Superoxide dismutase 2 expression can predict prognosis of renal cell carcinoma patients

Abstract

BACKGROUND AND OBJECTIVE:

Renal cell carcinoma (RCC) is the urological malignancy with the highest mortality rate and is increasing in incidence. The prognostic and predictive biomarkers are highly desired. This study aims to investigate the significance of superoxide dismutase 2 (SOD2) as a clinical biomarker in patients with renal cell carcinomas.

METHODS:

A cohort of 97 patients with RCC was analyzed retrospectively using various clinical parameters and SOD2 expression by immunohistochemistry.

RESULTS:

Cases with stronger SOD2 positivity of the tumor in comparison to the adjacent normal renal tubule by immunohistochemistry were categorized as high SOD2 and were associated with worse overall survivals ( $p=$ 0.005). In particular, in cases with metastatic RCC, high SOD2 expression in the tumors was significantly associated with a worse overall survival ( $p=$ 0.001), and the maximum critical risk. Treatment with current molecular targeting therapies did not improve the prognoses of cases with metastatic or recurrent RCC.

CONCLUSIONS:

High SOD2 expression can be predictive of a poor clinical outcome and be clinically useful in the follow-up of metastatic RCC. Therapeutics for metastatic RCCs require further improvement, such as supplementary administration of agents targeting mitochondrial SOD2.

1. Introduction

Renal cell carcinoma (RCC) is the urological malignancy with the highest mortality rate and is increasing in incidence [1]. The increased use of abdominal imaging, such as computed tomography and ultrasonography, has led to detection of incidental renal masses and consequently, patients with localized RCCs are in surge. Although radical nephrectomy remains a successful curative treatment for localized RCC, more than 30% of newly diagnosed cases are regionally advanced or eventually develop distant metastases [1]. Currently, advanced RCC cases with metastasis or recurrence are often treated with several molecular targeting drugs, which inhibit tyrosine kinase, vascular endothelial cell growth factor receptor (VEGFR), mammalian target of rapamycin (mTOR) kinase, etc. [2]. However, their therapeutic efficacy is poor. The common mechanism of action of these molecular targeting agents is based on nutrient starvation. The previous study investigated a novel target causing therapeutic resistance in “starvation-resistant” RCC cells, the study found that starvation-resistant cells possess high mitochondrial performance with sustained expression of superoxide dismutase 2 and mitochondrial function supported by stored carbon sources. In such RCCs, the current molecular targeting treatments based on nutrient starvation are not a rational therapeutic approach, however inhibition of the mitochondria can offer an effective therapeutic option for advanced RCC cases [3, 4, 5].

SOD2 is an enzyme that scavenges reactive oxygen species in the mitochondrial matrix. High SOD2 expression can inhibit cancer cell proliferation [6], but promote epithelial to mesenchymal transition and/or cancer cell migration [7]. At advanced stages, high SOD2 levels can positively contribute to metastatic tumor progression [8].

Here, we investigated the relationship betweenSOD2 expression by immunohistochemistry and various clinical parameters in a cohort of RCC patients. High SOD2 expression could predict a worse clinical outcome of RCCs cases.

2. Materials and methods

2.1 Patients

Previously, it was reported that higher SOD2 mRNA levels were significantly associated with poor survivals in metastatic RCCs [4]. In the present study, SOD2 expression was analyzed by immunohistochemistry in an independent cohort of patients with RCCs.

