Many patients are elderly, frail, or experience renal impairment following nephroureterectomy, rendering them ineligible for cytotoxic or immunotherapy. Thus, there is an urgent clinical need for a simple, objective, and reproducible prognostic model to stratify recurrence risk and personalize postoperative management.
Development of the JIKEI-YAYOI Score
To address this gap, we established the JIKEI-YAYOI collaborative group, encompassing 16 affiliated institutions of The Jikei University School of Medicine in Japan. We analyzed 700 patients who underwent radical nephroureterectomy between 2012 and 2021 and validated our model using an external cohort of 405 patients from the Mayo Clinic, USA. Patients who had received neoadjuvant or adjuvant chemotherapy were excluded to create a pure surgical population, allowing us to focus on intrinsic tumor biology.
We evaluated widely available clinicopathologic variables—such as pT stage, pN stage, tumor grade, and lymphovascular invasion (LVI)—and derived a score-based prognostic model termed the JIKEI-YAYOI score. Each factor was assigned a weighted point value according to its hazard ratio for recurrence, yielding a total score ranging from 0 to 7. Patients were then stratified into low-, intermediate-, and high-risk groups based on predicted 3-year recurrence probabilities (<20%, 20–49%, and ≥50%, respectively).
Key Findings and Validation
In the development cohort, the JIKEI-YAYOI model demonstrated excellent discrimination, with a C-index of 0.815, and retained robust performance after internal bootstrapping (corrected C-index 0.811). External validation at the Mayo Clinic confirmed its generalizability, showing clear separation of Kaplan-Meier curves for recurrence-free survival (RFS) across all risk groups (p < 0.001). Importantly, calibration plots showed strong agreement between predicted and observed recurrence rates at both 3 and 5 years, underscoring the model’s accuracy and reproducibility.
The 3-year RFS rates for the low-, intermediate-, and high-risk groups in the development cohort were 93.9%, 77.5%, and 41.3%, respectively. These findings were nearly identical in the validation cohort (94.5%, 59.0%, and 12.1%, respectively), confirming the model’s cross-population reliability.
Why the JIKEI-YAYOI Score Matters
Our model is unique in two major aspects.
First, it subclassifies pT3 disease into pT3a and pT3b according to peripelvic and/or periureteral fat invasion—an important but often overlooked detail in existing prediction tools. This distinction revealed a significant prognostic difference, particularly for renal pelvic versus ureteral tumors, refining recurrence risk estimation.
Second, we emphasized LVI as an independent and powerful prognostic marker. Although LVI is commonly recognized in urothelial carcinoma, its clinical relevance has been underexplored in recent adjuvant immunotherapy trials such as CheckMate 274 and IMvigor 010. In our cohort, approximately one-third of patients eligible for adjuvant immunotherapy exhibited LVI positivity, suggesting that incorporating this variable could improve the precision of patient selection in future perioperative trials.
Clinical and Research Implications
From a practical standpoint, the JIKEI-YAYOI score offers several advantages:
- Universality — It relies solely on pathological variables routinely available worldwide, making it applicable even in community settings.
- Ease of Use — The scoring system can be calculated immediately after pathological review without requiring molecular assays.
- Guidance for Adjuvant Therapy — It identifies patients at the highest risk of recurrence who may derive meaningful benefit from adjuvant systemic therapy, while sparing low-risk individuals from unnecessary toxicity.
- Framework for Future Integration — The model establishes a foundation for incorporating emerging biomarkers, such as genomic alterations or extracellular vesicle signatures, into hybrid prediction systems.
While our study provides a strong foundation, several challenges remain. The model was derived from retrospective cohorts, and prospective validation—particularly in populations receiving modern adjuvant therapies—is essential. Furthermore, as UTUC management increasingly incorporates perioperative immunotherapy and targeted agents, future work should explore how clinical scores like the JIKEI-YAYOI model interact with molecular predictors to refine individualized treatment pathways.
Ultimately, our goal is to move beyond simple risk stratification toward precision perioperative management for UTUC. By enabling clinicians to predict recurrence risk at the time of surgery, the JIKEI-YAYOI score bridges the gap between clinical evidence and real-world decision-making, helping to ensure that every patient receives treatment proportionate to their recurrence risk and overall health status.
Written by: Fumihiko Urabe, MD, PhD, Department of Urology, The Jikei University School of Medicine, Tokyo, Japan
Read the Abstract