Esophageal cancer is one of the most common malignant tumors in the upper digestive tract of the chest [1,2]. Esophageal squamous cell carcinoma (ESCC) is among the most common types of upper gastrointestinal cancer [3,4]. Approximately 300,000 people die of esophageal cancer annually worldwide, with more than half in China [5]. Moreover, the mortality of esophageal cancer is ranked fourth in China [5]. Recurrence and metastasis of ESCC are still the main causes for death and treatment failure, and have crucial impact on the prognosis of ESCC [6,7]. Although considerable progress has been made in diagnosis and treatment, the relevant predictors for early diagnosis and prognosis remain lacking. Therefore, identification of molecular biomarkers will promote the development of new diagnostic and prognostic predictors for early diagnosis and personalized treatment of patients with ESCC.

Mitochondrial fission regulator 2 (MTFR2), also known as similarity 54 family member A (FAM54A), belongs to the MTFR1/FAM54 family [8,9]. It has two subtypes and is produced by selective shearing. MTFR2 plays important roles in mitochondrial division and aerobic respiration [10], while few studies have assessed the effects of MTFR2 expression on clinicopathological characteristics and prognosis of ESCC. Some previous studies found that the high expression of MTFR2 was correlated with lymph node metastasis, age and prognosis of breast cancer [11,12]. Currently, some studies have shown that MTFR2 is overexpressed in glioblastoma (GBM) and lung cancer, and this was correlated with sex, age, smoking history, cancer stage, histological subtype, and TP53 mutation status of lung cancer [9,13]. However, the role of MTFR2 expression in ESCC remains unclear.

In this study, we analyzed the expression of MTFR2 in six paired ESCC and normal specimens and 115 paraffin embedded ESCC tumor samples by immunohistochemistry (IHC). We evaluated the associations of MTFR2 expression with clinicopathological characteristics of ESCC patients, and further explored the effect of MTFR2 expression on clinical outcome of ESCC.

Preparation of patients' tissue specimens

Besides the 6-pairs of ESCC patients, additional 115 paraffin-embedded tumor specimens from patients, pathologically, diagnosed with ESCC between 2005 and 2010 at Yantai Yuhuangding Hospital, were included in this study. Detailed clinical information of the patients is shown in Table 1. Written informed consents were given to all patients before undertaking the study which was approved by the ethic committee of Yantai Yuhuangding Hospital (IRB No. 2017-204).

Table 1 . Clinicopathological characteristics of patients and MTFR2 expression in ESCC

Characteristics	No. of cases (%)
Sex
Female	6 (5.22)
Male	109 (94.78)
Age (yr)
> 60	56 (48.70)
≤ 60	59 (51.30)
Clinical stage
I/II	79 (68.70)
III/IV	36 (31.30)
T stage
1	20 (17.39)
2	36 (31.30)
3	40 (34.78)
4	19 (16.52)
N stage
0	76 (66.09)
1	33 (28.70)
2	6 (5.22)
M stage
0	110 (95.65)
1	5 (4.35)
Vital status
Alive	53 (46.09)
Dead	62 (53.91)
Histological differentiation
High	22 (19.13)
Moderate	67 (58.26)
Low	26 (22.61)
Expression of MTFR2
Low	77 (67.00)
High	38 (33.00)

All procedures performed in studies involving human participants were in accordance with the ethical standards of the Ethics Committee of the Yantai Yuhuangding Hospital Affiliated Hospital of Qingdao University and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Measurement of MTFR2 expression by IHC

MTFR2 expression was measured in 115 paraffin embedded ESCC specimens and 6 pairs of ESCC and normal tissue samples by IHC, as previously described [14]. In IHC, we used rabbit monoclonal antibody targeting MTFR2 (1:200, Sigma^® Life Science). Samples without any antibody were used as the corresponding blank control [14]. In qualitative analysis, we selected 6 samples, which contained cancer and adjacent normal tissues, for comparison [15]. Two independent pathologists who were not aware of the relevant clinical information evaluated the results of IHC staining of tumor sections [16]. The positive staining score of cancer cells was divided into the following five grades: 0 (0), 1 (< 10%), 2 (11% to 50%), 3 (51% to 75%), and 4 (> 75%). The staining intensity was graded as follows: 0 (negative = no staining), 1 (weak = light yellow), 2 (moderate = yellow brown), and 3 (strong = brown). The staining index (SI) was defined as the staining intensity score × the proportion of positive staining tumor cells (ranging from 0 to 12). Simultaneously, we defined SI ≥ 6 as high expression and SI < 6 as low expression.

Statistical analysis

The relationships between MTFR2 expression levels and sex, age, T stage, lymphatic metastasis, distant metastasis (tumor-node-metastasis [TNM] stage) and clinical stage were analyzed using the Spearman correlation method. The effect of MTFR2 expression on clinicopathological characteristics was assessed using the χ² test or the Fisher’s exact test. For survival analysis, the effect of clinicopathological characteristics and MTFR2 expression on prognosis of ESCC were examined by the Kaplan–Meier analysis and the log rank test, and further evaluated by univariate and multivariate Cox regression analyses. All statistical analyses were performed using SPSS software version 20 (SPSS Inc., Chicago, IL, USA). Statistical significance was set at P < 0.05.

