A Time-to-Equivalent Risk (TiTER) for decision-making in deceased donor kidney transplant candidates
Article information
Abstract
Background
A United States study proposed the concept of Time-to-Equivalent Risk (TiTER) as a decision-making tool for accepting organs from expanded-criteria donors. While this study demonstrated a higher initial cumulative incidence of deceased donor kidney transplantation (DDKT) than death, the applicability and pattern of TiTER may differ in regions with extreme organ scarcity, such as Korea.
Methods
Using comprehensive, nationwide cohort data from the Korean Network for Organ Sharing registry and National Health Insurance (2008–2022), we analyzed the cumulative incidence and competing risks of DDKT versus waitlist mortality, estimating the Korean TiTER (K-TiTER). Analyses were stratified by recipient age, Korean Estimated Post-Transplant Survival (K-EPTS), blood type, and donor quality.
Results
Among 51,427 waitlisted patients, 10,603 (20.6%) received DDKT. The cumulative incidence of DDKT was initially lower than that of death, in the opposite pattern to the United States study. Therefore, K-TiTER was defined as the earliest time when DDKT probability exceeds waitlist mortality. Among candidates aged ≥65 years, those with high K-EPTS scores (75–100), or blood type O, K-TiTER was frequently not reached, as the waitlist mortality consistently exceeded transplant probability. Notably, sensitivity analyses showed that expanding donor acceptance criteria shortened the time to probability crossover. This finding reframes suboptimal donor organs as a strategic accelerator for mitigating waitlist mortality, especially in candidates with poor prognosis.
Conclusion
This descriptive study proposed preliminary hypotheses for the K-TiTER framework, indicating its potential to guide the identification of vulnerable candidates and support tailored allocation strategies or enrollment in innovative clinical trials for high-risk candidates.
Introduction
Kidney transplantation remains the optimal treatment for end-stage kidney disease (ESKD), offering superior survival and quality of life compared to maintaining dialysis therapy [1]. However, a chronic imbalance between organ supply and demand continues to challenge modern transplantation systems. As of 2025, more than 100,000 patients remain on the kidney transplant waitlist in the United States, with an additional 20,000 to 30,000 ESKD patients newly added to the waitlist annually [2,3].
Across Asian countries, deceased donor organ donation rates remain markedly lower than those in Western countries despite a comparatively higher incidence of ESKD [4]. This persistent supply-demand disparity translates into prolonged waiting times for deceased donor kidney transplantation (DDKT) in Asian countries [5,6], including Korea, often exceeding 5 to 10 years. Consequently, candidates face an increased risk of waitlist mortality before transplantation. This gap underscores the urgent need for a data-driven allocation framework that balances equitable access with optimal organ utility.
The 2014 United States Kidney Allocation System (KAS) adopted “longevity matching” between donor kidney quality (Kidney Donor Profile Index, KDPI) and recipient prognosis (Estimated Post-Transplant Survival, EPTS) to maximize utility, while awarding priority points to highly sensitized candidates based on calculated panel reactive antibody (PRA) levels to promote equity [7]. Multiple subsequent studies have demonstrated that the 2014 KAS successfully promoted both efficient organ utilization and equitable access to transplantation through a single, integrated framework [8–11]. Comparable equity-utility balancing models have been adopted in Eurotransplant, the United Kingdom, and Australia/New Zealand, incorporating sensitization-aware or age-related matching algorithms that reduce disparities in wait times without diminishing transplant outcomes [12,13].
In parallel, the incorporation of longevity matching using Korean version of donor and recipient risk scores (K-KDPI and K-EPTS, respectively) has provided a complementary efficiency mechanism. Retrospective analysis demonstrated that matching lower-risk kidneys to recipients with superior expected survival had better long-term outcomes, consistent with international experience. Moreover, for low K-EPTS recipients, selective acceptance of higher-risk (expanded-criteria) donor kidneys curtails waiting time without impairing patient and graft survival [14]. This suggests that carefully expanded acceptance criteria can simultaneously alleviate waitlist pressure and preserve clinical utility. Because the expansion of deceased donation remains stagnant, policy interventions that enhance the probability of timely transplantation (thereby reducing waiting time) or selectively prioritize candidates at high risk of waitlist mortality through calibrated access to expanded-criteria kidneys have become pivotal for reducing waitlist mortality.
