Prognostic relevance of preoperative circulating CD8-positive lymphocytes in the urinary bladder recurrence of urothelial carcinoma

Urologic Oncology: Seminars and Original Investigations 30 (2012) 680 – 687

Original article

Prognostic relevance of preoperative circulating CD8-positive lymphocytes in the urinary bladder recurrence of urothelial carcinoma夡 Chang-Te Lin, M.D.a, Chun-Liang Tung, M.D.b, Yuh-Shyan Tsai, M.D., Ph.D.c, Cheng-Huang Shen, M.D.a, Yeong-Chin Jou, M.D.a, Min-Tsung Yu, M.S.d, Shu-Fen Wu, Ph.D.d,* a

Department of Urology, Chia-Yi Christian Hospital, Chia-Yi, Taiwan Department of Pathology, Chia-Yi Christian Hospital, Chia-Yi, Taiwan c Department of Urology, College of Medicine, Hospital and Douliou branch, National Cheng Kung University, Tainan, Taiwan d Department of Life Science and Institute of molecular Biology National Chung Cheng University, Chia-Yi, Taiwan b

Received 26 May 2010; received in revised form 28 July 2010; accepted 11 August 2010

Abstract Objective: Tumor-infiltrate lymphocytes (TIL) have been associated with favorable outcomes in various tumors including urothelial carcinoma (UC). There is little literature about peripheral blood lymphocytes (PBL). Our objective is to investigate the clinical significance and relevance of PBL on the outcomes of UC patients. Materials and methods: Ninety UC treated patients at Chia-Yi Christian hospital were enrolled. Preoperative PBLs were collected and analyzed for the percentage of each subpopulation of lymphocyte using flow cytometry. The prognostic values were calculated by using Kaplan-Meier curve and Cox progression model for univariate and multivariate analyses, respectively. Furthermore, available tumor specimens from 27 patients were further analyzed for number of CD8⫹ tumor infiltrating lymphocytes (TIL) using immunohistochemistry. The correlation between percentage of CD8⫹ PBL and number of CD8⫹ TIL was analyzed using a linear regression model. Results: The log-rank test showed that tumor location (urinary bladder vs. upper urinary tract), enrolled status (primary or recurrent), and CD8⫹ PBL were significant prognostic indicators of recurrence (P values, 0.043, 0.039, and 0.018, respectively). Cox analyses showed that CD8⫹ PBL was the sole independent prognostic indicator for recurrence-free survival (P ⫽ 0.048). The results using a linear regression analysis showed there was a reverse correlation between CD8⫹ TIL and PBL (r2 ⫽ 0.635, P ⬍ 0.0001). Conclusions: In our investigation, preoperative CD8⫹ PBL was an independent predictor for bladder recurrence. The percentages of CD8⫹ PBL were reversely correlated with the number of TIL. Such findings may benefit in the decision for subsequent intravesical therapy after surgery. © 2012 Elsevier Inc. All rights reserved. Keywords: CD8; Prognosis; Recurrence; Tumor infiltrating lymphocytes; Urothelial neoplasms

1. Introduction Urothelial carcinoma (UC) can be defined as neoplasms that arise from the epithelial lining of the urinary tract from minor calyces to the urinary bladder, and even to the pros夡 This study was supported by Chia-Yi Christian Hospital (grant GR96-7). This work was supported in part by the National Science Council (Taiwan, Republic of China; NSC 99-2321-B-194-001). * Corresponding author. Tel.: ⫹886-5-276-5041; fax: ⫹886-5-277-4511. E-mail address: [email protected] (S.-F. Wu).

1078-1439/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.urolonc.2010.08.009

tatic urethra. The majority of UC are bladder cancer and transitional cell carcinoma (TCC) in histology [1]. Among human malignancies in the United States, bladder cancer is the fourth most common cancer in men and the ninth in women [2]. Although less than 5% of UCs occur in the upper urinary tract (UUT) in most parts of the world, there is a higher frequency of UUT TCCs in southern Taiwan (more than 20%) [3]. The behavior of urothelial tumors of the UUT display tumor similar to those of tumors arising in the bladder, such as tumor recurrence in the bladder and residual UUT, disease progression, or death [1,4]. There-