Ninety-seven patients who underwent radical or partial nephrectomy for the treatment of RCCs at the Shiga University of Medical Science Hospital from 1999 to 2015 were analyzed in this study. The median age of the patients was 62 years (34–82 years). There were 77 (79%) men and 20 (21%) females. The median follow-up period was 30 months (12–32 months) and the endpoint of the observation period was defined as death or the last visit to the hospital. Pathological types of RCC were 76 clear cell, 12 papillary, 8 chromophobe carcinomas, and 1 unclassified. T stages included 68 pT1, 4 pT2, 19 pT3, and 6 pT4. Vascular invasion and higher grading (G3) were identified in 67 and 18 cases, respectively. The pathological diagnosis and histological grading were made according to the WHO Histological Classification system [9], while the clinical stage was determined by the UICC TNM classification [10]. Radical and partial nephrectomies were performed in 51 (53%) and 46 (47%) patients, respectively. Surgery was performed in 70 (72%) patients as a curative measure whereas 27 (28%) patients underwent nephrectomy to reduce tumor size. A total of 27 (28%) patients had distant metastasis at the time of RCC diagnosis. A total of 25 (26%) patients were treated with molecular targeting drugs including sorafenib ( $n=$ 4), sunitinib ( $n=$ 17), axitinib ( $n=$ 14), everolimus ( $n=$ 8), and pazopanib ( $n=$ 6). Maximum treatment responses were partial response ( $n=$ 10), stable disease ( $n=$ 13), progressive disease ( $n=$ 1), and not available ( $n=$ 1). None of the patients showed complete remission. Clinical and pathological data were obtained from medical records with written informed consent from individual patients and approval by the Ethics Committee of the institute (No. 27–192).

2.2 Immunohistochemistry

Surgical specimens were transferred to 10% buffered formalin and fixed overnight. Fixed samples were embedded in paraffin and serially sliced into 5 $\mu$ m sections. After dewaxing, sections were autoclaved at 120 ${}^{\circ }$ C for 1 min in 10 mM sodium citrate buffer (pH 6.0) and immersed in 0.3% H ${}_2$ O ${}_2$ . They were then incubated overnight at 4 ${}^{\circ }$ C with primary antibodies against SOD2 (diluted 1:1000, #06-984, Millipore Corporation, Billerica, MA, USA). Sections were rinsed with phosphate-buffered saline and incubated with secondary antibody conjugated with horseradish peroxidase (Simple Stain MAX-PO, Nichirei, Tokyo, Japan) at room temperature for 1 hour. Sections were then stained with 3.3’-diaminobenzidine tetrahydrochloride and counter-stained with hematoxylin.

Figure 1.

Representative micrographs showing SOD2 expression by immunohistochemistry. (A–C) “High SOD2 expression” case. (A) SOD2 staining of renal cell carcinoma cells. (B) is stronger than the staining of adjacently normal proximal renal tubules. (C) in the same specimen. (D–F) “Low SOD2 expression” case. (D) the positivity of renal cell carcinoma cells. (E) is less or similar than normal renal tubules. (F) The multiple punctate stainings of SOD2 observed in the cytoplasmic regions could strongly suggest its mitochondrial expression in both tumorous and normal tubular cells. The scale bars correspond to 100 $\mu$ m.

2.3 Microscopic evaluation

SOD2 expression of each RCC was determined by immunohistochemistry by relative comparison to SOD2 expression of adjacently normal proximal renal tubule as an internal control on the same slide. The multiple punctate stainings of SOD2 observed in the cytoplasmic regions could strongly suggest its mitochondrial expression in both tumorous and normal tubular cells. We could define two qualitative grades for specimens according to the intensity of SOD2 expression including: i) in “high SOD2” samples the staining intensity of the tumor was apparently stronger than that of normal proximal renal tubule; and ii) in “low SOD2” the intensity of the tumor was less or similar than the proximal tubules. These semi-quantitatively immunohistochemical comparison could offer us a non-ambiguous evaluation for tumorous SOD2 expression (Fig. 1).

Grading by immunohistochemistry was performed by two independent researchers (R.K., T.C.) who had no information of the clinical parameters.

2.4 Statistical analysis

The disease-specific overall survival interval was defined as the period from the point of diagnosis to the point of death as the result of RCC. All statistical analyses were performed using SPSS 22.0 (IBM Inc., Chicago, IL, USA). All tests for statistical significance were two-sided. A $p$ -value $<$ 0.05 was considered statistically significant. The association between covariables and SOD2 expression was analyzed by the Chi-square and Fisher’s exact tests. Survival curves were estimated by the Kaplan-Meier method with a log-rank and Chi-square tests were used to assess significance. The univariate and multivariate Cox proportional hazards regression models were used to evaluate any independent prognostic effect of the variables with a 95% confidence interval. Overall survival (OS) intervals were used as the indicator for the hazard ratio (HR).

