Complete blood count-based inflammatory score (CBCS) is a novel prognostic marker for gastric cancer patients after curative resection

Gastric cancer (GC), one of the most common malignancies, is the third most common cause of cancer-related death worldwide [1]. Despite improvements in surgical techniques and therapeutic modalities, survival of GC remains poor [2]. The role of inflammation in the development of tumors was first described in the nineteenth century [3]. Currently, there is accumulating evidence that the host inflammatory response plays an important role in the development and progression of cancer [4]. Complete blood count (CBC)-based biomarkers are a series of inflammatory indicators based on blood cells [5]. Pretreatment CBC-based biomarkers, including blood neutrophil, lymphocyte, monocyte, and platelet counts; hemoglobin (Hb) levels; and their combinations, such as the neutrophil-lymphocyte ratio (NLR), lymphocyte-monocyte ratio (LMR) and platelet-lymphocyte ratio (PLR), have been reported to reflect systemic and local inflammation associated with cancer progression and prognosis [6‐10].

The American Joint Committee on Cancer (AJCC) staging system is the most widely used system to assess prognosis [11], but survival can vary in patients with GC even when they have the same TNM stage. Therefore, it is necessary to improve the individual prognostic prediction for GC by combining the AJCC staging system with other prognostic indicators. Previous studies have shown that preoperative CBC-based biomarkers can be used as part of a prognostic model to predict the prognosis of tumors more accurately [12, 13]. However, there is likely some degree of overlap between these CBC-based biomarkers, and some may be redundant. Therefore, for future studies and potential clinical use, a more simple and effective prognostic inflammatory biomarker is required. The aim of this study was to further evaluate the prognostic value of these CBC-based biomarkers and to establish a simple inflammatory scoring system based on CBC biomarkers to efficiently predict long-term outcomes for GC patients after surgery.

Currently, accumulating evidence suggests that the systemic inflammatory response plays an important role in tumor progression and metastasis [27, 28]. There has been growing interest in using CBC-based measures as biomarkers for GC [7, 8, 10]. However, there is an overlap between these indicators, and the actual prognostic accuracy of some indicators is poor. Therefore, we sought to evaluate which of these CBC-based biomarkers ultimately display the greatest potential in GC patients. The AUC is an index that can be used to compare the prognostic abilities of different factors. The higher the AUC value is, the stronger the predictive ability of the prognostic factor; this metric has been widely used in previous studies [5, 29, 30]. In this study, we compared the AUC values of similar CBC-based biomarkers, and the CBC-based biomarkers with higher AUC values were retained for further evaluation. Finally, the PLR, LMR and Hb level were retained for subsequent analyses. Multivariate analysis revealed that the preoperative LMR and Hb level were independent CBC-based predictors of OS for patients with GC undergoing curative surgical resection, which is consistent with the findings of previous studies [6, 7, 17]. Furthermore, we developed a novel CBC-based prognostic score, called the CBCS, based on the combination of the LMR and Hb level after dichotomization to more accurately and easily predict the long-term prognosis of GC patients.