The Time-to-Equivalent Risk (TiTER) framework offers a transparent method to quantify when the probability of receiving DDKT is overtaken by the risk of death under competing risk dynamics [15]. Operationally, TiTER is implemented by estimating cumulative incidence functions (CIFs) for both events and identifying their intersection with confidence intervals derived from intersecting CIF confidence bands. In the United States analyses, TiTER shifted intuitively with donor-side interventions: for older candidates, accepting a broader range of donor quality (KDPI 0%–100%) including suboptimal donors, prolonged the time until mortality outpaced transplant probability compared to restricting to lower-risk kidneys (e.g., KDPI <50%), with similar patterns observed using donor age as the risk indicator. These findings illustrate how TiTER can connect acceptance strategies to real-world timelines, in which transplantation remains more likely than death on the waitlist.
Despite its relevance in extended waiting time settings, no study has quantified TiTER for Korean candidates. To address this gap, we developed the Korean TiTER (K-TiTER), which is defined as the earliest post-listing time at which the cumulative incidence of DDKT equals that of waitlist mortality. Unlike the United States context, which is characterized by shorter waiting times and where TiTER identifies the point at which mortality begins to outweigh the probability of transplantation, K-TiTER identifies the critical juncture at which the probability of DDKT finally exceeds the high initial risk of waitlist mortality in the Korean setting. In a system characterized by extreme organ scarcity, this crossover point serves as a pivotal clinical metric for reassessing acceptance strategies for expanded-criteria donors (ECD) or prioritizing candidates for emerging therapeutic trials such as xenotransplantation, tailored to Korea’s persistently limited organ supply.
Methods
Data source
We retrospectively analyzed a nationwide cohort using data from the Korean Network for Organ Sharing (KONOS) and customized research datasets from the National Health Insurance Data Sharing Service (NHISS). The dataset included comprehensive inpatient and outpatient medical claims, demographic data, clinical data, diagnosis codes based on the International Classification of Diseases, 10th Revision (ICD-10), and treatment procedure codes.
Study design and population
This nationwide study evaluated the competing risks of DDKT and waitlist mortality among patients listed for DDKT in Korea between January 1, 2008 and December 31, 2022. The study population consisted of adult patients (aged ≥19 years) with ESKD who were registered on the solitary kidney transplant waitlist after 2008. Between 2008 and 2022, 53,650 patients were listed for DDKT. We excluded 372 aged <19 years and 1,851 candidates with invalid or missing data regarding diabetes status, viral status, serum creatinine level, or dialysis duration. The final study cohort comprised 51,427 adult candidates, of whom 10,603 received DDKT and 40,824 remained waitlisted (Fig. 1).
Study profiles of patients on the waiting list between 2008 and 2022.
DDKT, deceased donor kidney transplantation.
Individuals who did not experience the primary outcomes (DDKT or waitlist mortality) were censored at the administrative end of follow-up (December 31, 2022) or at the time of removal from the waitlist for other reasons. Censoring events included living donor kidney transplantation, delisting due to medical unsuitability, emigration, or withdrawal, whichever occurred first. The exact dates of death were meticulously ascertained by linking the KONOS registry and NHISS with the mortality database of the Ministry of the Interior and Safety. The DDKT recipients were identified from the KONOS registry and linked to corresponding NHISS claims data. For each candidate, the initial listing date served as the index date (time zero).
