Abstract
Background/Aim: To evaluate the difference in the clinical efficacy and safety of pembrolizumab between patients with metastatic upper tract urothelial carcinoma (UTUC), which includes renal pelvic urothelial carcinoma (UC) and ureteral UC, and those with metastatic lower tract urothelial carcinoma (LTUC). Patients and Methods: A total of 752 patients who received pembrolizumab for the treatment of chemoresistant UC were retrospectively analyzed. We compared progression-free survival (PFS), overall survival (OS) and adverse events (AEs) in patients with renal pelvic UC, ureteral UC, and LTUC. Results: The median follow-up period was 42.5 [interquartile range (IQR)=35.1-47.4] months. The primary tumor site was in the upper tract in 362 (48.1%) patients [renal pelvis, n=219 (60.5%); ureter, n=143 (39.5%)] and in the lower tract in 390 (51.9%) patients. The estimated glomerular filtration rate before pembrolizumab treatment in the UTUC group was significantly lower than that in the LTUC group (p<0.001). The median PFS in the UTUC and LTUC groups was 3.4 months, respectively (p=0.271). The median OS in the UTUC and LTUC groups was 10.1 months and 11.7 months, respectively (p=0.195). In an analysis of UTUC divided into renal pelvic UC, ureteral UC, and LTUC, patients with renal pelvic UC had a significantly poorer prognosis in comparison to the other two groups (p=0.041). The incidence of any-grade AEs (51.7% vs. 47.9%, p=0.343) and grade ≥3 AEs (12.2% vs. 12.8%, p=0.826) in the two groups was not statistically significantly different. Conclusion: No significant differences were found between the UTUC and LTUC groups with regard to the oncological outcomes and safety of pembrolizumab. Patients with renal pelvic UC had a significantly poorer prognosis than those with other ureteral UCs and LTUCs.
- Chemo-resistant urothelial carcinoma
- pembrolizumab
- upper tract urothelial carcinoma
- lower tract urothelial carcinoma
In 2022, urothelial carcinoma (UC) was the sixth most common type of cancer in developed countries (1, 2). According to the World Health Organization, more than 570,000 patients were diagnosed with bladder cancer worldwide in 2020 (3). Although bladder cancers account for ≥90% of UCs, upper tract urothelial carcinomas (UTUCs) are relatively uncommon and account for only approximately 5% of UCs (2). Approximately 50% of patients with UTUCs have progressive disease, and approximately 18% of patients with UTUCs present with stage 4 disease (4, 5). At the time of the initial diagnosis, UTUC is typically at a higher stage in comparison to bladder cancer (1).
Most of the upper tract cancers are histologically classified as UC, similar to bladder cancer, and the treatment of UTUC with metastasis is similar to that of bladder cancer according to the European Association of Urology and NCCN guidelines (1, 6). Although UTUC and bladder cancer have similar histological appearances, clinical and epidemiological differences suggest that UTUC may have a different treatment sensitivity from bladder cancer (7). In a recent study, differences in molecular and epigenetic profiles between UTUC and bladder cancer were noted (8-10). These results suggested that UTUC and bladder cancer may show different sensitivity to immune checkpoint inhibitors (ICIs) and raised the possibility that advanced UTUC would show better sensitivity to ICIs in comparison to bladder cancer. Although previous studies reported no difference in the response rate of platinum-based chemotherapy between UTUC and bladder cancer, there are no reliable real-world data to show that the sensitivity of UTUC to ICIs is the same as that of bladder cancer (11).
When used as second-line treatment for advanced UC, pembrolizumab prolonged overall survival (OS) in comparison to chemotherapy in a large, randomized phase 3 trials (12, 13). Since then, pembrolizumab has been widely used as a second-line treatment for patients with advanced UC in many countries, including Japan, where pembrolizumab was approved in December 2017. To the best of our knowledge, no previous study has focused on differences in the efficacy and safety of pembrolizumab between UTUC and bladder cancer.
In this study, we used real-world and large-scale data to assess the difference in the clinical efficacy and safety of pembrolizumab between patients with metastatic UTUC, which included renal pelvic urothelial carcinoma (UC) and ureteral UC, and those with metastatic lower tract urothelial carcinoma (LTUC).
