The predicting role of serum tumor-specific growth factor for prognosis of esophageal squamous cell carcinoma

Esophageal squamous cell carcinoma (ESCC) is one of the most common malignancies involving the digestive system, and the leading cause of cancer-related deaths worldwide owing to its aggressiveness and adverse survival probability. In China, ESCC is the principal subtype of esophageal cancer and remains the sixth leading cause of cancer incidence and the fourth leading cause of cancer mortality [1]. In recent years, the clinical approaches to overcoming ESCC have focused on the combination therapy comprising operative treatment, chemotherapy, and radiotherapy based on patient background and TNM staging by the American Joint Committee on Cancer (AJCC)/International Union Against Cancer (UICC) clinical stage [2]. Currently, emerging immunotherapy is an additional treatment method [3]. However, tumor development dominated by diverse tumor environments and complex heterogeneity. Despite the more extensive efforts to defeat ESCC and improve the prognosis, it is indispensable to find a reliable clinically beneficial biomarker to trace the progression of ESCC.

Tumor-specific growth factor (TSGF) is a collective name that includes multiple substances related to the malignant growth of tumors. Based on previous efforts, accumulating studies have revealed the importance of serum TSGF for the clinical identification and assessment of treatment effects in various malignancies, such as breast cancer, and digestive tract carcinoma [4‐17]. Of note, combining TSGF and conventional biomarkers improved essential clinical utility. The combination of color Doppler ultrasound with serum carcinoembryonic antigen (CEA), carbohydrate antigen 153 (CA15-3), and TSGF detection enhanced the diagnostic efficiency of breast cancer [4]. TSGF and other tumor markers combined with 18 F-FDG-PET achieved approximately 97.3% of diagnostic accuracy, and yielded 100% specificity for discriminating prostate cancer [11]. The joint detection of serum CEA, carbohydrate antigen 72-4 (CA72-4), carbohydrate antigen 19-9 (CA19-9), and TSGF has shown robust discriminatory power in gastric cancer (AUC = 0.913, sensitivity: 88.9%) [6]. Similarly, the combined assay with serum TSGF, carbohydrate antigen 242 (CA242), and CA19-9 appeared to be able to identify pancreatic cancer [7]. Meanwhile, three serum markers, consisting of AC007271.3, squamous cell carcinoma antigen (SCCA), and TSGF, resulted in an AUC of 0.917 with a sensitivity of 80.0% and specificity of 93.1% in oral squamous cell carcinoma [14]. TSGF predicts the survival consequence of pancreatic cancer patients undergoing cryoablation treatment [8], utilizes to monitor the treatment response in colon cancer [9], and determines the survival of bladder cancer patients undergoing robot-assisted radical cystectomy [10]. For primary hepatocellular carcinoma, patients carrying increased TSGF levels obtained a reduced 3-year survival rate, showing its value for evaluating survival benefit of transcatheter arterial chemoembolization (TACE) [12]. For osteosarcoma, higher TSGF levels decreased 5-year survival rate [15]. Yang L et al. performed a study with 753 retecal carcinomas and demonstrated that decline of TSGF in 5th postoperative day (POD) less than 10 U/mL showed a positive correlation with anastomotic leakage (AL) after anterior resection for rectal cancer with a double stapling technique [16]. Altogether, these results demonstrate that serum TSGF plays a crucial role in the screening, prognosis prediction, and treatment outcome of cancer. However, its role in ESCC is unclear.

Unexpectedly, only a few studies are currently available that have examined the role of TSGF in esophageal carcinoma. The combined detection of AgNOR, SCCA, CEA, and TSGF was favorable for screening esophageal carcinoma [18]. Therefore, we retrospectively reviewed 562 ESCC cases to assess their clinicopathological relevance and potential role as prognostic indicators for ESCC.

In our present research, we investigated the predictive property of preoperative serum TSGF and explored its clinical utility in ESCC. We analyzed the correlation between serum TSGF and some clinical parameters, including the survival of ESCC. We found that it correlated with sex, tumor location, smoking and drinking history, TNM stage, and lymph node involvement.

Epidemiologically, smoking and alcohol drinking enhanced the risk of ESCC [19]. Even the studies depicted tobacco and alcohol may play a positive synergistic effect to increase 3-fold of the risk of suffering ESCC compared with the predicted risk of any single factor alone(only 20–30%) [20]. In Non-small cell lung cancer, the median and positive rates of TSGF were 64(56–67)μ/mL and 10.14%(28/276). However, TSGF level showed no statistical difference between smokers(median: 63, range: 53–66) and non-smokers(median: 62, range: 51–67) (P > 0.05) [21]. To date, there was no report about its relationship with drinking alcohol. We firstly indicated that TSGF was associated with smoking and drinking history. And higher TSGF in the smoking and alcohol-drinking groups predicted poor clinical outcome of ESCC. More importantly, habitually consuming tobacco, alcohol, and areca nuts can influence the age-onset of male ESCC [22].