Baseline demographic and clinical characteristics were derived from the KONOS registry and NHISS claims data, including age, sex, comorbidities (diabetes, hypertension, cerebrovascular events), dialysis duration, viral serostatus, ABO blood types, and PRA sensitization. Recipient risk profiles were quantified using the Korean Estimated Post-Transplant Survival (K-EPTS) score [14,16]. We defined a negative PRA as having a value of 0% for both PRA class I and class II. Conversely, we defined a positive PRA as a case where either class I or class II showed a PRA value greater than 0%. The K-EPTS score was developed and calculated using four recipient factors: recipient age, presence of diabetes mellitus, hepatitis C virus (HCV) status, and duration of dialysis prior to transplantation [16]. Utilizing weighting coefficients specifically derived from Korean national data to accurately reflect the posttransplant survival characteristics of the Korean population. It differs from the existing United States EPTS. While both scoring systems include recipient age, diabetes, and dialysis duration, the United States EPTS includes “prior solid organ transplant” as its fourth variable, whereas the K-EPTS includes “HCV status.” Donor characteristics included age, sex, body mass index, cause of death, and the Korean Kidney Donor Profile Index (K-KDPI) [14,17]. The K-KDPI score was developed using five donor factors: age, height, presence of diabetes mellitus, serum creatinine levels, and HCV status [17].
Outcomes
The primary events were (1) receiving DDKT, and (2) waitlist mortality. Given that these events are mutually exclusive, analyses were conducted using a competing risk framework. The CIFs for DDKT and waitlist mortality were estimated nonparametrically using the Aalen-Johansen estimator to account for competing risks and evaluate the outcomes over the study period [18]. Given the competing risk framework, a unified risk set was used to calculate the cumulative incidences for both DDKT and waitlist mortality. Candidates who had experienced neither event were censored. In addition to the overall estimates, we calculated the 12-year cumulative incidence of DDKT and waitlist mortality stratified by clinically relevant strata (age, sex, PRA, dialysis duration, blood type, K-EPTS category, diabetes, hypertension, and transplant history).
Korean Time-to-Equivalent Risk
The primary analytical concept was TiTER, defined as the time (in months) from waitlist registration to the point where the cumulative incidence of receiving a DDKT equals that of waitlist mortality [15]. Conceptually, in the Korean context of prolonged waiting, the K-TiTER reflects the juncture beyond which the probability of transplantation outweighs the initial risk of waitlist mortality, representing a threshold of ascending transplant opportunity. The K-TiTER was used as the principal decision-support value. K-TiTER was defined as the earliest time point t >0 after waitlist registration at which the cumulative incidence of DDKT exceeds or equals that of waitlist mortality: K-TiTER = min {t > 0, CIFDDKT(t) ≥ CIFDeath(t)}. If the cumulative incidence of DDKT exceeded the waitlist mortality risk from the waitlist placement, the K-TiTER was defined as 0. If the waitlist mortality risk remained dominant and the two curves did not intersect within the observation period, K-TiTER was reported as “not reached.” The K-TiTER crossover point was operationally identified by evaluating the curves at all observed discrete event times. When the difference between the CIFs changed sign, linear interpolation was applied between the adjacent time points to calculate the exact crossover time. In cases of multiple crossings, the earliest crossing time was selected. To quantify uncertainty, 95% confidence intervals were estimated using the bootstrap percentile method with 1,000 resamples. In each resampled dataset, the CIFs were recalculated, and the corresponding crossover times were identified. The follow-up duration was measured in months from the initial KONOS registration to DDKT, death, or censoring, whichever occurred first. We additionally estimated the conditional K-TiTER stratified by (1) candidates’ age, (2) diabetes mellitus, (3) K-EPTS score, (4) blood type, and (5) PRA sensitization.
Statistical analysis
Continuous variables are presented as mean ± standard deviation or median (interquartile range), and categorical variables are presented as frequencies and percentages. Group comparisons between transplanted and waitlisted cohorts were performed using the Student t tests or Wilcoxon rank-sum tests for continuous variables and chi-square tests for categorical variables. We censored living donor kidney transplantation events, recognizing that this decision may introduce non-random censoring in the population, as our intent was to evaluate prognoses specifically for DDKT candidates.
Statistical analyses were performed using SAS EG version 7.1 (SAS Institute Inc.), and R version 4.0.3 in RStudio (RStudio Team, PBC). The manuscript adhered to the STROBE (STrengthening the Reporting of OBservational studies in Epidemiology) guidelines [19].
