Bladder cancer is the most common malignancy of the urinary tract (Siegel et al., 2011), with nonmuscle–invasive urothelial carcinoma (NMIUC) being its most frequent type. Approximately, 70% of patients treated by transurethral resection (TUR) experience a disease relapse, and 15–25% progress to muscle-invasive cancer (Amling, 2001). Thus, the assessment of the risk for tumor recurrence/progression is important to choose the appropriate adjuvant treatment (Babjuk et al., 2008; Witjes and Hendricksen, 2008).
Tumor biology, including tumor progression and therapy response, are influenced by the tumor microenviroment (tumor stroma) (Sica et al., 2008; Qian and Pollard, 2010) with its stromal cells, blood vessels, and infiltrating leukocytes (Sica et al., 2008). Tumor-associated macrophages (TAMs) are a major component of the tumor stroma that was shown to contribute to tumor progression in several types of cancer (Mantovani et al., 2002; Sica et al., 2008). In addition, the induction of an effective antitumor response relies on the innate and adaptive immunity coordinated by dendritic cells (DCs). DCs, after recognizing tumor molecules, become activated, capture tumor antigens, and present them to T-cells. The activated antigen-specific T-cells then migrate to the tumor sites to kill the transformed cells (Pinzon-Charry et al., 2005).
Under the influence of tumor-derived factors, DC dysfunction occurs with resultant generation and accumulation of immature macrophages, immature DC, and apoptotic DC within the tumor tissue. The immature macrophages cannot present antigens, and suppress tumor-specific T-cell activation through the production of reactive oxygen species, while the immature and apoptotic DCs can present antigens but fail to provide adequate costimulatory signals to the T-cells. This induces immunologic tolerance (anergy) or results in abortive proliferation of effector T-cells, thereby favouring tumor evasion (Pinzon-Charry et al., 2005).
Immunotherapy is a main modality for treatment of bladder carcinoma (Bohle and Bock, 2004; Brandau and Suttmann, 2007; Babjuk et al., 2013). Intravesical instillations of bacillus Calmette–Guérin (BCG) proved to be most effective to prevent recurrence and progression of high-risk NMIUC, as well as to treat carcinoma in situ (Brandau and Suttmann, 2007), and yet, 30–40% of patients failed to respond to BCG immunotherapy (Amling, 2000). Moreover, DC-based cancer immunotherapy was proposed as an additional treatment against advanced bladder cancer (Nishiyama et al., 2001).
BCG induces a massive influx of various leukocyte types into the bladder wall, including macrophages and DCs. The increased macrophages in the bladder wall present the BCG antigen to T-cells, and also function as tumoricidal cells upon activation by BCG (Luo et al., 2006; Luo et al., 2010). Thus, an immunocompetent host response appears to be mandatory for BCG to apply its action (Ayari et al., 2009), and yet, excess TAMs could stimulate phagocytosis with resultant BCG elimination. Moreover, Takayama et al. (2009) reported that through the induction of an immunosuppressive cytokine pattern, TAMs could interfere directly with BCG treatment response. Thus, the study of immunologic markers appears to be a promising tool to predict BCG treatment response (Saint et al., 2003).
On the basis of previous reports stating that the level of bladder tumor infiltration by immune cells before BCG treatment might have an impact on the response to therapy (Ayari et al., 2009) the current study was performed. Our aim was to semiquantitatively assess the level of tumor infiltration by CD68+ TAMs and CD1a+ DCs in TUR samples of papillary NMIUC before BCG immunotherapy, and to study their potential prognostic significance. This was done by correlating the outcome of BCG immunotherapy (success vs. failure) with the pretreatment level of tumor infiltration by macrophages and DCs, in a trial to identify patients with recurrent tumors who require adjuvant treatment.
Patients and methods
Cohort of patients
This retrospective study started with the data of 90 patients who presented with solitary papillary NMIUC during the period between January 2013 and December 2014 and underwent initial TUR followed by BCG immunotherapy, which is the standard treatment regimen for superficial bladder carcinoma, at the Department of Urogenital Surgery, Main University Hospital, Alexandria Faculty of Medicine. The study was approved by Alexandria University, Faculty of Medicine Research Ethics Committee.
