality checks (Fig. 1), 4% of samples (n = 35) were ultimately excluded due to failing sample QC, resulting in 923 samples and 851 unique donors [720 DBD donors, 131 DCD donors]. Data quality framework, which included data quality checks and reporting strategies, Blue boxes represent the primary steps of the UDS overtime in a linear fashion, beginning with OPO sample collection and ending with data validation. Specific dates and times for these steps are monitored via the DCF, REDCap, and the DCC’s Samples Inventory Management System (SIMS). Donor enrollment across Organ Procurement Organizations (OPO) over time, Total OPO donor enrollment over time is represented by the black line, with OPO #1 and OPO #2 enrollment trends delineated by the red and blue lines, respectively. Samples that did not pass QCs or were withdrawn from the study are not included in this figure. Of the 35 excluded samples, 18 were identified through QCs as being left unfrozen beyond acceptable protocol ranges8. 5 additional samples had compromised integrity, 7 samples were collected in violation of protocol, and 2 samples were processed incorrectly or violated protocol. Other exclusions included lack of consent from the participant or legal representative (n = 1) and a missing or compromised data collection form (DCF) (n = 2). Ultimately, 923 samples passed QC. The median time from sample collection to sample processing by OPO personnel was 11 (IQR 3–22) hours (Fig. 3D). The bulk of this is sample storage with a median storage time of 11 (4–23) hours (Fig. 3B). The other two workflow steps, sample collection to storage and sample processing to last LFD measurement, were of relatively shorter duration, with median times of 36 (9–305) minutes and 46 (40–55) minutes, respectively (Figs. 3A and 3C). QC of LFD measurement documented invalid biomarker readings in 5.1% of overall cases (Table 1). Repeated biomarker measurement in 2% (n = 17) samples, occurring five to twenty minutes following the completion of the first biomarker measurement, were excluded from analysis. Quality check measurements performed to ensure sample integrity, This figure represents the quality checks performed to ensure sample integrity during (A) sample collection to storage, (B) Storage to processing, (C) Processing to last LFD measurement and (D) Collection to Processing. Collection to processing time refers to the time of collection of the sample to the time of processing of the sample in the lab. Table 1Quality check report for failed urinary biomarker measurements using lateral flow device.Reader issueTest nameOPO #1OPO #2Invalid reading (%)UMOD9.18.9OPN0.92.6YKL-404.93.6Overall5.15.1Invalid reading rates are determined as a percentage of all readings, since the original lateral flow device may have been re-run, or a new lateral flow device ran instead. Quality check report for failed urinary biomarker measurements using lateral flow device. Invalid reading rates are determined as a percentage of all readings, since the original lateral flow device may have been re-run, or a new lateral flow device ran instead. The implementation of LFDs heavily relied on personnel performance which was enabled by remote personnel training and frequency of their LFD use over time (Fig. 4). Remote personnel training over time. Discussion Current research for diagnostic nephrology uses novel approaches such as artificial intelligence (AI)9, precision medicine11, proteomics, and genomics10. While there are ongoing efforts to integrate these methods into clinical practice, challenges remain in developing standardized protocols and gaining widespread acceptance among clinicians12–14. As a result, the full potential of these innovations has yet to be realized in clinical practice. Our study demonstrates the feasibility of integrating point-of-care protein biomarker measurements into routine OPO practice through LFDs. This work involved detailed protocols and multiple QCs to assess, modify, and improve steps or processes. LFDs offer a cheaper and more accessible alternative to traditional methods of protein measurement while additional strong QC measures ensure the validity and usefulness of LFD biomarker measurements. To our knowledge, there are no previous studies that have developed a quality check framework integrating real-time urinary biomarker measurements into OPO clinical workflow. Our team previously demonstrated in the laboratory that refrigeration at 4 °C before sample processing and freezing the sample at -80 °C after its processing retains the integrity of biomarker measurements8. The existing protocol allows for flexibility in either immediate processing or sample storage at 4 °C before processing and final storage at -80 °C. A major strength of our protocol is derived from its malleability: the time-sensitive nature of organ procurement often presents unexpected complications in managing sample quality and measurement. Data monitoring by the coordinating center and subsequent monthly meetings with each OPO generated a continuous dialogue which readily addressed said issues with protocol adjustments. For instance, normothermic regional perfusion (NRP) was introduced to the donation after circulatory death (DCD) transplant workflow at both OPOs. With the QC framework, sample integrity was not compromised despite the updated process. The prolonged sample processing times captured by monitoring of DCFs, led to the creation of an OPO workgroup. This workgroup discussed re-training requirements for laboratory members and potential improvements to the study protocol. After the development of this work group, sample processing times decreased overall. This demonstrates transplant investigators and OPOs may harmoniously work in tandem to advance translational research. The study protocol was novel in two key ways: 1. LFD measurements would be conducted by OPO staff alongside the organ procurement and allocation process, allowing for biomarker measurement of fresh sample. 