Table 1

Correlation between SOD2 expression and clinicopathological characteristics in 97 renal cell carcinomas

Variables		SOD2 expression, $n$
	Cases	High	Low	$p$
	$n$	( $n=$ 26)	( $n=$ 71)	value ${}^{†}$
Sex
Male	77	21	56	0.838
Female	20	5	15
Age, years
$⩽$ 62	45	12	33	0.977
62 $<$	52	14	38
Pathological type
Clear cell	76	17	59	0.061
Non-clear cell	21	9	12
T classification
1, 2	72	18	54	0.496
3, 4	25	8	17
Vascular invasion
Yes	67	20	47	0.311
No	30	6	24
Histological grade
G1–2	79	18	61	0.061
G3	18	8	10
Nephrectomy
Radical	51	13	38	0.758
Partial	46	13	33
Metastasis at diagnosis
Yes	27	8	19	0.522
No	70	18	52
Molecular targeting therapy
Yes	25	9	16	0.228
No	72	17	55

${}^{†}$ Chi-square and Fisher’s exact tests

3. Results

3.1 Correlation between SOD2 expression and clinicopathological characteristics of renal cell carcinomas

Representative cases of high and low expressions of SOD2 are shown in Fig. 1A–C and D–F, respectively. SOD2 reactivity of normal renal tubules was used as an internal control of each histological section. The samples of cases with stronger SOD2 positivity of the tumor compared to the adjacent normal renal tubules were categorized as high SOD2 cases. A total of 26 (27%) and 71 (73%) cases were categorized as high and low SOD2, respectively. The correlation between SOD2 expression and clinicopathological characteristics is shown in Table 1. Although the pathological type ( $p=$ 0.061) and histological grade ( $p=$ 0.061) tended to correlate with SOD2 expression, other characteristics such as age, sex, T stage, microvascular invasion, metastasis at diagnosis, and implementation of molecular targeting therapy were not significantly correlated with SOD2 expression.

Figure 2.

Kaplan-Meier plot displaying overall survival of RCC patients with high and low SOD2 expression. Cases with high SOD2 expression had significantly shorter survival periods than cases with low expression (log-rank test, $p=$ 0.005, Chi-square value $=$ 8.017).

Table 2

Prognostic evaluation of clinicopathological parameters for overall survival in 97 renal cell carcinomas by univariate and multivariate analyses

		Univariate			Multivariate
	$n$	HR (95% CI)		$p$ value ${}^{†}$	HR (95% CI)		$p$ value ${}^{†}$
Sex (male/female)	77/20	0.85	(0.24–3.02)	0.807
Age, years ( $>$ 62/ $⩽$ 62)	52/45	0.75	(0.27–2.04)	0.569
Pathological type (non-clear/clear)	21/76	1.30	(0.36–4.75)	0.690
T classification (3,4/1,2)	25/72	8.76	(2.90–26.42)	$<$ 0.001	3.11	(0.84–11.61)	0.091
Vascular invasion (yes/no)	67/30	36.65	(0.41–3276)	0.116
Histological grade (3/1,2)	18/79	12.83	(3.93–41.91)	$<$ 0.001	2.97	(0.81–10.80)	0.099
Nephrectomy (radical/partial)	51/46	10.27	(1.32–80.06)	0.026	1.96	(0.21–18.49)	0.556
Metastasis at diagnosis (yes/no)	27/70	23.31	(5.23–103.8)	$<$ 0.01	15.19	(2.55–90.58)	0.003
Molecular targeting therapy (yes/no)	25/72	5.19	(1.88–14.32)	0.001	0.64	(0.16–2.60)	0.536
SOD2 expression (high/low)	26/71	3.98	(1.43–11.07)	0.008	6.95	(1.66–29.14)	0.008

${}^{†}$ Cox proportional hazards regression models. HR: hazard ratio; CI: confidence interval.

