CD8-positive T cells and CD204-positive M2-like macrophages predict postoperative prognosis of very high-risk prostate cancer

To stratify the heterogeneity of prostate cancer (PCa) with seminal vesicle invasion (SVI) immunologically after radical prostatectomy focusing on the tumor microenvironment. We retrospectively reviewed the clinicopathological data of 71 PCa patients with SVI, which is known as a factor of very high-risk PCa. Preoperative clinical variables and postoperative pathological variables were evaluated as predictors of biochemical recurrence (BCR) with a multivariate logistic regression. Immune cell infiltration including the CD8-positive cell (CD8+ cell) and CD204-positive M2-like macrophage (CD204+ cell) was investigated by immunohistochemistry. The cumulative incidence and risk of BCR were assessed with a Kaplan–Meier analysis and competing risks regression. A higher CD8+ cell count in the SVI area significantly indicated a favorable prognosis for cancers with SVI (p = 0.004). A lower CD204+ cell count in the SVI area also significantly indicated a favorable prognosis for cancers with SVI (p = 0.004). Furthermore, the combination of the CD8+ and CD204+ cell infiltration ratio of the SVI area to the main tumor area was a significant factor for BCR in the patients with the PCa with SVI (p = 0.001). In PCa patients with SVI, the combination of CD8+ and CD204+ cell infiltration is useful to predict the prognosis.

Pathological analysis. The Gleason scores after the RP were cited from pathological reports. In patients who received neoadjuvant androgen deprivation therapy, Gleason scores obtained from a needle biopsy were used instead of prostatectomy Gleason scores. All PCa cases were histologically diagnosed before RP by ultrasound-guided needle biopsy via the transrectal or transperineal approach. After RP, all specimens were fixed in 10% formalin and embedded in paraffin. Whole-mount section pathology was performed, and all specimens were cut into thin slices perpendicular to the urethra from the apex to the base after removing the seminal vesicle. All seminal vesicles were cut longitudinally. All slides were stained with hematoxylin and eosin. The Gleason score of the PCa and the other pathological parameters, including SVI, were evaluated in each section. SVI was defined as cancer invasion into the extraprostatic portion of the seminal vesicles.
Immunohistochemistry. Immunohistochemistry was performed after the sections were formalin-fixed and paraffin-embedded. The sections were deparaffinized in xylene and then rehydrated in graded alcohols and distilled water. After antigen retrieval with citric acid (pH 6.0), endogenous peroxidase activity was blocked with 1% hydrogen peroxide for 30 min, followed by washing with distilled water. To bind nonspecific antigens, the sections were incubated with 5% skim milk for 15 min. The sections were incubated with an anti-CD8 rabbit polyclonal antibodies (1:100 dilation, Abcam, Cambridge, MA, USA) and anti-CD204 rabbit polyclonal antibodies (1:100 dilation, Abcam, Cambridge, MA, USA) at room temperature for one hour, followed by conjugation to the secondary antibody and DAB staining.

Evaluation of immunostaining.
To evaluate CD8 and CD204 staining, the cells with positive staining were counted in five representative fields (200 μm 2 /field) using light microscopy. The median number of the CD8 + and CD204 + cell was estimated for each main tumor area and SVI area. Values above median number of the CD8 + or CD204 + cells were considered as CD8-high or CD204-high. Values under median number of the CD8 + or CD204 + cells were considered as CD8-low or CD204-low. The experienced urologic pathologists, blinded to the patients' clinical data, performed the counting. SVM (the Seminal Vesicle area to the Main area) Score. Focusing on the distribution of the CD8 + and CD204 + cells in each tumor slide, we calculated the ratio of the CD8 + and CD204 + cell counts in the SVI area to those in the main tumor area for each patient. We defined the ratio as "SVM (the Seminal Vesicle area to the Main area) Score". We calculated estimates (with confidence intervals) of the best-fit receiver operating characteristic (ROC) curve and the corresponding area under the ROC curve, and then calculated the cutoff value of each SVM Score for CD8 + and CD204 + cells, respectively.
Statistical analysis. Differences in continuous variables between groups were evaluated using the Mann-Whitney U test. The Chi-squared test was used to analyze the difference in the number of patients between two groups. To identify factors predictive of BCR, univariate and multivariate analyses were performed using the Cox proportional hazards model with stepwise forward selection. Kaplan-Meier curves were drawn to evaluate postoperative BCR-free survival. All reported p-values were two-sided, and statistical significance was set at 0.05. The statistical analyses were performed using the R Statistical Language version 3.5.3 program (https:// www.r-proje ct. org) and the SPSS version 25.0 statistical software package (https:// www. ibm. com/ analy tics/ spssstati stics-softw are).
Ethics approval and consent to participate. Written informed consent was obtained from all patients included in the study. This study was approved by the Institutional Review Board of Keio University and Saitama City Hospital and was performed in accordance with the Declaration of Helsinki.

