Prognosis and risk factors of patients with upper urinary tract urothelial carcinoma and postoperative recurrence of bladder cancer in central China

Upper urinary tract urothelial carcinoma (UTUC) is a relatively rare tumor, accounting for approximately 10% of all renal tumors and only 5% of all urothelial carcinomas [1, 2], with an estimated annual incidence of almost two cases per 100,000 inhabitants in Western countries. The epidemiology and disease presentation of UTUCs in the Chinese population, however, are quite different from those in Western populations. First, UTUCs account for 20–30% of all transitional cell carcinomas (TCCs) in China, which is more common than that in Western countries [3, 4]. Second, UTUCs are more common in women than in men, with a ratio of 1.3:1 [4]. Third, the incidence of renal pelvis urothelial carcinoma is twice more among Chinese populations than among Western populations. TCC of the ureter accounts for more than half of UTUCs [5, 6] and about 25% of all renal carcinomas in China [3]. Finally, contrary to findings in Western countries, UTUC with an advanced stage, large tumor size, and lymph node metastasis (LNM) are less likely to progress in female Chinese patients than in male Chinese patients [7]. This may be associated with diverse environmental factors, such as pure-arsenic exposure in drinking water and consumption of aristolochic acid in Chinese herbs [3, 8].

However, the epidemiology and biological characteristics of UTUCs in the Chinese population have not gained research focus, and data are limited. Moreover, it remains unclear whether the following factors are related to prognosis and postoperative recurrence of bladder cancer (BC) in UTUC patients: diabetes mellitus (DM), metformin use in DM patients, preoperative ureteroscopy, and location of UTUC. Currently, radical nephroureterectomy (RNU) is the gold standard for treating UTUCs [9]. The incidence of intravesical tumor recurrence after operation is 20–50% [10]. Notably, 44% of patients with BC recurrence have invasive disease (≥pT1 stage), and 82–89% of BC recurrences occur within 2 years after RNU [11]. Hence, identifying high risk factors of BC recurrence after RNU is a major concern for UTUC patients.

Although a few studies have assessed predictive or risk factors of UTUCs and/or BC recurrence in Chinese people [3, 12], data were mainly obtained from the population of the North and Southeast of China, not from the central areas of China. Since the epidemiology and disease characteristics may be related to different areas, along with factors influencing survival and tumor recurrence remain controversial, we developed a database of UTUC in our region to investigate the prognostic risk factors of UTUC and BC recurrence in central China. We hope that our research will contribute to UTUC treatment options and reduce the risk of recurrence of BC after RNU.

Overall, 439 patients were included in the final statistical analysis. Among the 439 patients, 236 were males (53.7%) and 203 were females (46.3%), and the average age was 66.7 years. The follow-up period ranged from 18 to 84 months, with an average of 62.5 months. Factors included in the analysis are shown in Table 1. Tumor-associated death, BC recurrence rate, survival rate, risk factors for UTUCs, and BC recurrence were analyzed to predict the prognosis of patients with UTUC.

In this study, we found that smoking, consuming Chinese medicine containing aristolochic acid, history of BC, old age, advanced tumor stage, presence of high-grade tumor, larger tumor size, and LNM were predictive factors for worse survival of UTUC. In addition, tumor stage, tumor grade, preoperative ureteroscopy, UUC, and DM without metformin use were risk factors of BC recurrence, whereas regular postoperative bladder perfusion chemotherapy and DM with metformin use were protective factors for BC relapse. Because the accurate prediction of the prognosis of UTUC patients could contribute to risk stratification and devising therapeutic options for urologists or oncologists, we analyzed each predictor in detail.

The incidence of UTUCs has been increasing each year, and 60% of UTUCs are found to have already developed into invasive tumors at diagnosis compared with 15–25% of bladder tumors [1, 13]. UTUCs have a peak incidence in individuals aged 70–90 years and are three times more common in men in Western countries [1, 14]; these findings differed from those of our study in which the male-to-female ratio was 1:1.16. Previous studies showed that the survival rate of UTUC patients was related to the tumor stage and grade [4]. A multifactorial analysis of 252 UTUC cases by Hall et al. [15] revealed that tumor stage was the only indicator of postoperative survival. A study of 434 patients with UTUCs by Munoz and Ellison [16] showed that the 5-year survival rate of patients with Tis tumors was 95.1% and that of patients with local tumors was 88.9%; however, the 5-year survival rate of patients with distant metastasis was only 16.5%. In our study, tumor stage and grade were risk factors influencing the survival of postoperative UTUC patients. A significant difference in 5-year survival rate occurred between low tumor staging (T1–T2) and high tumor staging (T3–T4), as well as between different tumor grades (G1–G2 and G3–G4). This suggests that early diagnosis and treatment with periodic follow-up are crucial for improving the survival rate.

