Abstract
Objective
To elucidate the clinical, pathological, and prognostic characteristics of invasive pyelocaliceal urothelial carcinoma (pUC) with venous thrombus (pUC-VT).
Methods
The study included 23 pUC-VT patients and 88 pUC patients without venous thrombus (VT) from March 2016 to April 2022. The features of pUC-VT were detailed and were compared with those without VT. Normally distributed continuous variables were presented as mean±standard deviation; non-normally distributed variables were expressed as median (interquartile range); categorical variables were described as n (%). Propensity scores were calculated by age-adjusted Charlson comorbidity index level and tumor stage. After matching, differences in variables were compared using the paired t-test, Wilcoxon signed-rank test, and McNemar test. The Kaplan–Meier method and Gehan–Breslow test were used to compare overall survival (OS), cancer-specific survival (CSS), and progression-free survival (PFS) curves of pUC-VT patients and pUC patients without VT. Univariate and multivariate Cox proportional hazards regression models were used to determine whether VT was an independent risk factor. The landmark analysis was applied to explore the role of VT in different time periods.
Results
After matching, all 23 pUC-VT patients were paired with pUC patients without VT, revealing significant differences only in open surgery rates (48% vs. 4.3%, p=0.004), median intraoperative blood loss (400.0 mL vs. 80.0 mL, p=0.009), median red blood cell transfusion volume (400.0 mL vs. 0 mL, p=0.001), and median tumor diameter (7.5 cm vs. 5.0 cm, p<0.001). The median OS, CSS, and PFS for pUC-VT patients were 7 (95% confidence interval [CI] 3–20) months, 9 (95% CI 3–28) months, and 3 (95% CI 1–11) months, respectively. The pUC-VT patients had significantly higher risks for shorter OS (hazard ratio [HR] 2.120, p=0.008), CSS (HR 2.163, p=0.009), and PFS (HR 3.070, p<0.001) in the first 18 months.
Conclusion
Patients with pUC-VT face a higher risk of early death and recurrence, necessitating multimodal treatment strategies to improve prognosis.
Keywords: Upper tract urothelial carcinoma, Venous thrombus, Radical nephroureterectomy, Propensity-matched, Prognosis
1. Introduction
Upper urinary tract urothelial carcinoma (UTUC) comprises 5%–10% of urothelial carcinoma (UC), including pyelocaliceal UC (pUC) and ureteral UC. Patients with UTUC may benefit from radical nephroureterectomy, but the 5-year cancer-specific survival (CSS) is less than 50% for pT2/pT3 and under 10% for pT4 UTUC, with pUC patients generally having a better prognosis than those with ureteral UC [[1], [2], [3]]. Invasive pUC, classified into pT3 and pT4 stages, can infiltrate the renal parenchyma and the venous system, though pUC with venous thrombus (pUC-VT) is quite rare. The prevalence and pathogenesis of pUC-VT are not yet clear. Tian et al. [4] reported a case series on pUC-VT, concluding that radical nephroureterectomy with thrombectomy is a viable approach, despite the potential for a poor prognosis. To date, no study has assessed post-surgical survival in pUC-VT patients.
This study aimed to outline the clinical, surgical, pathological, and prognostic features of pUC-VT. We conducted a propensity score-matched retrospective cohort study comparing oncological outcomes between pUC-VT patients and those with pUC only. We hypothesized that pUC-VT patients exhibit unique prognostic characteristics compared to pUC patients without venous thrombus (VT).
2. Patients and methods
2.1. Data source and study cohort
We retrospectively reviewed medical records of pUC-VT patients and pUC patients without VT from March 2016 to April 2022 in the Department of Urology, Peking University Third Hospital. Demographic, perioperative, and follow-up data were meticulously recorded. Inclusion criteria were: (1) pathologically confirmed, initially diagnosed pUC with or without VT; (2) absence of hemorrhagic disease or severe liver dysfunction. Exclusion criteria were: (1) prior history of UC or concomitant UC in the bladder; (2) end-stage renal disease (defined as baseline estimated glomerular filtration rate <15 mL/min/1.73 m2 or baseline creatinine >442 μmol/L). All patients were highly suspected of UTUC through preoperative imaging procedures or urine cytology. This study received ethics approval from the Institutional Review Board of Peking University Third Hospital (approval number M2019404).
2.2. Surgical procedure
Nephroureterectomy was performed in all patients, while venous thrombectomy was done in those with renal vein or inferior vena cava (IVC) tumor thrombus. Ciancio et al. [5] previously reported using a liver transplantation approach for managing pUC-VT. At Peking University Third Hospital, we adapted the surgical techniques typically used for renal cell carcinoma (RCC) with VT to treat pUC-VT. Specifically, the renal artery was dissected first to reduce blood loss and facilitate kidney mobilization. We then skeletonized the renal vein or IVC to achieve full vascular control before performing the thrombectomy. Detailed thrombus management procedures in our center were thoroughly described by Tian et al. [4]; the lymph node dissection was done in those with preoperative or intraoperative suspected metastatic lymph nodes, and the dissecting areas were referred to standard templates [6]. Dissection regions included hilar, paracaval, retrocaval, interaortocaval, and para-aortic lymph nodes. The above procedures were performed using an open, laparoscopic, or robot-assisted laparoscopic (minimally invasive) approach.