Table 3

Prognostic evaluation of clinicopathological parameters for overall survival in 27 metastatic renal cell carcinomas by univariate and multivariate analyses

		Univariate			Multivariate
	$n$	HR (95% CI)		$p$ value ${}^{†}$	HR (95% CI)		$p$ value ${}^{†}$
Sex (male/female)	21/6	0.68	(0.18-2.56)	0.569
Age, years ( $>$ 62/ $⩽$ 62)	15/12	0.50	(0.16-1.60)	0.243
Pathological type (non-clear/clear)	3/24	3.73	(0.93-15.03)	0.064
T classification (3,4/1,2)	17/10	1.47	(0.48-4.54)	0.502
Vascular invasion (yes/no)	26/1	21.74	(0.00-1.2E $+$ 10)	0.735
Histological grade (3/1,2)	11/16	3.85	(1.15-12.91)	0.029	2.28	(0.59-8.82)	0.233
Nephrectomy (radical/partial)	26/1	0.211	(0.024-1.89)	0.164
Molecular targeting therapy (yes/no)	18/9	1.99	(0.54-7.32)	0.300
SOD2 expression (high/low)	8/19	7.78	(1.88-32.13)	0.005	5.20	(1.13-24.01)	0.034

${}^{†}$ Cox proportional hazards regression models. HR: hazard ratio. CI: confidence interval.

3.2 SOD2 is an important prognostic marker for OS in all renal cell carcinomas

The 5-year OS of high and low SOD2 groups were 55.8% and 85.0%, respectively. The high SOD2 group had a worse survival curve than that of the low SOD2 group (log-rank test $p=$ 0.005, Chi-square value $=$ 8.017; Fig. 2). Using univariate Cox proportional hazards analysis, T3/4, histological grade 3, presence of metastasis at diagnosis, implementation of molecular targeting therapy, and high SOD2 expression showed as independently worse prognostic factors. Multivariate Cox proportional hazards models indicated presence of metastasis at diagnosis (HR $=$ 15.19, 95% CI $=$ 2.55–90.58, $p=$ 0.003) and high SOD2 expression (HR $=$ 6.95, 95% CI $=$ 1.66–29.14, $p=$ 0.008) as significant risk factors for OS (Table 2).

3.3 SOD2 posses an especially predictive value in metastatic renal cell carcinomas

Among the 27 patients who had distant metastasis at diagnosis, high and low SOD2 expression RCC cases were 8 and 19, respectively. Regarding OS, the high SOD2 group showed a significantly lower survival rate than the low group (log-rank test $p=$ 0.001, Chi-square value $=$ 10.980; Fig. 3). Univariate Cox proportional hazards analysis suggested that histological grade 3 and high SOD2 expression were significantly worse prognostic factors, however multivariate Cox proportional hazards models statistically indicated high SOD2 expression as the maximum and only significant risk factor for OS (HR $=$ 5.20, 95% CI $=$ 1.13–24.01, $p=$ 0.034; Table 3).

Figure 3.

Kaplan-Meier plot displaying overall survival of 27 metastatic RCC patients with high and low SOD2 expression. Cases with high SOD2 expression had significantly shorter survival times than cases with low expression (log-rank test, $p=$ 0.001, Chi-square value $=$ 10.980).

In addition, 18 metastatic and 7 recurrent RCCs were analyzed in order to evaluate the efficacy of currently molecular targeting therapies in advanced RCC patients. The treatment resulted in partial responses in 10 cases, but no remission. Therefore, implementation of current molecular targeting therapies did not improve OS intervals in cases displaying high SOD2 RCCs expression (log-rank test $p=$ 0.003, Chi-square value $=$ 8.666; Fig. 4).