Clinical data of all cases were obtained from the patients’ files. The mean age of the patients was 58.7 (range: 40–84) years, with a male-to-female ratio of 55 : 8. All patients received induction bacillus Calmette–Guérin (iBCG) therapy for 6 consecutive weeks, starting 3 weeks after surgery. Patients were subsequently scheduled to receive maintenance bacillus Calmette–Guérin (mBCG) therapy. The schedule implimented in our institute is iBCG followed by mBCG protocol with instillations every 3 months for 2 years. Compliance was variable, with 29 patients receiving only induction BCG, 14 receiving the induction and a single mBCG, and only 20 receiving more than one of the scheduled maintenance cycles.
The follow-up of patients was scheduled as control cystoscopy with or without urine cytology every 3 months during the first 2 years after TUR, and after that biannually. Recurrence was defined as a tumor at control cystoscopy after the operation, localized away from the primary tumor bed and the area of the initial resection. BCG treatment failure, as opposed to BCG success, was defined as patients submitted to BCG treatment with tumor recurrence. Finally, recurrence-free survival (RFS) was defined as the period between the iBCG treatment and either recurrence or the most recent tumor-free cystoscopy and cytology.
Exclusion criteria included cases with incomplete clinical and follow-up data, cases with multiple tumors, cases that received adjuvant chemotherapy, or any other medical intervention before the initial TUR, recurrent cases with no available tumor tissue blocks before BCG therapy, and cases with insufficient tumor tissue for immunostaining. Following these exclusion criteria, only 63 cases remained to be enrolled in the current study.
All microscopic slides of each case were reviewed to confirm the diagnosis. Tumor grading and staging were carried out according to the guidelines of WHO/ISUP (2004) (Eble et al., 2004) and UICC-TNM (2002) (Rosai, 2004), respectively. Accordingly, cases were classified as papillary urothelial carcinomas of low (n=31) or high (n=32) grade. Patients were then divided in two groups: patients who developed recurrent disease (with or without progression) (n=38) and patients who did not develop recurrent disease during the follow-up for a minimum of 6 months (n=25). For each case the most representative section with the maximum viable tumor bulk was selected, and its formalin-fixed paraffin embedded TUR block was retrieved from the archives of the Pathology Department, Alexandria Faculty of Medicine, for immunohistochemical staining.
For immunohistochemical staining, the selected tumor tissue blocks were cut into 5-μm thick sections on Superfrost/Plus slides (Thermo Scientific, Fremont, USA). Following deparaffinization, antigen retrieval was performed by heating the slides in 0.01 mol/l citrate buffer (pH 6.0) for CD68, and Hier buffer for CD1a in microwave for 20 min endogenous peroxidase activity was blocked using 3% H2O2. Primary antibodies against human CD68 Ab-3 (mouse monoclonal, clone KP1; Thermo Scientific; Fremont, USA, 1 : 100), and CD1a (mouse monoclonal, ready to use, clone 010; Dako, California, USA), were then applied overnight. Antigen visualization was performed with Thermo Scientific UltraVision LP Detection System. Immunohistochemical reactions were developed with diaminobenzidine and sections were counterstained with Harrris hematoxylin. All immunostains were manually processed. Appropriate positive (tonsillar tissue for CD68 and CD1a) and negative (omission of the primary antibody) controls were included for each immunohistochemical run.
Scoring of immunostained cells in the papillary axis, in the stroma, in the lymphoid aggregates, and infiltrating into tumor masses was carried out by two independent observers (S.E. and H.F.) using an Olympus microscope.
For CD1a the number of CD1a+ TIDCs per ×40 microscopic field was scored as follows: 0 (no cells), 1 (weak/1–5 cells), 2 (moderate/6–10 cells), and 3 (high/≥10 cells). A scoring index of the average score of positive cells within the diverse structures, taking into account the presence or absence of each scored structure, was then calculated for each sample, and a high (2 or 3) or low (0 or 1) infiltration score was established (sum of individual scores divided by the number of structures present) (Ayari et al., 2009).
As regards CD68, the CD68+ macrophages infiltrating the tumor stroma and tumor areas were counted independently by S.E. and H.F. Each specimen was screened at low magnification (×10), and the five areas (×40) with highest number of positively stained cells (hot-spot area) were selected. Macrophages were then counted in the tumor stroma, which included the papillary axis, lymphoid aggregates, and stroma, and in tumor islets. Macrophage counts were classified as low or high. The criteria used for macrophage-specific counting were as follows: (i) cells must present the shape of a macrophage or exhibit the macrophage characteristic staining pattern; (ii) must present cell nucleus; and (iii) be birefringent if the size is small (Lima et al., 2014). The expression of CD68 within tumor cells was also scored as positive or negative.
