Prognostic impact of tumor-associated immune cell infiltrates at radical cystectomy for bladder cancer

ARTICLE IN PRESS

Urologic Oncology: Seminars and Original Investigations 000 (2019) 1−9

Clinical-Bladder cancer

Prognostic impact of tumor-associated immune cell infiltrates at radical cystectomy for bladder cancer Tina Schubert, M.D.a,b,1, Markus Renninger, M.D.b,1, Manuel Alexander Schmidb, Fahmy Nabil Hassana,b, Ioannis Sokolakis, M.D.a, Omar Fahmy, M.D.b,c, Georgios Hatzichristodoulou, M.D., Ph.D.a, Arnulf Stenzl, M.D., Professorb, Georgios Gakis, M.D., Professora,b,* a

Department of Urology and Pediatric Urology, University Hospital of Wuerzburg, Julius-Maximilians University, Wuerzburg, Germany b Department of Urology, University Hospital of Tuebingen, Eberhard-Karls University, Tuebingen, Germany c Department of Urology, University Putra Malaysia (UPM), Serdang, Selangor, Malaysia Received 16 April 2019; received in revised form 10 August 2019; accepted 19 August 2019

Abstract Objectives: To assess whether the presence and location of tumor-associated immune cell infiltrates (TAIC) on histological slides obtained from cystectomy specimens impacts on oncological outcomes of patients with bladder cancer (BC). Material and methods: A total of 320 consecutive patients staged with cM0 bladder cancer underwent radical cystectomy (RC) between 2004 and 2013. The presence of TAIC (either located peritumorally [PIC] and/or intratumorally [IIC]) on histological slides was retrospectively assessed and correlated with outcomes. Kaplan−Meier analyses were used to estimate the impact of TAIC on recurrencefree (RFS), cancer-specific (CSS), and overall survival (OS). Multivariable Cox-regression analysis was carried out to evaluate risk factors of recurrence. The median follow-up was 37 months (IQR: 10−55). Results: Of the 320 patients, 42 (13.1%) exhibited IIC, 141 (44.1%) PIC and 137 (42.8%) no TAIC in the cystectomy specimens. Absence of TAIC was associated with higher ECOG performance status (P = 0.042), histologically advanced tumor stage (≥pT3a; P < 0.001), lymph node tumor involvement (pN+; P = 0.022), positive soft tissue surgical margins (P = 0.006), lymphovascular invasion (P < 0.001), and elevated serum C-reactive protein levels (P < 0.001). The rate of never smokers was significantly higher in the IIC-group (64.3%) compared to the PIC-group (39.7%, P = 0.007) and those without TAIC (35.8%, P = 0.001). The 3-year RFS/CSS/OS was 73.9%/88.5%/76.7% for patients with IIC, 69.4%/85.2%/70.1% for PIC and 47.6%/68.5%/56.1% for patients without TAIC (P < 0.001/<0.001/0.001 for TAIC vs. no TAIC). In multivariable analysis, adjusted for all significant parameters of univariable analysis, histologically advanced tumor stage (P = 0.003), node-positive disease (P = 0.002), and the absence of TAIC (P = 0.035) were independent prognosticators for recurrence. Conclusions: In this analysis, the presence and location of TAIC in cystectomy specimens was a strong prognosticator for RFS after RC. This finding suggests that the capability of immune cells to migrate into the tumor at the time of RC is prognostically important in invasive bladder cancer. Ó 2019 Elsevier Inc. All rights reserved.

Keywords: Bladder cancer; Immune cell; Infiltrate; Radical cystectomy; Recurrence; Survival

1. Introduction Up to 30% of bladder cancer (BC) patients are diagnosed with muscle-invasive disease (MIBC) at initial presentation

1 Both authors have equally contributed to the manuscript. *Corresponding author. Tel.: +49-931 20132091; fax: +49-931-201-32013. E-mail address: [email protected] (G. Gakis).

https://doi.org/10.1016/j.urolonc.2019.08.013 1078-1439/Ó 2019 Elsevier Inc. All rights reserved.

[1,2]. Radical cystectomy (RC) with pelvic lymph node dissection constitutes the mainstay of treatment for patients with MIBC and non−muscle-invasive disease (NMIBC) progressing on or after intravesical instillation therapy [1,2]. Large studies have consistently demonstrated that tumor stage, lymph node involvement, soft tissue surgical margin status (STSM), and lymphovascular invasion are the major pathological determinants for survival after RC [3−5]. However, despite the absence of these well-established pathologic