Ethics statements
This study was approved by the Institutional Review Board of Severance Hospital (No. 4-2021-1358). The requirement for informed consent was waived because only deidentified data were retrospectively analyzed. This study was conducted in accordance with the Declaration of Helsinki [20] and the Declaration of Istanbul on Organ Trafficking and Transplant Tourism [21].
Results
Baseline characteristics
The demographic and clinical characteristics at the time of listing are summarized in Table 1. Between 2008 and 2022, 51,427 adult candidates were listed for DDKT. During the follow-up period, 10,603 waitlisted patients (20.6%) underwent DDKT, whereas 40,824 (79.4%) remained on the waitlist. DDKT recipients were younger than waitlisted candidates (mean age, 50.6 years vs. 57.0 years; p < 0.001). Male predominance was observed in both groups, although the proportion was slightly lower in DDKT recipients (61.3% vs. 63.3%, p < 0.001). Distinct patterns of blood type distribution emerged. The proportion of candidates with blood type O was lower among DDKT recipients (23.7% vs. 28.0%), whereas those with blood type AB comprised a larger proportion (14.8% vs. 10.5%). Comorbidities were highly prevalent, with the prevalence of diabetes mellitus and hypertension exceeding 80% in both groups. DDKT recipients showed a lower prevalence of PRA positivity (20.5% vs. 29.9%, p < 0.001) and a substantially lower mean K-EPTS score than waitlisted patients (48.3 vs. 65.8, p < 0.001).
Baseline demographic and clinical characteristics of the study population
In the DDKT group, the mean donor age was 46.8 years, and the mean K-KDPI—a measure of donor organ quality—was 60.7. Cerebrovascular accidents were the leading cause of donor death (43.9%). The mean cold-ischemic time was 268.8 minutes, with an average of 3.7 human leukocyte antigen mismatches. Regarding outcomes, DDKT recipients experienced a significantly lower crude mortality rate during the study period than waitlisted patients (14.0% vs. 24.8%, p < 0.001). Overall, death-censored graft failure occurred in 7.1% of recipients.
Cumulative incidence of deceased donor kidney transplantation and waitlist mortality
Table 2 summarizes the 12-year cumulative incidence of DDKT and waitlist mortality stratified by key baseline characteristics. The overall outcomes demonstrated distinct variations according to recipient age and risk profiles. Younger candidates (aged 19–39 years) had a higher cumulative incidence of DDKT (24.5%) than of mortality (11.4%). In contrast, older candidates (≥65 years) had a markedly higher risk of waitlist mortality (41.2%) than those who received a transplant (11.4%). Regarding dialysis duration, the cumulative incidence of mortality was notably higher in the initial 0–3 years group. Conversely, the probability of receiving a DDKT gradually increased in proportion to longer dialysis durations.
Twelve-year cumulative incidence of mortality and DDKT (n = 51,427)
Similar disparities were observed across K-EPTS categories (Table 2). Candidates with the most favorable prognosis (K-EPTS, 0%–24%) had a higher incidence of DDKT (16.9%) than those who experienced waitlist mortality (9.7%). Conversely, those in the highest-risk group (K-EPTS, 75%–100%) experienced a waitlist mortality rate (38.5%), which was more than double the incidence of transplantation (17.2%).
Notably, a profound structural inequity was observed according to the ABO blood types (Table 2). While the 12-year cumulative incidence of waitlist mortality remained relatively homogenous across blood groups (ranging from 24.5% to 28.2%), the probability of DDKT exhibited substantial divergence. Candidates with blood type O faced the lowest access to transplantation (15.4%), whereas those with blood type AB experienced the highest incidence (27.5%), representing a nearly two-fold disparity in transplant opportunity.
Korean Time-to-Equivalent Risk estimates
Competing risk analyses revealed distinct temporal patterns in the probabilities of transplantation versus waitlist mortality. Fig. 2 illustrates the CIFs of DDKT and waitlist mortality according to candidate age at listing. Graded prolongation of the K-TiTER score was observed with increasing age. For candidates aged 19–39 years, the cumulative incidence of DDKT surpassed the risk of waitlist mortality at 111.4 months (K-TiTER) post-listing. This crossover point was delayed to 126.2 months (K-TiTER) for candidates aged 40–49 years and further extended to 158.4 months (K-TiTER) for those aged 50–64 years. Notably, for candidates aged ≥65 years, waitlist mortality consistently exceeded DDKT throughout the observation period. Consequently, K-TiTER was “not reached.”
