Transitional Cell Carcinoma of the Ureter: Prognostic Factors Influencing Progression and Survival

Transcription

1 european urology 51 (2007) available at journal homepage: Urothelial Cancer Transitional Cell Carcinoma of the Ureter: Prognostic Factors Influencing Progression and Survival Jan Lehmann 1, Henrik Suttmann 1, *, Iva Kovač, Markus Hack, Jörn Kamradt, Stefan Siemer, Bernd Wullich, Ulrike Zwergel, Michael Stöckle Department of Urology and Pediatric Urology, Saarland University, Homburg/Saar, Germany Article info Article history: Accepted November 9, 2006 Published online ahead of print on November 17, 2006 Keywords: Transitional cell carcinoma Ureter Upper urinary tract tumour Urothelial cancer Prognostic factors Abstract Objectives: We retrospectively evaluated prognostic factors for progression-free and disease-specific survival in a large cohort of patients with transitional cell carcinoma (TCC) of the ureter. Methods: A single-centre series of 145 consecutive patients treated with partial resection of the ureter or nephroureterectomy between 1975 and 2004 was evaluated. Median follow-up was 96 mo. Routine preoperative laboratory parameters as well as clinical and tumour-specific data were assessed. Univariate and multivariate statistical analyses were performed. Results: Five-year disease-specific survival ranged from 96.1% for stages pta to 28.6% for pt4. Grade1 tumours were associated with 5-yr diseasespecific survival rates of 100% compared with 84% for G2, and 51.9% for G3 tumours, respectively. Univariate analyses identified pt stage and grade, tumour diameter, cm and pn categories, weight loss, the presence of synchronous tumour in the renal pelvis as well as elevated levels for humoral factors such as serum alkaline phosphatase (AP), white blood cell (WBC) count, platelet count, g-glutamyl transferase, creatinin, and blood urea nitrogen as prognostic factors. In multivariate analyses, tumour grade and WBC counts were predictive for low progression-free survival rates, whilst simultaneous tumour in the renal pelvis, high AP levels, and WBC counts were correlated with worse disease-specific survival. Conclusions: Our retrospective analysis provides clinical factors to identify patients with TCC of the ureter at high risk for progression and death of disease. Interestingly, humoral factors such as elevated serum AP levels and high WBC counts were demonstrated to be of paramount prognostic significance besides tumour stage, grade and multifocality. # 2006 European Association of Urology. Published by Elsevier B.V. All rights reserved. * Corresponding author. Department of Urology and Pediatric Urology, Saarland University, Kirrbergerstraße, Homburg/Saar, Germany. Tel. +49 (0) ; Fax: +49 (0) address: henrik.suttmann@uniklinikum-saarland.de (H. Suttmann). 1 J. Lehmann and H. Suttmann contributed equally to this manuscript /$ see back matter # 2006 European Association of Urology. Published by Elsevier B.V. All rights reserved. doi: /j.eururo

2 1282 european urology 51 (2007) Introduction 2. Patients and methods Upper urinary tract transitional cell carcinoma (UUT-TCC) is a rare disease, accounting for about 5% of all urothelial tumours [1,2]. The incidence accounts for more than 2500 new cases in the United States in 2005, with a male to female ratio of approximately 2:1 [3]. Amongst the known risk factors for the development of UUT-TCC are cigarette smoking, abuse of analgetics, occupational factors, chronic infection and stone disease, as well as antineoplastic agents such as cyclophosphamide [4]. UUT-TCC is often a multifocal disease. Whilst the incidence of bilateral tumours ranges between 2% to 8%, the ipsilateral upper urinary tract is affected in 27 36% by multifocal disease [1,4]. Only2 4%ofUUT-TCCsoccurafter the primary diagnosis of TCC of the urinary bladder [4,5]. In contrast, secondary TCC of the bladder after treatment of UUT-TCC occurs in up to 75% of cases [1,4,6]. Of all UUT-TCCs, about 75% are located in the collecting system of the kidney, whilst 25% occur in the ureter [1,7]. Seventy-three percent of ureteral malignancies involve the distal ureter, whilst 24% and 3% are primarily located in the middle and proximal ureter, respectively [2]. Because of the low incidence of the disease, most studies analyzing prognostic factors have