UTUCs are prone to relapse and have a recurrence rate of 16–58% after surgery; thus, the current standard treatment of UTUCs is nephroureterectomy with bladder cuff resection [16]. The study also showed that the surgical approach (laparoscopic or open surgery) is not an influencing factor of postoperative survival. Mufti [17] also proposed that the surgical method (laparoscopy or open surgery) was not a prognostic factor, which is consistent with our results.

For most people, age is a risk factor of UTUC [18]. Raman et al. [19] investigated 13,800 patients with upper urinary tract tumors and showed that mortality owing to UTUCs increased with age. Similarly, we showed that compared with patients aged < 50 years, the relative risk of postoperative mortality increased among patients aged > 50 years. It is hypothesized that as the patient’s age increases, the biological behavior of the tumor changes, with a decline in immune system function. LNM can be used to predict prognosis, especially for patients without lymph node dissection. Multicenter studies confirmed that LNM is closely related to higher tumor invasiveness, such as higher stage, higher grade, and distant metastasis. LNM can independently affect tumor recurrence and survival rate; hence, it is an independent risk factor for UTUC prognosis [20]. Our study also suggested that LNM was significantly associated with poor prognosis, which is consistent with the result of previous research. Tumor size has been also confirmed as a prognostic factor in some malignant tumors. The results of Simone’s study [21] on UTUCs revealed that metastasis-free survival was closely related with tumor size: when the tumor diameter was < 3 cm, there was no metastasis within 5 years, whereas when it was > 3 cm, the 5-year metastasis-free survival rate was 67%. Our study also revealed that tumor size was a risk factor of UTUC, wherein a larger tumor diameter indicated worse prognosis.

It is known that smoking is a prognostic factor of UTUC. A study [22] revealed that smoking is the main risk factor of UTUC and lower urinary tract urothelial cancer. The incidence of urothelial carcinoma in smokers is three times greater than that in nonsmokers, probably owing to a mutation in tumor protein p53, chromosomal changes, immune regulation, etc. Our study also confirmed that smoking was a risk factor of poor prognosis in UTUC patients. Moreover, we found that consumption of herbal medicines such as Longdanxiegan pills, Paishi granule, Paishi decoction, and caulis aristolochiae manshuriensis, which contain aristolochic acid, also affected the prognosis of UTUC patients. DNA compounds can be formed in vivo under the influence of aristolochic acid [23], leading to the A-T base-pair proto-oncogene mutation, activation of RAS, and dysfunction of the cancer suppressor gene P53. Thus, smoking may play an important role in the occurrence of UTUCs. Additionally, our study confirmed that BC was a prognostic factor for UTUC, which is consistent with the findings of Nuhn [24]. This may be due to the similar pathogenesis of UTUC and BC or a missed diagnosis of UTUC in some BC patients (especially for ureteral TCCs). For the latter reason, urologists who examine patients with hematuria may often be satisfied with a BC diagnosis and omit ureteral TCCs that could also lead to BC. This explains our finding that a prior history of muscle-invasive urothelial carcinoma of the bladder was significantly associated with an increased risk of disease recurrence and cancer-specific death in UTUC patients. Therefore, further adjuvant treatment, ureteroscopy, CT urography, and close follow-up should be performed after BC surgery.

It is generally believed that UTUC patients should be treated with radical RNU with the resection of bladder cuff resection [25]. However, the recurrence rate of BC is very high even after radical surgery. The incidence of BC relapse after operation in patients with TCC of the urinary tract is 30–70% [26], while the rate in our study was slightly lower at 20.3%. In the case of postoperative BC recurrence, a tumor located in the ureter, especially at the lower end of the duct, is considered a high-risk factor for BC recurrence. Zigeuner [27] suggested that a ureteral tumor is prone to spread into the bladder because of its close anatomical position. This could be due to the mechanical stress caused by a higher urinary flow rate and larger chamber pressure, facilitating tumor cell metastasis. Furthermore, in recent years, studies worldwide have found that the risk of BC relapse in ureteral carcinoma is significantly higher than that of a renal pelvis carcinoma after RNU [28], which corresponds to the findings of our study. This, in some degree, supports the cancer cell implantation theory as the mechanism of BC recurrence. Fang et al. [28] analyzed the risk factors of BC recurrence after primary UTUC radical resection in 438 cases and found that high-grade ureteral cancer and multiple tumors were high-risk factors of BC recurrence. Therefore, a close follow-up with cystoscopy is necessary for patients with a high-grade, high-stage tumor and ureteral cancer (especially those with lower urinary tract ureteral cancer). These results indicate that tumor size and LNM are not independent risk factors of BC recurrence in UTUC patients after operation. However, there is limited research on whether tumor size and LNM affect bladder tumor recurrence; thus, more samples and multicenter studies are required.