Cumulative incidence of mortality or DDKT according to candidate age groups at listing.
(A) Age 19–39 years, (B) 40–49 years, (C) 50–64 years, and (D) ≥65 years. The blue and red lines indicate the probability of receiving a DDKT and the cumulative incidence of waitlist mortality, respectively. The yellow circles in both the main panel and the enlarged view in the right upper panel indicate the Korean Time-to-Equivalent Risk (K-TiTER) point, defined as the specific time intersection where the probability of transplantation finally overcomes the accumulated risk of death. The number at risk represents the total number of event-free candidates remaining on the waitlist, as the risk set is identical for both competing events.
DDKT, deceased donor kidney transplantation.
Similarly, stratification by K-EPTS score confirmed that recipient prognosis significantly influenced the probability of crossover (Fig. 3). Candidates with K-EPTS scores of 0–24 achieved K-TiTER at 125.1 months, closely followed by those with K-EPTS scores of 25–49 at 128.4 months. For candidates with a K-EPTS score of 50–74, the probability of transplantation did not surpass the mortality risk until 151.2 months. Candidates in the highest-risk category (K-EPTS score, 75–100) showed no crossover, as waitlist mortality risk remained higher than transplantation throughout the observation period.
Cumulative incidence of mortality or DDKT according to K-EPTS quartile groups at listing.
(A) K-EPTS 0–24; (B) K-EPTS 25–49; (C) K-EPTS 50–74; (D) K-EPTS 75–100. The blue and red lines indicate the probability of receiving a DDKT and the cumulative incidence of waitlist mortality, respectively. The yellow circles in both the main panel and the enlarged view in the right upper panel indicate the Korean Time-to-Equivalent Risk (K-TiTER) point, defined as the specific time intersection where the probability of transplantation finally overcomes the accumulated risk of death. The number at risk represents the total number of event-free candidates remaining on the waitlist, as the risk set is identical for both competing events.
DDKT, deceased donor kidney transplantation; K-EPTS, Korean Estimated Post-Transplant Survival.
The ABO blood type further differentiated the timing of the probability crossover (Fig. 4). Candidates with the AB blood type achieved the earliest K-TiTER at 136.6 months. In contrast, patients with blood type O experienced the longest delay, reaching the crossover point over 180.0 months. Candidates with blood types A and B exhibited intermediate K-TiTER values of 162.6 months and 163.1 months, respectively.
Cumulative incidence of mortality or DDKT according to ABO blood type.
(A) Blood type A, (B) B, (C) AB, and (D) O. The blue and red lines indicate the probability of receiving a DDKT and the cumulative incidence of waitlist mortality, respectively. The yellow circles in both the main panel and the enlarged view in the right upper panel indicate the Korean Time-to-Equivalent Risk (K-TiTER) point, defined as the specific time intersection where the probability of transplantation finally overcomes the accumulated risk of death. The number at risk represents the total number of event-free candidates remaining on the waitlist, as the risk set is identical for both competing events.
DDKT, deceased donor kidney transplantation.
Korean Time-to-Equivalent Risk estimates in high-risk groups
Table 3 presents the estimated K-TiTER for candidates aged 50 years or older, stratified by the presence of key high-risk characteristics: diabetes mellitus, high K-EPTS score (75–100), blood type O, and positive PRA. Among candidates aged 50–64 years, the K-TiTER for the overall cohort was 158.4 months (95% CI, 155.2–160.7). The presence of individual risk factors significantly prolonged the time to crossover. Specifically, K-TiTER was delayed to 183.1 months in sensitized candidates, 158.5 months in those with diabetes, and 178.7 months in candidates with blood type O. Notably, candidates with a K-EPTS score of 75–100 failed to reach the crossover point within the observation period. The presence of multiple risk factors further exacerbated transplant inaccessibility. In candidates aged 50–64 years with both diabetes and blood type O, K-TiTER was extended to 178.6 months, and those with diabetes and positive PRA showed a delayed K-TiTER of 183.4 months. For the most vulnerable subgroups, such as those with high K-EPTS scores or combinations of three or more risk factors, K-TiTER was “not reached.” Furthermore, for candidates aged ≥65 years, K-TiTER was reached neither for the overall cohort nor for any specific risk subgroup. Regardless of the clinical profile, older adult candidates faced a waitlist mortality risk that remained higher than the probability of DDKT throughout the follow-up period.