Patients and Methods
Patient characteristics (Table I). This retrospective noninterventional cohort study was approved by the Institutional Review Board (IRB) at Kyoto University Graduate School of Medicine (approval number R1783) and the local IRB at each participating institute of the Japan Urological Oncology Group framework. The design of this study was previously reported (14). In brief, the cutoff date for data inclusion was April 22, 2022. We recruited a total of 755 patients from 59 institutes in the present cohort. To analyze the efficacy of pembrolizumab, three cases with unknown primary sites of UC were excluded from this study. This study analyzed a total of 752 patients who were diagnosed with UC, including 147 patients with variant histology.
Patient characteristics (n=752).
Statistical analyses. For the current study, patients were separated into two groups (UTUC and LTUC) to analyze progression-free survival (PFS), OS, and AEs. PFS and OS were analyzed by dividing UTUC into renal pelvic UC and ureteral UC. The clinicopathological features were summarized using frequencies and percentages for categorical variables and the median, range, mean, and standard deviation for continuous variables. Student’s t-test was used for the comparison of continuous variables, while Fisher’s exact probability test and the chi-squared test were used for the comparison of categorical variables. PFS and OS were defined as the time from the initiation of pembrolizumab therapy to disease progression and death from any cause, respectively. PFS and OS curves were estimated using the Kaplan–Meier method and compared using the log-rank test. The response evaluation criteria (according to the Response Evaluation Criteria in Solid Tumors version 1.1) (15) were used to evaluate the objective response rate (ORR), defined as the proportion of patients who achieved a complete response (CR) or partial response (PR). The classification of AEs was conducted according to the Common Terminology Criteria for Adverse Events version 5.0. The ORR and incidence of AEs in the groups were compared using Fisher’s exact probability. p-Values of <0.05 were considered to indicate statistical significance. All statistical analyses were performed using the SPSS software program (version 26.0 SPSS, SPSS Japan Inc., Tokyo, Japan).
Results
Patients. The median follow-up period was 42.5 [interquartile range (IQR)=35.1-47.4] months at the data cutoff point. The median cycle count of pembrolizumab was 5.0 (IQR=3.0-14.0). Of the 752 patients, the primary tumor site was in the upper tract in 362 (48.1%) patients and in the lower tract in 390 (51.9%) patients. Among UTUC patients, tumors were located in the renal pelvis in 219 (60.5%) patients, and in the ureter in 143 (39.5%) patients. The estimated glomerular filtration rate (eGFR) before pembrolizumab treatment in the UTUC group (50.4 ml/min/1.73 m2; IQR=39.7-63.3) was significantly lower than that in the LTUC group (56.8 ml/min/1.73 m2; IQR=41.4-71.8) (p<0.001). The number of male patients (p=0.022) and histological variants (p=0.034) in the UTUC group were significantly lower than those in the LTUC group (Table I).
The patient characteristics of renal pelvic UC and ureteral UC are shown in Table II. In a comparison of renal pelvic UC and ureteral UC, the rates of visceral and liver metastasis in the renal pelvic UC group (50.2% and 25.6%, respectively) were significantly higher than those in the ureteral UC group (36.4% and 18.2%) (p<0.001). Conversely, nephroureterectomy was significantly more common in ureteral UC patients (74.8%) than in renal pelvic UC patients (48.4%) (p<0.001).
Characteristics of patients with upper tract urothelial carcinoma (n=303).
At the data cutoff point, pembrolizumab treatment was ongoing in 12.0% of patients (n=90), whereas it had been discontinued due to progressive disease in 67.3% of patients (n=506), AEs in 13.0% (n=98), good disease control 3.7% (n=28), or other reasons in 4.0% (n=30).
Treatment efficacy. The median PFS and OS of all patients were estimated to be 3.4 months [95% confidence interval (CI)=3.1-3.8] and 10.5 months (95%CI=9.0-12.0), respectively. The number of deaths was 537 (71.4%). The median PFS in the UTUC and LTUC groups was 3.4 months (95%CI=2.8-4.1) and 3.4 months (95%CI=2.9-3.9), respectively. The median OS in the UTUC and LTUC groups was 10.1 months (95%CI=8.0-12.1) and 11.7 months (95%CI=9.5-13.9), respectively. There was no significant difference in PFS (p=0.271) or OS (p=0.195) between the two groups. The ORRs in the UTUC and LTUC groups were 26.2% and 26.4%, respectively. The CR rates in the UTUC and LTUC groups were 5.0% and 9.0%, respectively. The CR rate in the UTUC group was significantly lower than that in the LTUC group (p=0.033), but there was no significant difference in ORR between the two groups (p=1.000).