ARTICLE IN PRESS 2

T. Schubert et al. / Urologic Oncology: Seminars and Original Investigations 00 (2019) 1−9

risk factors, up to 20% to 30% of patients with organ-confined BC will experience disease recurrence after RC [1,6,7]. Therefore, identification of patients at high risk of recurrence after radical treatment might form the basis of advocating more selectively adjuvant treatment modalities. Generally, in oncology, the impact of immune cells on local cancer control and treatment response has been subject of debate for decades. It was reported that the presence or absence of immune cells may lead to changes within the tumor microenvironment and impact on tumor cell growth and progression [8]. A couple of studies have shown that tumor-infiltrating immune cells (TICs) are capable of eliminating tumor cells [9,10]. The most comprehensive of these studies were conducted in colorectal (CRC) and breast cancer (BRC). In these cancer entities, the density and location of immune cells were not only found to be more accurate in the prediction of prognosis but also in the response to treatment than the currently used tumor node metastasis (TNM) staging system [11,12]. Bladder cancer, like CRC and BRC, is a highly immunogenic malignancy, with an ability to evade the immune system by inhibiting its cytotoxic function and maintaining an immunosuppressive microenvironment [13]. Recently, a positive correlation between the presence of tumor-infiltrating immune cells and treatment response to BCG in patients with NMIBC was demonstrated [14]. Due to these promising correlations, we aimed to assess whether the presence of tumor-associated immune cell infiltrates (TAIC) assessed on cystectomy specimens impacts on the oncological outcomes of patients treated with RC. 2. Material and methods 2.1. Patients This retrospective study was conducted in accordance to the declaration of Helsinki and the provisions of the local ethics committee Tuebingen (approval number: 417/2010A). Our cystectomy database was searched for patients who underwent RC for cM0 staged BC between 2004 and 2013. Records were reviewed for clinical and histopathological characteristics as well as individual treatment response. Patients who underwent neoadjuvant chemotherapy during this time period were excluded (N = 9). Patients with BCGrefractory or relapsing Ta-T1 high-grade bladder cancer were also included in this study since it has been shown that the majority of these tumors show similar genetic profiles like muscle-invasive tumors [15]. A total of 320 consecutive cM0 patients were finally identified and considered for analysis. 2.2. Clinical and histological assessment The presence of TAIC on cystectomy specimens was correlated with the following clinical and pathological parameters: age at RC, gender, Eastern Cooperative Oncology Group (ECOG) performance status (PS), Charlson comorbidity index

(CCI), smoking status, median time between last transurethral bladder tumor resection (TUR-BT) and RC, number of TUR-BTs prior to RC, TAIC at last TUR-BT prior to RC, clinical and histopathological tumor and nodal stage, STSMs, histological entity of BC, tumor grade, estimated tumor size, preoperative administration of intravesical immuno-/chemotherapy, hydronephrosis at RC, and receipt of preoperative and postoperative systemic chemotherapy. In terms of postoperative chemotherapy, any administration of chemotherapy for >=pT3a and/or pN+ disease up to 4 months after radical cystectomy was defined as adjuvant chemotherapy [16]. By contrast, any chemotherapy delivered for any recurrent disease (either in the lymph nodes, local recurrence, or systemic disease) after RC was defined as palliative chemotherapy. In addition, preoperative concentrations of serum creatinine, serum C-reactive protein (CRP), neutrophile-to-lymphocyte ratio, hemoglobin, and thrombocytes were recorded [7,17,18]. Laboratory values were assessed 1 to 3 days prior to RC. 2.3. Histological analysis The histological assessment was based on the TNM classification as approved by the UICC and WHO grading system of 1973 [19,20]. During the investigated time period, cystectomy specimens were routinely macro- and microscopically assessed according to standardized protocols based on H&E and immunohistochemical staining to identify the presence of urothelial and nonurothelial histology [21]. As a unique part of this protocol, any presence of a TAIC was routinely reported in cystectomy reports. Location of the TAIC was further reported as peritumorally located round cell infiltrate (PIC) which describes the formation of a TAIC either at the advancing margin of the tumor, in the stroma or in the normal tissue adjacent to the tumor. By contrast, an intratumoral immune cell infiltrate (IIC) describes the presence of immune cells within the mass of malignant cells with direct proximity between cancer and immune cells (see Fig. 1a and b). The presence of TAIC, either located peritumorally and/ or intratumorally, on histological slides was retrospectively assessed as outlined on the histopathological report. Since quantification of immune cells was not standardized in histopathological reports over the investigated period, we decided to define the absence of any circumscript immune cell infiltrate in the histological report as a negative result. Conversely, any presence of an immune cell infiltrate, either located peritumorally or intratumorally, was considered as a positive result. Since granulomatous inflammatory components are often related to surgery, they were not considered as tumor-specific immune pattern per se. Lymphovascular invasion was defined as the presence of malignant cells within an endothelial lining. Surgical margins were considered positive in case of malignant cells at any soft-tissue margin of the specimen [21]. Histological re-evaluation of cystectomy slides was not part of this study.

ARTICLE IN PRESS T. Schubert et al. / Urologic Oncology: Seminars and Original Investigations 00 (2019) 1−9

3

t test. Kaplan−Meier analyses with log-rank test were used to estimate the impact of TAIC on recurrence-free (RFS), cancer-specific (CSS) and overall survival (OS). Univariable Cox-proportional hazard models were built to evaluate risk factors of recurrence, cancer-specific death or overall death. A multivariable Cox-regression model was constructed for risk factors of recurrence, cancer-specific death and death of any cause based on significant parameters associated with survival in univariable analysis. JMP 14.2 was used for statistical analysis (Cary, NC). 3. Results

Fig. 1. Histopathological slides of tumor associated immune cell infiltrates (TAICs) either located peritumorally (PIC; 1a) or intratumorally (IIC; 1b).