C.-T. Lin et al. / Urologic Oncology: Seminars and Original Investigations 30 (2012) 680 – 687

fore, regardless of tumor location, it is recommended that patients with UC have several examinations at regular follow-up, including urine cytology, cystoscopy, and imaging studies. Due to the lack of a simple, noninvasive, ideal tumor marker for bladder cancer surveillance, a number of clinical or molecular studies are researching this area [5]. Tumor infiltrating lymphocytes (TIL) are lymphocytes of the host immune system that have been observed within tumor sites; presumably they migrate to the tumor site in order to combat the growing malignant cells. They are normally activated T cells, natural killer (NK) cells, and non-T or non-B lymphocytes. These lymphocytes can be physically characterized by cluster differentiation and certain surface-antigen groupings. Migration of peripheral blood lymphocytes (PBL) to the tumor site might imply that the host immune system is capable of initiating an antitumor response. Although most bladder cancer patients were immunodeficient [6], bladder cancer is the only solid tumor other than melanoma for which immunotherapy confers a survival benefit [7]. In fact, regional antitumor immunity within the urinary bladder could be enhanced in UC patients after intravesical immunotherapy of bacillus Calmette-Guerin (BCG), which might be associated with the presence of neutrophil, release of tumor necrosis factorrelated apoptosis-inducing ligand, subsequent cytokine release, and lymphocyte enrollment [8]. Nevertheless, there is only limited data regarding the role of lymphocytes (either TIL or PBL) in bladder cancer [7,9]. Two decades ago, Carballido and colleagues reported that a decreased circulating NK cell activity in UC patients was observed compared with healthy controls, and disease recurrence was associated with decreased NK cell activity [10]. Tsujihashi et al. reported that both TIL and PBL could augment immune response when culturing with immunomodulating cytokine, such as interleukin-2. However, TIL could produce a tumor-specific response, and PBL could not [11]. Moreover, in a large cohort of 514 bladder cancer patients with over 9-year follow-up, Lipponen et al. determined that TIL predicted a favorable outcome in invasive bladder cancer, representing as a sign of efficient host defense mechanisms [12]. Recently, Sharma et al., in a small cohort of 69 UC patients, reported that the number of CD8⫹ TIL can significantly influence the overall survival in muscle-invasive UC patients. Such a correlation was associated the expression of major histocompatibility complex (MHC)-I and NY-ESO-1, a tumor-associated antigen [9]. Therefore, preclinical studies frequently reported that the ability of treatment modalities to increase the number of TIL or enhance the activity of TIL was associated with therapeutic efficacy [13,14]. Although such methods can induce effective TIL against the tumor, these immunologic responses are not easily found if the tumor specimen could be obtained. Therefore, it is worth investigating the clinical significance of PBL on clinical outcomes of UC patients according to T cell subpopulations, particularly CD4⫹,

681

CD8⫹, and CD56⫹ lymphocytes, on the bladder tumor recurrence of UC patients after surgery.

2. Materials and methods 2.1. Patient populations From November 2007 to December 2008, a total of 103 consecutive patients with UC underwent transurethral resection of bladder tumor (TUR-BT) or nephroureterectomy plus bladder cuff resection (NUBCR) at Chia-Yi Christian hospital (Chia-Yi, Taiwan), and except patients who died within 1 year or patients who went to another hospital when recurrent, most of patients who had been followed-up for more than 1 year were included in the study. To be free of active infection, patients who had abnormal preoperative c-reactive protein (⬎1.0 mg/L), total white blood cell count (⬎10,000 cells/␮l), or preoperative chemotherapy within 6 months were excluded from this study. Tumors were staged according to the 2002 TNM classification and graded using the 2004 WHO classification. Regular follow-up studies consisted of physical examination, urinalysis, urine cytology, and image study. Cystoscopy was performed every 3 months for the first 2 years, and every 6 months for the next 2 years. The study was undertaken with the approval and institutional oversight of the Institutional Review Board for the Protection of Human Subjects at our hospital. 2.2. Flow cytometric analysis of peripheral blood mononuclear cells (PBMC) Fresh venous blood was obtained immediately before surgery. The peripheral blood mononuclear cells were isolated by the density-gradient centrifugation method. Analysis of cell surface markers was performed using monoclonal antibodies in direct immunofluorescence assays. A FACS Calibur flow cytometer (Becton Dickinson, Mountain View, CA) were used in this study. 2.3. Immunohistochemistry (IHC) Tumor specimens were fixed with formalin, embedded with paraffin. Briefly, paraffin sections were deparaffinized, rehydrated, and treated with some procedures before incubation with a 1:250 CD8 monoclonal antibody (DAKO, Glostrup, Denmark). The immunoreactivity was visualized with a BioGenex IHC kit (BioGenex Laboratories Inc., San Ramon, CA) and then counterstained with hematoxylin for microscopic examination. TIL CD8⫹ were calculated as the Sharma’s method [9]. For each specimen, four independent, 0.0328-mm2 areas with the most abundant CD8⫹ lymphocytes cells were selected and digitally imaged with a Nikon

