Esophageal cancer is a representative disease among malignancies in the digestive system. In 2020, esophageal cancer had the seventh highest incidence and the sixth highest mortality in the world (1). The age-standardized incidence rate of esophageal cancer is highest in Eastern Asia (1). Squamous cell carcinoma (SqCC) is the most common histologic subtype of esophageal cancer, not only in Eastern Asia, but also globally (2). Surgery is the most radical and most common treatment option for esophageal cancer (3). According to stage or physician’s opinion, neoadjuvant concurrent chemoradiotherapy (CCRT) can be performed before surgery (4). After surgery, chemotherapy or radiotherapy can be conducted as adjuvant therapy. Definitive CCRT is another treatment option for patients who cannot be candidates for surgery (5). As most esophageal cancer patients undergo radical operations, it is important to predict the prognosis after surgery.

F-18 fluorodeoxyglucose positron emission tomography/computed tomography (FDG PET/CT) is a robust imaging modality to diagnose and evaluate stages of esophageal cancer (6, 7). It provides additional information of nodal staging and distant metastasis (8). Image findings and parameters of FDG PET/CT are well known to predict the prognosis of esophageal cancer. A high standardized uptake value (SUV) of the primary tumor was a significant risk factor in esophageal cancer based on a meta-analysis (9). Although there have been many studies, they included small numbers of patients or applied meta-analysis designs to overcome the limitations of low subject numbers. Furthermore, there were no previous studies dealing with prognostic values according to all discrete substages (e.g., IA and IB) due to the small numbers of subjects. Therefore, it was not clear whether the FDG PET/CT parameters of the primary tumor were independent prognostic factors over pathological substages or stages.

The tumor–node–metastasis (TNM) cancer staging system suggested by the American Joint Committee on Cancer (AJCC)/Union for International Cancer Control (UICC) is the most used system to evaluate the progression of malignancies including esophageal cancer. In 2017, the 8th edition of the AJCC/UICC staging system was released (10). A remarkable change between the 7th and 8th editions is the subdivision of the T1 stage into T1a and T1b according to invasion depth (11). Staging N3 as IVA and re-defining esophagogastric junction cancer are other changes (11). The classification criteria for the N stage have not been changed. In survival analysis, staging based on the 8th edition of the AJCC/UICC staging system still showed good prognostic stratification performance. However, no previous study has explored the independent prognostic value of FDG PET/CT compared to staging based on the 8th edition of the AJCC/UICC staging system.

In this study, the prognostic value of FDG PET/CT was investigated in a large number of subjects with esophageal SqCC according to the 7th and 8th editions of the AJCC/UICC staging system. The significance of SUVmax as a reference to stratify patients of same pathological stage was also explored.

The present study showed that the SUVmax of the primary tumor on FDG PET/CT was a significant prognostic factor on both continuous and discrete scales in univariate analysis. In addition, it was selected as an independent prognostic factor as well as a pathological TNM substage/stage based on both the 7th and 8th editions of the AJCC/UICC staging system in multivariate analysis. When subjects were grouped into our proposed new staging system including the 8th pathological stage and SUVmax, the new staging system was revealed to discriminate prognosis better than the conventional pathological staging system.

SUV is a semi-quantitative value for FDG PET/CT, reflecting the glucose metabolism of tissue. It is very useful to diagnose malignancy, evaluate aggressiveness, and predict the prognosis of cancer (14). Also, it is utilized in many radiogenomics approaches to investigate tumor biology non-invasively (15). There are other quantitative parameters derived from FDG PET/CT, such as metabolic tumor volume (MTV) and total lesion glycolysis (TLG). There have been several reports that have suggested the prognostic roles of MTV and TLG from pre-treatment FDG PET/CT in esophageal cancer (16–19). However, MTV and TLG have lesser reproducibility than SUVmax in terms of various and arbitrary tumor segmentation methods. Furthermore, an additional process is needed to delineate tumors and calculate MTV and TLG so that they are difficult to utilize in clinical fields. Compared to other volumetric parameters, SUVmax is the most common parameter in terms of its high reproducibility and simplicity and is easily clinically applicable in routine practice.

In esophageal cancer, the usefulness of FDG PET/CT was investigated in many previous reports (20, 21). Among them, a few studies showed that SUV was an independent predictor for OS after surgery (22–25). However, most of the previous studies recruited less than 100 subjects; Cerfolio et al. employed 89 patients (22), Cheze-Le Rest et al. used 47 patients (23), and Sepesi et al. utilized 72 patients (24). Kato et al. enrolled 184 patients and analyzed the prognostic role of peak SUV with TNM stage (e.g., I and II) based on the 6th edition of the AJCC/UICC staging system (25). Other scholars performed meta-analyses to include larger numbers of patients. Pan et al. incorporated 542 patients from 10 studies (9). However, the study included heterogeneous histologic subtypes and a threshold definition of SUV. Han et al. employed 1,294 patients from 16 studies to explore the prognostic value of FDG PET/CT parameters (16). However, CCRT without operation was performed in the majority of study subjects. To the best of our knowledge, this is the first study to explore the prognostic power of FDG PET/CT with a large number of subjects from a single institute. It is significant in that all patients were diagnosed with SqCC and underwent radical operations. Therefore, this study strongly suggests that SUVmax has a prognostic role in patients with esophageal SqCC undergoing curative surgery.

