Donor-Derived Cell-Free DNA Assessment in a Predominantly Hispanic/Latino Lung Transplant Population: Single-Center Analysis

Lara Jones ¹, David J. Ross ^2,*, Andrew Doud ², Navchetan Kaur ², Sangeeta Bhorade ², Holly Keyt ¹

1. University of Texas Health Sciences Center, San Antonio, TX , USA

2. Natera, Inc. Austin, TX , USA

* Correspondence: David J. Ross

Academic Editor: Kamyar Afshar

Received: March 31, 2025 | Accepted: July 14, 2025 | Published: July 21, 2025

OBM Transplantation 2025, Volume 9, Issue 3, doi:10.21926/obm.transplant.2503254

Recommended citation: Jones L, Ross DJ, Doud A, Kaur N, Bhorade S, Keyt H. Donor-Derived Cell-Free DNA Assessment in a Predominantly Hispanic/Latino Lung Transplant Population: Single-Center Analysis. OBM Transplantation 2025; 9(3): 254; doi:10.21926/obm.transplant.2503254.

© 2025 by the authors. This is an open access article distributed under the conditions of the Creative Commons by Attribution License, which permits unrestricted use, distribution, and reproduction in any medium or format, provided the original work is correctly cited.

1. Introduction

Lung transplantation (LT) is a potentially life-saving option for patients with myriad end-stage cardiopulmonary diseases. In 2022, nearly 3200 patients were added and approximately 4200 patients were actively wait listed for LT with the United Network Organ Sharing (UNOS) [1]. The demographics of LT candidates in the United States are evolving with an increasing representation of self-identified racial/ethnic minority groups, recently reported as 4.8% Asian, 11.3% Black, 14.7% Hispanic/Latino, and 68.4% White [1]. U.S. Congress recently requested that the National Institutes of Health (NIH) sponsor a National Academics of Sciences, Engineering, and Medicine study to review equity in the US organ transplant system. Marked variation and inconsistency in performance was observed, with demonstrable inequity in the dimensions of race and ethnicity, geographic location, and socioeconomic position [2]. In line with these findings, reports of U.S. trends in survival after lung transplant have consistently shown differences across race and ethnicity [1,3]. Although socioeconomic and geographic differences are frequently cited in the construct of an area deprivation index with less access to healthcare, the contributing factors to lung transplant recipient (LTR) outcomes are unarguably quite complex across the intersection of race, age, sex, and primary lung diagnosis [2,4,5].

Herein, we sought to analyze, from our single-center experience, potential differences in status of allograft health due to racial/ethnic disparity in a self-described predominantly Hispanic/Latino (HL) lung transplant patient population compared to Non-Latino/White (NLW). The fraction of donor-derived cell-free DNA (dd-cfDNA%) has been validated for detection of acute rejection after LT and represents a noninvasive plasma biomarker of allograft injury [6,7,8]. We hypothesized that recipient minority race/ethnicity status may be associated with levels of allograft injury, measurable using dd-cfDNA, allowing us to assess for potential bias in treatment status in this Hispanic/Latino LT cohort.

2. Methods

2.1 Study Design

Dd-cfDNA% had been implemented on protocol and as standard-of-practice (SOP) at the University of Texas Health Sciences Center, San Antonio (UTHSCSA) Lung Transplant Program in July 2022. This retrospective analysis of deidentified data was approved by the UTHSCSA Institutional Review Board (STUDY00000109). We analyzed electronic medical record (EMR) data for LTR demographics including patient self-reported race/ethnicity [9,10]. The patient demographics at UTHSCSA for UNOS wait list registered patients (2023) demonstrated 42.4% White, 6.1% African American, 42.4% Hispanic/Latino, 3.0% Asian, and 6.1% “Other” [11]. The result of dd-cfDNA% testing was compared for the UTHSCSA HL vs NLW LTR cohorts. The study was conducted according to the principles of the Declaration of Helsinki and the Declaration of Istanbul.