682

C.-T. Lin et al. / Urologic Oncology: Seminars and Original Investigations 30 (2012) 680 – 687

(Tokyo, Japan) Eclipse 80i microscope and Nikon Digital Sight DS-L2 camera with standard commercial software. CD8⫹ TIL were counted manually from the digital images displayed on a monitor. All counts were repeated 3 times by the same pathologist, and the average of the repeat counts was used for statistical analyses.

ues reported are 2-sided and considered significant if P ⬍ 0.05.

3. Results 3.1. Patient cohort and characteristics

2.4. Statistical analysis Flow cytometric results of PBMC, representing the percentages of CD4⫹, CD8⫹, and CD56⫹ lymphocytes, were normal distributed and analyzed for statistical significance using unpaired Student t-test according to the clinic-pathologic characteristics, including age, gender, body mass index (BMI), preoperative glomerular filtration rate (GFR), post-operative intravesical therapy (BCG, Mitomycin C, epirubicin), tumor size, enrolled status (primary or recurrent), tumor location, tumor multiplicity, pathological stage, and tumor grade. Also, the correlation between intratumoral and circulating CD8⫹ lymphocytes was analyzed using a linear regression model. Most of patients with UC underwent TUR-BT received one, immediate, post-operative intravesical instillation of chemotherapy (Mitomycin C, epirubicin) except cases of overt or suspected intra- or extra-peritoneal perforation. Furthermore, 31 patients underwent TUR-BT with higher recurrent risk (high grade, multiple tumors, concomitant CIS) received further intravesical therapy (successive Mitomycin C, BCG, epirubicin), and we took them as group of “intravesical therapy” for further analyses. The European Organization for Research and Treatment of Cancer (EORTC) developed a scoring system and risk tables [15], and took tumor diameter more than 3 cm as a recurrent risk factor. So we separated tumor size into 2 groups, as ⬎3 cm or ⬍3 cm. We estimated GFR with Modification of Diet in Renal Disease (MDRD) formula and took GFR 60 ml/min as a standard of whether renal function was fair or not (as The Kidney Disease Outcomes Quality Initiative of the National Kidney Foundation defines chronic kidney disease as GFR of less than 60 ml/min). The clinical outcomes analyzed in the study were the events of tumor recurrence in the urinary bladder. Recurrence-free survival (RFS) was calculated from TUR-BT or NUBCR to the date of the first documented tumor recurrence in the urinary bladder. Statistical analysis was performed using Statistical Package for Social Sciences, ver. 12.0, software (SPSS Inc., Chicago, IL). The relationship between the percentages of CD4⫹, CD8⫹, or CD56⫹ lymphocytes, or clinicopathologic factors, and clinical outcome was analyzed with Kaplan-Meier plots, the log-rank test, and the multivariate Cox regression model. When the results from the rerun analyses stratified by any form of clinic-pathologic characteristics remained consistent and the patterns and P values were essentially unchanged, the results were presented with all patients combined. All P val-