2.4. Follow-up Follow-up was based on electronic hospital charts and physician records. In general, patients were seen postoperatively at least every 3 to 4 months for the first year, semiannually for the second and third year, and annually thereafter. Follow-up examinations included cross-sectional imaging with computed tomography scans and/or magnetic resonance imaging. In addition to physical examination with laboratory testing, intravenous pyelography, cystoscopy, urine cytology, urethral washings, and bone scintigraphy were carried out, if indicated [1,2]. Recurrence was defined as disease recurrence locally or in distant organs. Patients who did not experience recurrence, cancer-specific death or death were censored at last follow-up. The median follow-up was 37 months (interquartile range: 10−55). 2.5. Statistics Statistical analyses were conducted for correlation of the presence and location (IIC vs. PIC) of TAIC with various prognostic parameters using Fisher exact, Pearson x2 test or

Of the 320 patients, 43 (13.1%) exhibited IIC, 141 (44.1%) PIC and 137 (42.8%) no TAIC in the cystectomy specimens. Absence of TAICs was associated with higher ECOG PS (P = 0.042), histologically advanced tumor stage (≥pT3a; P < 0.001), lymph node tumor involvement (pN+; P = 0.022), positive STSMs (P = 0.006), lymphovascular invasion (P < 0.001), and elevated serum CRP levels (P < 0.001). The rate of never smokers was significantly higher in the IIC-group compared to the PIC-group (P = 0.007) and to those without TAIC (P = 0.001). No further significant associations were found between the IICand PIC-group as listed in Table 1. Of the 320 patients, a total of 108 (33.8%), 58 (18.9%), and 122 (38.1%) experienced recurrence, cancer-specific death, and all cause death during the follow-up period, respectively. In Kaplan−Meier analysis, patients who received adjuvant chemotherapy had superior 3-year RFS (P = 0.026), but not CSS (P = 0.76) and OS (P = 0.97) compared to patients who received palliative chemotherapy. The 3-year RFS was 73.9% for patients with IIC, 69.4% for PIC, and 47.6% for patients without TAIC (P < 0.001 for TAIC vs. no TAIC; see Fig. 2). The 3-year CSS was 88.5% for patients with IIC, 85.2% for PIC, and 68.5% for patients without TAIC (P < 0.001 for TAIC vs. no TAIC). Likewise, the 3-year OS was 76.7% for patients with IIC, 70.1% for PIC, and 56.1% for patients without TAIC (P = 0.001 for TAIC vs. no TAIC). In univariable Cox-regression analysis, RFS, CSS, and OS rates were significantly lower in patients with absence of TAIC, histopathologically advanced tumor stage and lymph node positive disease, positive STSMs, and lymphovascular invasion (P < 0.001 for all parameters and survival endpoints). In addition, patients with higher serum CRP concentration exhibited also significantly lower RFS (P = 0.005), CSS (P = 0.002) and OS (P < 0.001). A significantly lower OS was also noted for patients with higher ECOG PS (P < 0.001) and age at RC (P = 0.001) see Table 2). In multivariable analysis, adjusted for all significant parameters of univariable analysis, histologically advanced tumor stage (P = 0.003), node-positive disease (P = 0.002), and the absence of TAIC (P = 0.035) were independent prognosticators for recurrence (Table 2). Lymph node

ARTICLE IN PRESS 4

T. Schubert et al. / Urologic Oncology: Seminars and Original Investigations 00 (2019) 1−9

Table 1 Clinical and histopathological characteristics of patients subdivided according to the presence and location of immune cell infiltrates in cystectomy specimens. Parameter

Number of patients (%) Gender Male Female Smoking Current and/or previous smoker Never Mean age at RC [a] Median IQR Mean time between last TUR-BT and RC [d] Median IQR Mean number of TUR-BTs before RC Median IQR TAICs at TUR-BT prior to RC Absent Present Not reported Charlson Comorbidity Index Mean Median IQR ECOG PS at RC 0 1 2 3 Clinical tumor stage ≥cT3 ≤cT2 cTX pT-stage ≥pT3a ≤pT2b pTX pT0 pTa pTis pT1 pT2a pT2b pT3a pT3b pT4a pT4b Clinical nodal stage cN0 cN1 cN2 cN3 Histopathological nodal stage pN+ pN0 pNX Surgical margins Positive

TAIC

No TAIC

P

IIC

PIC

P

42 (13.1)

141 (44.1)

27 (64.3) 15 (35.7)

109 (77.3) 32 (22.7)

0.10

102 (74.5) 35 (25.5)

1.0

15 (35.7) 27 (64.3) 68 68 62−73

85 (60.3) 56 (39.7) 66 68 59−73

0.007

88 (64.3) 49 (35.8) 69 70 61−76

0.08

34 31 20−45 2.1 1 1−3

48 28 19−48 1.9 1 1−2

59 35 21−53 2.1 1 1−3

0.12

33 (78.6) 9 (21.4) 0 (0)

120 (85.1) 20 (14.2) 1 (0.7)

0.33

123 (89.8) 14 (10.2) 0 (0)

0.18

4.2 4 2−6

4.9 4 2−6

0.90

5.4 5 3−7

0.64

41 (97.6) 1 (2.4) 0 (0) 0 (0)

122 (86.6) 16 (11.4) 2 (1.4) 1 (0.7)

0.24

106 (77.4) 26 (19.0) 4 (2.9) 1 (0.7)

0.042

9 (21.4) 32 (76.2) 1 (2.4)

38 (27.0) 101 (71.6) 2 (1.4)

0.54

38 (27.7) 97 (70.8) 2 (1.5)

0.70

14 (33.3) 28 (66.7) 0 (0) 0 (0) 1 (0.2) 5 (11.9) 4 (9.5) 11 (26.2) 6 (14.3) 6 (14.3) 7 (16.6) 2 (4.8) 0 (0)

61 (43.3) 80 (56.7) 0 (0) 1 (0.7) 7 (5.0) 13 (9.2) 20 (14.2) 16 (11.3) 24 (17.0) 25 (17.7) 21 (14.9) 13 (9.2) 1 (0.7)

