Point-of-care detection of fibrosis in liver transplant surgery using near-infrared spectroscopy and machine learning.

Introduction: Visual assessment and imaging of the donor liver are inaccurate in predicting fibrosis and remain surrogates for histopathology. We demonstrate that 3-s scans using a handheld near-infrared-spectroscopy (NIRS) instrument can identify and quantify fibrosis in fresh human liver samples. Methods: We undertook NIRS scans on 107 samples from 27 patients, 88 from 23 patients with liver disease, and 19 from four organ donors. Results: Liver disease patients had a median immature fibrosis of 40% (interquartile range [IQR] 20-60) and mature fibrosis of 30% (10%-50%) on histopathology. The organ donor livers had a median fibrosis (both mature and immature) of 10% (IQR 5%-15%). Using machine learning, this study detected presence of cirrhosis and METAVIR grade of fibrosis with a classification accuracy of 96.3% and 97.2%, precision of 96.3% and 97.0%, recall of 96.3% and 97.2%, specificity of 95.4% and 98.0% and area under receiver operator curve of 0.977 and 0.999, respectively. Using partial-least square regression machine learning, this study predicted the percentage of both immature (R 2 = 0.842) and mature (R 2 = 0.837) with a low margin of error (root mean square of error of 9.76% and 7.96%, respectively). Conclusion: This study demonstrates that a point-of-care NIRS instrument can accurately detect, quantify and classify liver fibrosis using machine learning.

Australian Donation and Transplantation Biobank: A Research Biobank Integrated Within a Deceased Organ and Tissue Donation Program.

We aimed to facilitate the donation of tissue samples for research by establishing a centralized system integrated in the organ donation program for collection, storage, and distribution of samples (the Australian Donation and Transplantation Biobank [ADTB]). Methods: Feasibility of a research biobank integrated within the deceased organ and tissue donation program was assessed. DonateLife Victoria sought consent for ADTB donation after consent was received for organ donation for transplantation from the donor's senior available next of kin. ADTB samples were collected during donation surgery and distributed fresh to researchers or stored for future research. The main outcome measures were ADTB donation rates, ADTB sample collection, ADTB sample use, and to identify ethical considerations. Results: Over 2 y, samples were collected for the ADTB from 69 donors (28% of 249 donors). Samples were obtained from the spleen (n = 59, 86%), colon (n = 57, 83%), ileum (n = 56, 82%), duodenum (n = 55, 80%), blood (n = 55, 80%), bone marrow (n = 55, 80%), skin (n = 54, 78%), mesenteric lymph nodes (n = 56, 81%), liver (n = 21, 30%), lung (n = 29, 42%), and lung-draining lymph node (n = 29, 42%). Heart (n = 20), breast (n = 1), and lower urinary tract (n = 1) samples were obtained in the second year. Five hundred fifty-six samples were used in 19 ethics-approved research projects spanning the fields of immunology, microbiology, oncology, anatomy, physiology, and surgery. Conclusions: The integration of routine deceased donation and transplantation activities with a coordinated system for retrieval and allocation of donor samples for use in a range of research projects is feasible and valuable.

Low Levels of Hepatocyte-Specific Methylation in Cell-Free DNA Are a Strong Negative Predictor for Acute T Cell-Mediated Rejection Requiring Treatment Following Liver Transplantation.

Graft-derived cell-free DNA (gdcfDNA) quantification is a promising, minimally invasive tool for detecting acute T cell-mediated rejection (ATCMR) following liver transplantation (LT). We investigated the utility of measuring hepatocyte-specific methylation in cfDNA (HS-cfDNA) to quantify gdcfDNA, examining its accuracy in detecting ATCMR in a prospective, cross-sectional study. Blood was collected from LT recipients immediately prior to graft biopsy for suspected rejection. HS-cfDNA was quantified using droplet-digital polymerase chain reaction. Prebiopsy liver function tests (LFTs) and HS-cfDNA levels were correlated with biopsy results and the primary outcome of treated biopsy-proven acute rejection (tBPAR). A total of 51 patients were recruited; 37 had evidence of rejection on biopsy and 20 required treatment. As much as 11 patients needed inpatient treatment for rejection. HS-cfDNA significantly outperformed LFTs in identifying patients with tBPAR, particularly those needing inpatient treatment (area under the curve, 73.0%; 95% confidence interval, 55.4%-90.6%; P = 0.01). At a threshold of <33.5% of the total cfDNA fraction, HS-cfDNA had a specificity of 97%, correctly excluding tBPAR in 30/31 patients. Quantifying graft-specific methylation in cfDNA has a major advantage over previous gdcfDNA techniques: it does not require genotyping/sequencing, lending it greater feasibility for translation into transplantation care. Low levels of HS-cfDNA were a strong negative predictor for tBPAR (negative predictive value, 86%) and may have a future role in triaging patients prior to invasive graft biopsies.

The Measurement of Donor-Specific Cell-Free DNA Identifies Recipients With Biopsy-Proven Acute Rejection Requiring Treatment After Liver Transplantation.

BACKGROUND: Assessment of donor-specific cell-free DNA (dscfDNA) in the recipient is emerging as a noninvasive biomarker of organ rejection after transplantation. We previously developed a digital polymerase chain reaction (PCR)-based approach that readily measures dscfDNA within clinically relevant turnaround times. Using this approach, we characterized the dynamics and evaluated the clinical utility of dscfDNA after liver transplantation (LT). METHODS: Deletion/insertion polymorphisms were used to distinguish donor-specific DNA from recipient-specific DNA. Posttransplant dscfDNA was measured in the plasma of the recipients. In the longitudinal cohort, dscfDNA was serially measured at days 3, 7, 14, 28, and 42 in 20 recipients. In the cross-sectional cohort, dscfDNA was measured in 4 clinically stable recipients (>1-y posttransplant) and 16 recipients (>1-mo posttransplant) who were undergoing liver biopsies. RESULTS: Recipients who underwent LT without complications demonstrated an exponential decline in dscfDNA. Median levels at days 3, 7, 14, 28, and 42 were 1936, 1015, 247, 90, and 66 copies/mL, respectively. dscfDNA was higher in recipients with treated biopsy-proven acute rejection (tBPAR) when compared to those without. The area under the receiver operator characteristic curve of dscfDNA was higher than that of routine liver function tests for tBPAR (dscfDNA: 98.8% with 95% confidence interval, 95.8%-100%; alanine aminotransferase: 85.7%; alkaline phosphatase: 66.4%; gamma-glutamyl transferase: 80.1%; and bilirubin: 35.4%). CONCLUSIONS: dscfDNA as measured by probe-free droplet digital PCR methodology was reflective of organ health after LT. Our findings demonstrate the potential utility of dscfDNA as a diagnostic tool of tBPAR.