Of the 90 patients enrolled, there were 67 (74.4%) male and 23 (25.6%) female patients in this study, whose average age was 67.42 year (range: 26 –90 years) with a median follow-up of 19 months (range: 3–28).There were 63 patients with urinary bladder TCC, 22 with UUT TCC, and 5 simultaneous with both tumors. Fifty-nine (65.6%) tumors were non-muscle-invasive and 31 (34.4%) were muscle invasive. Twenty-four (26.7%) tumors were low-grade and 66 (73.3%) were high-grade. About two-thirds of all 90 tumors were single in tumor number and half were newly diagnosed. Forty (44.4%) patients were fair renal function (GFR ⱖ 60 ml/min) and 50 (55.6%) patients were decreased or poor renal function (GFR ⬍ 60 ml/min) (Table 1). 3.2. Association of CD4⫹, CD8⫹, and CD56⫹ PBL with clinicopathologic factors, especially the CD8 and GFR The statistical correlations between the percentages of CD4⫹, CD8⫹, and CD56⫹ PBLs and several clinico-pathologic factors are displayed on Table 1. Among 90 patients with UC in this study, the percentage of CD8⫹ PBL was significantly associated with the preoperative GFR and tumor multiplicity (P ⫽ 0.01 and 0.03, respectively), rather than with other factors. UC patients with either a preoperative GFR of less than 60 ml/s or with multiple tumors exhibited significantly higher percentages of CD8⫹ PBL than those with preoperative GFR of more than 60 ml/s or with single tumors. However, there was no statistically significant association between the percentages of CD4⫹ and CD56⫹ PBL and other parameters (all P values ⬎ 0.05, unpaired t-test). We also found the phenomenon that male, older, BMI more than 24 kg/m2, higher stage, tumor diameter more than 3 cm, and UUT TCC exhibited higher percentages of CD8⫹ PBL but no statistical significance. 3.3. CD8 was associated with the urinary bladder recurrence At the time of the analyses, there had been 34 bladder tumor recurrences, 2 disease progressions, and 8 deaths from the disease or other causes. Due to a low number, the association with disease progression or death was not measured. Among 34 bladder tumor recurrence, 24 patients were bladder cancer and 10 were UUT TCC. The average interval between surgery and bladder recurrence was 6.35 months with a range of 3 to18. Table 2 shows the results of both univariate and multivariate analyses of variables associated with RFS. Univariate analysis with the log-rank test

C.-T. Lin et al. / Urologic Oncology: Seminars and Original Investigations 30 (2012) 680 – 687

683

Table 1 Percentages of CD4⫹, CD8⫹, and CD56⫹ PBMC in 90 UC patients (mean ⫾ SD) Variable

Number (%)

CD4⫹

Total UC Sex Male Female Age ⱖ67 years ⬍67 years BMI ⱖ24 (kg/m2) ⬍24 (kg/m2) GFR ⬍60 ml/s ⱖ60 ml/s Stage TaT1Tis T2–4 Grade Low High Location UB UUT Multiplicity Single Multiple Enroll status Primary Recurrent Chemotherapy Yes No Tumor size ⬍3 cm ⱖ3 cm

90

34.4 ⫾ 11.0

67 (74.4) 23 (25.6)

34.8 ⫾ 11.7 33.1 ⫾ 8.8

0.52

21.4 ⫾ 10.0 19.2 ⫾ 9.2

0.35

26.0 ⫾ 13.8 24.7 ⫾ 12.8

0.69

50 (55.6) 40 (44.4)

32.4 ⫾ 12.1 36.9 ⫾ 9.1

0.06

21.4 ⫾ 11.8 20.1 ⫾ 6.4

0.54

27.4 ⫾ 14.9 23.5 ⫾ 11.3

0.18

41 (45.6) 49 (54.4)

34.4 ⫾ 10.4 34.4 ⫾ 11.8

0.98

21.3 ⫾ 11.0 20.2 ⫾ 8.1

0.58

26.5 ⫾ 14.0 24.7 ⫾ 12.9

0.53

50 (55.6) 40 (44.4)

32.5 ⫾ 10.9 36.7 ⫾ 10.9

0.07

23.2 ⫾ 11.0 17.8 ⫾ 7.1

0.01

26.5 ⫾ 13.4 24.6 ⫾ 13.7

0.50

59 (65.6) 31 (34.4)

33.9 ⫾ 10.3 35.3 ⫾ 12.5

0.55

20.1 ⫾ 9.3 22.1 ⫾ 10.7

0.36

24.6 ⫾ 12.8 27.7 ⫾ 14.7

0.30

24 (26.7) 66 (73.3)

32.6 ⫾ 10.6 35.0 ⫾ 11.2

0.35

22.7 ⫾ 10.8 20.1 ⫾ 9.4

0.27

27.1 ⫾ 12.9 25.1 ⫾ 13.9

0.54

68 (75.6) 22 (24.4)

34.6 ⫾ 10.9 33.7 ⫾ 13.4

0.74

20.9 ⫾ 8.4 20.5 ⫾ 13.4

0.87

25.2 ⫾ 11.8 27.1 ⫾ 17.9

0.59

60 (66.7) 30 (33.3)

34.2 ⫾ 10.4 34.7 ⫾ 12.3

0.84

19.2 ⫾ 9.8 24.0 ⫾ 9.1

0.03

26.3 ⫾ 13.1 24.0 ⫾ 14.3

0.55

45 (50.0) 45 (50.0)