Data were fed to the computer using the statistical package for the social science (SPSS) program for statistical analysis (ver 21). Qualitative data were described using number and percentage. Quantitative data were described using measures of central tendency (mean and median) and dispersion (SD, minimum, and maximum). Association between categorical variables was tested using the χ2-test. When more than 20% of the cells had an expected count less than 5, correction for χ2 was conducted using the Monte Carlo correction. The distributions of quantitative variables were tested for normality using the Kolmogorov–Smirnov test. Parametric statistical tests were applied when either the sample size exceeded 30 or the variable distribution was not significantly different from normal. The independent t-test (parametric) or Mann–Whitney test (nonparametric) were used to compare quantitative variables between the two groups, whereas the linear correlation between two quantitative variables was tested using Pearson’s correlation test.
The diagnostic performance of CD1a+ TIDCs to discriminate recurrent from nonrecurrent cases was evaluated using the receiver operating characteristic (ROC) curve analysis. Furthermore, sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, positive predictive value, and negative predictive value were calculated. The cut-off points at which the highest diagnostic accuracy was achieved was estimated using the Youden index.
Survival analysis was used to determine the relation between different clinicopathologic variables, as well as CD68+ TAM and CD1a+ TIDC infiltration and BCG treatment outcome. Kaplan–Meier plots were used to estimate the survival function among the different groups. The log-rank test was used to compare survival distributions among different samples. As regards studying the impact of CD68+ TAMs and CD1a+ TIDCs infiltration and the occurrence of tumor recurrence, univariate Cox regression was used. Both hazard rate and 95% confidence interval (CI) were computed to determine the effect of quantitative variables on the occurrence of the event. This was followed by multivariate Cox regression analysis to identify independent predictors of survival of the studied bladder carcinoma patients. Its application followed the enter method. The model as a whole was assessed using model χ2. The contribution of different predictors was assessed using adjusted hazard rate and 95% CI.
Pattern of CD68+ TAM and CD1a+ TIDC infiltration
We evaluated the localization of TAMs and TIDC within all studied tumor specimens. The CD68+ macrophages and CD1a+ DC infiltration were counted in both tumor stroma (including the papillary axis, stroma, and lymphoid aggregates) and in the tumor islets. A high CD68+ TAM infiltration score was observed in 93.7% (n=59) of cases, whereas only 6.3% (n=4) of cases revealed a low infiltration score. Expression of CD68 was also observed in the cytoplasm of tumor cells in 51 (81%) cases. In these, the proportion of positive cells ranged from 10 to 100% (Fig. 1a–c).
A high CD1a+ TIDC infiltration score was observed in 49.2% (n=31) of cases, whereas 50.8% (n=32) of tumors revealed a low CD1a+ TIDC infiltration score. (Fig. 1d–f). The tumor cells in all cases did not react with the CD1a. The mean count of CD68+ macrophages was 20.12, whereas for CD1a+ DCs it was 6.39.
The infiltration of CD68+ TAMs in tumor stroma (papillary axis, stroma, and lymphoid aggregates) correlated significantly with that infiltrating into the tumor epithelium (Kendall’s τb=0.306, P=0.001). Similary, CD1a+ TIDC infiltration in tumor stroma and tumor islets revealed a significant correlation (Kendall’s τb=0.449, P=0.000). Interestingly, the overall infiltration by CD68+ TAMs and CD1a+ TIDCs revealed neither a significant correlation (Kendall’s τb=0.088, P=0.330) nor an agreement [weighted κ=0.123 (95% CI: 0.00684–0.240)].
Relation of CD68+ TAMs and CD1a+ TIDCs counts with clinicopathologic parameters
The relation between clinicopathologic variables and the infiltration scores (stratified as low and high) of CD68+ TAM and CD1a+ TIDC are shown in Table 1.