2. The multiple QCs included in the data collection forms also served as instructions for conducting accurate LFD measurements and proper storage of samples for downstream laboratory testing. Our study is limited to two OPOs, both of which have a dedicated donor center with an in-house laboratory. While the sample collection protocol may need revisions to accommodate different workflows, the developed check list is broad enough to be applied to non-transplant research studies that interface point-of-care testing with biosample collection. This report showcases how a combination of QCs and improved accessibility through LFDs may pragmatically integrate biomarker measurements into OPO workflows. The success of our framework may benefit other scientific investigators who are starting up clinical studies seeking to collect biological samples for point-of-care testing and are seeking guidance on QC measures to mitigate the risks of a sample compromise in these intricate settings. Our QC platform helped maintain sample integrity which, in turn, improved the accuracy of our repair biomarker measurements. By measuring these biomarkers via point-of-care testing, these results have the potential to supplement current criteria for deceased donor kidney quality assessment and thereby improve organ utilization. Materials and methods We developed a data quality framework (Fig. 1) with multiple QCs across the entire workflow, including OPO sample collection, transportation and storage, processing, sample shipment back to the data coordinating center (DCC), and data transcription. These key data elements were recorded in Data Collection Form (DCF), (Figure S2) allowing us to monitor data discrepancies and sample quality issues. This study was conducted through a partnership between the DCC (Johns Hopkins University) and two OPOs: Gift of Life Michigan [Ann Arbor, MI] and Mid-America Transplant [St. Louis, MO]. The protocol instructs OPOs to obtain urine samples both on-site (donor care unit) and off-site (donor hospital) from DBD and DCD donors. To meet inclusion criteria, participants had to be kidney donors undergoing organ procurement who were at least 18 years of age, had a minimum of 30 mL of urine obtained via an indwelling urinary catheter, and could provide broad research consent via a legally authorized representative. Our study protocol did not change any clinical protocols at the OPOs, nor did it alter how the personnel managed donors prior to organ procurement. All methods were carried out in accordance with relevant guidelines and regulations. The scientific review committees of the participating OPOs (Gift of Life Michigan [Ann Arbor, MI] and Mid-America Transplant [St. Louis, MO]) and the Hopkins institutional review board approved the research protocol. (IRB00172529). Informed consent was obtained from next of kin and family. The study did not involve donors who were minor in age. Sample collection to shipment Sample collection Once donor eligibility was determined, clinical staff at the OPO obtained urine prior to kidney procurement using clinical kits supplied by the DCC (Figure S1A). Clinical staff obtaining urine recorded the sample collection date and time, volume collected, and the source of the sample (on-site vs. off-site) on the DCF for further monitoring by the DCC. As the study progressed, DCD donors required two urine samples: one collection around the time of life support withdrawal followed by emptying the foley bag and a second collection shortly after cross-clamping. With input from OPOs, we later expanded our protocol to include these modifications and revised the DCF accordingly. The quality check required OPO staff to document on DCFs (Figure S2) whether the DCD participant’s Foley bag was emptied immediately after collection of the first sample. Samples where the foley bag was not emptied immediately were discarded. Sample transport/Storage Each OPO had an on-site laboratory wherein urine samples were transported to the laboratory at ambient temperature within 30 min of urine collection. This limited the duration for which the urine samples were left at room temperature, thus limiting the number of compromised samples8. Samples left at room temperature for more than 4 h were identified in the DCF and subsequently excluded. Urine samples were then stored in a 4 °C refrigerator for processing at a permissible time or handed to laboratory personnel for immediate processing. Off-site sample collection required the personnel to transport the samples on ice or gel packs and indicate its use in the DCF for the inclusion of the sample in the study. Sample storage personnel recorded storage date/time – providing another time point to aid in the DCC’s identification of protocol deviations. The DCC met monthly with each OPO to discuss potential points of improvement for decreasing the time from urine sample collection to storage. Processing of samples All urine samples were linked to a United Network for Organ Sharing (UNOS) ID and assigned a specific UDS ID found within the laboratory kit (Figure S1B). Collected urine was then split into two 15 mL labeled blue conical tubes: one uncentrifuged tube for urinalysis and the other centrifuged at 2000 g for 12 min for LFD biomarker measurements. Centrifuged urine was also decanted into a 15 mL red conical tube and transferred to the LFD kits for testing of UMOD, OPN, and YKL-40 biomarkers via a transfer pipette and dilution vial (Figure S1C). Details such as processing date and times of centrifugation, urinalysis, were also recorded in the DCF. Personnel initials