Introduction

Bladder cancer (BC), as one of the most prevalent genitourinary malignancies, accounts for an estimated 500 000 new cases and 200 000 deaths annually worldwide1. About 75% patients are initially diagnosed as non-muscle invasive bladder cancer (NMIBC) with mucosal or submucosal confined disease2. In spite of undergoing transurethral resection of the bladder tumor (TURBT) and adjuvant intravesical therapy including chemotherapy or immunotherapy, disease recurrence and progression remain prevalent, which aggravate the burden of health care system and prompt researchers’ widespread attention3.

Various predictive models have been established to stratify recurrence and progression risk, among which the European Organisation for Research and Treatment of Cancer (EORTC) and Club Urologico Espanol de Tratamiento Oncologico(CUETO) scoring systems are widely adopted, taking into account risk factors including tumor size, tumor multiplicity, age, carcinoma in situ (CIS), tumor grade, tumor stage, and previous recurrence3,4,5,6.

Upper tract urothelial carcinoma (UTUC) stands as a relatively uncommon malignancy, encompassing roughly 5–10% of urothelial carcinomas7. Patients with UTUC history exhibit an augmented risk of intravesical recurrence(IVR), with subsequent BCa occurring in 15–50% of patients who undergo radical nephroureterectomy8. However, the impact of UTUC history on the prognosis of NMIBC remains unclear and overlooked in most predictive models. As a result, its treatment strategy is still based on the guidelines of primary bladder cancer.

While Bacillus Calmette-Guérin (BCG) is the gold standard treatment in high-risk patients, its application is limited by severe toxicity, high price and persistent shortage, especially in developing countries9,10,11. Intravesical chemotherapy (IC) is considered as a reasonable alternative in such scenario. Due to aforementioned reasons, IC remains the most adopted adjuvant intravesical instillation strategy during past decades in our institution.

The goal of our research was to evaluate the impact of UTUC history on the prognosis of primary NMIBC patients receiving adjuvant IC. Through our study, we hope to improve current postoperative treatment strategy in NMIBC patients with prior UTUC.

Patients and methods

After obtaining approval from the Ethics Committee, a retrospective analysis was conducted on the medical records of patients who underwent TURBT at Peking University People’s Hospital between 2000 and 2015. Patients with primary NMIBC who underwent ‘adequate’ IC were selected. ‘Adequate’ IC was defined as at least weekly induction for the first 8 weeks and then monthly up to 6 months. The follow-up for patients after TURBT was usually every 3 months for the first 2 years, every 6 months for the next 3 years, and annually thereafter. Follow-up protocols were devised and executed in accordance with the risk stratification conducted by the EAU guidelines.

Patients were divided into two groups: those with primary NMIBC after UTUC (UTUC-NMIBC), and those without prior UTUC. Patients’ clinical characteristics and pathological features were collected: age, sex, body mass index, smoking history, hypertension, diabetes, pathological grade, tumor stage, tumor number and tumor size. Recurrence was characterized as reoccurrence of bladder tumor irrespective of stage or grade. Progression was defined as defined as progression at tumor stage or grade. Patients with previous UTUC that didn’t undergo curative treatment or suffered UTUC recurrence were excluded.

Continuous variables were reported as the median and interquartile range (IQR), while categorical variables were described as proportion of events. The clinicopathological characteristics were compared using Student’s t-test, the Wilcoxon rank test and the chi-squared test as appropriate. Kaplan-Meier survival analysis was utilized for all survival analyses. The associations between clinicopathological parameters and oncologic outcomes were evaluated through univariate and multivariate Cox proportional hazard regression analysis. Multivariate analysis was performed only for parameters with P < 0.1 in univariate analysis. Statistical significance was defined at P ≤ 0.05. IBM SPSS Statistics version 25.0 (IBM, Armonk, NY, USA) was used for all statistical analysis.