K-TiTERs for patients initially waitlisted aged 50 years or higher
Table 4 further describes the impact of recipient prognosis by jointly stratifying K-TiTER according to age and K-EPTS groups. Within the younger cohorts (aged 19–39 and 40–49 years), K-TiTER was achievable for candidates with low-to-moderate K-EPTS scores (98.2 months for age 19–39 years/K-EPTS 0–24; 106.2 months for age 40–49 years/K-EPTS 25–49). For candidates aged 50–64 years, crossover was observed across a broader range of K-EPTS scores (0–74), with values ranging from 143.2 to 234.5 months, whereas those with the highest K-EPTS scores (75–100) did not achieve crossover. Consistent with the findings in Table 3, candidates aged ≥65 years showed no identifiable K-TiTER across all K-EPTS strata.
K-TiTER according to K-EPTS and the candidate age group
Sensitivity analyses
To evaluate the potential impact of donor selection on waiting times, we performed sensitivity analyses of waitlist candidates and DDKT recipients stratified by donor age and K-KDPI (Supplementary Tables 1, 2; available online). When older donors were included, K-TiTER became shorter (Supplementary Table 1, available online). For example, K-TiTER values for candidates aged between 40 and 49 years were 172.6, 167.5, and 155.1 months for donors aged 19–49 years, 50–64 years, and 19–64 years, respectively. Similarly, donors with K-KDPI between 75% and 100% had a shorter K-TiTER than those with K-KDPI in other quartiles (Supplementary Table 2, available online). A broad range of donors—such as donor age of 19–64 years or K-KDPI of 0%–100% including high-risk donors—showed a shorter TiTER compared with donors with relatively high quality, such as donor age of 19–49 years, K-KDPI of 0%–24%, or K-KDPI of 0%–74%. Candidates aged 40–49 or 50–64 years reached K-TiTER only when accepting a broad range of donors, including suboptimal organs, whereas candidates aged ≥65 years could not reach K-TiTER irrespective of donor quality.
Discussion
This nationwide study provides the first comprehensive estimation of the TiTER for kidney transplant candidates in Korea, a region characterized by extreme organ scarcity and prolonged waiting times. The principal finding is that for high-risk candidates, specifically those with advanced age, high K-EPTS scores, or blood type O, the cumulative incidence of waitlist mortality persistently exceeds the probability of receiving a DDKT. Consequently, a crossover point representing a probability shift is frequently “not reached” in these vulnerable subgroups. The primary clinical and structural goal of identifying a “not reached” K-TiTER is to explicitly distinguish high-risk populations whose waitlist mortality consistently dominates their probability of transplant throughout the waiting period.
It is imperative to recognize that this “not reached” phenomenon stems from two distinct mechanisms: biological vulnerability and structural disparity. For instance, as shown in Table 2, the incidence of waitlist mortality does not substantially differ across ABO blood types, yet the K-TiTER for blood type O remains “not reached.” As demonstrated in our previous study, this is driven by the fact that blood type O has a 30% lower chance for DDKT allocation due to systemic imbalances, rather than an inherently higher biological mortality [22] (Supplementary Table 3, available online). Conversely, higher K-EPTS scores and older age inherently drive inferior waitlist mortality due to true biological risk. Furthermore, while biological risks like high K-EPTS are generally associated with poorer posttransplant survival, structural disadvantages such as blood type O or PRA sensitization do not significantly impair posttransplant outcomes [23] (Supplementary Table 3, available online). However, ‘not reached’ should not be interpreted as evidence of no crossing, but rather as an indication that any potential intersection occurred beyond the maximum observation window (264 months). Taken together, these findings signify that under the current allocation system, the promise of transplantation remains statistically elusive for both the most medically urgent and structurally disadvantaged candidates, leaving them exposed to a dominant risk of waitlist mortality throughout their waiting period.