for each step of this process were recorded in DCF for stringent monitoring. The DCC provided remote training for laboratory personnel at each OPO, in addition to an annual refresher course, to aid in sample processing, streamlining workflow, and troubleshooting both the assay and data reader. Recording the frequency of LFD measurements performed by each trained laboratory staff member, as well as biomarker readings associated with each personnel, ensured that potential errors could be swiftly identified within specific staff for individualized training. Urinalysis Uncentrifuged urine underwent urinalysis using the Siemens CLINITEK Status® + Analyzer and Siemens Multistix® 10 SG Reagent Strips. Laboratory personnel were asked to attach the paper urinalysis report to the DCF to confirm that the test date/time and uncentrifuged urinalysis measurements aligned with the operator’s transcribed values. This secondary check using the analyzer’s printed time, performed by the DCC, helped detect transcription errors. LFD biomarker measurements Once the diluted urine was placed on each LFD’s sample port, a twenty-minute timer was set. Once time elapsed, the operator read the biomarker results within five minutes using the Detekt RDS-2500 Reader. If the results could not be read within five minutes, the assay was repeated and documented in the DCF. Biomarker measurements were recorded on both the DCF and Detekt’s Cloud Data Manager, a secure platform that allowed the DCC access to measurements over any internet connection. The cloud recorded every assay measurement performed, making it an excellent resource for identifying errors and validating data in conjunction with the DCF. The DCF was critical in manually documenting instances of invalid readings due to incorrect LFD positioning, lack of control line development, out of frame test/control line development, and the operator’s resolution [assay repeated with new LFD or re-run with original LFD]. Detekt’s cloud included the exact date and time assays were run with the reader, which enabled the DCC to develop QC reports identifying re-runs which deviated from established protocol and were thus excluded from future analyses. Samples frozen at -80 °C & shipped to DCC Once all biomarker results were read by the Detekt reader, the 15 mL red conical tube containing sample supernatant was placed in a freezer box and frozen at -80 °C, and shipped back to the DCC. Sample storage personnel initials, sample freezing date/time, and sample location within the freezer box were recorded in the DCF to facilitate easier retrieval of samples. Data transcription Data from the DCF was transcribed into the Research Electronic Data Capture (REDCap) Platform – a secure, online system allowing authorized users from the DCC to access data, generate reports, and record discrepancies or data quality rules violated. Two operator checks for data transcription were performed by the DCC’s research coordinator and program manager. Any QC rules potentially violated were automatically flagged and generated a query within the REDCap system. Queries, errors, and missing information were documented by the DCC and resolved through regular communication with the OPOs. If critical information could not be obtained that may have compromised sample integrity, the affected sample was excluded from future analyses. The DCC’s biostatistician then used the DCF values entered in the REDCap to generate an OPO report summarizing any data outliers or discrepancies. These points would then be brought to the OPOs for further clarification during monthly discussions. Supplementary Information Supplementary Information. Acknowledgements This manuscript, a result of funding in whole or in part by the National Institutes of Health (NIH) awarded to Dr Chirag Parikh (R01DK-93770). Through acceptance of this federal funding, NIH has been given a right to make this manuscript publicly available in PubMed Central upon the Official Date of Publication, as defined by NIH. The content is solely the responsibility of the authors and does not necessarily represent the official views of NIH. Dr. Peter Reese, director of Vanderbilt University’s Center for Transplant Science and one of the collaborators for the Urine Dipstick Study, was instrumental in smooth functioning of this project through his valuable suggestions and technical inputs. A part of the work was presented as a poster at the World Transplant Congress, 2025. Author contributions Conceptualization : CP, WO, DH; Data curation : DH, HTP, MB, WO Formal analysis : HTP, DH, WO; Investigation : MG, EG, GM, NO, WW, WH, MD Funding acquisition, Project administration, Resources, Software : CP Supervision : CP, WO, NO, GM; Validation : WO, DH; Visualization : DH Writing – original draft: MB, SD, HTP Writing – review & editing: All authors. Data availability All data generated or analysed during this study are included in this published article (and its Supplementary Information files). Declarations Competing interests The authors declare no competing interests. Author disclosure statement MO, SD, WO, DH, MD, SX, MG, EG, GM, WW, WH, HT have no disclosures. SM is a consultant for Sanofi, Visterra, Takeda and Deputy Editor for KI reports; NO at GOLM received travel support from International Institute for Advancement of Medicine) for the AOPO conference; CP has received funding support from NIH/NIDDK (R01DK-93770). Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. These authors are jointly supervised to this work: Mitali Barik and Serena DSouza. References Associated Data Supplementary Materials Supplementary Information. Data Availability Statement All data generated or analysed during this study are included in this published article (and its Supplementary Information files).