35.0 ⫾ 11.4 33.8 ⫾ 10.8

0.61

20.7 ⫾ 11.4 21.0 ⫾ 8.1

0.87

26.4 ⫾ 14.2 25.0 ⫾ 12.9

0.64

31 (34.4) 59 (65.6)

34.3 ⫾ 8.2 34.5 ⫾ 12.4

0.94

21.1 ⫾ 9.2 20.7 ⫾ 10.2

0.88

24.9 ⫾ 9.3 26.1 ⫾ 15.3

0.64

60 (66.7) 30 (33.3)

35.6 ⫾ 10.6 32.0 ⫾ 11.7

0.15

19.7 ⫾ 7.4 23.1 ⫾ 13.0

0.2

23.7 ⫾ 11.9 29.5 ⫾ 15.6

0.06

Pa

CD8⫹

Pa

20.8 ⫾ 9.8

CD56⫹

Pa

25.7 ⫾ 13.5

BMI ⫽ body mass index; GFR ⫽ glomerular filtrate rate; UB ⫽ urinary bladder; UUT ⫽ upper urinary tract; Chemotherapy ⫽ intravesical chemotherapy. a Student’s t-test.

showed that tumor location (urinary bladder vs. UUT), enrolled status (primary or recurrent), and CD8⫹ PBL were significantly prognostic indicators (P values, 0.043, 0.039, and 0.018, respectively), but other parameters were not. Cox’s analyses showed that CD8⫹ PBL was the sole independent prognostic indicator for RFS (HRs, 0.458; 95% CI, 0.212– 0.992; P ⫽ 0.048) (Table 2 and Fig. 1). Furthermore, we separated the data of bladder UC from that of UUT UC and re-analyzed the data of bladder UC alone; lower part of Table 2 showed the results of both univariate and multivariate analyses of variables associated with RFS in bladder UC. Univariate analysis with log-rank test and multivariate analyses with Cox’s analyses showed that CD8⫹ PBL was the sole independent prognostic indicator for RFS. (HRs, 0.425; 95% CI, 0.185– 0.977; P ⫽ 0.044 and HRs, 0.406; 95% CI, 0.175– 0.941; P ⫽ 0.036, respectively).

During follow-up, 21 of 45 (46.7%) patients with a higher percentage of CD8⫹ PBL (more than median) developed bladder recurrence, whereas only 10 of 45 (22.2%) patients whose the percentage of CD8⫹ PBL was less than median developed recurrence. Moreover, only 8 (8.9%) patients were dead at the end of this study; the only predictive factor of overall survival was stage (P ⫽ 0.044). 3.4. Reverse correlation between CD8⫹ TIL and PBL in UC patients To further clarify why patients with a higher percentage of CD8⫹ PBL exhibited higher bladder recurrence rate, we performed immunohistochemical (IHC) staining study to measure the intratumoral CD8⫹ lymphocytes. Twentyseven tumors were retrieved for IHC with available tissue, including 15 bladder tumors and 12 UUT TCCs. The results from the immunohistochemical staining showed that the

684

C.-T. Lin et al. / Urologic Oncology: Seminars and Original Investigations 30 (2012) 680 – 687

Table 2 Prognostic significance of CD4⫹, CD8⫹, CD56⫹, and clinicopathologic factors using univariate and multivariate analyses according to tumor recurrence or not Variable

Total patients (n ⫽ 90) Age Sex BMI GFR Stage Grade Location Multiplicity CD4⫹ CD8⫹ Intravesicaltherapy Primary/recurrent Tumor size Bladder cancer (n ⫽ 68) Age Sex BMI GFR Stage Grade Multiplicity CD4⫹ CD8⫹ CD56⫹ Intravesical therapy Primary/recurrent Tumor size

Univariate*

Multivariate**

HR

95% CI

P

HR

95% CI

P

1.527 1.041 1.104 0.569 1.014 0.937 0.292 0.596 1.724 0.397 1.763 0.441 2.738

0.746–3.125 0.448–2.417 0.543–2.245 0.275–1.176 0.450–2.283 0.431–2.037 0.089–0.962 0.287–1.240 0.837–3.552 0.185–0.854 0.857–3.516 0.203–0.959 0.957–7.835

0.247 0.926 0.785 0.128 0.973 0.869 0.043 0.166 0.140 0.018 0.125 0.039 0.060

0.451

0.126–1.609

0.220

0.459

0.212–0.992

0.048

0.596

0.262–1.358

0.218

1.377 1.534 1.761 0.730 0.636 0.702 0.653 1.150 0.425 0.788 1.278 0.523 1.621

0.649–2.921 0.607–3.880 0.819–3.785 0.344–1.548 0.250–1.617 0.308–1.602 0.297–1.434 0.545–2.426 0.185–0.977 0.369–1.658 0.608–2.686 0.212–1.293 0.381–6.901