Our findings stand in stark contrast to the TiTER estimates reported in the United States In the United States context, where initial transplant rates are high, TiTER marks the closure of a transplant opportunity window [15]. Conversely, due to extreme organ scarcity in Korea, candidates begin their wait with a mortality risk that already outweighs the probability of DDKT. Thus, K-TiTER represents the critical opening of a DDKT opportunity that finally overcomes the accumulated risk of waitlist mortality. In this sense, transferring a United States-derived construct to Korea with a substantially different allocation environment would be one of the limitations of this study.
In the Korean context, a shorter K-TiTER signifies an earlier liberation from the dominant risk of waitlist mortality, which is the desired condition. Critically, our sensitivity analyses demonstrated the potential benefit of broadening the donor acceptance criteria to include suboptimal donors or ECD. When the definition of acceptable donor quality was expanded to include higher-risk kidneys (e.g., K-KDPI 0%–100%, older donors), the probability of reaching a K-TiTER improved. Although recipients of high-KDPI kidneys in the current system still experience long waiting times, the fact that they ultimately achieve transplantation contrasts sharply with the lack of opportunity for candidates who remain on the waitlist to receive DDKT. This suggests that the willingness to accept ECD and, by extension, the strategic utilization of Donation after Circulatory Death (DCD) is a viable strategy to shorten the infinite K-TiTER for high-risk candidates.
These findings highlight the imperative to revise allocation policies to promote the appropriate use of suboptimal donors and improve the efficiency of scarce organ utilization. To address this issue, our findings highlight the potential benefits of considering a “fast-track” or “old-for-old” allocation program, similar to the Eurotransplant Senior Program (ESP) [24] or the United Kingdom Fast-Track system [25,26]. For instance, the ESP prioritized the local allocation of elderly donor kidneys to elderly recipients to minimize cold-ischemic time and maximize utility [24]. Similarly, the United Kingdom Fast-Track system allows kidneys to be rapidly declined by standard centers to be rapidly offered to a wider pool, thereby reducing discard rates and offering a lifeline to high-risk candidates [25–27]. This framework is particularly relevant to the upcoming implementation of DCD in Korea. Rather than distributing DCD organs into the general pool, prioritizing them for candidates with “not reached” K-TiTER offers a critical survival advantage. For candidates with poor prognosis, the benefit of receiving an ECD or DCD graft—even with its potential risks—significantly outweighs the certainty of waitlist mortality associated with indefinite maintenance dialysis. This strategy is supported by a recently published clinical practice guideline, which advocates the utilization of kidneys at risk of discard, including those from ECD, DCD, and donors with acute kidney injury [28]. The guideline explicitly suggests that for selected high-risk candidates, transplantation with these suboptimal organs confers a survival benefit superior to remaining on the waitlist with dialysis [28]. Studies analyzing the outcomes of kidneys with a high KDPI or ECD support this approach. While these grafts are associated with higher rates of biopsy-proven acute rejection or delayed graft function than standard-criteria donor (SCD) kidneys [25,29,30], they confer a significant survival advantage over continued maintenance dialysis [31,32]. Crucially, when allocated to appropriately age-matched recipients, the long-term patient and death-censored graft survival rates are comparable to those of SCD recipients [33]. We propose the establishment of a proactive pathway in which high-risk kidneys (older donors, high K-KDPI, and DCD) are preferentially and rapidly allocated to candidates with risk factors (old age, high K-EPTS, and blood type O) for long K-TiTER, especially multiple risk factors, early in their listing period. K-TiTER provides a data-driven rationale for implementing such systems. Using K-TiTER as a trigger, a Korean Fast-Track policy could automatically identify candidates for whom standard waiting is futile and prioritize them for these strategic organ pools. This approach would effectively convert an unreachable high-risk subgroup into a quantifiable survival opportunity while simultaneously reducing the discard rate of suboptimal kidneys and enhancing the overall utility of the national donor pool. However, future prospective policy simulation studies are required to quantify the specific impact of the proposed fast-track strategies before implementation, since observational data alone cannot establish causal policy effects.