In a subanalysis of UTUC divided into renal pelvic UC and ureteral UC, the median PFS in the renal pelvic UC and ureteral UC groups were 2.8 months (95%CI=2.1-3.4) and 4.1 months (95%CI=2.8-5.5), respectively. The median OS in the renal pelvic UC and ureteral UC groups was 8.3 months (95%CI=6.2-10.4) and 12.1 months (95%CI=8.7-15.6), respectively. Although there were no significant differences in PFS among the renal pelvic UC, ureteral UC and LTUC groups (p=0.187), patients with renal pelvic UC had a significantly poorer prognosis in comparison to the other two groups (p=0.041) (Figure 1).
In the overall population (n=752), Kaplan–Meier curves for progression-free survival (PFS) of patients treated with pembrolizumab in the upper tract urothelial carcinoma and lower tract urothelial carcinoma groups (A) and in the renal pelvic urothelial carcinoma, ureteral urothelial carcinoma, and lower tract urothelial carcinoma groups (B). Kaplan–Meier curves for overall survival (OS) of patients treated with pembrolizumab in the upper tract urothelial carcinoma and lower tract urothelial carcinoma groups (C), and in the renal pelvic urothelial carcinoma, ureteral urothelial carcinoma, and lower tract urothelial carcinoma groups (D).
Analysis of pure UC (n=605). Of the 605 patients with pure UC, the primary tumor site was in the upper tract in 303 (50.1%) patients and in the lower tract in 302 (49.9%) patients. Among UTUC patients, the tumor was located in the renal pelvis in 187 (61.7%) patients and in the ureter in 116 (38.3%) patients. The median PFS in the UTUC and LTUC groups was 3.4 months (95%CI=2.8-4.1) and 3.4 months (95%CI=2.9-4.0), respectively. The median OS in the UTUC and LTUC groups was 9.4 months (95%CI=7.1-11.8) and 11.3 months (95%CI=8.5-14.0), respectively. There was no significant difference in PFS (p=0.285) or OS (p=0.259) between the two groups.
The median PFS in the renal pelvic UC and ureteral UC groups was 2.8 months (95%CI=2.3-3.2) and 4.1 months (95%CI=2.8-5.5), respectively. The median OS in the renal pelvic UC and ureteral UC groups was 8.3 months (95%CI=6.1-10.6) and 12.1 months (95%CI=6.4-17.9), respectively. There was no significant difference in PFS (p=0.136) or OS (p=0.110) between the renal pelvic UC and ureteral UC groups. However, similar to the results of all patients, renal pelvic UC tended to be associated with a poor prognosis (Figure 2).
In pure urothelial carcinoma patients (n=605), Kaplan–Meier curves for progression-free survival (PFS) of patients treated with pembrolizumab in the upper tract urothelial carcinoma and lower tract urothelial carcinoma groups (A), and in the renal pelvic urothelial carcinoma, ureteral urothelial carcinoma, and lower tract urothelial carcinoma groups (B). Kaplan–Meier curves for overall survival (OS) of patients treated with pembrolizumab in the upper tract urothelial carcinoma and lower tract urothelial carcinoma groups (C) and in the renal pelvic urothelial carcinoma, ureteral urothelial carcinoma, and lower tract urothelial carcinoma groups (D).
Adverse events (AEs). Table III and Figure 3 summarize the treatment-related AEs. Any-grade AEs were observed in 49.7% (n=374). Grade ≥3 AEs were observed in 12.5% (n=94). There was no significant difference in the rate of any-grade AEs between the UTUC and LTUC groups (51.7% vs. 47.9%, p=0.343, Fisher’s exact probability). Although the difference was not statistically significant, AEs tended to be more frequent in UTUC than in LTUC. However, there was no significant difference in the frequency of grade ≥3 AEs between the two groups (12.2% vs. 12.8%, p=0.826, Fisher’s exact probability). With regard to individual AEs of any grade, there were no significant differences between the groups in the rates of individual grade ≥3 AEs.
Treatment-related adverse events.
Frequency of immune-related adverse events in patients treated with pembrolizumab in the upper tract urothelial carcinoma and lower tract urothelial carcinoma groups.