included tumours of the ureter and the renal pelvis. Radical nephroureterectomy with a bladder cuff remains the standard treatment for UUT-TCC [4,7]. Survival at 5 yr following nephroureterectomy depends primarily on tumour grade and stage, and varies from 95% and 90%, respectively, for Ta/T1 tumours to 85% for T2/T3 tumours and 38% for T4 tumours [4,6]. Further prognostic factors with negative impact on survival are multiple tumours, renal insufficiency, and synchronous bladder tumours [8]. Additionally, tumour location has been demonstrated to be of prognostic importance in UUT-TCC with tumours of the renal pelvis having a better prognosis than ureteral tumours [9]. Tumours of the ureter seem to have a higher incidence of local recurrences and distant metastases compared with tumours of the renal pelvis [9]. Only a few studies have investigated prognostic factors in TCC of the ureter separately. We retrospectively evaluated a large cohort of 145 consecutive patients treated for TCC of the ureter in our institution, and performed univariate and multivariate analyses to identify prognostic factors influencing progression and survival Patients We identified 145 consecutive patients who underwent open surgery for TCC of the ureter in our department between 1975 and Patient charts were assessed for histopathologic data including tumour volume, World Health Organization grading and TNM classification [10]. Parameters such as blood count, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), serum calcium, creatinin, lactate dehydrogenase (LDH), blood urea nitrogen, alanine aminotransferase (ALT), aspartate aminotransferase (AST), g-glutamyl transferase (ggt), bilirubin, alkaline phosphatase (AP), blood sugar, serum protein, and routine coagulation were assessed from patient charts. Furthermore, demographic and clinical data were evaluated, including preoperative body weight, size, and temperature, tumour-related symptoms such as haematuria, weight loss and/or pain, and blood pressure. Routine follow-up consisted of physical examination, ultrasound, cystoscopy, and urine cytology four times per year for the first 2 yr, twice for the third and fourth years, and once yearly thereafter. Additional studies such as intravenous urograms, computed or magnetic resonance tomography, bone scintigraphy, chest x-ray, and/or ureteroscopy were scheduled according to individual patients risk profiles at the discretion of the responsible physician Statistical analyses Progression-free survival and disease-specific survival from time of surgery were defined as end points for this retrospective analysis. Distribution of event times was calculated separately for each of the prognostic factors with the univariate product-limit method by Kaplan-Meier. Continuous variables including tumour size, ESR, blood count, CRP, serum calcium, creatinin, LDH, blood urea nitrogen, ALT, AST, ggt, bilirubin, AP, blood sugar, serum protein, and routine coagulation parameters were dichotomized for this purpose according to the best cutoff value obtained from corresponding, single-factor receiver operator characteristic (ROC) curves with respect to the end points as reported previously [11]. If categoric factors were to be analyzed, all categories were introduced simultaneously into the statistical analysis, except when one category either did not contain or comprised only censored observations. Case censoring was applied in the analysis of progression-free analysis and disease-specific survival when the patient had no signs of recurrence or when death occurred unrelated to tumour during the observation period, respectively. The prognostic significance of each variable was tested univariately with the log-rank test. The simultaneous effects of multiple prognostic factors were estimated by multiple regression analysis by using the Cox proportional hazards model in a forward-selection strategy. Only factors that had a prognostic impact at a significance level ( p) of 0.05 according to the univariate analyses were entered into the multiple regression model. The proportional hazard assumption of these factors was met. Categoric values such as T classification