Our results showed that the overall CD68+ macrophage counts (within tumor stroma and tumor nests) correlated significantly with patients’ age and tumor grade (r=0.402, P=0.000 and r=0.347, P=0.005, respectively). Interestingly, the significant correlation with age was maintained when separately testing the total stromal CD68+ TAM infiltration with patients’ age but was lost when testing the tumor epithelial CD68+ TAM infiltration. Regarding the tumor grade, the significant correlation was lost when separately testing the CD68+ TAM infiltration into tumor stroma and tumor epithelium. The CD68+ TAM infiltration did not correlate with sex, tumor stage, lymphovascular invasion.
As for the CD1a+ TIDC infiltration, no significant correlation was noted with patients’ age, sex, tumor grade, tumor stage, lymphovascular invasion.
Association of clinicopathologic characteristics with BCG treatment outcome
Approximately 60.3% (n=38) of the patients experienced the event of tumor recurrence, with a median recurrence time of 5 (range: 2–36) months. The median RFS was 10 months. No significant differences were observed between recurrent and nonrecurrent cases regarding patients’ age, sex, tumor grade, tumor stage, lymphovascular invasion, CIS presence, mBCG, and risk stratification. Among the 38 recurrent cases only seven cases experienced tumor progression.
CD68+ TAMs and CD1a+ TIDCs and BCG treatment outcome
We evaluated the CD68+ TAM and CD1a+ TIDC infiltration in the context of BCG treatment outcome, both as continous variables and after stratifying the counts into low and high.
No significant difference was noted between the CD68+ TAM infiltration score in the recurrent versus nonrecurrent cases (Z=1.744, P=0.081); however, after stratifying cases into low and high infiltration scores, patients with high CD68+ TAM infiltration revealed significantly higher recurrence rates compared with patients with low infiltration scores (Yate’s χ2=4.080, Yate’s P=0.043). Moreover, patients with CD68-positive staining of tumor cells did not exhibit significant differences regarding BCG treatment failure compared with those with negative staining. The seven cases that progressed revealed high CD68+ TAM infiltration scores.
A significant difference was observed between recurrent and nonrecurrent cases regarding the overall CD1a+ TIDC infiltration scores (Z=3.107, P=0.002), and this significance was maintained when separately testing the epithelial and stromal CD1a+ TIDC infiltration (Z=2.108, P=0.035 and Z=3.343, P=0.010, respectively). Moreover, after stratifying cases into low and high infiltration scores, patients with high overall CD1a+ TIDC infiltration revealed significantly higher recurrence rates compared with patients with low overall infiltration scores (Yate’s χ2=10.536, P=0.001). Regarding the CD1a+ TIDC score in the seven cases that progressed, three cases revealed high and four cases revealed low infiltration scores.
The ROC analysis showed that the overall CD68+ TAM score could not discriminate the recurrent from nonrecurrent cases (area under the curve=0.631, 95% CI: 0.500–0.749, P=0.0861). Conversely, the overall CD1a+ TIDC score could significantly discriminate the recurrent from nonrecurrent cases (area under the curve=0.732, 95% CI: 0.605, 0.835, P=0.0003). Using the Youden index, the value (>7) was determined as a cut-off point of the CD1a+ TIDC count. At a cut-off point more than 7 for the CD1a+ TIDC count, the patient will be predicted to experience a recurrence with a positive predictive value of 84.6%, and a negative predictive value of 56.8%, sensitivity=57.89 and specificity=84.00 (Fig. 2).
Only the CD68+ TAM counts revealed a significant inverse correlation with RFS, (r=−0.505, P=0.000), as higher CD68+ TAM scores correlated with shorter RFS, (Fig. 3). The median survival (last time to see the patient) was significantly higher in patients with low CD68+ TAM infiltration compared with patients with high CD68+ TAM scores, (Z=2.798, P=0.002). Conversely, no significant difference was observed in median survival between patients with low and high CD1a+ TIDC infiltration, (Z=1.572, P=0.116).
A Kaplan–Meier analysis was performed to estimate the influence of high CD68+ TAM counts in terms of RFS after BCG treatment, and significant differences were noted in the median RFS between patients with high and low CD68+ TAM scores (log-rank P=0.026) (Fig. 4a). Similarly, the impact of high CD1a+ TIDC on RFS after BCG treatment was tested, and was significant (log-rank P=0.010) (Fig. 4b). Patients with high CD1a TIDC scores revealed a median RFS of 7 months compared with a median RFS of 24 months for patients with low scores.