In this study, seven discrete substages (e.g., IA and IB) were used in univariate and multivariate Cox regression analyses. Notably, the prognostic power of SUV was significant with substage/stage not only based on the 7th edition of the AJCC/UICC staging system but also based on the 8th edition. In the 8th edition, it is remarkable that the T1 stage was classified more precisely according to the invasion depth of a tumor and that N3 was classified into stage IVA. One institute demonstrated that the staging system based on the 8th edition of the AJCC TNM staging system was superior to that based on the 7th edition in the prognosis of OS (26). Despite the improved prognostic power of the 8th-edition staging, this study showed that SUVmax still has its prognostic role. The present study may be the first study to investigate the prognostic role of FDG PET/CT based on both the 7th- and 8th-edition staging systems including all substages.

Herein, a new staging system including the 8th-edition staging system and the SUVmax of the primary tumor was proposed. The proposed new staging system demonstrated better survival curves than the conventional staging system. Notably, stage II and stage III patients with low SUVmax could be downstaged into the new stage I and stage II, respectively, in terms of prognostic stratification. Stage I patients with high SUVmax could be upstaged into the new stage II. This study suggests that SUVmax of the primary tumor deserves to be included in the next edition of staging system for esophageal SqCC to discriminate prognosis more accurately. Specifically, 3.4 or 4.7 is suggested as a reference value to upstage or downstage patients. In contrast, subjects with stage IV were not adjusted because there was no prognostic difference regardless of SUVmax reference. This indicates that FDG PET/CT may have a better prognostic role in the early stages of esophageal SqCC. In the current staging system, patients without lymph node metastasis (N0) are classified into an early stage below stage II. Therefore, it is presumed that biologic characteristics of the primary tumor affect prognosis more significantly in early stages than in advanced stages.

In the clinical field, the conventional TNM stage is commonly used to select appropriate treatment options and predict prognoses. The National Comprehensive Cancer Network guideline recommends esophagectomy as a primary treatment option for patients with low-risk stage II esophageal SqCC, as compared to preoperative or definitive chemoradiation for those with high-risk stage II esophageal SqCC (27). The guideline suggests the presence of lymphovascular invasion, size of tumor, and grade of differentiation as criteria for risk classification. Based on the results of this study, we suggest the SUVmax of the primary tumor as another favorable criterion to select high-risk stage II patients in terms of its good prognostic prediction power, the non-invasiveness of FDG PET/CT examination, and the simplicity of SUVmax. After radical operation, the surveillance follow-up duration is recommended as 3–6 months for 1–2 years. The result of this study implies that follow-up intervals for those with high FDG uptake should be shorter than the intervals for those with low FDG uptake considering the high risk for DFS and OS. In brief, the present study suggests that patients with the same TNM stage can be treated with different treatment options and follow-up plans according to the FDG uptake of primary tumor. Further study is warranted to evaluate the role of FDG PET/CT in selecting an appropriate treatment option in esophageal SqCC patients, especially those in stage II.

This study has several limitations. First, two different kinds of PET/CT scanners were used in this study, which might affect the measurement of SUVmax. Nevertheless, one study showed that SUVs from different instruments were not significantly different in a phantom study (28). Therefore, the analysis was performed based on the hypothesis that SUVs from different instruments can be analyzed without a specific integration process. Furthermore, subgroup analysis according to the PET/CT scanner model showed that SUVmax had a similar prognostic power irrespective of the PET/CT scanner model (Supplementary Tables 2–7). Second, only subjects undergoing curative surgery as initial treatment were included in this study for accurate pathological staging. Therefore, relatively small numbers of subjects with advanced stages were included. Further study including all esophageal cancer patients treated with various kinds of therapy is warranted. Third, our proposed new staging system was designed only for pathological stage, not for pathological substage. There was a lack of subjects in specific substages for subgroup analysis to compare HRs according to SUVmax. Further study with more patients could substantiate the value of SUVmax for upstaging or downstaging in pathological substages. Finally, due to the retrospective study design, subjects who underwent therapy based on the 6th- or 7th-edition staging system were also included. A further prospective study including subjects undergoing appropriate treatment according to the 8th-edition staging system is necessary.

In conclusion, the FDG uptake of the primary tumor on FDG PET/CT was found to be an independent prognostic factor along with pathological stage based on both 7th- and 8th-edition staging systems in patients with esophageal SqCC undergoing curative surgery. The proposed new staging system including SUVmax may be better for predicting prognoses than the conventional staging system.

The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding author.

The studies involving human participants were reviewed and approved by the Samsung Medical Center Institutional Review Board. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

HL and JC designed the study. YM, HK, JZ, and YS contributed to data collection. HL and JC performed data analysis and interpretation. HL drafted the article. KL and JC provided critical revision of the article. All authors contributed to the article and approved the submitted version.

This work was supported by the National R & D Program for Cancer Control of the Ministry of Health and Welfare of Korea (#1720180) and Future Medicine 20*30 Project of the Samsung Medical Center (#SMO1220071).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.