3 european urology 51 (2007) and tumour grade were assigned as score values, whereas continuous variables entered the model as actual values. All reported p values were based on two-sided tests, and the threshold for significance was Statistical analyses were performed with the Statistical Package for the Social Sciences, version 10.0 (SPSS Inc, Chicago, IL, USA). 3. Results 3.1. Patients and follow-up For the 145 patients with TCC of the ureter, median age was 68 yr with a range of yr. The median follow-up was 96 mo. Table 1 provides further details on patient characteristics. In addition to surgical treatment, 12 patients (8%) received some form of adjuvant systemic chemotherapy. Median overall survival was 70 mo. Seventy-six of 145 patients were dead at the time of analysis. Thirtyone of 145 patients (21%) had died because of tumour progression. Disease-specific survival at 10 yr was 87% for patients with superficial pta/t1 tumours and 36% for patients with muscle-invasive pt2 tumours. For patients undergoing partial ureterectomy, 10-yr progression-free survival was 80%. Of patients undergoing partial ureterectomy, about 10% (5 of 51) experienced an ipsilateral tumour recurrence during the later course Univariate analysis Patient characteristics, and histopathologic and preoperative humoral parameters were subjected to univariate analysis in relation to disease-specific and progression-free survival. Amongst the parameters tested (see Patients and methods), a statistically significant progression-free survival benefit was found for the following (Table 2): tumour diameter (<3.1 cm), serum creatinin (<1.005 mg/dl), alkaline phosphatase (<116 U/l; normal values: <180 U/l for males and <160 U/l for females), total WBC count (< /l); blood urea nitrogen (<39 mg/dl), and platelet count (< /l). Lower pt stages and grades differed favourably between pta versus all other pt stages, except pt1; pt1 versus pt2 and pt1 versus pt3; G2 versus G3 and G1 versus G3. Compared with progression-free survival, the following parameters predicted a statistically significant better outcome for disease-specific survival (Table 3): serum creatinin, tumour diameter, total WBC count, and platelet count (same cutoff as above). Additionally, better survival was correlated with low AP (<127 U/l), ggt (<15.5 U/l), and the absence of positive lymph nodes, distant metastases, weight loss, and a synchronous tumour Table 1 Patient characteristics No. of pts % pts Total Men Women Associated TCC of renal pelvis Side affected Left Right Unknown 7 5 Grade Not classified 12 8 pt category Partial ureterectomy Ta T T2 9 6 T3 3 2 T4 5 3 Not classified 2 1 Nephroureterectomy Ta T T T T4 2 1 Not classified 3 2 PN category pn pn+ 9 6 cm category cm cm Symptomatic disease Yes No Type of surgery Nephroureterectomy Partial resection of ureter Endoscopic laser coagulation 1 1 Simultaneous cystectomy 2 1 Recurrences Ipsilateral after partial ureterectomy 5/51 10 Contralateral 8/145 6 Associated bladder cancer None Before After Before and after 7 5 Pt = patient; TCC = transitional cell carcinoma. of the renal pelvis. Again, various pt stages and grades were highly predictive for a positive outcome: pta versus all other pt stages except pt1; pt1 versus pt3 and pt1 versus pt4; all other pt stages versus pt4; G1 versus G2 and G1 versus G3, as well as

4 1284 european urology 51 (2007) Table 2 Prognostic factors for progression-free survival (univariate analysis) Parameter/cutoff 5-yr progression-free survival (%) SE Median survival (mo) HR 95%CI p value pt category PTa (vs. pt2) n.r pt n.r. pt2 (vs. pt1) pt3 (vs. pt1) pt Grade G1 100 n.r. G2 (vs. G3) n.r G3 (vs. G1) >100 >100 <0.001 Tumor Ø <3.1 cm n.r. >3.1 cm Creatinin <1.005 mg/dl n.r. >1.005 mg/dl n.r AP <116 U/l n.r. >116 U/l n.r WBC count < /l n.r. > /l Blood urea nitrogen <39 mg/dl n.r. >39 mg/dl Platelet count < /l n.r. > /l n.r SE = standard error; HR = hazard ratio; CI = confidence interval; n.r. = not yet reached; AP = alkaline phosphatase; WBC = white blood cell. G2 versus G3 (see Fig. 1A and B for Kaplan-Meier analyses of disease-specific survival according to pt stage and grade). Interestingly, there was no statistically significant benefit for patients undergoing partial resection of the ureter versus those receiving nephroureterectomy (hazard ratio (HR): 0.532; 95% confidence interval [95%CI], ; p = 0.271) Multivariate analysis All significant univariate factors were entered in a stepwise, forward-selection, multivariate Cox regression analysis (Table 4). Low histologic grading was identified as an independent prognostic factor for better progression-free survival. The presence of synchronous tumour in the renal pelvis at the time of surgery and high serum AP were independent negative prognostic factors for disease-specific survival. The only independent prognostic parameter for both progression-free and disease-specific survival was total WBC count (Fig. 1C andd). 