Cox regression analysis
Univariate Cox regression analysis revealed that patients with tumors presenting high overall or high stromal CD68+ TAM infiltration scores exhibited an increased risk trend for recurrence after BCG treatment [hazard ratio (HR)=1.035; 95% CI: 1.011–1.060; P=0.005 and HR=1.025; 95% CI: 1.004–1.047; P=0.020, respectively]. High epithelial CD68+ TAM infiltration, and expression of CD68 by tumor cells did not significantly increase the recurrence risk, Table 2. Regarding the CD1a+ TIDC infiltration, only high stromal TIDC score increased the risk for recurrence (HR=1.080; 95% CI: 1.009–1.155; P=0.027). High overall and high tumor epithelial infiltration scores did not significantly increase the risk for recurrence after treatment. To assess the individual effects of CD68+ TAM and CD1a+ TIDC infiltration on BCG treatment outcome, a multivariate analysis was carried out. A significant increase in the risk for recurrence was observed in patients with high overall CD68+ TAM infiltration scores (HR=1.031; 95% CI: 1.005–1.057; P=0.019), and for patients with high stromal CD1a+ TIDC (Table 2).
Up to 40% of patients with NMIUC have been reported to fail intravesical BCG therapy (Zlotta et al., 2009; Babjuk et al., 2013). The majority of low-grade NMIUC are liable to recur but rarely progress. Failure to respond to BCG in high-risk T1 bladder cancer and/or carcinoma in-situ is more problematic, as those tumors are prone to progress to muscle invasion. In these cases, radical cystectomy remains the mainstay after BCG failure (Zlotta et al., 2009). Therefore, the identification of biomarkers that can predict treatment failure to allow early identification of those patients is crucial for the patients’ management (Alexandroff et al., 2010).
The presence of TAMs and TIDCs in bladder tumors before BCG treatment has been suggested to affect the response to BCG immunotherapy, (Ayari et al., 2009; Takayama et al., 2009; Ajili et al., 2013); however, more studies are still needed to validate this (Lima et al., 2014). The current study was undertaken to assess the prognostic significance of CD68+ TAMs and CD1a+ TIDC infiltration in NMIUC before BCG treatment. Knowing that the tumor microenvironment plays a significant role in tumor biology, we, therefore, evaluated TAMs and TIDCs independently in the tumor and in the tumor stroma.
We started by studying the correlation between CD68+ TAM and CD1a+ TIDC counts and the clinicopathological variables. The CD68+ TAM infiltration scores correlated with patients’ age (when tested as overall and as total stromal) and higher tumor grade (only as overall score). This implies that the correlation between TAM counts and patients’ age arises mainly from the stromal rather than the epithelial infiltration. Similar results were observed by other authors for bladder cancer using CD68 (Ayari et al., 2009; Lima et al., 2014). Conversly, similar to others (Ayari et al., 2009) the CD1a+ TIDC scores did not correlate with any of the studied clinicopathologic parameters.
Macrophages play an important role in inflammation, but the biological significance of tumor infiltration by TAMs, recruited from peripheral blood monocytes, remains unclear. TAMs represent a pivotal component of stromal cells, and extensive TAM infiltration was reported to be associated with angiogenesis and poor prognosis in several human cancers (Hanada et al., 2000; Ishigami et al., 2003). TAMs, however, exhibit both positive and negative effects on tumor growth. Thereby, the biological context is suggested to represent an important role in determining the TAM function (Bingle et al., 2002; Ishigami et al., 2003). The efficacy of BCG treatment in responsive patients relies upon the host macrophages, and a high pretreatment level of TAMs could prevent BCG from inducing a long-term local inflammation by being phagocytized and eliminated by the high number of macrophages (Ayari et al., 2009).
In agreement with this, we observed a significantly higher recurrence rate (BCG failure) in patients with high TAM infiltration compared with patients with low scores. These findings contradict the findings of Lima et al. (2014), and yet augment those of other studies (Ayari et al., 2009; Takayama et al., 2009; Ajili et al., 2013). Lima et al. (2014) even reported that high density of M2-polarized macrophage counts in the stroma but not in the tumor related to BCG treatment failure, and stated that a high density of macrophages in the tumor presented a more favourable outcome similar to those presenting an overall low density of M2 macrophages. These results suggest that M2-macrophages may be influencing treatment outcome in different ways possibly because of the influence of differentiated micro environmental stimuli in the stroma and the tumor. Moreover, among our cases significant differences were noted in the median RFS between patients with high and those with low CD68+ TAM counts, which is in agreement with our observation of the presence of a significant negative correlation between the CD68+ TAM count and the duration of survival, a finding that was reported in other cancers (Ayari et al., 2009).