Beyond systemic reform, K-TiTER holds profound utility in clinical counseling and ethical research. First, in a clinical setting, K-TiTER necessitates a paradigm shift in shared decision-making. Clinicians can utilize these estimates to transform abstract risk into actionable insight. Informing high-risk candidates that their K-TiTER is not reached creates an objective imperative to pivot immediately toward alternative strategies—such as living donor transplantation, or immediate ECD listing—rather than futilely waiting for SCD under dialysis. Second, the K-TiTER establishes a crucial ethical framework for emerging innovations. As the field advances toward novel therapies like xenotransplantation, candidates for whom waitlist mortality persistently dominates transplant probability represent the most ethically sound candidates for clinical trials. Prioritizing these patients with unachievable K-TiTER results maximizes potential benefits while mitigating the ethical dilemmas regarding the risks associated with experimental procedures.
This study is strengthened by the use of a comprehensive nationwide cohort linking registry, administrative data, and claims data to ensure the capture of all mortality events. The application of a competing risk framework provides a more accurate estimation of real-world probabilities than standard survival analyses. However, several limitations should be considered. First, although treating living donor transplants and delisting as censoring events reflects the real-world dropout dynamics of the DDKT waitlist, it may introduce informative censoring bias, as candidates receiving living donor kidney transplantation often possess better baseline health. Second, unmeasured confounders, such as dynamic changes in clinical status after listing, were not modeled. Therefore, findings in this study should be interpreted as descriptive and potentially affected by time-dependent confounding. Third, independent predictors for “not reached TiTER” were not analyzed beyond subgroup analysis. Interpreting CIFs in small subgroups is limited by statistical power due to small sample sizes. Furthermore, subgroup analysis in this study should be interpreted as descriptive findings rather than as evidence of independent effects due to the lack of adjusted competing risk regression analyses. Fourth, our study pooled data over a 15-year period (2008–2022), which may obscure temporal era effects such as policy shifts or the impact of the COVID-19 pandemic. Furthermore, the national registry data did not allow us to account for center-level practice variations. Lastly, due to the inherent limitations of the administrative claim dataset, accurately specifying the exact cause of death (e.g., cardiovascular death) was not feasible. Therefore, future studies incorporating detailed cardiovascular parameters and cause-specific mortality are needed to further refine the K-TiTER estimates.
In conclusion, this descriptive study proposed preliminary hypotheses for the K-TiTER framework within the Korean allocation system. K-TiTER effectively visualizes the extreme vulnerability of specific subgroups, highlighting that waiting for SCD is often futile for elderly and medically complex candidates. Using K-TiTER to explore tailored pathways—such as Fast-Track, ECD, and DCD programs—may benefit waitlisted patients with poor prognosis and contribute to a more efficient utilization of scarce organs. Furthermore, K-TiTER may be a practical instrument in prioritizing candidates for innovative trials.
Supplementary Materials
Supplementary data are available at Kidney Research and Clinical Practice online (https://doi.org/10.23876/j.krcp.26.031).
Notes
Conflicts of interest
All authors have no conflicts of interest to declare.
Funding
This study was supported by a grant from the National Institute of Organ, Tissue and Blood Management (20232400C1B-00), which was not involved in the design or analysis of the study.
Acknowledgments
We thank the Korean Network for Organ Sharing (KONOS) and the National Health Insurance Data Sharing Service (NHISS) for sharing their database. We also thank MID (Medical Illustration & Design), a member of the Medical Research Support Services of Yonsei University College of Medicine, for providing excellent support with medical illustration.
Data sharing statement
The data presented in this study are available from the corresponding author upon reasonable request.
Authors’ contributions
Conceptualization, Data curation, Investigation, Methodology: JHL, JY
Formal analysis: All authors
Funding acquisition, Supervision: JY
Project administration: JY
Visualization: JHL
Writing–original draft: JHL, JY
Writing–review & editing: JHL, JY
All authors read and approved the final manuscript.