4. Discussion TCC of the ureter accounts for only about 25% of all UUT-TCCs [1,7]. Because of this low incidence, most studies investigating prognostic factors in UUT-TCCs have analyzed tumours of the renal pelvis and ureter together. However, according to a recent publication by Park et al [9], TCC of the ureter seems to be associated with a worse prognosis compared with tumours of the renal pelvis. The few studies investigating prognostic factors of the ureter separately are old and hampered by the small number of cases reported per study, ranging between 43 and 73 [2,12 14]. In this retrospective study, we present the largest monocentre patient cohort published to date, in which we investigated prognostic factors in 145 consecutive cases of TCC of the ureter. Additionally, we provide the first analysis of the prognostic significance of various humoral factors in TCC of the ureter. With regards to progression and survival, our results compare favourably with the series of 73

5 european urology 51 (2007) Table 3 Prognostic factors for tumor-specific survival (univariate analysis) Parameter/cutoff 5-yr progression-free survival (%) SE Median survival (mo) HR 95%CI p value pt category pta (vs. pt2) n.r pt1 (vs. pt3) n.r pt2 (vs. pt4) pt pt4 (vs. pt3) Grade G1 (vs. G3) 100 n.r. <0.001 <0.001 <0.001 G2 (vs. G1) n.r. >1000 > G3 (vs. G2) Tumor Ø <3.1 cm n.r. >3.1 cm cm category cm n.r. cm <0.001 pn category pn n.r. pn <0.001 Weight loss No n.r. Yes n.r Creatinin <1.005 mg/dl n.r. >1.005 mg/dl AP <127 U/l n.r. >127 U/l WBC count < /l n.r. > /l <0.001 Platelet count < /l n.r. > /l GGT <15.5 U/l n.r. >15.5 U/l n.r SE = standard error; HR = hazard ratio; CI = confidence interval; n.r. = not yet reached; AP = alkaline phosphatase; WBC = white blood cell; ggt = g-glutamyl transferase. patients with TCC of the ureter published by Zungri et al [14]. In their study, 5-yr overall survival ranged between 100% for pta and 50% for pt2 3 tumours. In our study, 5-yr disease-specific survival ranged from 96.1% for stages pta to 28.6% for pt4. pt stage and grade were significantly correlated with both progression-free and disease-specific survival. Grade1 tumours were associated with a 5-yr diseasespecific survival rate of 100% compared with 84.0% and 51.9% for G2 and G3 tumours, respectively. None of the previous studies on TCC of the ureter analyzed enough patients to perform systematic univariate or multivariate analyses of prognostic factors [2,12 14]. However, they also found a trend towards better outcome for low pt stages and grades. In UUT-TCC including tumours of the ureter and renal pelvis, pt stage and tumour multifocality have recently been identified as prognostic indicators [4]. In addition to tumour stage and grade, we defined several prognostic factors influencing progression and/or survival such as tumour diameter, tumour positive lymph nodes, distant metastases, and the presence of a synchronous tumour in the renal pelvis. Weight loss was the only symptom associated with a negative outcome. A history of previous bladder cancer did not represent an unfavourable prognostic factor as found by others [5]. Furthermore, we identified several

6 1286 european urology 51 (2007) Fig. 1 Kaplan-Meier analyses for disease-specific survival in relation to (A) pt stage, (B) grade, (C) serum alkaline phosphatase (AP), and (D) white blood cell (WBC). HR = hazard ratio; CI = confidence interval. humoral factors that affect the prognosis of patients with TCC of the ureter. The importance of humoral factors for predicting the outcome in patients with solid tumours was acknowledged in 1995 by the College of American Pathologists [15]. Interestingly, in a recent report we were able to demonstrate that certain humoral factors are of superior prognostic value compared with molecular proliferation markers such as Ki-67 status in