2.3. Follow-up
Routine imaging and laboratory tests were conducted during follow-up visits, with data collected regularly by two data managers (Song Y and Sun Y). Follow-up visits or phone calls occurred every 3 months in the first year post-surgery, every 6 months up to the third year, and annually thereafter. Causes of death were determined from death certificates issued by local medical institutions or treating physicians if patients died during follow-up visits [7].
2.4. Variables and outcomes
Clinical, surgical, and pathological features and oncological outcomes of patients were described. The age-adjusted Charlson Comorbidity Index (aCCI) was calculated to evaluate underlying disease burden [8]. The American Society of Anesthesiologists Physical Status (ASA-PS) classification was used to estimate the operative risks [9]. The Mayo Clinic Classification System was used to assess thrombus level with reference to its application in RCC [10]. Postoperative complications were graded using the Clavien–Dindo system [11]. Complications within 30 days post-surgery were evaluated. Overall survival (OS), CSS, and progression-free survival (PFS) represented oncologic outcomes. pUC patients without VT served as a control group to elucidate prognostic features of pUC-VT patients.
2.5. Statistical analysis
Baseline characteristics were presented as categorical and continuous variables. Missing data were imputed using a random forest method. Categorical variables were presented as numbers and percentages. The Shapiro–Wilk test was used for the normal distribution test. Normally distributed continuous variables were presented as mean±standard deviation, or presented as median (interquartile range) in cases of non-normally distributed data. Differences in normally distributed continuous data were compared using the t-test. Non-normally distributed continuous data were assessed using the Wilcoxon rank-sum test. Categorical data were compared using the X2 test or Fisher's exact test. Propensity scores were calculated using a multivariable logistic regression model, with 1:1 matching performed without replacement and a caliper width of 0.02. Matching criteria included the aCCI level (level 1: 0–4, level 2: ≥5) and tumor stage (the American Joint Committee on Cancer 8th edition, stage III: T3N0M0, stage IV: T4N0M0, TanyN1–2M0, or TanyNanyM1). Matching factors were considered to be balanced in the matched cohort if the standardized mean difference value was less than 0.1. Differences of variables in the matched cohort were compared using the paired t-test, Wilcoxon signed-rank test, or McNemar test. The survival analysis was conducted using the Kaplan–Meier method in the matched cohort. As survival curves of the two groups were separated more rapidly at the initial phase, the Gehan–Breslow test rather than the log-rank test was used to compare differences in the survival curves. The univariate and multivariate Cox proportional hazards regression analyses with a robust sandwich estimator [12] were performed to determine whether VT was an independent risk factor for worse survival outcomes in pUC patients. The proportional hazards assumption was tested for VT (exposure factor) and confounding factors included in the multivariate Cox regression analysis. Confounding factors were selected from those still showing significant differences after propensity score matching (PSM) and the selection followed the 10 events per variable principle [13]. Though survival curves of the pUC-VT group decreased more rapidly at the beginning, those of the two groups eventually crossed over. Thus, a landmark analysis was performed [14]. The landmark was selected according to the crossing points of the curves and set to 18 months. All statistical tests were performed using R version 4.2.1 (Posit Inc., Boston, MA, USA). All tests were two-sided with a significance level set at p-value <0.05
3. Results
A total of 111 patients were included in the analysis, comprising 23 patients with pUC-VT and 88 patients with pUC only. After PSM, all 23 pUC-VT patients were matched with pUC patients without VT with a 1:1 matching ratio (Fig. 1). The two matching factors were both balanced between the two groups in the matched cohort, with standard mean differences less than 0.001 (Supplementary Table 1). Typical clinical imaging and pathological findings of a patient with pUC-VT and ipsilateral adrenal metastasis were demonstrated in Figure 2, Figure 3. Baseline characteristics of pUC-VT patients and pUC patients without VT were summarized in Table 1.
Figure 1.
Schematic diagram of inclusion and exclusion criteria. UC, urothelial carcinoma; pUC, pyelocaliceal UC; VT, venous thrombus; pUC-VT, pUC with VT; PSM, propensity score matching.
Figure 2.
Contrast-enhanced CT images from a patient with left pUC and Mayo grade II inferior vena cava thrombus. (A) The axial unenhanced CT scan showed a poorly defined isodense mass in the renal medulla (attenuation: 30 HU, within the yellow circle); (B) The contrast-enhanced CT scan (corticomedullary phase) showed mild enhancement (yellow circle in A) less than that of the cortex and medulla (67 HU); (C) The tumor region (yellow circle in A) increased to 92 HU on the nephrogenic phase; (D) The tumor region decreased to 64 HU on the delayed excretory phase; (E) The coronal enhanced CT scan showed the inferior vena cava thrombus (red arrow) and adrenal metastasis (red arrowhead). pUC, pyelocaliceal urothelial carcinoma; HU, Hounsfield Units.
