Metabolic score shows modest promise for predicting pancreas-kidney transplant outcomes in diabetes

Patients with type 2 diabetes and kidney failure who received a simultaneous pancreas-kidney transplant had substantially fewer deaths, kidney graft failures and major cardiovascular events than comparable patients who received a kidney transplant alone, according to a single-center exploratory study conducted in China between 2015 and 2025. The research, published in BMC Nephrology, also attempted to build a prognostic score from pretransplant metabolic measurements, but the authors themselves caution that the score achieved only modest discrimination and is not ready for clinical use.

The cohort included 218 patients with type 2 diabetes mellitus and kidney failure. Of these, 128 received a simultaneous pancreas-kidney transplantation (SPK), while 90 received a kidney transplant alone (KTA). Over a median follow-up of 4.8 years, the primary endpoint—a composite of all-cause death, death-censored kidney graft failure, or major adverse cardiovascular events—occurred in 22 of 128 SPK recipients (17.2 percent) compared with 57 of 90 KTA recipients (63.3 percent). The unadjusted hazard ratio was 0.25, with a 95 percent confidence interval of 0.15 to 0.41 and a P value below 0.001. Because these comparisons were not randomized, they are associative and do not prove that SPK caused the better outcomes.

Glycemic control also differed markedly between the two groups. Among SPK recipients, 78.9 percent achieved an HbA1c below 7.0 percent, compared with 31.1 percent of KTA recipients, a difference that was statistically significant at P < 0.001.

The study’s central innovation was its attempt at metabolic phenotyping. The researchers measured dynamic C-peptide levels, insulin resistance assessed by the HOMA-IR index, and visceral fat area (VFA). The motivation was that selection criteria for SPK in patients with type 2 diabetes have historically been extrapolated from experience with type 1 diabetes, which may not capture the metabolic heterogeneity of type 2 disease, particularly in East Asian populations.

Using LASSO Cox regression, the team derived what they named the Chinese SPK-T2DM Prognostic Score (CSTPS), validated internally through nested cross-validation and bootstrap correction. Five variables were retained in the final model: a C-peptide level at 120 minutes of 1.5 ng/mL or higher, HOMA-IR below 4.5, visceral fat area below 100 cm², diabetes duration below 15 years, and age below 55 years.

However, the score’s performance fell short of what would be needed for clinical decision-making. Discrimination was modest, with a C-statistic of 0.537 (95 percent CI 0.396 to 0.683), barely better than chance. The score was not significantly associated with the SPK-specific endpoint, showing a hazard ratio of 0.92 per 10-point increase with P = 0.34. Crude event rates across score strata were 8.3 percent, 28.9 percent and 44.4 percent, with a log-rank P value of 0.046, and an interaction test suggested heterogeneity in the treatment association (P = 0.032), hinting that the benefit associated with SPK may vary across metabolic subgroups.

The authors are explicit about these limitations. In their conclusions, they state that the score achieved only modest stratification and no validated discrimination, and that few events, outcome-driven thresholds and the absence of external validation preclude its use for prediction or as an indication tool. They emphasize that independent external validation is required before any clinical application could be considered.

The study was registered retrospectively on ClinicalTrials.gov as NCT05497232 on 11 August 2022. As a single-center analysis, its findings may not generalize to other populations or transplant programs, and the retrospective registration means the analysis was designed after patient care had begun. The published version is an early-release, peer-reviewed accepted manuscript that remains subject to further editorial edits before the final version of record.

But until the score is validated externally, the authors’ own verdict stands: the phenotyping approach is promising as a research direction, yet the resulting tool should not guide patient selection or prediction in practice.

Subject of Research: A prognostic score based on metabolic phenotyping for simultaneous pancreas-kidney transplantation in type 2 diabetes with kidney failure

Article Title: Metabolic phenotyping and a prognostic score for simultaneous pancreas-kidney transplantation in type 2 diabetes mellitus with kidney failure: a single-center exploratory study based on the Chinese population

Article References: Chen, R., Wu, J., Li, G., Fang, J., Guo, Y., Liu, L., & Chen, Z. (2026). Metabolic phenotyping and a prognostic score for simultaneous pancreas-kidney transplantation in type 2 diabetes mellitus with kidney failure: a single-center exploratory study based on the Chinese population. BMC Nephrology. https://doi.org/10.1186/s12882-026-05466-z

Image Credits: AI Generated

DOI: 10.1186/s12882-026-05466-z

Keywords: simultaneous pancreas-kidney transplantation, type 2 diabetes, kidney failure, kidney transplant alone, metabolic phenotyping, C-peptide, HOMA-IR, visceral fat area, prognostic score, HbA1c, BMC Nephrology, external validation