This study assessed the associations between preoperative CBCS and clinicopathological factors. Our results revealed that an elevated CBCS was associated with a number of variables that were previously shown to be predictive of poor outcomes. These variables include tumor location, tumor diameter, vascular invasion, and tumor stage. In addition, multivariate analysis demonstrated that the CBCS is an independent prognostic factor for GC patients. At present, there is no consensus on how the AUC value can be used in clinical practice [9, 31, 32]. In our study, the AUC value of CBCS was 0.627, which was significantly higher than those of the LMR and Hb level according to the Delong test [33]. Our study indicated that the CBCS has better discriminatory ability than its components in terms of determining the prognosis of GC patients. As an inflammatory scoring system based on the LMR and Hb level, the biological rationale behind the prognostic value of the CBCS might involve the function of monocytes, lymphocytes and Hb. Circulating monocytes may contribute to both tumor growth and reduced immunosurveillance, which is supported by previous findings [34]. In addition, there is mounting evidence that tumor-associated macrophages (TAMs) primarily exert protumoral activity, including the promotion of metastasis, immunosuppression, and tumor angiogenesis [35]. Therefore, an increase in peripheral blood monocytes is associated with poor prognosis in patients. Lymphocytes are basic components of the adaptive and innate immune systems and form the cellular basis of immunosurveillance and immunoediting [36]. Due to tumor-infiltrating lymphocyte-induced antitumor activity and the inhibition of angiogenesis, the presence of tumor-infiltrating lymphocytes is associated with improved survival in various cancers [37]. Lymphopenia has been associated with poor prognosis in cancer patients [38, 39]. In addition, Zhang et al. demonstrated that anemia is an independent risk factor for advanced GC [40]. Anemia may have an impact on the quality of life, performance status, treatment tolerance, clinical symptoms, recovery from surgery and even outcomes [41, 42]. Therefore, the CBCS, which is based on both the LMR and the Hb level, may enable a better understanding of the effects of the tumor on both ongoing systemic inflammation and the functional state of patients. We also found that the CBCS could be used for further risk stratification in each TNM stage, suggesting that the CBCS might provide additional prognostic information to complement postoperative pathological staging.

In recent years, a number of inflammatory scoring systems, such as the SII, PNI, mGPS, and CRP/Alb, have been established to predict the prognosis of GC [9, 18, 19, 32]. In this study, we used t-ROC analyses to compare the prognostic values of CBCS and other inflammatory scoring systems, including the SII, PNI, mGPS and CRP/Alb. The advantage of this method is that it can assess the impact of individual prognostic factors and enables the analysis of survival data with censoring using ROC curves [25]. In addition, we found that the t-ROC curve for the CBCS was consistently superior to those for the SII and PNI after surgery. We also attempted to clarify the utility of the mGPS and CRP/Alb in comparison with CBCS. In our center, CRP was not a routine parameter tested in GC patients in the past. However, in recent years, it has been reported that preoperative CRP levels may affect the prognosis of GC. Thereafter, we routinely performed preoperative CRP examinations. Thus, CRP levels were only available in some patients. Our analysis demonstrated that the CBCS yielded a better t-ROC curve than did mGPS and CRP/Alb. Thus, as a novel inflammatory prognostic factor, CBCS is a superior predictor of OS compared with other inflammatory scoring systems. Furthermore, we found that compared with the traditional pathological stage, the model established by combining the CBCS and pTNM stages can predict the long-term survival of patients with GC more effectively. Thus, the CBCS can be used as a supplement to the traditional pathological stage in clinical practice to better stratify patients and provide a more accurate basis for guiding postoperative follow-up and individualized treatment.

Nevertheless, there were several limitations in our study. First, because of its retrospective nature, our study may have been subject to selection bias. For example, only some patients had preoperative CRP values. Additionally, because of the retrospective nature of the study, the time between drawing the blood and analysis of the blood sample was not noted. Second, we excluded patients with neoadjuvant therapy to ensure that all patients were in the same state before blood sampling. It is common practice in the West to give neoadjuvant treatment to patients with locally advanced disease, but this is not common practice in the East [43]. In China, most patients do not receive pre-adjuvant therapy, so there are relatively fewer patients with of neoadjuvant therapy in this study, and most patients with neoadjuvant chemotherapy have more advanced disease, usually stage T4b or metastatic disease. Therefore, the results of this study are not applicable to GC patients undergoing neoadjuvant therapy. Third, due to the lack of data from other centers, we could not validate the results externally. Last, the AUC values of CBCS are low, but after combining the CBCS with the traditional staging system, we found that the CBCS can improve the accuracy of the prognostic evaluation for patients with GC.

Our study is the first to develop the CBCS and demonstrate that the preoperative CBCS, based on the LMR and the Hb level, is the most efficient marker for predicting OS in GC patients. The findings of this study can help clinicians select, as part of individualized GC treatment strategies, the most effective inflammatory markers for preoperative risk stratification and postoperative follow-up.