0.404 0.366 0.147 0.411 0.342 0.400 0.288 0.714 0.044 0.540 0.517 0.161 0.514

1.863

0.864–4.018

0.113

0.406

0.175–0.941

0.036

BMI ⫽ body mass index; GFR ⫽ glomerular filtration rate; HR ⫽ hazard ratio; CI ⫽ confidence interval. * Log-rank test. ** Cox’s proportional hazards model.

median count of CD8⫹ TIL was 3.25 (Fig. 2A and B). Also, flow cytometric analysis in the same patient showed the median percentage of CD8⫹ PBL was 18 (Fig. 2C and E).

The result using a linear regression analysis showed there was a reverse correlation between CD8⫹ TIL and PBL (r2 ⫽ 0.635, P ⬍ 0.0001) (Fig. 2F).

4. Discussion

Fig. 1. Kaplan-Meier plots of recurrence-free survival (RFS) according to the percentage of CD8⫹ peripheral blood lymphocytes (PBL).

In this study we demonstrated that preoperative CD8⫹ PBL was an independent predictor for bladder recurrence, meaning that higher CD8⫹ PBL percentage was significantly associated with shorter time to bladder cancer recurrence in UC patients following surgery. In contrast, CD4⫹ and CD56⫹ PBL were not. In exploring the possible mechanism, we found that a reverse correlation between percentage of preoperative CD8⫹ PBL and the number of CD8⫹ TIL. Furthermore, the reverse correlation between CD8⫹ PBL and CD8⫹ TIL was consistent with clinical outcomes, that is, both lower preoperative CD8⫹ PBL and higher CD8⫹ TIL indicated favorable outcomes. Several studies have reported that the number of TIL has a favorable prognostic factor for clinical outcome in patients with some human malignancies [16 –19], as well as in those with UC [9,12]. Taken together, such

C.-T. Lin et al. / Urologic Oncology: Seminars and Original Investigations 30 (2012) 680 – 687

685

Fig. 2. A reverse relationship between the percentage of CD8⫹ peripheral blood lymphocytes (PBL) and the number of CD8⫹ tumor-infiltrate lymphocytes (TIL). (A) and (C) immunohistochemical staining for CD8⫹ TIL in 1 patient with higher percentage (27.98%) of CD8⫹ PBL (A) ⫻200; (B) and (D) immunohistochemical staining for CD8⫹ TIL in 1 patient with lower percentage (14.80%) of CD8⫹ PBL (C) ⫻200; (E) CD8⫹ TIL infiltrating in tumor area(⫻400; (F) the result using a linear regression analysis showed a reverse correlation between percentage of CD8⫹ PBL and number of CD8⫹ TIL (P ⬍ 0.0001, r2 ⫽ 0.635). (Color version of figure is available online.)

findings might imply that UC patients with susceptible bladder cancer recurrence may have some defects in the aspects of CD8⫹ PBL migration into the tumor site. Little information is known how the processes of lymphocyte differentiation, migration, or activation are influenced when the immune system encounters the development or progression of UC. Sharma et al. reported that CD8⫹ TIL was an independent prognostic factor for disease survival in UC patients [9]. Despite of the lack of any prognostic significance for tumor associated antigen NYESO-1 expression in this series, they found that 1 patient

with a higher number of CD8⫹ TIL had a clonal T cell population specific for antigen NY-ESO-1 [9]. Also, TIL rather than PBL, harvested from the same patient, had a tumorspecific immune response [11]. These data might imply that some UC patients exhibit obstacles to obtain a tumor-specific, Th1-associated immune response, such as effective lymphocyte activation against tumor antigen, e.g., NY-ESO-1. In addition, these findings can prompt us a direction in exploring the host defense immunity in UC patients, particularly in the aspects of migration of CD8⫹ PBL into the tumor site or functional activation of CD8⫹ PBL.