0.28

86 (62.8) 51 (37.2) 1 (0.7) 13 (9.5) 3 (2.2) 4 (2.9) 5 (3.7) 11 (8.0) 14 (10.2) 27 (19.7) 33 (24.1) 17 (12.4) 9 (6.6)

<0.001

37 (88.1) 1 (2.4) 3 (7.1) 1 (2.4)

115 (81.6) 9 (6.4) 14 (10.0) 3 (2.1)

0.70

114 (83.2) 7 (5.1) 15 (11.0) 1 (0.7)

0.72

5 (11.9) 36 (85.7) 1 (2.3)

33 (23.4) 102 (72.3) 6 (4.3)

0.20

44 (32.1) 83 (60.6) 10 (7.3)

0.022

5 (11.9)

11 (7.8)

0.41

27 (19.7)

0.006

137 (42.8)

0.38

0.17

0.54

0.08

0.70

(continued)

ARTICLE IN PRESS T. Schubert et al. / Urologic Oncology: Seminars and Original Investigations 00 (2019) 1−9

5

Table 1 (Continued) Parameter

Negative Not assessible Lymphovascular invasion LVI LV0 LVX Estimated tumor size [cm] Mean Median IQR Tumor grade G1 G2 G3 GX Hydronephrosis at RC Present Absent Data not available Nonpure UC pathology at RC Present Absent Pure SCC Pure AC Pure sarcomatoid UC/AC UC/SCC UC/Sarcomatoid UC/Micropapillary UC/Small cell UC/nested SCC/nested Preop. serum creatinine [mg/dl] Mean Median IQR Preop. neutrophile-to-lymphocyte ratio (NLR) Mean Median IQR Preop. serum C-reactive protein [mg/dl] Mean Median IQR Preop. hemoglobin [mg/dl] Mean Median IQR Preop. thrombocytes (x103/ml) Mean Median IQR Intravesical BCG and/or chemotherapy Performed Not performed Postoperative systemic chemotherapy Adjuvant Palliative

TAIC

No TAIC

P

IIC

PIC

P

37 (88.1) 0 (0)

129 (91.5) 1 (0.7)

8 (19.0) 34 (81.0) 0 (0)

41 (29.1) 89 (63.1) 11 (7.8)

0.16

62 (45.3) 66 (48.2) 9 (6.6)

<0.001

2.8 2.5 1.5−3.5

3.2 3 2−4

0.12

3.0 3 1.7−4.0

0.65

0 (0) 14 (33.3) 27 (64.3) 1 (2.4)

1 (0.7) 45 (31.9) 88 (62.4) 3 (2.1)

0.74

1 (0.7) 29 (21.2) 93 (67.9) 14 (10.2)

0.12

8 (19.0) 34 (81.0) 0 (0)

27 (19.4) 113 (80.1) 1 (0.7)

1.0

28 (20.4) 109 (79.6) 0 (0)

0.88

5 (11.9) 37 (88.1) 1 (2.4) 0 (0) 0 (0) 1 (2.4) 0 (0) 0 (0) 1 (2.4) 1 (2.4) 1 (2.4) 0 (0)

27 (19.2) 115 (81.6) 2 (1.4) 0 (0) 0 (0) 1 (0.7) 5 (3.6) 6 (4.3) 4 (2.8) 1 (0.7) 8 (5.7) 0 (0)

0.48

31 (22.6) 106 (77.4) 5 (3.6) 1 (0.7) 1 (0.7) 0 (0) 6 (4.4) 2 (1.5) 4 (2.9) 2 (1.5) 9 (6.6) 1 (0.7)

0.53

1.0 1.0 0.8−1.1

1.0 1.1 0.8−1.1

0.76

1.0 1.0 0.8−1.2

0.33

2.5 2.5 1.9−3.2

3.6 3.3 2.3−3.9

0.14

5.4 3.0 2.5−4.6

0.08

0.2 0.5 0.1−0.6

1.0 0.4 0.1−1.1

0.06

2.9 0.5 0.1−1.8

<0.001

13.3 13.4 12.3−14.8

13.3 13.5 12.4−14.6

0.94

12.9 13.3 11.4−14.4

0.09

294 292 229−347

309 294 237−357

0.38

311 278 224−384

0.64

12 (28.6) 30 (71.4)

38 (27.0) 103 (73.0)

0.84

40 (29.2) 97 (70.8)

0.80

2 (4.7) 3 (7.0)

5 (3.6) 22 (15.6)

0.48

5 (3.7) 29 (21.1)

0.12

105 (76.6) 5 (3.7)

Abbreviations: a = year; AC = adenocarcinoma; BCG = Bacille−Calmette−Guerin; CCI = Charlson-comorbidity index; d = days; ECOG PS = Eastern Cooperative Oncology Group performance status; IIC = intratumoral immune cells; IQR = interquartile range; MIBC = muscle-invasive bladder cancer; mo = months; P = P value; PIC = peritumoral immune cells; preop. = preoperative; RC = radical cystectomy; SCC = squamous cell carcinoma; TAIC = tumorassociated immune cells; TUR-BT = transurethral bladder tumor resection; UC = urothelial carcinoma; bold values indicate a statistically significant difference.