As DCs are central to the development and maintenance of an immune response, it is therefore plausible to hypothesize that tumor-infiltrating DCs are essential for a cellular antitumor immune response. A correlation between the quantity of TIDCs and clinical outcome in different tumor types has been reported (Goldman et al., 1998; Murdoch et al., 2004). CD1a is a nonconventional antigen presenting molecule expressed on DC subtypes. It is was regarded as a marker for immature DCs, which becomes downregulated on maturation. However, different studies have shown that CD1a can be expressed on both mature and immature DCs (Hanada et al., 2000; Murdoch et al., 2004).
In our study, we used CD1a to semiquantify TIDCs. The CD1a+ TIDC counts revealed a significant difference between patients who experienced tumor recurrence and those who did not, and thereby patients with high infiltration scores exhibited a higher recurrence rate compared with patients with low scores. Although no significant difference was noted in the median survival between patients with low and high CD1a+ TIDC scores, a high DC infiltration score using the Kaplan–Meier analysis was shown to have a significant impact on the RFS after BCG treatment.
Cox regression analysis revealed that patients with high overall and high stromal CD68+ TAM scores exhibited an increased risk for recurrence after BCG treatment. Conversely, high overall and high tumor epithelial DC infiltration scores did not increase the risk for BCG failure, as opposed to high stromal DC infiltration, which significantly increased the risk for BCG treatment failure. Our results indicate that discrimination of the stromal CD1a+TIDC infiltration is a better indicator of BCG treatment failure than is the overall CD1a+ TIDC infiltration. These findings point to the fact that not only the number of infiltrating dendritic but also the location of infiltration is of importance. Our findings support the observations of other studies on breast cancer (Bell et al., 1999) and colorectal cancer (Suzuki et al., 2002).
The explanation of the negative correlation between stromal TIDCs and prognosis is speculative. A hypothesis is that antigen-capturing and antigen-processing (CD1a) DCs on their way to the tumor cells become dysfunctioning after exposure to inhibitory factors (cytokines, chemokines, etc.) in the tumor environment, whereas maturing DCs in the tumor epithelium are hindered on their way to migrate out to secondary lymphoid tissue after antigen capture and activation possibly by similar inhibitory factors. Thus, these large numbers of stromal DCs actually represent dysfunctioning DCs that are not capabale to help mounting an antitumor immune response (Sandel et al., 2005).
To assess the individual effects of CD68+ TAM and CD1a+ TIDC infiltration on BCG treatment outcome, a multivariate analysis was carried out. When adjusted to potential confounders, such age, sex, number of mBCG, and tumor stage, a significant increase in the risk for recurrence was observed in patients with high overall CD68+ TAM infiltration scores and for patients with high stromal CD1a+ TIDC. An almost 1.1-fold increase in the risk for recurrence was observed for high stromal CD1a+ TIDC infiltration.
We also observed expression of CD68 in tumor cells in 81% of tumors. In line with the findings of several other studies (Ayari et al., 2009), CD68 expression in tumors showed a trend to an association with higher risk for recurrence. CD68 as a transmembrane lysosomal glycoprotein could play a role in the phagocytic activities of tissue macrophages (Holness and Simmons, 1993). Thus, the expression of CD68 by urothelial carcinoma cells and its association with recurrence are consistent with the observed immune properties of urothelial cells (Lin et al., 2003; Bevers et al., 2004). Moreover, recent studies support the involvement of urothelial cells in BCG immunotherapy, as BCG is internalized both by professional antigen-presenting cells and by urothelial cells, resulting in presentation of BCG antigens and cytokine secretion by both types of cells (Bevers et al., 2004).
Although our results highlighted that high overall and high stromal CD68+ TAM counts represent predictors of recurrence after BCG treatment, some limitations need to be overcome. We stratified our cases into low and high infiltration scores following previous publications. However, before validating the use of this biomarker in clinical practice, developing an automatic counting software to reach a standard counting technique with established cut-off values would seem useful. Moreover, our study revealed that only high stromal CD1a+ TIDC scores significantly increased the risk for recurrence. We thereby support the concept that not only high DC infiltration correlate with prognosis of NMIUC patients treated with BCG but also the location of this infiltration with its underlying implications, in differences in the functional subsets of TIDCs, affect the local antitumor immune respone. Using the ROC curve analysis and Youden index, we established a cut-off value for CD1a+ TIDC (>7) that needs to be validated in a larger series and in different cohorts.