2.2 dd-cfDNA% Testing Protocol

The dd-cfDNA% surveillance protocol was developed to mirror that previously described by NHLBI in Genomic Research Alliance for Transplantation (GRAfT) ⎯ testing performed approximately monthly during the initial year and every 3 months thereafter or as adjusted by the treating provider [12]. Protocol flexible bronchoscopy with bronchoalveolar lavage (BAL) and trans-bronchial biopsies (TBB) are performed at UTHSCSA as SOP at 1, 2, 3, 4, 6-months post-transplant. The protocol procedures may be omitted at months 1 and 4 at the discretion of the provider in the setting of a “low risk” dd-cfDNA% result along with other SOP clinical assessments. Peripheral blood for dd-cfDNA% testing (The Prospera™ Lung test) was collected in 10-mL cell-free Streck™ tubes and sent overnight at room temperature to the Natera, Inc. (Austin, TX) Clinical Laboratory Improvement Amendments (CLIA)-certified, College of American Pathologists (CAP)-accredited laboratory in San Carlos, CA. After extraction, cell-free DNA (cfDNA) was amplified using a massively multiplexed polymerase chain reaction assay targeting a curated panel of >13,000 single-nucleotide polymorphisms designed to maximize variant frequency across ethnicities [13]. For each sample, amplicons were sequenced by using the Illumina™ NovaSeq™ sequencer with an average of 14 to 15 million reads per sample. Sequencing data was processed to estimate the fraction of donor-derived cfDNA (expressed as a percentage; dd-cfDNA%). The Prospera™ test for single LT patients incorporates a correction factor (2×) for resulted dd-cfDNA% [14].

2.3 Analysis

Analysis of dd-cfDNA% levels by race/ethnicity, sex (M/F), time (years) post-LT, single (SLT) vs double lung transplant (DLT) recipient, and proportion of “High Risk” test results (defined as dd-cfDNA ≥1.0%) were compared by Chi-square or Kruskal-Wallis tests with Benjamini-Hochberg to adjust p-values (p < 0.05). To assess intra-patient variability for dd-cfDNA, we analyzed robust coefficient of variation (CV_i%) for patients with ≥3 longitudinal tests performed. Data is expressed as median (25-75% interquartile range, IQR).

3. Results

Between September 2022 and September 2024, the UTHSCSA data set included 245 LTR with N = 1,149 dd-cfDNA results (Range: 1-10 tests/patient). In Table 1, demographics demonstrated that median age was younger for HL but there were no differences for the distributions of sex or type of LT procedure. In Figure 1, the proportion of dd-cfDNA% samples obtained during the first-year post-transplant was higher for HL (45.3%) than NLW (34.3%, p = 0.044), and the proportion of dd-cfDNA% samples obtained >10 years post-transplant was higher for NLW (12.6%) than HL (3.7%, p < 0.0001). No significant differences were observed in the number of samples drawn 1-5 years (p = 0.341) and 6-10 years (p = 0.096) post-transplant. In Table 1 and depicted graphically in Figure 2, when analyzing for “High Risk” dd-cfDNA% results of HL vs NLW, there was a lower proportion of patients with suspected allograft injury in the HL cohort − 33.4% vs 39.6% (p = 0.0079), respectively. Similarly, there was a lesser proportion of “High Risk” results in HL (29.1%) than for NLW (47.9%) for the single lung transplant cohort (p = 7.45 × 10^-8). Analysis of CV_i% for 75 patients with N = 661 longitudinal samples (from 1-month to 3-years post-LT) obtained for HL vs NLW, showed no significant difference in median intra-patient variability− 0.379 (IQR: 0.193-0.540) vs 0.408% (0.254-0.503, p = 0.087), respectively.

Table 1 Cohort Demographics.

Figure 1 Sample frequency for HL (green) and NLW (blue) by time after lung transplantation. The proportion of dd-cfDNA% samples obtained during the first year was higher for HL (45.3%) than NLW (34.3%, p = 0.044), but no differences were observed during the later period of 1-5 years and 6-10 years.