Figure 3.
Hematoxylin-eosin staining images from a patient with left pUC and Mayo grade II inferior vena cava thrombus. (A) The microscopic analysis showed that cancer cells infiltrated the renal parenchyma; (B) The tumor thrombus involved the inferior vena cava wall (yellow arrow) and infiltrated the muscularis (yellow rectangular box); (C) Enlarged image of the yellow rectangular box in (B) showed tumor thrombus infiltrating the muscularis of the venous wall; (D) Ipsilateral adrenal metastasis was confirmed microscopically. pUC, pyelocaliceal urothelial carcinoma.
Table 1.
Baseline characteristics of pUC patients with VT and non-VT before and after propensity matching.
| Baseline characteristic | Before PSM |
After PSM |
||||
|---|---|---|---|---|---|---|
| With VT (n=23) | Non-VT (n=88) | p-Value | With VT (n=23) | Non-VT (n=23) | p-Value | |
| Clinical feature | ||||||
| Age, mean±SD, year | 63.9±12.3 | 70.4±9.6 | 0.008∗∗ | 63.9±12.3 | 68.2±8.8 | 0.12 |
| Gender, n (%) | 0.084 | 0.3 | ||||
| Male | 7 (30) | 47 (53) | 7 (30) | 12 (52) | ||
| Female | 16 (70) | 41 (47) | 16 (70) | 11 (48) | ||
| BMI, n (%) | 1 | 0.6 | ||||
| <24 kg/m2 | 11 (48) | 42 (48) | 11 (48) | 8 (35) | ||
| ≥24 kg/m2 | 12 (52) | 46 (52) | 12 (52) | 15 (65) | ||
| aCCI, median (IQR) | 3.0 (3.0–4.0) | 4.0 (3.0–4.0) | 0.2 | 3.0 (3.0–4.0) | 4.0 (3.0–4.0) | 0.2 |
| ASA-PS class, n (%) | 1 | 1 | ||||
| 1–2 | 3 (13) | 12 (14) | 3 (13) | 4 (17) | ||
| 3 | 20 (87) | 76 (86) | 20 (87) | 19 (83) | ||
| Preoperative SCR, median (IQR), μmol/L | 110.0 (92.0–121.5) | 88.5 (75.8–113.8) | 0.018∗ | 110.0 (92.0–121.5) | 95.0 (77.0–119.5) | 0.073 |
| Laterality, n (%) | 0.045∗ | 0.096 | ||||
| Left | 5 (22) | 42 (48) | 5 (22) | 12 (52) | ||
| Right | 18 (78) | 46 (52) | 18 (78) | 11 (48) | ||
| Mayo grade, n (%) | NA | NA | ||||
| 0 | 12 (52) | NA | 12 (52) | NA | ||
| I | 3 (13) | NA | 3 (13) | NA | ||
| II | 8 (35) | NA | 8 (35) | NA | ||
| Surgical feature | ||||||
| Surgical approach, n (%) | <0.001∗∗∗ | 0.004∗∗ | ||||
| Minimally invasive | 12 (52) | 86 (98) | 12 (52) | 22 (96) | ||
| Open | 11 (48)a | 2 (2.3) | 11 (48)a | 1 (4.3) | ||
| Operative time, median (IQR), min | 284.0 (210.0–332.0) | 197.0 (166.8–257.8) | 0.001∗∗ | 284.0 (210.0–332.0) | 265.0 (205.5–298.0) | 0.14 |
| Blood loss, median (IQR), mL | 400.0 (200.0–800.0) | 50.0 (20.0–100.0) | <0.001∗∗∗ | 400.0 (200.0–800.0) | 80.0 (50.0–250.0) | 0.009∗∗ |
| RBC transfusion, median (IQR), mL | 400.0 (0–800.0) | 0 (0–0) | <0.001∗∗∗ | 400.0 (0–800.0) | 0 (0–0) | 0.001∗∗ |
| Complication grade, n (%) | 0.078 | 0.8 | ||||
| 0–I | 16 (70) | 77 (88) | 16 (70) | 18 (78) | ||
| II–IVa | 7 (30) | 11 (12) | 7 (30) | 5 (22) | ||
| Postoperative hospital stay, median (IQR), day | 8.0 (6.0–9.5) | 6.0 (5.0–8.0) | 0.029∗ | 8.0 (6.0–9.5) | 7.0 (6.0–8.5) | 0.3 |
| Pathological feature | ||||||
| Tumor diameter, median (IQR), cm | 7.5 (6.2–9.1) | 3.6 (2.5–5.0) | <0.001∗∗∗ | 7.5 (6.2–9.1) | 5.0 (3.3–6.0) | <0.001∗∗∗ |
| pT stage, n (%) | <0.001∗∗∗ | 0.11 | ||||
| pT3 | 5 (22) | 76 (86) | 5 (22) | 11 (48) | ||
| pT4 | 18 (78) | 12 (14) | 18 (78) | 12 (52) | ||
| pN+, n (%) | 10 (43) | 15 (17) | 0.015∗ | 10 (43) | 14 (61) | 0.3 |
| M1 stage, n (%) | 6 (26) | 7 (8.0) | 0.041∗ | 6 (26) | 7 (30) | 1 |
| Nuclear grade, n (%) | 0.6 | 1 | ||||
| Low | 0 (0) | 5 (5.7) | 0 (0) | 1 (4.3) | ||
| High | 23 (100) | 83 (94) | 23 (100) | 22 (96) | ||
| Squamous differentiation, n (%) | 3 (13) | 10 (11) | 1 | 3 (13) | 4 (17) | 1 |
| Sarcomatoid differentiation, n (%) | 6 (26) | 4 (4.5) | 0.005∗∗ | 6 (26) | 1 (4.3) | 0.13 |
| Intratumoral necrosis, n (%) | 9 (39) | 21 (24) | 0.2 | 9 (39) | 10 (43) | 1 |
pUC, pyelocaliceal urothelial carcinoma; VT, venous thrombus; non-VT, without venous thrombus; BMI, body mass index; aCCI, age-adjusted Charlson Comorbidity Index; RBC, red blood cell; SCR, serum creatinine; pN+, patients with pathologically confirmed metastatic lymph nodes; PSM, propensity score matching; M1, distant metastasis; NA, not applicable.