Our study revealed that patients with low infiltration scores of either immune cells in NMIUCs at the time of initial TUR exhibited significantly lower rates of BCG treatment failure compared with patients with high scores. These observations, after being validated in larger cohorts, pave the way for providing long-awaited predictor biomarkers of the response to BCG immunotherapy. Moreover, our data suggest that not only the pretreatment level of infiltration by these immune cells but also the location of this infiltration (especially for DCs) might possibly be used to select the appropriate adjuvant treatment.
Conflicts of interest
There are no conflicts of interest.
Ajili F, Kourda N, Darouiche A, Chebil M, Boubaker S (2013). Prognostic value of tumor-associated macrophages count in human non-muscle-invasive bladder cancer treated by BCG immunotherapy. Ultrastruct Pathol 37:56–61.
Alexandroff AB, Nicholson S, Patel PM, Jackson AM (2010). Recent advances in bacillus Calmette–Guerin immunotherapy in bladder cancer. Immunotherapy 2:551–560.
Amling CL (2000). Diagnosis and management of superficial bladder Lamm DL. Efficacy and safety of bacille Calmette–Guérin immunotherapy in superficial bladder cancer. Clin Infect Dis 31:86–90.
Amling CL (2001). Diagnosis and management of superficial bladder cancer. Curr Probl Cancer 25:219–278.
Ayari C, La Rue H, Hovington H, Decobert M, Harel F, Bergeron A, et al (2009). Bladder tumor infiltrating mature dendritic cells and macrophages as predictors of response to bacillus Calmette–Guérin immunotherapy. Eur Urol 55:1386–1395.
Babjuk M, Oosterlinck W, Sylvester R, Kaasinen E, Böhle A, Palou-Redorta J (2008). EAU guidelines on non muscle-invasive urothelial carcinoma of the bladder. Eur Urol 54:303–314.
Babjuk M, Burger M, Zigeuner R, Shariat SF, van Rhijn BW, Comperat E, et al (2013). EAU guidelines on non-muscle-invasive urothelial carcinoma of the bladder: update 2013. Eur Urol 64:639–653.
Bell D, Chomarat P, Broyles D, Netto G, Harb GM, Lebecque S, et al (1999). In breast carcinoma tissue, immature dendritic cells reside within the tumor, whereas mature dendritic cells are located in peritumoral areas. J Exp Med 190:1417–1426.
Bevers RF, Kurth KH, Schamhart DH (2004). Role of urothelial cells in BCG immunotherapy for superficial bladder cancer. Br J Cancer 91:607–612.
Bingle L, Brown NJ, Lewis CE (2002). The role of tumour-associated macrophages in tumour progression: implications for new anticancer therapies. J Pathol 196:254–265.
Bohle A, Bock PR (2004). Intravesical bacillus Calmette–Guerin versus mitomycin C in superficial bladder cancer: formal meta-analysis of comparative studies on tumour progression. Urology 63:682–686.
Brandau S, Suttmann H (2007). Thirty years of BCG immunotherapy for non-muscle invasive bladder cancer: a success story with room for improvement. Biomed Pharmacother 61:299–305.
Eble JN, Sauter G, Epstein JI, Sesterhenn IA. World Health Organization classification of tumours: pathology and genetics of tumours of the urinary system and male genital organs. Lyon, France: IARC Press; 2004;90–109.
Goldman SA, Baker E, Weyant RJ, Clarke MR, Myers JN, Lotze MT (1998). Peritumoral CD1a-positive dendritic cells are associated with improved survival in patients with tongue carcinoma. Arch Otolaryngol Head Neck Surg 124:641–646.
Hanada T, Nakagawa M, Emoto A, Nomura T, Nasu N, Nomura Y (2000). Prognostic value of tumor-associated macrophage count in human bladder cancer. Int J Urol 7:263–269.
Holness CL, Simmons DL (1993). Molecular cloning of CD68, a human macrophage marker related to lysosomal glycoproteins. Blood 81:1607–1613.