Marchioni et al. [29] confirmed that preoperative ureteroscopy increases the risk of postoperative BC recurrence, which is consistent with the results our study. If the cancer cell implantation theory as the main mechanism of BC recurrence is accepted, there is greater risk of cancer cell exfoliation and implantation in the bladder after ureteroscopy examination. Thus, preoperative ureteroscopy should not be the primary and routine method for diagnosing UTUCs if imaging diagnosis is relatively clear; rather, a safer imaging method should be performed.

Because UTUC is closely related to BC and is associated with a higher BC recurrence rate, postoperative intravesical instillation chemotherapy has been widely used to prevent BC recurrence in postoperative UTUC patients. Many studies showed that postoperative bladder perfusion chemotherapy can effectively reduce the recurrence rate of BC after UTUC surgery [30]. Our study showed that postoperative bladder perfusion chemotherapy was a protective factor of BC recurrence after RNU. Therefore, UTUC patients could benefit from early diagnosis and regular postoperative intravesical instillation chemotherapy.

Recent studies have demonstrated that the incidence and mortality of tumor patients with DM were significantly higher than those of patients without DM. In this study, we found that UTUC patients with DM were more likely to experience BC recurrence. Among postoperative UTUC patients, the BC recurrence rate in those with DM was approximately four times that in patients without DM, and the result was significantly different. However, the specific mechanism remains unknown, which could be closely associated with DM itself and the associated hyperglycemia, hyperinsulinemia, and lipid metabolism disorder. In a study of 251 people with non-muscle-invasive bladder carcinoma, DM was an independent risk factor for disease recurrence [31]. Liu’s in vitro experiments showed that high doses of insulin can promote the proliferation of urinary tract epithelial cells [32]. DM is often accompanied by insulin resistance and hyperinsulinemia; thus, Liu’s study results may be a plausible explanation for easily recrudescent BC after surgery in UTUC patients with DM. Metformin is currently a first-line oral hypoglycemic drug for the treatment of type 2 DM. Currently, research has shown that metformin can reduce the risk of cancer by inhibiting the proliferation of tumor cells in vivo and in vitro [33, 34]. Our study found that after RNU, the BC recurrence rate in UTUC patients with DM who did not take metformin was significantly higher than that in UTUC patients with DM who took metformin. Our study supported the fact that metformin can reduce the recurrence rate of BC in patients with UTUC and DM, which has not been reported thus far. Metformin mainly activates the AMPK pathway, promotes the expression of the P53 gene, and inhibits the mTOR pathway, thereby inhibiting tumor cell proliferation and reducing insulin and insulin-like growth factor levels [35]. However, whether these mechanisms also apply to the occurrence and progression of UTUCs in DM patients remain under study.

This study has some limitations. First, the research data represent a retrospective review of findings at a single center. Second, patients who were not surgically treated were not included in the analysis. Finally, lack of information on molecular biomarkers may reduce the strength of the findings. Therefore, further studies are necessary to confirm the role of molecular biomarkers as predictors for worse pathological outcomes of UTUCs.

In this study, the 5-year survival rate of UTUC patients from central China was not high compared with that of UTUC patients from other regions. We found that advanced tumor stage and presence of a high-grade tumor were risk factors for not only poor UTUC prognosis but also BC recurrence. In addition, preoperative ureteroscopy and DM without metformin use were high risk factors for BC recurrence, whereas postoperative bladder perfusion chemotherapy was a protective factor. Therefore, we suggest that UTUC patients with the above risk factors for BC recurrence should compulsorily undergo regular bladder perfusion chemotherapy. In addition, we suggest that preoperative ureteroscopy should not be exercised except when preoperative imaging diagnosis is difficult.