Figure 2 Violin plots for dd-cfDNA% on log₁₀ scale for all samples, Non-Latino White (NLW) samples, and Hispanic/Latino (HL) lung transplant patient samples. The proportion of samples with “High Risk” dd-cfDNA% result (green) was less for the HL (33.4%) vs NLW (39.6%) race/ethnicity cohort (p = 0.0079).

4. Discussion

In our study, we endeavored to assess whether any potential racial/ethnic disparity would be reflected in the fraction of donor-derived cell-free DNA, a noninvasive plasma biomarker of allograft injury [6,7]. Our analysis of a Hispanic/Latino LT population demonstrated several key findings. First, a significantly higher proportion of dd-cfDNA% tests were performed during the initial year in the HL (45.3%) than NLW (34.3%) cohort, and a significantly higher proportion of dd-cfDNA% tests were performed >10 years in the NLW (12.6%) than HL (3.7%) cohort. The focus of post-transplant care is to mitigate risks of acute rejection, infection, and allograft dysfunction through vigilance and personalized medicine. Although our surveillance protocol had been developed with the intent of uniform implementation, we questioned whether the higher proportion of early testing in the HL population may represent an element of provider bias. We speculate that increased travel distance from the transplant center, which is typical for the HL cohort, may have led to more frequent early dd-cfDNA% testing by remote phlebotomy. Counterintuitively, this may have led to better outcomes, contradicting past studies that have shown an increase in travel distance being associated with worse outcomes [15]. Regardless, this does not support the concern that HL patients may not receive equivalent care to NLW patients. Second, despite more frequent testing during the first year in HL, there was no detected difference in the median %dd-cfDNA between race/ethnic cohorts when matched for time after LT. Third, we hypothesized that an increased proportion of “High Risk” dd-cfDNA% results may be evident due to potential inherent differences or disparity in healthcare; however to the contrary, the HL cohort was characterized by a lower proportion of high-risk results (33.4%) than for NLW (39.6%). This could be, in part, due to the younger median age of the HL cohort (64) as compared to the NLW cohort (69). Finally, we analyzed for potential differences in intra-patient variability for dd-cfDNA% results as stratified by our race/ethnicity cohorts. An association between dd-cfDNA% variability with mortality and development of de novo HLA donor-specific antibodies was previously described in heart transplant recipients [14]. In our HL population analysis, however, no difference was detected in median intra-patient robust coefficient of variation for dd-cfDNA%, compared to the NLW cohort.

The limitations of our current study are several, and principally attributed to the limited data available for analysis. As the study was designed as a Quality Assurance and Performance Improvement (QAPI) Initiative investigation of potential inequity based on race/ethnicity, it did not include granular patient clinical and outcomes data. dd-cfDNA% test results were leveraged to thereby assess for inequity based on frequency of biomarker testing and evidence of disparate population results reflecting allograft injury. Additionally, our study would have been enhanced by performing multiple comparisons across different minority racial/ethnic groups; however, the small sample sizes prohibited this analysis. Further, potential errors could be introduced by the demographic designations for patient self-reported race/ethnicity in the EMR. Notwithstanding, to our knowledge this represents the first study in LT that endeavors to examine dd-cfDNA% across different races/ethnicity groups. In our single-center experience and “birds eye view”, we attempted to analyze for potential disparity attributable to minority race/ethnicity in a predominantly Hispanic/Latino LT population. Although these findings must be considered in the context of the study design limitations, these data suggested similar time-dependent levels in a blood-based injury biomarker and further supported the clinical utility of dd-cfDNA% in the Hispanic/Latino LTR population. Further analyses are forthcoming to compare appropriately matched race/ethnicity cohorts with an increased patient-centric granular data with longitudinal clinical outcomes.

Abbreviations

Author Contributions

Study design: LJ, DR, HK, SB; Sample collection, Data entry: LJ, DR, HK. Data analysis: AD, NK, LJ, HK, DR. Manuscript development: DR, LJ, HK, NK. SB. All authors reviewed and approved the final manuscript.

Competing Interests

DJR, AD, NK and SB are employees of Natera, and may hold stock in Natera. LJ and HK have nothing to disclose.