686

C.-T. Lin et al. / Urologic Oncology: Seminars and Original Investigations 30 (2012) 680 – 687

Our study also showed that patients with GFR less that 60 ml/s exhibited higher percentage of CD8⫹ PBL when compared with those with GFR more than 60 ml/s. Indeed, many reports have demonstrated that patients with renal insufficiency have a higher risk of developing UC [20 –22]. Our findings might provide some support from the viewpoint of percentage of CD8⫹ PBL. Tumor-infiltrate lymphocytes (TIL) have been indicated with the association of improved disease outcome and better survival in various tumors including UC. However, the quantification of TIL from tumor samples was not easily conducted with small or fragment tissues. Besides, PBL is more easily accessible and relatively prospective to predict the UC recurrence. In addition to CD8⫹ TIL, CD8⫹ PBL may become a potential predictor for UC recurrence. Second TUR of T1 tumor is an important issue, because significant risk of residual tumor after the initial TUR of T1 lesions has been demonstrated [23,24]. Furthermore, the tumor may be under-staged by the initial resection [25,26]. It has been reported that a second TUR can increase recurrence-free survival [27–29]. In our study, we tried our best to eradicate tumor completely in initial TUR. There were 17 T1 UCs, and second TUR was performed in 11 cases. In our country, the common people cannot accept second TUR generally, but we do our best to perform second TUR in T1 tumor, especially in high risk patients, such as large and/or multiple tumors, high grade tumors, and no muscle contained in specimens reported by pathologist. Although bladder UC and UUT UC arise from urothelial epithelium and have the same morphology, the clinical courses exhibit differences. We have separated the data of bladder cancer from that of UUT carcinoma and re-analyzed the data of bladder UC alone. Univariate analysis with the log-rank test and multivariate analyses with Cox analyses also showed that CD8⫹ PBL was the sole independent prognostic indicator for RFS. Even though the patient number was small, CD8⫹ PBL already exhibited significance in UC recurrence. Therefore, we took CD8⫹ PBL as an important parameter for UC recurrence. To our surprise, many important variables for recurrence were not statistically significant in our study, and we suppose this is due to our limitations. The limitations of our study include lack of long-term follow-up, the small number of patients studied, and little further mechanism that influenced on CD8⫹ PBL and TIL. Therefore, one should interpret the results with caution. Given the exploratory nature of this study, further investigation is necessary to verify our findings. That is, we expect that there will be more patients enrolled in the future studies to verify the clinical application of CD8⫹ PBL in UC recurrence.

5. Conclusions In this study, we demonstrated that preoperative CD8⫹ PBL was an independent predictor for bladder recurrence.

The percentages of CD8⫹ PBL are reversely correlated with the number of TIL, and the result is consistent with clinical outcomes. Such findings might provide an important hint for understanding the host defense immune response when encountering or escaping from subsequent bladder cancer recurrence in UC patients following surgery.

Acknowledgments The authors thank Dr. Cheng-Da Hsu for helpful comments and suggestions, Hsiao-Yen Hsieh and Syue-Yi Chen for technical assistance, and En-Shu Chiang for assistance with specimen management.

References [1] Flanigan RC. Urothelial tumors of the upper urinary tract. In: Wein AJ, Kavoussi LR, Novick AC, et al., editors. Campbell-Walsh Urology, 9th ed., Vol. 2. Philadelphia: Saunders, Elsevier, 2007 [chapter 48]. p. 1638 –52. [2] Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2008. CA Cancer J Clin 2008;58:71–96. [3] Chou YH, Huang CH. Unusual clinical presentation of upper urothelial carcinoma in Taiwan. Cancer 1999;85:1342– 4. [4] 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; 98:1620 – 6. [5] Shen CH, Wang YH, Wang WC, et al. Inducible nitric oxide synthase promoter polymorphism, cigarette smoking, and urothelial carcinoma risk. Urology 2007;69:1001– 6. [6] Olsson CA, Rao CN, Menzoian JO, et al. Immunologic unreactivity in bladder cancer patients. J Urol 1972;107:607–9. [7] Liakou CI, Narayanan S, Ng Tang D, et al. Focus on TILs: Prognostic significance of tumor infiltrating lymphocytes in human bladder cancer. Cancer Immun 2007;7:10. [8] Ludwig AT, Moore JM, Luo Y, et al. Tumor necrosis factor-related apoptosis-inducing ligand: A novel mechanism for bacillus CalmetteGuerin-induced antitumor activity. Cancer Res 2004;64:3386 –90. [9] Sharma P, Shen Y, Wen S, et al. Cd8 tumor-infiltrating lymphocytes are predictive of survival in muscle-invasive urothelial carcinoma. Proc Natl Acad Sci USA 2007;104:3967–72. [10] Carballido J, Alvarez-Mon M, Solovera OJ, et al. Clinical significance of natural killer activity in patients with transitional cell carcinoma of the bladder. J Urol 1990;143:29 –33. [11] Tsujihashi H, Matsuda H, Uejima S, et al. Immunocompetence of tissue infiltrating lymphocytes in bladder tumors. J Urol 1988;140: 890 – 4. [12] Lipponen PK, Eskelinen MJ, Jauhiainen K, et al. Tumor infiltrating lymphocytes as an independent prognostic factor in transitional cell bladder cancer. Eur J Cancer 1992;29A:69 –75. [13] Yen MC, Lin CC, Chen YL, et al. A novel cancer therapy by skin delivery of indoleamine 2,3-dioxygenase siRNA. Clin Cancer Res 2009;15:641–9. [14] Shieh GS, Shiau AL, Yo YT, et al. Low-dose etoposide enhances telomerase-dependent adenovirus-mediated cytosine deaminase gene therapy through augmentation of adenoviral infection and transgene expression in a syngeneic bladder tumor model. Cancer Res 2006; 66:9957– 66. [15] Sylvester RJ, van der Meijden AP, Oosterlinck W, et al. Predicting recurrence and progression in individual patients with stage TaT1