MTFR2 is highly expressed in ESCC tissues

After IHC in 115 specimens, we found that MTFR2 was highly expressed in ESCC tissues. To validate the expression of MTFR2 in ESCC, we further analyzed the expression of MTFR2 in 6 pairs of ESCC tumor and adjacent normal tissues. Compared to that in normal tissues, the expression of MTFR2 in ESCC tissues was significantly upregulated as shown in Figure 1. In addition, to evaluate the association between MTFR2 expression and clinicopathological characteristics of patients with ESCC, we analyzed such associations in remaining 115 ESCC patients. These included 25 patients with stage I, 53 patients with stage II, 32 patients with stage III, and 5 patients with stage IV. The IHC staining showed that 38 cases (33.00%) had high MTFR2 protein expression and 77 cases (67.00%) had low or no expression (Table 1).

Figure 1. Upregulation of MTFR2 in ESCC tumor tissues.

Association between MTFR2 expression and clinicopathologic characteristics of ESCC

As shown in Table 2, there was no significant difference in MTFR2 expression for age and sex (P > 0.05), while the significant differences in MTFR2 expression were found for clinical stage (P < 0.001), pathological grade (P = 0.029), T/N stage (P < 0.001) and M classification (P = 0.040) (Table 2). These results suggested that MTFR2 expression was significantly associated with advanced tumors. Additionally, we also analyzed the correlations between MTFR2 expression and clinicopathologic characteristics. As shown in Table 3, there was statistically significant correlation between MTFR2 expression and clinical stage (r = 0.460, P < 0.001) (Fig. 2), T stage (r = 0.318, P = 0.001), N stage (r = 0.373, P < 0.001), M stage (r = 0.213, P = 0.040), and histological differentiation (r = 0.192, P = 0.038). Therefore, MTFR2 was significantly related to overall clinical stage, TNM stage, and poor pathological differentiation of ESCC, supporting a role of MTFR2 in progression of ESCC.

Table 2 . The association of MTFR2 expression with clinicopathological characteristics of ESCC patients

Characteristics	No. of cases	MTFR2 expression		χ2 test P	Fisher’s exact test P
		Low (%)	High (%)
Sex				0.351	0.267
Female	6	5 (83.33)	1 (16.77)
Male	109	72 (66.06)	37 (33.94)
Age (yr)				0.213	0.097
≤ 60	56	40 (71.43)	16 (28.57)
> 60	59	37 (62.71)	22 (37.29)
Clinical stage				< 0.001	< 0.001
I/II	79	65 (82.29)	14 (17.71)
III/IV	36	12 (33.33)	24 (66.67)
T stage				< 0.001	< 0.001
1	20	17 (85.00)	3 (15.00)
2	36	27 (75.00)	9 (25.00)
3	40	27 (67.50)	13 (32.50)
4	19	6 (31.58)	13 (68.42)
N stage				< 0.001	< 0.001
0	76	60 (78.95)	16 (21.05)
1	33	16 (48.48)	17 (51.52)
2	6	1 (16.67)	5 (83.33)
M stage				0.040	0.037
0	110	76 (69.09)	34 (30.91)
1	5	1 (20.00)	4 (80.00)
Histological differentiation				0.029	0.015
High	22	17 (77.27)	5 (22.73)
Moderate	67	47 (70.15)	20 (29.85)
Low	26	13 (50.00)	13 (50.00)

Table 3 . Spearman correlation analysis between MTFR2 expression and clinicopathologic characteristics

Variables	MTFR2 expression level Spearman’s correlation	P
Clinical stage	0.460	< 0.001
T stage	0.318	0.001
N stage	0.373	< 0.001
M stage	0.213	0.040
Vital status	0.427	< 0.001
Histological differentiation	0.192	0.038

Figure 2. MTFR2 expression in normal esophagus tissues and ESCC tumor specimens at different clinical stages.

MTFR2 expression is inversely associated with survival of ESCC

In the patients with low MTFR2 expression, the rate of cumulative 5-year survival was 61.0% (95% CI, 60.6%, 75.9%), while that was only 15.8% (95% CI, 15.2%, 28.1%) in patients with high MTFR2 expression.

Furthermore, we found that the survival time of ESCC patients was negatively correlated with MTFR2 expression (r = –0.580, P < 0.001), and the patients with high MTFR2 expression had worse survival than those with low expression (log-rank: P < 0.001, Fig. 3A). Our stratified analysis also showed the survival differences in several subgroups. Thus, there were significant differences in survival between MTFR2 expression in patients with early disease stage (stage I/II) (log-rank, P < 0.001, Fig. 3B) and those with advanced disease (stage III/IV) (P < 0.001, Fig. 3C). Similarly, there were significant survival differences between the MTFR2 expression and cumulative survival of patients with T1-2, T3-4, positive lymph nodes, negative lymph nodes, non-distant metastasis or distant metastasis groups (P < 0.001, Fig. 4). These findings in the subgroups further indicated that patients with high expression of MTFR2 had shorter survival than those with low expression of MTFR2, supporting the role of MTFR2 as a potential prognostic biomarker for ESCC.