Results
Patients characteristics of seminal vesicle invasion. The clinicopathological data of the 71 patients with SVI are summarized in Table 1. The median PSA value at biopsy was 9.1 ± 8.6 ng/ml. The median prostate volume at biopsy was 30.0 ± 12.2 ml. The median PSA density at biopsy was 0.36 ± 0.51 ng/ml/ml. SVI was not detected (clinical T3b) in any patients before RP.  Fig. 1a). However, the statistically significant difference was observed in BCR between the higher and lower CD8 + cell count in the SVI area, which suggested that a higher CD8 + cell count in the main tumor area predicted a favorable prognosis for PCa with SVI (p = 0.004, Fig. 1b).

Scientific Reports
The median CD204 + cell count in the main tumor area and the SVI area was 300.0 ± 139.9 (95% CI: 50.0-540.0) cells/mm2 and 300.0 ± 163.1 (95% CI: 25.0-600.0) cells/mm 2 , respectively. The statistically significant difference was observed in BCR between the higher and lower CD204 + cell count in the main tumor area, which suggested that a higher CD204 + cell count in the main tumor area predicted a worse prognosis for PCa with SVI (p = 0.042, Fig. 1c). In the SVI area, the statistically significant difference was also observed in BCR between the higher and lower CD204 + cell count (p = 0.004, Fig. 1d).

Combination of immune cell infiltration.
The combination of the CD8 + and CD204 + cells in the main tumor area and SVI area was evaluated in each patients. Each patients were classified into four groups according to the amount of the CD8 + and CD204 + cells. Group I included patients with a higher CD8 + cell count and a lower CD204 + cell count. Group II included patients with a higher CD8 + cell count and a higher CD204 + cell count. Group III included patients with a lower CD8 + cell count and a lower CD204 + cell count. Group IV included patients with a lower CD8 + cell count and a higher CD204 + cell count. No significant difference among four groups was observed in the main tumor area (p = 0.150, Fig. 2a). However, a significant difference between four groups was observed in the SVI area (p < 0.001, Fig. 2b Logistic regression analysis. Univariate Cox regression revealed the serum PSA level, the CD8 + cell count in the SVI area and the CD204 + cell count in the SVI area had significant impact on BCR (p = 0.001, p = 0.002 and p = 0.036, respectively, Table 2). Among the clinicopathologic parameters analyzed, multivariate Cox regression revealed that the higher serum PSA level (Hazard Ratio (HR) = 2.870, p = 0.004), the lower CD8 + cell count (HR = 3.004, p = 0.018) and the higher CD204 + cell count in the SVI area (HR = 3.290, p = 0.013) are an indication of the unfavorable prognosis of cancers with SVI (Table 2).

SVM (the Seminal Vesicle area to the Main area) Score
Each cut-off value of SVM Score of CD8 + and CD204 + cell was 1.5 and 0.6, respectively. A statistically significant difference in the SVM Score of the CD8 + cell was observed (p = 0.012, Fig. 3a). It suggested that the higher SVM Score of the CD8 + cell predicted a favorable prognosis of PCa with SVI. A statistically significant difference in the SVM Score of the CD204 + cell was also observed (p = 0.037, Fig. 3b). It suggested that the lower SVM Score of the CD204 + cell predicted a favorable prognosis of PCa with SVI. A statistically significant difference in the combination of the SVM Score of the CD8 + and CD204 + cell was observed among the four groups (p = 0.002, Fig. 3c).