Commonly, metastasis leads to an unfavorable clinical outcome of ESCC. For ESCC, lymph metastasis showed a remarkable correlation with malignant features such as tumor differentiation, perineural invasion, advanced-stage tumor, and venous invasion. It predicted a worse prognosis and could act as an independent prognostic factor for ESCC [23]. In our cohort, 246 of 302(43.77%) ESCC cases harbored lymph node involvement, which was slightly higher than the previous report with 36.8% lymph node metastasis [23]. Min BH et al. demonstrated that the different LNM rates may be associated with tumor invasion depth [24]. The LNM rates of tumors invading the lamina propria, muscularis mucosa, and SM1 layers were 3.7%, 15.5%, and 40.7%, respectively. To date, there have been few available studies on the predictive role of the serum level of TSGF for lymph node involvement in cancer. TSGF is a polypeptide that triggers initiation and facilitates metastasis of cancer. TSGF levels have been related to tumor differentiation in pancreatic cancer [8] and lymph node involvement in colon cancer [9]. Our previous study revealed that increased serum TSGF may enhance the emergence of multifocality and the development of PTC, suggesting that TSGF may be qualified as an excellent promising marker for predicting lymphatic involvement in PTC [25]. Our results focused on the clinical utility of TSGF in cancer. Nevertheless, its role in ESCC for predicting prognosis was not uncovered. Our analysis supported that increased TSGF was likely to develop metastasis involving the lymph node in ESCC. This result suggested that serum TSGF may play a distinct role in cancer progression.

In general, the TNM stage represents the tumor burden in tumor-bearing patients. Advanced TNM stage not only results in increased metastasis probability, including more lymph node involvement, but also contributes to poor survival. In the past several years, the therapeutic benefit of ESCC patients was limited by the late diagnosis and asymptomatic early disease [1]. In our present study, we identified that the cases were predominantly at stage II, and 239 of 562 (41.72%) had advanced disease. Moreover, the ESCC patients with high TSGF levels were prone to be at advanced TNM stages. And TNM stage was a predictor of lymph node involvement and poor prognosis [26]. Our result was indicative of its pro-tumor effect in ESCC progression. In addition, there was no significant correlation with other traits, such as age, tumor grade, or pathologic T stage.

To date, whether serum TSGF has a connection with cancer-related deaths has been previously described in several types of cancer. It has been reported that serum TSGF exhibits a superior early detection capability [4‐7, 11, 13], prognosis prediction role [12, 15], and response to the therapeutic outcomes [8‐10, 12, 14‐16] of malignancies. Based on the previous exploration, combining serum TSGF with traditional biomarkers or other detection tools caused better diagnostic accuracy [5]. Higher levels appeared to correlated with worse clinical outcomes of cancer [12, 15]. Our findings indicated that ESCC patients with decreased serum TSGF had a better prognosis than those with high TSGF. This finding seemed to be consistent with the predictive value of hepatocellular carcinoma [12] and osteosarcoma [15]. Furthermore, we conducted the strata analysis to achieve the satisfactory survival prediction for the special ESCC patients. We found that younger-aged ESCC patients with elevated serum TSGF had worse survival. There were similar results in the ESCC subgroup with male cases, middle esophagus, tobacco and alcohol consumption, well and moderate differentiation, advanced TNM staging and pathologic T stage, and lymph node metastasis. These findings might aid clinicians in determining the population at risk for exposure for precise therapeutic intervention and clinical management.

Finally, the univariate analysis showed that higher serum TSGF levels, some malignant features including TNM stage, lymph node status, pathologic T stage, and ESCC lifestyle factors such as tobacco consumption and alcohol exposure could lead to unfavorable clinical outcomes in patients with ESCC. However, multivariate analysis showed that only advanced pathologic T stage, smoking habit, and increased serum TSGF increased the risk of ESCC-related prognosis. Our results suggested it’s necessary to explore the clinical implications of serum TSGF in ESCC.

There were also some limitations in this study. Noticeably, the cases at stage II accounted for most of our cohorts, while the minority were in stage IV. In our cohort, not all subjects had received the same post-operation treatment, and due to the relatively small sample size, it was unfavorable to perform a subgroup analysis based on different treatment options. In addition, the sample size from the single center was limited, although it was larger than the previous cohort. Therefore, it’s necessary to enlarge the sample from multiple centers to obtain a better performance of TSGF.

Collectively, serum TSGF was relevant to malignant features of ESCC. Increased TSGF predicted the adverse prognosis of ESCC, especially for male patients aged under 60 and with lymph node involvement, advanced disease stage, and smoking and drinking habits. Moreover, the serum TSGF was qualified as an independent predictor for the clinical outcome of ESCC by multivariate analysis. Since the assay of cytokine levels in ESCC patients is feasible and convenient in clinical practice, we will perform further observation to investigate its clinical utility and the molecular mechanisms underlying TSGF activity to shed light on its role in ESCC progression.