Discussion
A few studies have compared the chemosensitivity of UTUC and LTUC (11, 16). These studies showed no significant difference in the clinical response rate or survival between patients with UTUC and LTUC. However, there have been no reports comparing the efficacy of ICI treatment between UTUC and LTUC in a large population of patients with advanced UC. To the best of our knowledge, this is the first study investigating the difference in efficacy and toxicity of 2nd-line pembrolizumab between patients with UTUC [including renal pelvic urothelial carcinoma (UC) and ureteral UC] and patients with LTUC in a large multi-institutional cohort. Although we demonstrated that pembrolizumab provided UTUC patients with similar clinical outcomes to LTUC, patients with renal pelvic UC had a significantly poorer prognosis than those with ureteral UC and LTUC.
Recently, differences in epigenetics and genetic profiles between UTUC and LTUC have been noted (8-10). Based on a genomic analysis of 199 UTUC samples, the frequencies of several alterations were substantially different between UTUC and LTUC. CDKN2A and KMT2D were preferentially affected in UTUC, whereas ERBB2 was more frequently mutated in LTUC. This study revealed that UCs showed a distinct pattern of genetic alterations depending on the tumor location. Furthermore, the frequencies of genetic lesions differed between the renal pelvis and ureter. Ureteral UCs showed higher frequencies of KMT2D and TP53 mutations, whereas HRAS, KDM6A, and the TERT promoters were more commonly mutated in renal pelvic UCs (8). Fujimoto et al. performed single CpG resolution genome-wide DNA methylation screening to establish new criteria for the diagnosis of UTUC using tissue samples from patients without UCs and from UCs, including UTUC and LTUC. In this study, 2488 CpG sites showed significant differences in DNA methylation levels between normal urothelial and UTUC. Among these CpG sites, 10 were located within CpG islands or their shores and shelves included in genomic domains, which allowed for discrimination of UTUC from LTUC (10). Epigenetic alterations in cancers can be caused and/or affected by genetic alterations (17). Furthermore, epigenetic and genetic alterations could enhance the immunogenicity of cancer. These alterations were demonstrated to be linked to cancer immunogenicity and the outcome of patients treated with ICIs (18, 19). Thus, there are possible underlying mechanisms behind the differences in the response to ICIs that are observed among tumors located in the renal pelvis, ureter and bladder.
Previous clinical trials have shown several different findings with regard to the efficacy of ICI treatment for patients with UTUC and LTUC. In the KEYNOTE-052 phase II trial of pembrolizumab as first-line for cisplatin-ineligible patients, the ORR of patients with UTUC or LTUC was 26.1% (18 of 69) and 29.3% (88 of 300), respectively, while the median OS was 10.8 months and 11.5 months (20). In the IMvigor210 phase II trials on atezolizumab, the ORRs were 39% and 17% among UTUC and LTUC patients, respectively (21). The KEYNOTE-045 trial suggested that pembrolizumab prolonged OS in both UTUC [hazard ratio (HR)=0.53, 95%CI=0.28-1.01] and LTUC (HR=0.77, 95%CI=0.60-0.97) in comparison to chemotherapy (12). Esagian et al. reported real-world data on ICI treatment for 746 advanced UC patients (22). This study showed that patients with UTUC and LTUC had similar ORRs [24% vs. 28%; adjusted odds ratio (aOR)= 0.73, 95%CI=0.43-1.24] and OS [median 9.8 vs. 9.6 months; adjusted hazard ratio (aHR)= 0.93, 95%CI=0.73-1.19], although only 65% of the patients had received prior platinum-based chemotherapy (22). These results might indicate that tumor location in the urinary tract did not influence the efficacy of pembrolizumab.
In our study, in the analysis of pure UC, PFS and OS were not statistically significantly different among renal pelvic UC, ureteral UC and LTUC. However, renal pelvic UC showed a poorer prognosis in comparison to ureteral UC and LTUC in the overall population. Although no significant difference was found, it is possible that the poor prognosis of renal pelvic cancer in the overall population was due to the higher frequency of visceral and liver metastasis in comparison to the other two groups.