Figure 3. Effect of MTFR2 expression on survival in all patients (A), early-stage patients (B), and late-stage patients (C).

Figure 4. Effect of MTFR2 expression on survival of ESCC patients stratified by T classification, lymph node status, and metastasis status.

Finally, we performed both univariate and multivariate Cox regression analyses to assess the associations between MTFR2 expression and survival of patients with ESCC. Our results showed that the expression of MTFR2 was a significantly independent prognostic predictor for ESCC (adjusted hazard ratio [aHR], 3.04, 95%CI, 1.75 to 5.30; P < 0.001; Table 4).

Table 4 . Univariate and multivariable analysis on association between MTFR2 expression and survival in patients with ESCC

Variables	No. of patients	Univariate analysis				Multivariate analysis
		P	cHR†	95% CI		P	aHR*	95% CI
MTFR2 expression		< 0.001				< 0.001
Low	77		1.00				1.00
High	38		3.12	1.66-5.27			3.04	1.75-5.30

^†cHR, crude hazard ratio; *aHR, adjusted hazard ratio; HR was adjusted by age, sex, clinical stage, T/N/M stage, and histological differentiation.

In this study, MTFR2 expression was relatively higher in tumor lesions than that in normal tissues. MTFR2 was highly expressed in ESCC tumor specimens, and it was significantly correlated with clinical stage, T, N, M classification, histological differentiation, and metastasis of ESCC. The expression of MTFR2 increased with the progression of ESCC. Our current findings demonstrate that MTFR2 is an independent prognostic factor for the prognosis of patients with ESCC, and the patients with high MTFR2 expression have worse survival than those with corresponding low expression. Therefore, it is likely that MTFR2 expression may have a significant clinical value as a new prognostic predictor and one of potential novel targets for future targeted treatment of ESCC.

As a mitochondrion fission factor, MTFR2 participates in aerobic respiration and has antioxidant effects [8,17]. In addition, MTFR2 may play a key role in the induction of intrinsic apoptosis [18,19]. Compared to its role in normal cells, mitochondrion induces tumor-related programming and accelerates growth of tumor cells and blood vessels [20,21]. Some studies have shown that MTFR2 may transcriptionally regulate dual specificity protein kinase TTK [9]. MTFR2 is one of the key genes related to TTK, which can be regulated by activating the promoter of TTK [22,23]. Furthermore, it is involved in regulation of proliferation of glioma stem cells (GSCs) and thus may affect the prognosis of GBM [9,20]. Most recently, MTFR2 has been reported to promote the proliferation, migration, and invasion of oral squamous carcinoma [24], and MTFR2-dependent regulation of TTK was involved in maintaining GSCs in GBM and might serve as a potential novel druggable target for GBM. In addition to cell proliferation, TTK was also involved in centrosome duplication, DNA damage response, and organ development [25]. Another study showed that reduced TTK levels could induce abnormal mitoses, activate apoptosis and reduce survival of breast cancer [26].

TTK depletion might seriously impair the viability and clonogenic capability of triple-negative breast cancer cell lines [27]. Thus, TTK might be an independent predictor of prognosis, and it appears biologically plausible that MTFR2 might play important roles in regulation of TTK in ESCC as well as occurrence and development of ESCC. However, the exact mechanisms underlying such associations require further exploration.

Dastsooz et al. [10] showed that the expression of MTFR2 was related to poor prognosis, suggesting that MTFR2 might be involved in the progression and invasion of several cancers [28,29]. In recent studies [11,12], MTFR2 was highly expressed in breast cancer tissue. Its altered expression affected the proliferation and metastasis of breast cancer and inhibited epithelial mesenchymal transition. Other similar studies found that the high expression of MTFR2 was related to the clinicopathology, diagnosis and prognosis, and promoted the occurrence and cancer development through p53-dependent cell cycle, DNA replication, and homologous recombination in lung cancer [13]. Such findings may highlight the potential role of MTFR2 in ESCC as a novel therapeutic target for treatment.

Although we found that MTFR2 was overexpressed in ESCC, and was significantly associated with advanced disease and prognosis, which was consistent with the research on glioma [9], our research has some limitations. First, we only analyzed the expression levels of cell protein, while we did not specifically analyze other aspects of genetic alterations. Second, our sample size is relatively small, and all study patients are from a single hospital. Moreover, the selection bias may also exist. Nevertheless, the current preliminary findings from this pilot study might serve as a basis for future hypothesis generation and testing as well as validation in large well-designed prospective studies. Finally, more studies are needed to explore the exact molecular mechanisms underlying such associations.

In conclusion, in the current study, we found that MTFR2 is overexpressed in ESCC and correlated with ESCC progression. MTFR2 overexpression was inversely associated with survival of patients with ESCC, particularly in aggressive tumors. Thus, our current findings might support that MTFR2 could be an independent biomarker for prognosis and exploited as a potential new therapeutic target for ESCC.

This article is one of the results of the Key Research and Development Plan of Yantai City (2018SFGY110).

No potential conflicts of interest were disclosed.