C.-T. Lin et al. / Urologic Oncology: Seminars and Original Investigations 30 (2012) 680 – 687

[16]

[17]

[18]

[19]

[20] [21] [22] [23]

bladder cancer using EORTC risk tables: A combined analysis of 2,596 patients from seven EORTC trials. Eur Urol 2006;49:466 –77. Zhang L, Conejo-Garcia JR, Katsaros D, et al. Intratumoral t cells, recurrence, and survival in epithelial ovarian cancer. N Engl J Med 2003;348:203–13. Galon J, Costes A, Sanchez-Cabo F, et al. Type, density, and location of immune cells within human colorectal tumors predict clinical outcome. Science 2006;313:1960 – 4. Nakano O, Sato M, Naito Y, et al. Proliferative activity of intratumoral cd8(⫹) t-lymphocytes as a prognostic factor in human renal cell carcinoma: Clinicopathologic demonstration of antitumor immunity. Cancer Res 2001;61:5132– 6. Schumacher K, Haensch W, Roefzaad C, et al. Prognostic significance of activated cd8(⫹) t cell infiltrations within esophageal carcinomas. Cancer Res 2001;61:3932– 6. Yang MH, Chen KK, Yen CC, et al. Unusually high incidence of upper urinary tract urothelial carcinoma in Taiwan. Urology 2002;59:681–7. Petkovic SD. Treatment of bilateral renal pelvic and ureteral tumors. A review of 45 cases. Eur Urol 1978;4:397– 400. Jiaan BP, Yu CC, Lee YH, et al. Uremia with concomitant urothelial cancer. Br J Urol 1993;72:458 – 61. Brausi M, Collette L, Kurth K, et al. Sylvester R; EORTC GenitoUrinary Tract Cancer Collaborative Group. Variability in the recur-

[24]

[25] [26]

[27]

[28]

[29]

687

rence rate at first follow-up cystoscopy after TUR in stage Ta T1 transitional cell carcinoma of the bladder: A combined analysis of seven EORTC studies. Eur Urol 2002;41:523–31. Miladi M, Peyromaure M, Zerbib M, et al. The value of a second transurethral resection in evaluating patients with bladder tumours. Eur Urol 2003;43:241–5. Jakse G, Algaba F, Malmstrom PU, et al. A second-look TUR in T1 transitional cell carcinoma: why? Eur Urol 2004;45:539 – 46. Brauers A, Buettner R, Jakse G. Second resection and prognosis of primary high risk superficial bladder cancer: Is cystectomy often too early? J Urol 2001;165:808 –10. Grimm M-O, Steinhoff Ch, Simon X, et al. Effect of routine repeat transurethral resection for superficial bladder cancer: a long-term observational study. J Urol 2003;170(2 Pt 1):433–7. ¨ , Zorlu F, et al. The effect of repeat transDivrik RT, Yildirim U urethral resection on recurrence and progression rates in patients with T1 tumors of the bladder who received intravesical mitomycin: A prospective, randomized clinical trial. J Urol 2006;175: 1641– 4. Babjuk M, Oosterlinck W, Sylvester R, et al. EAU Guidelines on Non-Muscle-Invasive Urothelial Carcinoma of the Bladder. Eur Urol 2008;54:303–14.