A Study On Prognosis Factors For Upper Tract Urothelial Carcinoma

[Purpose]Upper tract urothelial carcinoma (UTUC) is a relatively rare malignancy, accounting for approximately 5% of genitourinary cancers and 10% of renal cancers. Improvement in imaging and endoscopic techniques has gradually led to an age of early diagnosis and management for UTUC and bladder cancer as well. Despite such a tendency of early-stage intervention, UTUC still keeps its characteristics of frequent recurrence, aggressiveness and poor prognosis. Our study retrospectively evaluated the impact and predictive significance of preoperative and surgical factors for prognosis of UTUC, and also the relationship between clinical, surgical and pathological parameters, including part I:preoperative hydronephrosis, part Ⅱ: preoperative hemoglobin level, part III:laparoscopic procedure and part IV:bladder cuff excision (BCE).[Materials and method]Study population:167 consecutive patients were clinically diagnosed as UTUC in Huashan Hospital, Fudan University between January 1999 and December 2012. According to the purpose of each part, different inclusion and exclusion criteria were identified, including 125 patients in part I,122 patients in part II and III,156 patients in part IV. The resected specimens were pathologically confirmed as UC and evaluated based on the UICC TNM classification system.Follow-up regimen:Examinations of follow-up after surgery comprised urine routine examination, ultrasonic examination of urinary system, cystoscopy, chest plain radiography, etc. Patients were followed up every 3 months for the first year after surgery, then every 6 months until the third year after surgery, and thereafter annually. Local relapse, intravesical recurrence (IVR), distant metastasis and death were observed. These oncological outcomes were defined as disease progression.Statistical analyses:In part I,125 patients were divided into two groups:presence or absence of preoperative hydronephrosis (74 vs.51), and subgroups were divided according to tumor location (renal pelvis 25 vs. ureter 48) in the group of hydronephrosis. In part II, continuous values of preoperative laboratory tests were expressed as the median (interquartile range, IQR). In part III,122 patients were divided into two groups according to surgical methods:open (ONU) or laparoscopic (LNU) radical nephroureterectomy (101 vs.21). In part IV, patients were divided into two groups according to surgical procedures:radical nephroureterectomy with (NUC) or without (NU) bladder cuff excision (122 vs.34). The Student t test, Wilcoxon signed rank test and x2 test were used to estimate differences of clinical and pathological characteristics between two groups. Relationship between preoperative laboratory parameters was analyzed by Spearman’s correlation in part II. Impact of study parameters on IVR, progress-free survival (PFS) and overall survival (OS) was interpreted through univariable and multiyariable Cox regression models, respectively. The Kaplan-Meier method was conducted for evaluation of PFS and OS rates after nephroureterectomy.[Results]The presence of preoperative hydronephrosis was associated with tumor located in ureter (p<0.001) and diseases without lympho vascular invasion (LVI, p=0.005). In multivariate analyses, hydronephrosis presented a significance of predicting progression-free survival (p=0.040), but was not associated with overall survival (p=0.754). In renal pelvis group, PFS of hydronephrosis subgroup was significantly worse than non-hydronephrosis subgroup (p=0.004), but no difference was found between two subgroups in ureter group (p=0.914). In univariate analysis, preoperative hemoglobin level was associated with PFS and OS (p=0.013, 0.010), whereas multivariate analysis presented that it was not an independent predictor of either outcome (p=0.053,0.055). Moreover, preoperative serum creatinine level was associated with PFS and OS (p=0.005,0.048) in univariate analysis, while preoperative white blood cell count was only associated with PFS (p=0.002). The Spearman’s correlation indicated that lower preoperative hemoglobin level was significantly correlated with positive lymph node status (p=0.040) and lower preoperative albumin level (p<0.001). In univariate analysis, the open or laparoscopic procedures exerted an impact neither on PFS (p=0.304) nor on OS (p=0.646). The Kaplan-Meier curves illustrated that patients undergoing ONU or LNU enjoyed equivalent PFS and OS rates (p=0.297,0.644). Univariate analysis showed that omission of BCE did not increase risk of postoperative IVR (p=0.526), but decreased OS significantly (p=0.005). In multivariate analysis, the performance of BCE maintained a significant impact on OS (p=0.002). The 1-,2-and 5-year OS rate of patients undergoing RNU with or without BCE were 92.6% vs.73.5%,87.7% vs.67.6%, 78.7% vs.58.8%, respectively, which illustrated that patients from the former group enjoyed a significantly better survival rate (p=0.004). Additionally, several clinicopathological parameters such as previous bladder cancer, lymphovascular invasion and positive lymph node, were detected to influence the oncological outcomes of UTUC.[Conclusions]Preoperative hydronephrosis has a predictive effect for postoperative PFS of patients with UTUC, but has no impact on OS. Preoperative hemoglobin level was associated with both PFS and OS, but not an independent prognosticator. Open or laparoscopic radical nephroureterectomy has no influence on either PFS or OS. Nephroureterectomy with BCE has capability to improve OS, but no significant impact on preventing IVR. Identifying the predictive effect of preoperative and surgical factors for prognosis of UTUC can assist guidance of treatment strategy and follow-up regimen, especially preoperative parameter determination for adjuvant treatment. Predictive effect of the parameters demands further multi-institutional studies with larger sample size.