In our cohort, the UTUC group included a significantly lower number of patients with histologically variant UC in comparison to the LTUC group (p=0.034). In our previous subanalysis of histological variants in patients treated with pembrolizumab, there were no significant differences in the objective response rate or disease control rate between patients with variant histological UC and those with pure UC (23). Another report described a multicenter retrospective study of 519 patients with advanced UCs who received ICIs. The ORR to ICI therapy in patients with pure US was comparable to that in patients with variant UC (28% vs. 29%, p=0.90), without a significant difference between histological variants and pure urothelial carcinoma. The median OS for patients with pure UC was 11.0 months, whereas that of patients with variant UC was 10.1 months (p=0.60) (24). These results suggest that histological variants did not affect the efficacy of pembrolizumab.
Various studies have shown that the renal function declines more significantly following surgical treatment in UTUC patients undergoing radical nephroureterectomy than in LTUC patients undergoing radical cystectomy (25-27). In a comparative study (11) of metastatic UTUC versus bladder cancer in patients treated with cisplatin-based chemotherapy, the eGFR in UTUC patients (56.8±20.4 ml/min/1.73 m2) was lower than that in LTUC patients (64.4±24.8 ml/min/1.73 m2), although the difference between the groups was not statistically significant. Thus, UTUC patients may have a more unfavorable renal function in comparison to LTUC patients, even in those with metastatic disease. In our study, the patients with UTUC had significantly lower eGFR values than those with LTUC. However, the difference does not seem to be affected by the response to pembrolizumab, as shown by the results from subgroup analyses in several clinical trials of ICIs, as mentioned previously. Furthermore, our previous study revealed the same safety profile and oncological outcomes between patients with creatinine clearance (CrCl) ≥30 and those with CrCl <30 (28). These results indicate that renal function does not affect the treatment efficacy or tolerability of ICIs, regardless of the location of the primary tumor.
The present study has limitations in addition to its retrospective nature and the fact that almost all of the patients were East Asian. Due to the multi-institutional nature of this retrospective study, there was no standardized follow-up schedule, and treatment decisions were not standardized across the study, such as when to discontinue pembrolizumab and third- or more-line treatment after pembrolizumab failure. AE profiles were retrospectively collected from medical charts at each institution, which might have caused an information bias. However, in Japan, only pembrolizumab has been approved with proven efficacy in second-line treatment for advanced UC, so almost all chemo-resistant UC patients who were eligible for ICI treatment received pembrolizumab as their next treatment. Despite these limitations, the results of this study are informative and helpful when pembrolizumab is considered for use in the treatment of UTUC or LTUC.
Conclusion
Our study showed no significant differences in the oncological outcomes of pembrolizumab between the UTUC and LTUC groups. However, patients with renal pelvic UC had a poorer prognosis than those with ureteral UC and LTUC. This result might be due to the high frequency of visceral and liver metastasis in patients with renal pelvic UC in comparison to the other two groups. Although patients with UTUC had a lower eGFR than patients with LTUC, the incidence of AEs appeared to be similar between the UTUC and LTUC groups. A large prospective study is needed to validate these results.
Footnotes
Authors’ Contributions
Study concepts: Naotaka Nishiyama, Takashi Kobayashi, Hiroshi Kitamura. Study design: Naotaka Nishiyama, Takashi Kobayashi, Hiroshi Kitamura. Data acquisition: Naotaka Nishiyama, Yuki Kita, Minoru Kato, Shingo Hatakeyama, Yuto Matsushita, Sei Naito, Makito Miyake, Shotaro Nakanishi, Yoichiro Kato, Tadamasa Shibuya, Tetsutaro Hayashi, Hiroaki Yasumoto, Takashi Yoshida, Motohide Uemura, Rikiya Taoka. Quality control of data and algorithms: Takashi Kobayashi. Data analysis and interpretation: Naotaka Nishiyama, Takashi Kobayashi, Yuki Kita. Statistical analysis: Naotaka Nishiyama Yuki Kita. Manuscript preparation: Naotaka Nishiyama, Takashi Kobayashi, Yuki Kita. Manuscript editing: Takashi Kobayashi, Hiroshi Kitamura. Manuscript review: Takashi Kobayashi, Hiroyuki Nishiyama.
Conflicts of Interest
Hiroshi Kitamura reports honoraria and research expenses from MSD. The other Authors have no conflicts of interest to declare in relation to this study.
- Received July 19, 2023.
- Revision received September 13, 2023.
- Accepted September 14, 2023.
- Copyright © 2023 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.