ARTICLE IN PRESS 6

T. Schubert et al. / Urologic Oncology: Seminars and Original Investigations 00 (2019) 1−9

Number of patients at risk of recurrence at given time intervals after RC Variable/Time

0

12

24

36

48

60

IIC

42

33

29

23

18

14

PIC

141

90

73

58

44

31

No TAIC

137

71

49

32

21

14

Fig. 2. Recurrence-free survival for patients with cM0 bladder cancer treated with RC (P < 0.001 between TAIC and no TAIC) according to the presence of intratumoral (IIC; green line) vs. peritumoral (PIC; blue line) vs. no tumor-associated immune cell infiltrate (TAIC; red line) in cystectomy specimens (Leg.: IIC: intratumor immune cell infiltrates; PIC: peritumoral immune cell infiltrates; TAIC: tumor-associated immune cell infiltrates (IIC+PIC); RC: radical cystectomy)).

tumor involvement (P < 0.001) was an independent risk factor for CSS while the absence of TAIC was not (P = 0.055). Nodal involvement (P < 0.001), positive STSMs (P = 0.016), age at RC (P = 0.017), and ECOG PS (P = 0.034) were independent predictors for OS. 4. Discussion There is an increasing body of evidence suggesting immunosurveillance as a significant contributor to cancer development. In the last years, checkpoint inhibitors targeting the PD-1 or PD-L1 axis have opened a new era for the treatment of metastatic urothelial cancer [22]. In this regard, the role of TICs has been the subject of debate for decades. A large number of studies have shown that TICs exert cytotoxic effects on tumor cells in a variety of cancer entities [23,24]. Despite effective local treatment options, MIBC is associated with an approximately 50% risk of recurrence, and the 5-year OS of only 10% to 15% is one of the lowest among all metastatic cancer entities [2]. Although many prognostic parameters, such as clinicopathologic factors and a variety of sophisticated biomarkers, have been investigated for the prediction of recurrence, their prognostic accuracy in the postcystectomy setting remains

controversial. Therefore, the use of molecular predictors for recurrence after RC is currently not recommended in daily clinical routine [2]. We aimed to investigate the prognostic significance of TAIC as a “simple” histopathological characteristic on cystectomy slides. Interestingly, we found RFS, CSS, and OS rates to be significantly lower in patients without TAIC in the cystectomy specimens. Moreover, we found a strong correlation between the absence of TAIC and histopathological risk factors like advanced tumor stage, positive lymph node involvement, positive surgical margins, and lymphovascular invasion. These findings suggest a prognostic role of TAIC in the postcystectomy setting. Our results are consistent with findings in CRC. A couple of independent studies with large cohorts demonstrated that a lower grade of TICs was independently associated with worse oncological outcomes after radical colorectal surgery [25−27]. A meta-analysis of 9 trials examining tumor inflammation in CRC confirmed an improved OS (hazard ratio: 0.59 (95% confidence interval: 0.48−0.72, P < 0.001)) as well as CSS (hazard ratio: 0.40 (95% confidence interval: 0.27−0.61, P < 0.001)) for patients with prominent TAIC compared to those without [26]. Moreover, there is evidence that TAIC are associated with improved prognostication than the TNM staging system for

ARTICLE IN PRESS

Abbreviations: a = year; CI = confidence interval; CRP = C-reactive protein; CSS = cancer-specific survival; LV0 = lymphovascular invasion absent; LVI = lymphovascular invasion present; OS = overall survival; RFS = recurrence-free survival; RR = risk ratio; TAICs = tumor-associated immune cells; bold values indicate a statistically significant difference.

0.017 1.02 (1.00−1.04) − − 1.02 (0.99−1.04) 0.61 1.01 (0.99−1.02)

0.26

1.03 (1.01−1.05)

0.001

−

−

0.034 − 0.32

TAICs absent vs. present ≥pT3a vs. ≤pT2b pN+ vs. pN0 Pos. vs. neg. surgical margins LVI vs. LV0 Serum CRP (cont. per unit [mg/ml]) ECOG PS at RC ≥ 1 vs. 0 Age at RC (cont.; per year)

1.26 (0.56−2.54) 0.96 0.90 (0.48−1.56)

0.54

2.04 (1.30−3.10)

0.002

0.74 (0.38−1.33)

−

1.69 (1.04−2.75)

0.75 0.37 1.08 (0.66−1.78) 1.03 (0.97−1.10) 0.22 0.55 1.57 (0.75−3.27) 1.02 (0.44−4.68) 0.56 0.35 1.15 (0.71−1.86) 1.60 (0.59−4.33) <0.001 <0.001 2.30 (1.60−3.30) 1.11 (1.06−1.16) 4.63 (2.71−8.21) 1.09 (1.01−1.16) <0.001 0.012 2.76 (1.88−4.04) 1.07 (1.01−1.13)

<0.001 0.029

0.15 <0.001 0.016 1.43 (0.87−2.35) 2.48 (1.50−4.11) 2.01 (1.14−3.57) 0.43 <0.001 0.12 1.34 (0.64−2.81) 4,43 (2.11−9.30) 1.75 (0.85−3.59) 0.003 0.002 0.16 2.06 (1.27−3.39) 2.13 (1.30−3.51) 1.50 (0.84−2.61) <0.001 <0.001 <0.001 5.71 (2.85−9.64) 3.31 (2.26−4.84) 3.97 (2.52−6.04) 5.10 (2.85−9.64) 7.91 (4.46−14.54) 5.38 (2.94−9.43) <0.001 <0.001 <0.001 3.73 (2.49−5.71) 3.70 (2.49−5.49) 3.36 (2.02−5.36)

<0.001 <0.001 <0.001

1.77 (0.99−3.19) 0.035 1.59 (1.03−2.44) <0.001 1.90 (1.33−2.72) 2.62 (1.56−4.52) <0.001 2.02 (1.38−2.96)

<0.001

RR (95% CI) RR (95% CI) RR (95% CI) RR (95% CI)

p

p

RR (95% CI)

p

RFS OS CSS RFS

Univariable Parameter

Table 2 Uni- and multivariable Cox- regression analysis for survival of patients with bladder cancer treated with radical cystectomy.