4. Discussion

It has been previously reported the presence of enhanced mitochondrial function accompanied with high SOD2 levels in dormant-like starvation-resistant RCC cells, which complicates the therapeutic of RCC. Moreover, higher SOD2 mRNA levels in metastatic RCC were associated with significantly shorter survival intervals [4]. Apart from the cohort of previous study [4], here, SOD2 protein expression was assessed by immunohistochemistry in another independent cohort of patients with RCCs. We found that high SOD2 expression in RCC was significantly associated with worse prognosis, especially in patients with metastatic RCCs, where SOD2 expression was the maximum critical risk factor. The follow-up of RCC case with abundant SOD expression should be under the medical control more frequently and longer than those of RCC cases including subtle SOD2.

Figure 4.

Kaplan-Meier plot displaying overall survival of 25 advanced RCC patients treated with molecular targeting therapies with high and low SOD2 expression. In this series, OS interval was defined as the period from the point of initial treatment to the point of RCC specific death. Cases with high SOD2 expression had significantly shorter survival periods than cases with low expression (log-rank test, $p=$ 0.003, Chi-square value $=$ 8.666).

SOD2 is one of the primary mitochondrial antioxidants in a network of detoxification enzymes that neutralize highly reactive superoxide ions ( $\bullet$ O ${}^{2-}$ ) to less reactive hydrogen peroxide (H ${}_2$ O ${}_2$ ) followed by its immediate conversion to H ${}_2$ O by glutathione peroxidases or catalases [11]. Sustained SOD2 expression causes a low level of mitochondrial reactive oxygen species (ROS) and reduces cell death related to toxic ROS accumulation [4]. High SOD2 levels can promote epithelial to mesenchymal transition and/or cancer cell migration [7] and contribute to metastatic tumor progression [8]. This is in accordance with our data where high SOD2 expression was the maximum critical risk of survival.

In this study, implementation of current molecular targeting therapies did not improve the prognoses of cases with high SOD2 tumor expression. The common mechanism of action of the currently applied molecular agents is based on nutrient starvation via the inhibition of tyrosine kinase, VEGFR, mTOR kinase, etc. [2]. RCCs are hypervascularized, and inhibiting angiogenesis is one of main objectives of anticancer molecular targeting therapies [12] and agents inhibiting tumor angiogenesis are widely used in the treatment of metastatic RCC. The mechanism of action of tyrosine kinase inhibitor (TKI) is based on nutrimental deprivation via the inhibition of tumor angiogenesis [12]. Although these newly developed agents have shown a certain degree of tumor volume reduction, complete remission is scarce, suggesting that current molecular targeting therapies are clinically problematic for the management of metastatic RCCs. It has been proposed that SOD2-abundant and starvation-resistant malignant cells can survive hostile environments like nutrimental starvation and one of the reasons for the low efficacy of molecular targeting techniques [13]. Therefore, treatment for metastatic and/or recurrent RCCs harboring cells with high SOD2 expression should be modified immediately. In the Checkmate 025 study [14], patients treated with nivolumab, an anti-PD-1 human immunogloblin [15], had a significantly better survival than the patients treated with everolimus, an mTOR inhibitor. This finding also suggests the need for second-line therapies and that nivolmab should become another therapeutic option after TKI treatment failure for metastatic RCCs [16, 17]. It has been previously demonstrated that two inhibitors targeting mitochondrial oxidative phosphorylation can become therapeutic options in SOD2-abundant RCC cells. Etomoxir, an inhibitor of beta-oxidation, and buformin, a biguanide that has been used for the treatment of diabetes mellitus, can inhibit mitochondrial oxidative phosphorylation and induce death of RCC cells [4]. Therefore, targeting mitochondrial activity with drugs such as etomoxir and biganide could be useful for the treatment of patients with poor prognosis due to SOD2-abundant RCCs and calls for drug repositioning.

In conclusion, we showed that high SOD2 expression is associated with worse clinical outcome and should be considered a useful marker in the follow-up of patients with primary and metastatic RCCs. Therapeutics for metastatic RCCs require further improvement, such as supplementary application of agents targeting mitochondrial SOD2.