Results

Baseline characteristics

The final cohort included 444 patients of NMIBC who received sufficient IC treatment. Among them, 34 individuals had a history of UTUC, with a median interval of 24.0 months (interquartile range [IQR]: 9.0-58.3months) between UTUC and NMIBC diagnoses. The clinicopathological characteristics of these patients are detailed in Table 1. Compared to the primary NMIBC, UTUC-NMIBC group are older [median age, 72.0 (65.0–81.0) vs. 66.0 (58–75) years; P = 0.007] and less likely to have smoking history. Additionally, the UTUC-NMIBC group had a higher incidence of multiple tumors (52.9% vs. 33.9%; P = 0.026). The pathological characteristics of prior UTUC are detailed in Table 2.

Table 1 General characteristics and tumor features of patients with NMIBC, stratified by UTUC history.
Full size table
Table 2 The pathological characteristics of prior UTUC.
Full size table

Entire cohort survival analysis

Throughout the follow-up period, the UTUC-NMIBC group exhibited a higher rate of recurrences (52.9% vs. 30.7%; P = 0.008) and any-grade/stage progressions, although the latter was on the verge of statistical significance (26.5% vs. 14.6%; P = 0.067). Importantly, the UTUC-NMIBC group was associated with significantly worse recurrence-free survival (median survival, 29.91 vs. 84.00 months, P < 0.001). However, there was no discernible difference in PFS between the two groups (P = 0.117), as depicted in Fig. 1.

Fig. 1
figure 1

Kaplan–Meier curves for recurrence-free survival and progression-free survival in UTUC-NMIBC group and primary NMIBC group.

Full size image

Univariate Cox regression analysis revealed that smoking status, tumor grade, pathological stage, multiplicity, and UTUC history were all associated with poorer recurrence-free survival. Further multivariate Cox analysis demonstrated that T1 stage, multifocal tumors, and UTUC history were still statistically significant. The presence of UTUC history was linked to a higher risk of recurrence (HR, 2.257; 95% CI, 1.368–3.725; P = 0.001). T1 stage (HR, 2.232; 95% CI, 1.405–3.545; P = 0.001) and multifocal tumors (HR, 1.564; 95% CI, 1.113–2.197; P = 0.01) were also associated with increased recurrence, with hazard ratio of 2.23 and 1.56, respectively (Table 3). However, UTUC history was not associated with an increased risk of progression in univariate and multivariate regression analysis.

Table 3 Univariate and multivariate Cox proportional hazard regression analysis to identify predictive factors for recurrence-free survival.
Full size table

Subgroup analysis in patients with NMIBC

Among patients with UTUC history, stratified by median interval between UTUC and NMIBC diagnoses, 19 individuals had shorter interval (≤ 24 months), while 15 had longer interval (> 24 months). Kaplan-Meier curve revealed a significant difference in recurrence free survival (P = 0.023; Fig. 2), with 73.7% of patients in the short interval group undergoing recurrence versus 26.7% of those in long interval (P = 0.014). Additionally, rates of progression, radical cystectomy and upgrading at cystectomy were not increased in the longer interval group compared with the longer interval group, showed in Table 4.

Fig. 2
figure 2

Kaplan–Meier curves for recurrence-free survival between different interval subgroups in patients with prior UTUC.

Full size image
Table 4 Prognosis in patients with UTUC history stratified by interval between UTUC and NMIBC diagnoses.
Full size table

Discussion

Extensive attention has focused on identifying risk factors that can forecast recurrence, progression, and treatment response in NMIBC patients. Various nomograms have been created to assist healthcare providers in stratifying patient risk, with the EORTC and CUETO scoring systems being the most prevalent3,4,5,6. However, it’s noteworthy that most models excluded patients with UTUC from their analyses. The incidence of NMIBC following UTUC, also defined as intravesical recurrence(IVR), is well-documented, with 15–50% of patients undergoing radical nephroureterectomy experiencing intravesical recurrence8.