Discussion
SVI is an independent prognostic factor of PCa, and in general, adjuvant radiotherapy is effective in patients with SVI 10 . However, this study revealed that the higher CD8 + cell count and the lower CD204 + cell count in the SVI area are an indication of the favorable prognosis of cancers with SVI. Several studies have been reported that CD8 + cell infiltration is associated with better prognosis 11,12 . In other reports, CD8 + cell was also associated with recurrence, progression, or lymph node invasion 13,14 . CD8 + cell may be associated with cancer progression by inhibiting the activity of effector T cells 15,16 . Thus, we had hypothesized that higher infiltration of CD8 + cell could be a favorable prognostic factor of BCR, however, no significant difference was observed in BCR between the higher and lower CD8 + cell count in the main tumor area (Fig. 1a). Then, we focused on the distribution of the CD8 + cell in each tumor slide. CD8 + cell secret cytotoxic molecules and cytokines, including perforin, granzyme, interferon-gamma, tumor necrosis factor-alpha and so on, which elicit anti-tumor effects. They are released only in the direction of the target cell by identifying cancer antigen-derived epitopes presented on MHC molecules on tumor cells by T cell receptors to avoid non-specific bystander damage  www.nature.com/scientificreports/ to normal tissue 17 . The mechanism might suggest our data that CD8 + cell increased in the SVI area rather than the main tumor area in each unfavorable tumors was correct (Fig. 1b).
On the other hand, tumor-associated macrophage (TAM) infiltration in the tumor microenvironment is directly associated with tumor invasion, nodal status, and clinical stage in some cancers, and TAMs also have been reported to regulate the growth of prostate cancer [18][19][20] . A previous study reported that M2 TAMs inhibit cytotoxic CD8 + cell, resulting in weakened anti-tumor immunity and increased tumor-infiltrating CD8 + cells 21 . We also focused on the distribution of CD204 + cell in each tumor slide as well as CD8 + cell. CD204 is not M2 specific, and CD163 might be suitable for M2-like marker 22 . In recent article, high number of CD163-positive cells was related to poor clinical course in prostate cancer, however, M1 cells were also associated to poor clinical course 23 . In breast cancer, the data that cancer-derived factor induced CD204 expression in cultured macrophages rather than CD163 and CD204 + cells predicted poor clinical course rather than CD163 indicated that CD204 might be a marker for protumor phenotype more suitable than CD163 in some organs 24 . Macrophages express PD-L1 and PD-L2, and macrophages in SVI can be considered positive for PD-L1 and L2, which might be related to immune suppression. We will evaluate PD-L1 and L2 in our future study. In our present study, a significant difference between the higher and lower CD204 + cell count in the main tumor area and the SVI area was observed in BCR ( Fig. 1c and 1d).
Patients with very high-risk PCa, including SVI are usually considered to be candidates for adjuvant therapy [25][26][27] . However, it is uncertain whether all patients need adjuvant therapy, as some of them do not appear to benefit. Therefore, the development of an improved prediction model of the very high-risk PCa is necessary and unmet needs. In this study, we assessed the relationship of the distribution of the CD8 + and CD204 + cell. The combination of those different immune cells could be more useful for predicting the prognosis of the very high-risk PCa than a single immune cell. This study was derived from a retrospective review of patients treated at two institutions and the specimens used in this study could not reflect all status of the tumor immune microenvironment. The small tumor volume in SVI is also difficult to evaluate. However, the findings obtained in this study could be useful for clinical assessment and decision-making for PCa patients with SVI after RP. It is highly worthwhile to verify these results focused on the different type of immune cells and the marker localization in future prospective studies.
In conclusion, in prostate cancer patients with SVI, the combination of the CD8 + and CD204 + cell infiltration in the SVI area is useful to predict the prognosis.