Multivariable

p

CSS

Multivariable

p

0.055

1.20 (0.78−1.83)

p RR (95% CI)

OS

Multivariable

0.40

T. Schubert et al. / Urologic Oncology: Seminars and Original Investigations 00 (2019) 1−9

7

CRC [27]. In ovarian and breast cancer, the presence of TAIC was associated with improved OS [28−30]. These tumor entities, including urothelial BC, are considered to be highly immunogenic [13]. Cancer cells express high levels of neoantigens and ligands that can inhibit immune cell function and activation by suppression T cell function [31]. This potentially empowers the tumor cells to evade the antitumor immunity cycle. In addition to the presence, the localization of immune cells seems to be of importance for an effective cancer control. In our study, we found an improved RFS, CSS, and OS in favor of patients exhibiting an intratumorally located immune cell infiltrate in their specimens albeit not statistically significant. Similar results were reported in patients with non−small lung cancer and melanoma [32,33]. Interestingly, patients who never smoked were more likely to exhibit intratumoral immune cell infiltrates. In this regard, a study reported that smoking, besides its genotoxic properties, induces a chronic and subclinical immune suppression which contributes to cancer progression [34]. Altogether, these findings suggest that the presence of immune cells within the tumor mass is of prognostic benefit and may also reflect a higher local activity of the immune system. A qualitative description of the interaction of tumor cells and TAIC has been termed the “immune contexture,” and includes the location and the type of TAIC as well as the cytokines and chemokines involved in this microenvironment [35]. Therefore, a large effort has been made to investigate the impact of the different immune cells subtypes on the oncological outcome of patients. Nevertheless, the exact interactions and functions remain unclear. Even within individual immune cell types, there are opposing functions as CD4+ T cells, CD8+ T cells, macrophages, and natural killer T cells have either tumor-suppressive or tumor-promoting properties, depending on the tissue context and surrounding cellular stimuli [36]. In general, a pro-inflammatory tumor microenvironment and infiltrating CD8+ T lymphocytes are both associated with improved clinical outcomes, whereas myeloid-derived suppressor cells and regulatory T cells may decrease immunologic functions resulting in worse outcome in a broad range of cancer entities [37]. In terms of nonmuscle invasive bladder cancer (NMIBC), a positive association between oncological outcome and a higher rate of infiltrating CD4+, CD8+, and CD3+ T lymphocytes was found [38]. Recently, a survival analysis for NMIBC treated with BCG demonstrated prolonged RFS in patients with an increased count of CD4+ and GATA3+ T cells, whereas TAMs, Tregs, and T-bet+ T cells were inversely correlated with RFS [14]. With respect to MIBC, conflicting results have been published [37]. Yet, routine status assessment of the tumor immune microenvironment may be important at the background of ongoing developments regarding checkpoint inhibitors and targeted immunotherapy. Mariathasan et al. reported that concurrent TGF- and PD-L1 signaling blocking promotes infiltration of immune

ARTICLE IN PRESS 8

T. Schubert et al. / Urologic Oncology: Seminars and Original Investigations 00 (2019) 1−9

cells into the tumor in tumors unresponsive to PD-L1 blockage [39]. Back in 2007, Sharma et al. reported that the level of CD8(+) immune cells in muscle-invasive bladder cancers correlated with outcomes [40]. Later, Sweis et al. showed that lack of PD-L1 response was related to exclusion of CD8(+) killer cells resulting in a non-T cell inflamed tumor microenvironment [41]. Sj€ odahl et al. reported that a low level CD3(+) tumor-infiltrating lymphocytes with concurrent high level of CD68(+) tumor-associated macrophages contributed to worse outcomes in MIBC [42]. Eckstein et al. reported further that high concentration of stromal tumor-infiltrating lymphocytes, which enrich in tertiary lymphoid structures, mediate antitumoral activity [43]. Bellmunt et al. reported on the expression levels of PD-L1 in the tumor cell membrane and in tumor infiltrating mononuclear cells [44]. By contrast to PD-L1 expression on tumor cells, expression on tumor infiltrating macrophages was predictive for survival. Altogether, it still remains questionable whether the efforts of a characterization will give a reliable prognostic benefit. Due to our retrospective study design, we cannot address the impact of the different immune cell subtypes. Nevertheless, our data confirm the prognostic benefit of TAIC suggesting that the capability of immune cells to migrate into the tumor at the time of RC may be prognostically important. Our study is limited by its retrospective design, the missing subcharacterization (i.e., PD-L1 staining) and quantification of the immune cells and lack of second review of slides by a single independent pathologist. We agree that reassessment by a single or 2 pathologists might be important but would still inherit the risk of intra- and interobserver variabilities. Nonetheless, in terms of standardization, we considered this report to be valid since the slides were assessed in a single university pathology department by experienced pathologists throughout the time period. Recent data show that neoadjuvant chemotherapy reinforces antitumoral immune response in MIBC [45]. Since our series consisted only of patients who were treated without neoadjuvant chemotherapy, we cannot provide data on how the presence and location of immune cell infiltrates impacts on outcomes of patients who underwent neoadjuvant treatment. Considering the slight prognostic advantage of intratumoral over peritumoral immune cell infiltrates, we considered any presence of an intratumoral location as an IIC, irrespective of whether an additional peritumoral immune cell infiltrate was presented. Therefore, our study cannot address possible prognostic differences between patients with combined IIC plus PIC and those with a solitary IIC. We decided to pool IIC and PIC together into 1 group for final analysis since there was no significant difference in survival outcomes albeit a trend toward superior survival outcomes was observed for IICs. Nevertheless, it can be assumed that peritumorally located immune cells may have different antitumoral activities compared to