Just as two mechanisms exist behind intravesical recurrence after transurethral resection of bladder cancer12, so are both behind intravesical recurrence after prior UTUC, i.e., tumor cells intracavity transplantation following UTUC surgery and field change cancerization effect.

UTUC has a proclivity to reoccur in proximity to the cystostomy tube wall or within the bladder neck, particularly in cases where the urethral duct has suffered damage13. This observation backs the theory that cancer cells, suspended within the bladder, predominantly adhere to the compromised urethra and subsequently manifest as recurring instances through intraluminal inoculation. Notably, Habuchi and colleagues unearthed a consistent pattern of the same distinct p53 mutation in the upper urinary tract and bladder tumors of the same patient14. In a comprehensive analysis of pertinent literature, Doeveren and collaborators systematically reviewed the potential clonal connection between UTUC and BC. Their findings indicated that approximately 94% of primary UTUC cases and intravesical recurrences (IVR) share a clonal relationship15. Field change cancerization effect refers to independent tumor origin, indicating a pan-urinary tract pre-cancerous state, as UTUC and bladder cancer share many risk factors, including smoking, dye exposure and gender.

Currently, the management of patients with IVR is based on the primary bladder tumor guideline. Meanwhile, with shortage and expensiveness of BCG limiting its accessibility to lots of patients, intravesical chemotherapy is still the standard of care in many circumstances. It is thus imperative to evaluate the ramification of previous UTUC on NMIBC prognosis to improve treatment strategy.

The current study revealed that patients with UTUC history exhibited poorer response to intravesical chemotherapy for NMIBC in contrast to those without prior UTUC. These findings are in line with prior studies, which concluded presence of UTUC prior to diagnosis of NMIBC was associated with poor prognosis after BCG therapy16,17. Interestingly, the influence of prior UTUC on muscle invasive bladder cancer (MIBC) seems quite different. Shigeta found MIBC developed after radical nephroureterectomy for UTUC was different from bladder primary MIBC, characterized by higher FGFR3 expression, a protective prognostic factor18. Further tissue sample analysis revealed that molecular expression in the bladder tumor specimen changed from CK20 high, CK5/6 low expression to CK20 low CK5/6 high expression as NMIBC following UTUC progressed to MIBC, indicating a luminal-basal shift19.

A possible explanation of poor IC response is that IVR represent a special subset of NMIBC lesions, with recent data suggesting they primarily exhibit luminal characteristics and are depleted in T-cells, possibly diminishing chemotherapy-induced anti-tumor immune response20.

The current study does come with certain limitations. Firstly, the utilization of a retrospective cohort introduces the potential of unaccounted-for confounding variables. Secondly, given the infrequent occurrence of UTUC, our analysis is constrained by the relatively limited number of UTUC patients in the cohort. Thirdly, treatment variations across different surgeons were not factored in, including differences in UTUC management, the administration of re-Turbt. Lastly, our analysis was confined to patients who received a relatively sufficient amount of intravesical chemotherapy to ensure a consistently treated cohort. However, this could potentially lead to selection bias.

In summary, among patients with history of UTUC, the biological and clinical trajectory of NMIBC may substantially diverge from those with primary NMIBC. This observation seems particularly significant for those facilities troubled by BCG shortage. Under such circumstances, BCG, widely considered to be more efficacious than chemotherapy drugs, should be administered priorly to those with previous UTUC. Other options, such as immune checkpoints inhibitors or even preemptive cystectomy, should also be taken into account. In addition, considering the high recurrence follow intravesical therapy, patients with previous UTUC should undergo more intensive follow-up to detect recurrent lesion at an early stage.

Conclusion

The presence of prior UTUC correlated with a significant increase in recurrence rates following intravesical chemotherapy. UTUC history should be considered when counselling patients and designing cohorts for clinical trials. Future prospective trials are warranted to optimize interventions strategy for this high-risk subgroup.