One pUC patient with VT was transferred from laparoscopic surgery to open surgery.
∗p<0.05, ∗∗p<0.01, ∗∗∗p<0.001.
3.1. Clinical features and thrombus status
No significant differences were observed in gender, body mass index, aCCI, or ASA-PS class both before and after PSM. Differences in age, tumor laterality, and preoperative serum creatinine were no longer significant after PSM.
In the pUC-VT group, VT in 52% of the patients was restricted within the renal vein (Mayo grade 0), while others protruded into the IVC; the median length of thrombus was 4.5 (interquartile range 2.3–6.4) cm; thrombus in eight (35%) patients invaded the venous wall; three of them (38%) had IVC thrombus and underwent IVC wall segmental resection, as the venous wall was circumferentially invaded by the thrombus (Supplementary Table 2).
3.2. Surgical and pathological features
There were a significantly higher proportion of open surgery, greater intraoperative blood loss, and higher red blood cell transfusion volume in pUC-VT patients compared with pUC patients without VT both before and after PSM. Operative time and postoperative hospital stay exhibited significant differences only before PSM and no significant differences after PSM, while complication grades showed no significant differences between the two groups both before and after PSM.
Except for tumor diameter (p<0.001), pathological features (pT stage [p=0.11], positive lymph node status [p=0.3], distant metastasis [p=1], nuclear grade [p=1], squamous differentiation [p=1], sarcomatoid differentiation [p=0.13], and intratumoral necrosis [p=1]) were all comparable between pUC-VT and pUC patients without VT after PSM, though most of these features showed significant differences before PSM.
A subgroup of patients with positive lymph nodes in pathological examination in the matched cohort (n=24) was analyzed (Supplementary Table 3). In this subgroup, the inter-aortocaval lymph node dissection ratio was significantly higher in pUC-VT patients (6/10) than in pUC patients without VT (0/14) (p=0.004). Moreover, pUC-VT patients had a greater number of metastatic lymph nodes (p<0.001) and higher proportion of pN2 stage than pUC patients without VT (p=0.002), both with significant differences.
3.3. Oncological outcomes
During the follow-up period, nine out of 23 (39%) patients in the pUC-VT group and the pUC without VT group (p=1) received adjuvant chemotherapy, radiotherapy, or immunotherapy (not listed in the tables). As most of these patients received adjuvant therapies outside our institution, the detailed regimen of treatment was not available.
For OS, the pUC-VT group demonstrated a median survival time of 7 (95% confidence interval [CI] 3–20) months with 1-year and 3-year survival rates of 35% and 6.5%, respectively. In contrast, the pUC without VT group showed a median OS of 12 (95% CI 6–18) months with 1-year and 3-year survival rates of 44% and 15%, respectively, after surgery. Regarding CSS, the pUC-VT group exhibited a median survival time of 9 (95% CI 3–28) months with 1-year and 3-year survival rates of 38% and 7.1%, respectively. The pUC without VT group maintained a median CSS time of 12 (95% CI 6–18) months accompanied by 46% and 16% survival rates at corresponding timepoints. For PFS, the pUC-VT group displayed a median PFS time of 3 (95% CI 1–11) months while that of the pUC without VT group was 10 (95% CI 5–13) months. The 1-year and 3-year PFS rates were 17% versus 32% and 8.7% versus 16% for pUC-VT versus pUC without VT, respectively (Supplementary Table 4). There were 30–40 events that occurred in OS, CSS, and PFS in the matched cohort, and thus three variables—VT, tumor diameter, and blood loss volume—were selected as factors in the multivariate analysis. Results of proportional hazards tests were summarized in Supplementary Tables 5 and 6.