Ishigami S, Natsugoe S, Tokuda K, Nakajo A, Okumura H, Matsumoto M, et al (2003). Tumor-associated macrophage (TAM) infiltration in gastric cancer. Anticancer Res 23:4079–4083.
Lima L, Oliveira D, Tavares A, Amaro T, Cruz R, Oliveira MJ, et al (2014). The predominance of M2-polarized macrophages in the stroma of low-hypoxic bladder tumors is associated with BCG immunotherapy failure. Urol Oncol 32:449–457.
Lin Z, Chen S, Ye C, Zhu S (2003). Nitric oxide synthase expressionin human bladder cancer and its relation to angiogenesis. Urol 31:232–235.
Luo Y, Yamada H, Evanoff DP, Chen X (2006). Role of Th1-stimulating cytokines in bacillus Calmette–Guérin (BCG)-induced macrophage cytotoxicity against mouse bladder cancer MBT-2 cells. Clin Exp Immunol 146:181–188.
Luo Y, Han R, Evanoff DP, Chen X (2010). Interleukin-10 inhibits Mycobacterium bovis bacillus Calmette–Guérin (BCG)-induced macrophage cytotoxicity against bladder cancer cells. Clin Exp Immunol 160:359–368.
Mantovani A, Sozzani S, Locati M, Allavena P, Sica A (2002). Macrophage polarization: tumor-associated macrophages as a paradigm for polarized M2 mononuclear phagocytes. Trends Immunol 23:549–555.
Murdoch C, Giannoudis A, Lewis CE (2004). Mechanisms regulating the recruitment of macrophages into hypoxic areas of tumors and other ischemic tissues. Blood 104:2224–2234.
Nishiyama T, Tachibana M, Horiguchi Y, Nakamura K, Ikeda Y, Takesako K, Murai M (2001). Immunotherapy of bladder cancer using autologous dendritic cells pulsed with human lymphocyte antigen-A24-specific MAGE-3 peptide. Clin Cancer Res 7:23.
Pinzon-Charry A, Maxwell T, López JA (2005). Dendritic cell dysfunction in cancer: a mechanism for immunosuppression. Immunol Cell Biol 83:451–461.
Qian BZ, Pollard JW (2010). Macrophage diversity enhances tumor progression and metastasis. Cell 141:39–51.
Rosai EJ. Rosai and Ackerman’s surgical pathology, 9th ed. China: Elsevier; 20041317–1359.
Saint F, Salomon L, Quintela R, Cicco A, Hoznek A, Abbou CC, et al (2003). Do prognostic parameters of remission versus relapse after bacillus Calmette–Guérin (BCG) immunotherapy exist? analysis of a quarter century of literature. Eur Urol 43:351–361.
Sandel MH, Dadabayev AR, Menon AG, Morreau H, Melief CJ, Offringa R, et al (2005). Prognostic value of tumor-infiltrating dendritic cells in colorectal cancer: role of maturation status and intratumoral localization. Clin Cancer Res 11:2576–2582.
Sica A, Larghi P, Mancino A, Rubino L, Porta C, Totaro MG, et al (2008). Macrophage polarization in tumour progression. Semin Cancer Biol 18:349–355.
Siegel R, Ward E, Brawley O, Jemal A (2011). Cancer statistics, 2011: the impact of eliminating socioeconomic and racial disparities on premature cancer deaths. CA Cancer J Clin 61:212–236.
Suzuki A, Masuda A, Nagata H, Kameoka S, Kikawada Y, Yamakawa M, et al (2002). Mature dendritic cells make clusters with T cells in the invasive margin of colorectal carcinoma. J Pathol 196:37–43.
Takayama H, Nishimura K, Tsujimura A, Nakai Y, Nakayama M, Aozasa K, et al (2009). Increased infiltration of tumor associated macrophages is associated with poor prognosis of bladder carcinoma in situ after intravesical bacillus Calmette–Guérin instillation. Urology 181:1894–1900.
Witjes JA, Hendricksen K (2008). Intravesical pharmacotherapy for non–muscle-invasive bladder cancer: a critical analysis of currently available drugs, treatment schedules, and long-term results. Eur Urol 53:45–52.
Zlotta AR, Fleshner NE, Jewett MA (2009). The management of BCG failure in non-muscle-invasive bladder cancer: an update. Can Urol Assoc J 3:199–205.