patients with renal cell carcinoma [11]. Thresholds for serum creatinin (>1.005 mg/dl), AP (>116 U/l), blood urea nitrogen (>39 mg/dl), WBC count (> /l), platelet count (> /l), and ggt (>15.5 U/l) dichotomized our patients with TCC of the ureter into a high-risk and a low-risk cohort. Even though these thresholds were within the normal range of our local laboratory values, they defined subgroups of patients with significantly lower progression-free and/or diseasespecific survival. Elevated levels of serum AP and high WBC counts were even found to be of prognostic impact in multivariate analyses. High AP levels have been shown to be of prognostic significance in other urologic malignancies such as renal cell or prostate cancer as well [11,16,17]. In prostate cancer patients, an elevated AP points to an accelerated bone metabolism and usually indicates the presence of skeletal metastases [17]. In contrast, in renal cell cancer patients an elevated AP often represents the early presence of a paraneoplastic liver dysfunction without necessarily elevating other liver enzymes also in the absence of bone metastases [16]. In our study, none of the patients had radiologically detectable bone metastases at the time of tumour diagnosis. Therefore, higher AP levels are very unlikely to result from occult bone lesions. Our results identify WBC counts > /l as an outstanding factor for predicting progressionfree and disease-specific survival in patients with TCC of the ureter. WBCs in the peripheral blood consist of about 70% neutrophil granulocytes and

7 european urology 51 (2007) Table 4 Prognostic factors for progression-free and tumor-specific survival (multivariate analysis) Parameter/cutoff Regression coefficient SE HR 95%CI of HR p value Progression-free survival Grade WBC count (< /l) <0.001 < Disease-specific survival Synchronous tumor in renal pelvis AP (<127 U/l) WBC (< /l) < SE = standard error; HR = hazard ratio; CI = confidence interval; AP = alkaline phosphatase; WBC = white blood cell. about 20% lymphocytes. Higher WBC counts usually result from a rapid release of immature neutrophils from the bone marrow as a sign of high cell turnover at distant inflammatory sites. In recent years, a newly discovered putative role of neutrophil granulocytes in malignant tumour growth and tumour immunology has been a matter of intensive debate. Whilst most studies suggest a tumour-promoting nursing function of neutrophils in advanced disease, under certain circumstances they also seem to exert tumour-inhibitory actions during early cancer cell growth [18,19]. The identification of higher WBC counts as an independent prognostic factor in TCC of the ureter might reflect an early futile activation of the innate immune system in response to local and/or metastatic tumour spread. Recent data on adjuvant systemic chemotherapy for locally advanced upper urinary tract cancer has been contradictory: Adjuvant chemotherapy for pt3n0m0 TCC has not proven useful in terms of improving survival [20]; however, another group from Korea has reported a significant survival benefit in 16 of 27 patients receiving cisplatinumbased chemotherapy [21]. Data on combined adjuvant radiochemotherapy has been even more limited. Nevertheless a group from Boston [22] presented data on 31 patients, concluding that the addition of concurrent cisplatin to adjuvant radiotherapy improves the ultimate outcome in patients with resected, locally advanced upper tract urothelial malignancies. They recommend that such a regimen be considered in patients with T3 4 and/or node-positive upper tract TCC. In our patient cohort, more than one third of patients received a partial resection of the affected ureteral segment with nephroureterectomy, representing the more prevalent surgical procedure. In fact, when comparing patients with pta or pt1 (cn0, cm0) ureteral tumours, no difference in survival was found for partial ureterectomy or nephroureterectomy (data not shown). Therefore, from our experience, conservative surgery is a valid option mainly in Ta/T1 lesions, which could be established from the beginning of our observation period. Whilst we did not detect a statistically significant difference for partial ureteral resection versus nephroureterectomy, we cannot provide insights from our experience on the ongoing debate about the prognostic impact of a conservative (endoscopic) versus an open surgical approach for TCC of the ureter. However, the data available to date suggest that intraluminal surgical techniques should only be applied in low-grade and low-stage tumours [4]. 