As the overall survival of 43/46 (93%) patients in the matched cohort was less than 3 years, and the remaining three patients (one in the pUC-VT group, follow-up time: 85 months; two in the pUC without VT group, follow-up time: 102 and 103 months) were still alive and did not show evidence of disease progression at the latest follow-up, a 3-year survival analysis was done. Fig. 4A–C illustrate the 3-year Kaplan–Meier survival curves for OS, CSS, and PFS. Only PFS curves of the two groups showed a significant difference (Gehan–Breslow test, p=0.030).
Figure 4.
Survival curves and landmark survival analyses of pUC patients with VT and non-VT in the matched cohort. (A–C) 3-year survival curves: OS (A), CSS (B), and PFS (C); Landmark survival curves (18-month follow-up cutoff): OS (D), CSS (E), and PFS (F). OS, overall survival; CSS, cancer-specific survival; PFS, progression-free survival; pUC, pyelocaliceal urothelial carcinoma; VT, venous thrombus; non-VT, without venous thrombus; HR, hazard ratio; CI, confidence interval; NA, not applicable.
In the univariate and multivariate analyses, VT was not an independent risk factor for worse survival outcome (p>0.05) (Table 2). However, in the landmark analysis, VT served as an independent risk factor for worse OS, CSS, and PFS in the first 18 months (Fig. 4D, E, and F, Table 3).
Table 2.
The univariate and multivariate Cox proportional hazards regression analyses in pUC patients with VT and non-VT in the matched cohort.
| Cox regression model | OS |
CSS |
PFS |
|||
|---|---|---|---|---|---|---|
| HR (95% CI) | p-Value | HR (95% CI) | p-Value | HR (95% CI) | p-Value | |
| Univariate model | ||||||
| VT vs. non-VT | 1.253 (0.679–2.314) | 0.5 | 1.255 (0.667–2.359) | 0.5 | NAa | NAa |
| Multivariate model | ||||||
| VT vs. non-VT | 0.839 (0.425–1.658) | 0.6 | 0.776 (0.403–1.496) | 0.5 | 1.212 (0.598–2.457) | 0.6 |
| Tumor diameter | 1.070 (0.987–1.160) | 0.10 | 1.084 (1.005–1.169) | 0.036∗ | 1.068 (0.968–1.179) | 0.19 |
| Blood loss | 1.001 (1.000–1.001) | 0.018∗ | 1.001 (1.000–1.001) | 0.002∗∗ | 1.001 (1.000–1.001) | 0.075 |
VT, venous thrombus; non-VT, without venous thrombus; pUC, pyelocaliceal urothelial carcinoma; OS, overall survival; CSS, cancer-specific survival; PFS, progression-free survival; HR, hazard ratio; CI, confidence interval; NA, not applicable.
Proportional hazards assumption was violated (proportional hazards test, p <0.05).
∗p<0.05, ∗∗p<0.01.
Table 3.
The univariate and multivariate Cox proportional hazards regression analyses in pUC patients with VT and non-VT in the matched cohort (Landmark analysis).
| Time period | Cox regression model | OS |
CSS |
PFS |
|||
|---|---|---|---|---|---|---|---|
| HR (95% CI) | p-Value | HR (95% CI) | p-Value | HR (95% CI) | p-Value | ||
| ≤18 months | • Univariate model | ||||||
| - VT vs. non-VT | 2.120 (1.212–3.707) | 0.008∗∗ | 2.163 (1.214–3.853) | 0.009∗∗ | 3.070 (1.587–5.938) | 0.001∗∗∗ | |
| • Multivariate model | |||||||
| - VT vs. non-VT | 2.006 (1.085–3.706) | 0.026∗ | 1.939 (1.071–3.511) | 0.029∗ | 3.103 (1.244–7.744) | 0.015∗ | |
| - Tumor diameter | 0.988 (0.898–1.086) | 0.8 | 0.999 (0.915–1.090) | 0.982 | NAa | NAa | |
| - Blood loss | 1.000 (1.000–1.001) | 0.4 | 1.000 (1.000–1.001) | 0.183 | 1.000 (1.000–1.001) | 0.5 | |
| >18 months | • Univariate model | ||||||
| - VT vs. non-VT | 3.034 (0.405–22.700) | 0.3 | 3.034 (0.405–22.700) | 0.3 | NAb | NA | |
| • Multivariate model | NAc | NAc | NAc | NA | |||
| - VT vs. non-VT | |||||||
| - Tumor diameter | |||||||
| - Blood loss | |||||||
VT, venous thrombus; non-VT, without venous thrombus; pUC, pyelocaliceal urothelial carcinoma; OS, overall survival; CSS, cancer-specific survival; PFS, progression-free survival; HR, hazard ratio; CI, confidence interval; NA, not applicable.
Proportional hazards assumption was violated (proportional hazards test, p <0.05).
The number of events was too small, leading to an exceedingly high HR value.
The number of events was <10. The multivariate analysis could not be executed according to 10 events per variable principle.
∗p<0.05, ∗∗p<0.01, ∗∗∗p<0.001.
4. Discussion
Radical nephroureterectomy with bladder cuff excision is the standard treatment for high-risk non-metastatic UTUC and is considered to improve prognosis [3]. Although UTUC can invade the venous system, VT is more commonly associated with RCC [10,15]. The presence of VT in patients with pUC is generally considered a poor prognostic factor. Previous case series have described surgical experiences with pUC-VT [4,5].
In this study, we characterized the clinical, pathological, and prognostic features of pUC-VT patients and compared them to pUC patients without VT. Notably, there were more female patients in the pUC-VT group. All tumor thrombi were below the level of the hepatic vein (Mayo grades 0–II). Most of the baseline variables that showed significant differences, including common risk factors such as age, clinical stage, and sarcomatoid differentiation, were comparable between the two groups after PSM. We found that pUC-VT patients presented with a higher open surgery rate, suffered more blood loss, and had a greater tumor size, indicating greater difficulty in their surgical procedures. pUC-VT patients were also subjected to a wider range of lymph node dissection and had more positive lymph nodes in pathological examination. However, the rate of high-grade postoperative complications (Clavien–Dindo grade ≥2) was comparable between the two groups after PSM (pUC-VT vs. pUC without VT: 30% vs. 22%, p=0.8).
One of the key challenges in managing pUC-VT is distinguishing it from RCC with VT. Compared to RCC, pUC typically presents with ill-defined margins, medullary involvement, and homogeneous enhancement that is less than that of the renal cortex and medulla across all CT phases [16]. Once diagnosed, radical nephroureterectomy with thrombectomy is the recommended treatment. In our study, although there were no Mayo grades III–IV thrombi that needed complicated liver mobilization and management of thrombus above the diaphragm, there was a high proportion of perirenal wall invasion (pT4 stage, 78%), venous wall invasion (35%), and large tumor size (7.5 cm), with heavy perirenal adhesion in most cases, which greatly added to surgical complexity. Nevertheless, the complication rate and length of postoperative hospital stay showed no significant differences between the two groups, demonstrating that nephroureterectomy with thrombectomy is safe and feasible, yielding acceptable surgical outcomes.
To date, the long-term survival of patients with pUC-VT has not been comprehensively evaluated. In our study, the 3-year CSS rate for pUC-VT patients was only about 7.1%. Though not statistically significant, pUC-VT patients had a shorter median OS, CSS, and PFS than pUC patients without VT. Cox regression models in the landmark survival analysis indicated that pUC-VT patients might have a worse oncological outcome in the early stage after surgery than pUC patients without VT. Several factors may contribute to this poor prognosis. First, VT likely results from tumor cells entering the venous system after lymphovascular invasion, which is an independent predictor of poor prognosis in UTUC patients [1,17]. Additionally, 80% of the patients in the pUC-VT group had perinephric fat invasion, defined as pT4. Besides, all of them were high-grade invasive carcinomas. Last but not least, not all patients received postoperative adjuvant therapy. The primary limitation of adjuvant chemotherapy in these cases was the reduced ability to tolerate cisplatin-based regimens due to impaired renal function and overall physical status following extensive surgery.
Multimodal treatment strategies are essential for pUC patients. Adjuvant chemotherapy has been shown to improve disease-free survival and OS in UTUC patients with pT2–T4 disease and/or lymph node metastasis [18,19]. In our study, only eight patients with pUC-VT received postoperative adjuvant therapy, likely reflecting the challenges posed by radical surgery, which can impair tolerance to subsequent chemotherapy. Based on our clinical observations, the primary reason for this limited use was contralateral renal dysfunction. Neoadjuvant chemotherapy has been shown to downstage tumors and may provide OS benefits in high-risk UTUC patients undergoing radical nephroureterectomy [20]. However, no studies to date have assessed the efficacy of neoadjuvant chemotherapy specifically in pUC-VT. At our center, we do not routinely administer neoadjuvant chemotherapy due to concerns about potential renal toxicity, but this approach warrants further investigation.
Another promising strategy that may improve prognosis is neoadjuvant immunotherapy. Although no data are currently available for UTUC, more than 100 clinical trials involving neoadjuvant anti-programmed death-1 or programmed death-ligand 1 therapy (either as monotherapy or in combination) are ongoing or planned. Neoadjuvant immunotherapy is thought to enhance the detection and elimination of micro-metastatic tumor deposits, which are often the source of postsurgical relapse. By leaving the primary tumor in place, neoadjuvant immunotherapy leverages higher levels of endogenous tumor antigens to prime T cells and generate a systemic anti-tumor response [21]. Although its efficacy in pUC-VT has yet to be established, we encourage its exploration in clinical trials.
To the best of our knowledge, our findings provide initial insights into this rare clinical presentation and highlight its poor oncological outcomes. However, our study has several limitations. The most significant limitation is its retrospective, non-randomized design. Details on some prognostic factors such as adjuvant therapy were not available for a great number of patients. As our institution is one of the largest centers for the surgical treatment of RCC with VT in China, rare pUC-VT patients, likely to be initially diagnosed as RCC-VT, had a higher probability of receiving surgical treatment in our department. That may partly contribute to a high percentage of pUC-VT patients in the whole study population, leading to selection bias. Although propensity matching was made between the two groups, to make sure all pUC-VT patients could be matched with pUC patients without VT, the matching procedure was only done at a 1:1 ratio and only two essential factors were used for matching, which led the study to be subject to other confounders that could affect the oncological outcomes and limit internal validity. Additionally, the sample size of the pUC-VT group was still small, limiting the ability to fully assess their features. Further studies with larger cohorts are needed to validate our findings.
5. Conclusion
Surgical resection remains a viable curative option for patients with pUC-VT. Nephroureterectomy with thrombectomy is feasible, with acceptable surgical outcomes. However, pUC-VT patients might have a higher risk of death and recurrence in the early period compared with pUC patients without VT. We propose that future clinical trials should explore the role of neoadjuvant chemotherapy or immunotherapy in this specific patient population.
Author contributions
Study concept and design: Zhanyi Zhang, Yu Zhang, Hongxian Zhang, LuLin Ma.
Data acquisition: Ruiyang Xie, Min Lu, Yimeng Song, Yuehao Sun.
Data analysis: Zhanyi Zhang, Shaohui Deng, Yu Zhang, Ruiyang Xie, Min Lu.
Drafting of the manuscript: Zhanyi Zhang, Shaohui Deng, Yu Zhang.
Critical revision of the manuscript: Hongxian Zhang, Shudong Zhang, LuLin Ma.
Conflicts of interest
The authors declare no conflict of interest.
Footnotes
Peer review under responsibility of Tongji University.
Supplementary data to this article can be found online at https://doi.org/10.1016/j.ajur.2025.03.019.
Contributor Information
Hongxian Zhang, Email: zwl2006@163.com.
Shudong Zhang, Email: zhangshudong@bjmu.edu.cn.
Appendix A. Supplementary data
The following is the Supplementary data to this article:
References
- 1.Lughezzani G., Burger M., Margulis V., Matin S.F., Novara G., Roupret M., et al. Prognostic factors in upper urinary tract urothelial carcinomas: a comprehensive review of the current literature. Eur Urol. 2012;62:100–114. doi: 10.1016/j.eururo.2012.02.030. [DOI] [PubMed] [Google Scholar]
- 2.Lughezzani G., Jeldres C., Isbarn H., Sun M., Shariat S.F., Alasker A., et al. Nephroureterectomy and segmental ureterectomy in the treatment of invasive upper tract urothelial carcinoma: a population-based study of 2299 patients. Eur J Cancer. 2009;45:3291–3297. doi: 10.1016/j.ejca.2009.06.016. [DOI] [PubMed] [Google Scholar]
- 3.Roupret M., Babjuk M., Burger M., Capoun O., Cohen D., Comperat E.M., et al. European Association of Urology guidelines on upper urinary tract urothelial carcinoma: 2020 update. Eur Urol. 2021;79:62–79. doi: 10.1016/j.eururo.2020.05.042. [DOI] [PubMed] [Google Scholar]
- 4.Tian X., Hong P., Tang S., Liu Z., Yang F., Zhang S., et al. Urothelial carcinoma of the renal pelvis with renal vein and inferior vena cava tumor thrombus: case series and literature review. Transl Androl Urol. 2021;10:2879–2888. doi: 10.21037/tau-21-253. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Ciancio G., Tabbara M.M., Martucci M., Gaynor J.J., Morsi M., Gonzalez J. Surgical management of upper urinary tract urothelial cell carcinoma with venous tumor thrombus: a liver transplant-based approach. J Clin Med. 2021;10:5964. doi: 10.3390/jcm10245964. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Yanagisawa T., Kawada T., von Deimling M., Laukhtina E., Kimura T., Shariat S.F. Need for and extent of lymph node dissection for upper tract urothelial carcinoma: an updated review in 2023. Curr Opin Urol. 2023;33:258–268. doi: 10.1097/MOU.0000000000001097. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Zhang Y., Bi H., Yan Y., Liu Z., Wang G., Song Y., et al. Comparative analysis of surgical and oncologic outcomes of robotic, laparoscopic and open radical nephrectomy with venous thrombectomy: a propensity-matched cohort study. Int J Clin Oncol. 2023;28:145–154. doi: 10.1007/s10147-022-02265-y. [DOI] [PubMed] [Google Scholar]
- 8.Kluth L.A., Xylinas E., Kent M., Hagiwara M., Kikuchi E., Ikeda M., et al. Predictors of survival in patients with disease recurrence after radical nephroureterectomy. BJU Int. 2014;113:911–917. doi: 10.1111/bju.12369. [DOI] [PubMed] [Google Scholar]
- 9.Hackett N.J., De Oliveira G.S., Jain U.K., Kim J.Y. ASA class is a reliable independent predictor of medical complications and mortality following surgery. Int J Surg. 2015;18:184–190. doi: 10.1016/j.ijsu.2015.04.079. [DOI] [PubMed] [Google Scholar]
- 10.Blute M.L., Leibovich B.C., Lohse C.M., Cheville J.C., Zincke H. The Mayo Clinic experience with surgical management, complications and outcome for patients with renal cell carcinoma and venous tumour thrombus. BJU Int. 2004;94:33–41. doi: 10.1111/j.1464-410X.2004.04897.x. [DOI] [PubMed] [Google Scholar]
- 11.Mitropoulos D., Artibani W., Graefen M., Remzi M., Roupret M., Truss M., et al. Reporting and grading of complications after urologic surgical procedures: an ad hoc EAU guidelines panel assessment and recommendations. Eur Urol. 2012;61:341–349. doi: 10.1016/j.eururo.2011.10.033. [DOI] [PubMed] [Google Scholar]
- 12.Qin J., Ning J., Liu H., Shen Y. Maximum likelihood estimations and EM algorithms with length-biased data. J Am Stat Assoc. 2011;106:1434–1449. doi: 10.1198/jasa.2011.tm10156. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Peduzzi P., Concato J., Kemper E., Holford T.R., Feinstein A.R. A simulation study of the number of events per variable in logistic regression analysis. J Clin Epidemiol. 1996;49:1373–1379. doi: 10.1016/s0895-4356(96)00236-3. [DOI] [PubMed] [Google Scholar]
- 14.Maeng M., Tilsted H.H., Jensen L.O., Krusell L.R., Kaltoft A., Kelbæk H., et al. Differential clinical outcomes after 1 year versus 5 years in a randomised comparison of zotarolimus-eluting and sirolimus-eluting coronary stents (the SORT OUT III study): a multicentre, open-label, randomised superiority trial. Lancet. 2014;383:2047–2056. doi: 10.1016/S0140-6736(14)60405-0. [DOI] [PubMed] [Google Scholar]
- 15.Zhang Y., Tian X., Bi H., Hong P., Liu Z., Yan Y., et al. Oncologic outcomes of renal cell carcinoma patients undergoing radical nephrectomy and venous thrombectomy: prospective follow-up from a single center. J Oncol. 2022;2022 doi: 10.1155/2022/9191659. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Zhu Q., Zhu W., Wu J., Chen W. Multidetector CT imaging features of invasive renal parenchyma urothelial carcinoma. Br J Radiol. 2016;89 doi: 10.1259/bjr.20151068. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Saito K., Kawakami S., Fujii Y., Sakura M., Masuda H., Kihara K. Lymphovascular invasion is independently associated with poor prognosis in patients with localized upper urinary tract urothelial carcinoma treated surgically. J Urol. 2007;178:2291–2296. doi: 10.1016/j.juro.2007.08.019. [DOI] [PubMed] [Google Scholar]
- 18.Birtle A., Johnson M., Chester J., Jones R., Dolling D., Bryan R.T., et al. Adjuvant chemotherapy in upper tract urothelial carcinoma (the POUT trial): a phase 3, open-label, randomised controlled trial. Lancet. 2020;395:1268–1277. doi: 10.1016/S0140-6736(20)30415-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Seisen T., Krasnow R.E., Bellmunt J., Roupret M., Leow J.J., Lipsitz S.R., et al. Effectiveness of adjuvant chemotherapy after radical nephroureterectomy for locally advanced and/or positive regional lymph node upper tract urothelial carcinoma. J Clin Oncol. 2017;35:852–860. doi: 10.1200/JCO.2016.69.4141. [DOI] [PubMed] [Google Scholar]
- 20.Leow J.J., Chong Y.L., Chang S.L., Valderrama B.P., Powles T., Bellmunt J. Neoadjuvant and adjuvant chemotherapy for upper tract urothelial carcinoma: a 2020 systematic review and meta-analysis, and future perspectives on systemic therapy. Eur Urol. 2021;79:635–654. doi: 10.1016/j.eururo.2020.07.003. [DOI] [PubMed] [Google Scholar]
- 21.Topalian S.L., Taube J.M., Pardoll D.M. Neoadjuvant checkpoint blockade for cancer immunotherapy. Science. 2020;367 doi: 10.1126/science.aax0182. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.