intratumoral immune cells for which further subcharacterization would be interesting. Yet, the most interesting question that needs to be resolved is whether subtyping of infiltrating lymphocytes really provides any additional prognostic information beyond the histopathological evaluation of the lymphocytic reaction patterns. To date there remains a lack of consensus whether the presence of TAICs in BC specimens should be explicitly reported the cystectomy reports. In 2015, an international working group developed a standardized methodology for visual assessment of TAIC on H&E sections in breast cancer [46]. This standardized approach includes 5 simple steps to assess the presence and location of TAIC allowing easy integration into daily practice. Nonetheless, validation of our results in a prospective cohort study is desirable. Based on the present body of evidence, including our data, histopathological reporting of TAIC offers the potential for improved prognostication and clinical decision-making for adjuvant treatment in invasive bladder cancer. 5. Conclusions In this study, the presence of TAIC in cystectomy specimens was found to be a strong and easily assessable prognosticator for improved RFS after RC. This finding suggests that the capability of immune cells to migrate into the tumor at the time of RC may exert a prognostic benefit in invasive bladder cancer. Therefore, routine histopathological reporting of TAIC may help to identify patients at high risk of recurrence and improve the clinical decision-making for adjuvant treatment. Acknowledgments We thank Marcus Scharpf, M.D., Institute of Pathology, University Hospital of Tuebingen, for providing the slides depicted in Fig. 1. References [1] Gakis G, Efstathiou J, Lerner SP, et al. ICUD-EAU International Consultation On Bladder Cancer 2012: radical cystectomy and bladder preservation for muscle-invasive urothelial carcinoma of the bladder. Eur Urol 2013;63:45–57. [2] Alfred Witjes J, Lebret T, Comperat EM, et al. Updated 2016 EAU guidelines on muscle-invasive and metastatic bladder cancer. Eur Urol 2017;71:462–75. [3] Stein JP, Lieskovsky G, Cote R, et al. Radical cystectomy in the treatment of invasive bladder cancer: long-term results in 1,054 patients. J Clin Oncol 2001;19:666–75. [4] Madersbacher S, Hochreiter W, Burkhard F, et al. Radical cystectomy for bladder cancer today−a homogeneous series without neoadjuvant therapy. J Clin Oncol 2003;21:690–6. [5] Novara G, Svatek RS, Karakiewicz PI, et al. Soft tissue surgical margin status is a powerful predictor of outcomes after radical cystectomy: a multicenter study of more than 4,400 patients. J Urol 2010;183:2165–70.

ARTICLE IN PRESS T. Schubert et al. / Urologic Oncology: Seminars and Original Investigations 00 (2019) 1−9 [6] Gakis G, Black PC, Bochner BH, et al. Systematic review on the fate of the remnant urothelium after radical cystectomy. Eur Urol 2017;71:545–57. [7] Schubert T, Todenhofer T, Mischinger J, et al. The prognostic role of pre-cystectomy hemoglobin levels in patients with invasive bladder cancer. World J Urol 2016;34:829–34. [8] Mantovani A, Allavena P, Sica A, Balkwill F. Cancer-related inflammation. Nature 2008;454:436–44. [9] Jiang B, Mason J, Jewett A, et al. Tumor-infiltrating immune cells: triggers for tumor capsule disruption and tumor progression? Int J Med Sci 2013;10:475–97. [10] Song G, Hsiao H, Wang JL, et al. Differential impact of tumorinfiltrating immune cells on basal and luminal cells: implications for tumor invasion and metastasis. Anticancer Res 2014;34:6363– 80. [11] Mlecnik B, Tosolini M, Kirilovsky A, et al. Histopathologic-based prognostic factors of colorectal cancers are associated with the state of the local immune reaction. J Clin Oncol 2011;29:610–8. [12] Garcia-Martinez E, Gil GL, Benito AC, et al. Tumor-infiltrating immune cell profiles and their change after neoadjuvant chemotherapy predict response and prognosis of breast cancer. Breast Cancer Res 2014;16:488. [13] Gakis G. The role of inflammation in bladder cancer. Adv Exp Med Biol 2014;816:183–96. [14] Pichler R, Fritz J, Zavadil C, et al. Tumor-infiltrating immune cell subpopulations influence the oncologic outcome after intravesical Bacillus Calmette-Guerin therapy in bladder cancer. Oncotarget 2016;7:39916–30. [15] Patschan O, Sjodahl G, Chebil G, et al. A molecular pathologic framework for risk stratification of stage T1 urothelial carcinoma. Eur Urol 2015;68:824–32:discussion 835-826. [16] Sternberg CN, Skoneczna I, Kerst JM, et al. Immediate versus deferred chemotherapy after radical cystectomy in patients with pT3pT4 or N+ M0 urothelial carcinoma of the bladder (EORTC 30994): an intergroup, open-label, randomised phase 3 trial. Lancet Oncol 2015;16:76–86. [17] Gakis G, Todenhofer T, Renninger M, et al. Development of a new outcome prediction model in carcinoma invading the bladder based on preoperative serum C-reactive protein and standard pathological risk factors: the TNR-C score. BJU Int 2011;108:1800–5. [18] Todenhofer T, Renninger M, Schwentner C, et al. A new prognostic model for cancer-specific survival after radical cystectomy including pretreatment thrombocytosis and standard pathological risk factors. BJU Int 2012;110:E533–40. [19] Edge SB, Compton CC. The American Joint Committee on Cancer: the 7th edition of the AJCC cancer staging manual and the future of TNM. Ann Surg Oncol 2010;17:1471–4. [20] Chen Z, Ding W, Xu K, et al. The 1973 WHO Classification is more suitable than the 2004 WHO Classification for predicting prognosis in non-muscle-invasive bladder cancer. PloS One 2012;7:e47199. [21] Cheng L, Montironi R, Davidson DD, Lopez-Beltran A. Staging and reporting of urothelial carcinoma of the urinary bladder. Modern Pathol 2009;22(Suppl 2):S70–95. [22] Mischinger J, Comperat E, Schwentner C, et al. Inflammation and cancer: what can we therapeutically expect from checkpoint inhibitors? Cur Urol Rep 2015;16:59. [23] Man YG, Stojadinovic A, Mason J, et al. Tumor-infiltrating immune cells promoting tumor invasion and metastasis: existing theories. J Cancer 2013;4:84–95. [24] Lipponen PK, Eskelinen MJ, Jauhiainen K, et al. Tumour infiltrating lymphocytes as an independent prognostic factor in transitional cell bladder cancer. Eur J Cancer 1992;29A:69–75. [25] Huh JW, Lee JH, Kim HR. Prognostic significance of tumor-infiltrating lymphocytes for patients with colorectal cancer. Arch Surg 2012;147:366–72.

9

[26] Mei Z, Liu Y, Liu C, et al. Tumour-infiltrating inflammation and prognosis in colorectal cancer: systematic review and meta-analysis. Brit J Cancer 2014;110:1595–605. [27] Jochems C, Schlom J. Tumor-infiltrating immune cells and prognosis: the potential link between conventional cancer therapy and immunity. Exp Biol Med 2011;236:567–79. [28] 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. [29] DeNardo DG, Coussens LM. Inflammation and breast cancer. Balancing immune response: crosstalk between adaptive and innate immune cells during breast cancer progression. Breast Cancer Res 2007;9:212. [30] Schmidt M, Bohm D, von Torne C, et al. The humoral immune system has a key prognostic impact in node-negative breast cancer. Cancer Res 2008;68:5405–13. [31] Schmielau J, Finn OJ. Activated granulocytes and granulocytederived hydrogen peroxide are the underlying mechanism of suppression of t-cell function in advanced cancer patients. Cancer Res 2001;61:4756–60. [32] Dieu-Nosjean MC, Antoine M, Danel C, et al. Long-term survival for patients with non-small-cell lung cancer with intratumoral lymphoid structures. J Clin Oncol 2008;26:4410–7. [33] Ladanyi A. Prognostic and predictive significance of immune cells infiltrating cutaneous melanoma. Pigment Cell Melanoma Res 2015; 28:490–500. [34] Clement JM, Duan F, Srivastava PK. Smoking-induced immune deviation contributes to progression of bladder and other cancers. Oncoimmunology 2015;4:e1019199. [35] Becht E, Giraldo NA, Dieu-Nosjean MC, et al. Cancer immune contexture and immunotherapy. Cur Opin Immunol 2016;39:7–13. [36] Grivennikov SI, Greten FR, Karin M. Immunity, inflammation, and cancer. Cell 2010;140:883–99. [37] Barnes TA, Amir E. HYPE or HOPE: the prognostic value of infiltrating immune cells in cancer. Brit J Cancer 2017;117:451–60. [38] Rayn KN, Hale GR, Grave GP, Agarwal PK. New therapies in nonmuscle invasive bladder cancer treatment. Ind J Urol 2018;34:11–9. [39] Mariathasan S, Turley SJ, Nickles D, et al. TGFbeta attenuates tumour response to PD-L1 blockade by contributing to exclusion of T cells. Nature 2018;554:544–8. [40] 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 U S A 2007;104:3967–72. [41] Sweis RF, Spranger S, Bao R, et al. Molecular drivers of the non-Tcell-inflamed tumor microenvironment in urothelial bladder cancer. Cancer Immunol Res 2016;4:563–8. [42] Sjodahl G, Lovgren K, Lauss M, et al. Infiltration of CD3(+) and CD68(+) cells in bladder cancer is subtype specific and affects the outcome of patients with muscle-invasive tumors. Urol Oncol 2014; 32:791–7. [43] Eckstein M, Wirtz RM, Pfannstil C, et al. A multicenter round robin test of PD-L1 expression assessment in urothelial bladder cancer by immunohistochemistry and RT-qPCR with emphasis on prognosis prediction after radical cystectomy. Oncotarget 2018;9:15001–14. [44] Bellmunt J, Mullane SA, Werner L, et al. Association of PD-L1 expression on tumor-infiltrating mononuclear cells and overall survival in patients with urothelial carcinoma. Ann Oncol 2015;26: 812–7. [45] Krantz D, Hartana CA, Winerdal ME, et al. Neoadjuvant chemotherapy reinforces antitumour T cell response in urothelial urinary bladder cancer. Eur Urol 2018;74:688–92. [46] Salgado R, Denkert C, Demaria S, et al. The evaluation of tumorinfiltrating lymphocytes (TILs) in breast cancer: recommendations by an International TILs Working Group 2014. Ann Oncol 2015;26: 259–71.