5. Conclusions We provide data on prognostic factors in TCC of the ureter from the largest monocentre patient cohort published to date. Tumour stage and grade, tumour diameter, metastatic and symptom status, and multifocality as well as various humoral factors such as elevated serum AP levels or WBC counts were demonstrated to predict clinical responses. Our results will help to identify patients with ureteral tumours at high risk for progression and might provide a tool to stratify patients into different risk groups for yet-to-be-established adjunctive treatment such as systemic chemotherapy. References [1] Hall MC, Womack S, Sagalowsky AI, et al. Prognostic factors, recurrence, and survival in transitional cell carcinoma of the upper urinary tract: a 30-year experience in 252 patients. Urology 1998;52: [2] Babaian RJ, Johnson DE. Primary carcinoma of the ureter. J Urol 1980;123: [3] Jemal A, Murray T, Ward E, et al. Cancer statistics, CA Cancer J Clin 2005;55: [4] Oosterlinck W, Solsona E, van der Meijden AP, et al. EAU guidelines on diagnosis and treatment of upper urinary tract transitional cell carcinoma. Eur Urol 2004;46:

8 1288 european urology 51 (2007) [5] Hurle R, Losa A, Manzetti A, et al. Upper urinary tract tumors developing after treatment of superficial bladder cancer: 7-year follow-up of 591 consecutive patients. Urology 1999;53: [6] Kakizoe T, Fujita J, Murase T, et al. Transitional cell carcinoma of the bladder in patients with renal pelvic and ureteral cancer. J Urol 1980;124:17 9. [7] Krogh J, Kvist E, Rye B. Transitional cell carcinoma of the upper urinary tract: prognostic variables and postoperative recurrences. Br J Urol 1991;67:32 6. [8] Kang CH, Yu TJ, Hsieh HH, et al. The development of bladder tumors and contralateral upper urinary tract tumors after primary transitional cell carcinoma of the upper urinary tract. Cancer 2003;98: [9] Park S, Hong B, Kim CS, et al. The impact of tumor location on prognosis of transitional cell carcinoma of the upper urinary tract. J Urol 2004;171: [10] Sobin LH, Fleming ID. TNM classification of malignant tumors. 5th ed. (1997) Union Internationale Contre le Cancer and the American Joint Committee on Cancer. Cancer 1997;80: [11] Lehmann J, Retz M, Nürnberg N, et al. The superior prognostic value of humoral factors compared with molecular proliferation markers in renal cell carcinoma. Cancer 2004;101: [12] Batata MA, Whitmore WF, Hilaris BS, et al. Primary carcinoma of the ureter: a prognostic study. Cancer 1975;35: [13] Anderstrom C, Johansson SL, Pettersson S, et al. Carcinoma of the ureter: a clinicopathologic study of 49 cases. J Urol 1989;142: [14] Zungri E, Chechile G, Algaba F, et al. Treatment of transitional cell carcinoma of the ureter: is the controversy justified? Eur Urol 1990;17: [15] Henson DE, Fielding LP, Grignon DJ, et al. College of American Pathologists Conference XXVI on clinical relevance of prognostic markers in solid tumors Summary. Members of the Cancer Committee. Arch Pathol Lab Med 1995;119: [16] Stauffer M. Nephrogenic hepatoslenomegaly. Gastroenterology 1961;40: [17] Jorgensen T, Kanagasingam Y, Kaalhus O, et al. Prognostic factors in patients with metastatic (stage D2) prostate cancer: experience from the Scandinavian Prostatic Cancer Group Study-2. J Urol 1997;158: [18] Coussens LM, Werb Z. Inflammation and cancer. Nature 2002;420: [19] Di Carlo E, Forni G, Musiani P. Neutrophils in the antitumoral immune response. In: Cassatella MA, editor. The neutrophil. Basel: Karger; p [20] Lee SE, Byun SS, Park YH, et al. Adjuvant chemotherapy in the management of pt3n0m0 transitional cell carcinoma of the upper urinary tract. Urol Int 2006;77:22 6. [21] Kwak C, Lee SE, Jeong IG, et al. Adjuvant systemic chemotherapy in the treatment of patients with invasive transitional cell carcinoma of the upper urinary tract. Urology 2006;68:53 7. [22] Czito B, Zietman A, Kaufman D, et al. Adjuvant radiotherapy with and without concurrent chemotherapy for locally advanced transitional cell carcinoma of the renal pelvis and ureter. J Urol 2004;172: