Abstract
Background: We investigated the utility of a novel combined indicator of both tumor depth and size in gastric cancer. Patients and Methods: A total of 938 patients with gastric cancer were analyzed. We tested tumor index (TI), calculated by T category × tumor size (mm) as a novel combined indicator of tumor depth and size. Results: Patients were classified into two groups using a cut-off value of 180 (p<0.0001) by the Kaplan–Meier method. TI was significantly positively correlated with older age, tumor located in the upper third of stomach, advanced type of macroscopic appearance, undifferentiated type, lymphatic invasion, venous invasion, lymph node metastasis, and recurrence. TI>180 only correlated with peritoneal recurrence (p=0.0394). A multivariate analysis identified TI>180 as an independent prognostic factor (hazard ratio=2.38, p=0.0004). Conclusion: TI may be a novel combined indicator of tumor status for predicting a poor prognosis and peritoneal recurrence in gastric cancer.
- Tumor size
- tumor depth
- prognosis
- gastric cancer
Gastric cancer is one of the most common causes of death from cancer worldwide (1). Recent advances have been achieved in diagnostic techniques, less-invasive treatment techniques and perioperative management, and mortality and morbidity rates have been reduced by the earlier detection of gastric cancer (2). However, tumors with advanced-stage disease have a greater depth of invasion and still present a poor prognostic outcome. The tumor status is the strongest predictor of the prognosis of gastric cancer, and treatment strategy based on the tumor status is the most important clinical issue (3).
Tumor size as horizontal tumor invasion is utilized as an important staging factor in head and neck cancer, as well as in lung and breast cancer (4). Although tumor depth has been identified as a strong prognostic and staging factor in gastric cancer (5, 6), several recent studies suggested that tumor size as horizontal tumor invasion was a prognostic factor in gastric cancer (7-12). However, the combination of tumor depth and invasion as an indicator has not yet been examined. Therefore, we herein tested the combined indicator of tumor depth and size using the Tumor Index (TI), which was calculated by multiplying the numerical T category by the tumor size (in millimeters), and hypothesized that TI may be a prognostic factor with the ability to stratify the prognosis of each stage in gastric cancer. The results of our study suggest that TI represents a better system for tumor staging and may be an indicator for future treatment strategies in gastric cancer.
Patients and Methods
Patients and surgical procedures. We retrospectively analyzed 938 consecutive patients who underwent curative gastrectomy for gastric cancer at the Division of Digestive Surgery, Kyoto Prefectural University of Medicine, between 1997 and 2011. Curative gastrectomy with lymphadenectomy was performed mainly based on the Japanese gastric cancer treatment guidelines (13, 14). Resected specimens were examined by pathologists and evaluated based on the 14th Japanese Classification of Gastric Carcinoma (JCGC) (15) and the 7th Tumor-Node-Metastasis (TNM) classification (4). Written informed consent was obtained from each patient prior to treatment initiation. The clinicopathological findings of these patients were examined retrospectively from hospital records (Table I). Surgical procedures comprised of distal gastrectomy in 601 patients, total gastrectomy in 262 patients, and proximal gastrectomy in 75 patients according to the preoperative stage and tumor location. Japanese style D2 lymphadenectomy or more was performed in 507 patients and less than D2 lymphadenectomy was performed in 431 patients. As a result, disease in 575 patients was classified as pStage I, in 171 as pStage II, and in 192 as pStage III. Patients with gastric cancer who underwent remnant gastrectomy and limited gastrectomy were excluded from this study.
Measurement of tumor diameter and calculation of TI. The resected stomach was opened, placed on a flat board with the mucosal side up, and fixed in 10% buffered formalin solution. After fixation, tumors in the resected stomach were sectioned on the maximum cross-sectional plane parallel to the lesser curvature line based on the general rules of the JCGC published by the Japanese Gastric Cancer Association (15). Tumors were generally sectioned in their entirety parallel to the reference line at 5-mm intervals. The resected specimens were embedded in paraffin and stained with hematoxylin and eosin. The longest tumor diameter was pathologically measured. TI was calculated by multiplication of the longest tumor size (mm) and pathological tumor (pT) stage (15); TI=pT category × tumor size (mm). Histological types were classified as differentiated (papillary adenocarcinoma or moderately or well-differentiated adenocarcinoma) or undifferentiated (poorly differentiated or undifferentiated adenocarcinoma, signet-ring cell carcinoma, or mucinous adenocarcinoma) based on the 14th JCGC (15).
Statistical analysis. The χ2 test and Fisher's exact probability test were performed for categorical variables, while the Student's t-test and Mann–Whitney U-test were performed for unpaired data of continuous variables to compare clinicopathological characteristics between the two groups. In order to analyze survival rates, survival curves were estimated using the Kaplan–Meier method and significant differences were examined using the log-rank test. Univariate and multivariate survival analyses were performed using the likelihood ratio test of the stratified Cox proportional hazards model. A p-value of less than 0.05 was considered significant. All statistical analyses were performed using the JMP 10.0 software program (Roppongi, Tokyo, Japan).
Results
Cut-off value of TI for stratifying prognosis. In order to properly classify the prognosis into four groups by TI, we performed detailed survival analyses using various cut-off interval values such as 10, 20, 30, 50, 70, and their multiples (Figure 1a-c). When increments of 30 or 90 were used as cut-off values, such as TI<90, 90≤TI<180, 180≤TI<270, and TI≥270, patients were more properly classified into four groups than when using other cut-off values (p<0.0001). The 5-year survival rates of patients with TI<90, 90≤TI<180, 180≤TI<270, and TI≥270 were 97.1%, 72.8%, 55.8% and 43.1%, respectively (Figure 1d). Furthermore, the cut-off value of 180 more significantly stratified the prognosis of gastric cancer patients into two groups (p<0.0001, 5-year survival rate; TI <180 vs. TI≥180: 92.7% vs. 48.5%) (Figure 2).
Comparison of clinicopathological factors between patients with TI<180 and TI≥180. We next compared clinicopathological factors between patients with TI<180 and TI≥180 (Table II). Older age (p=0.0326), tumor located in the upper third of stomach (p=0.0066), advanced type of macroscopic appearance (p<0.0001), undifferentiated type (p<0.0001), presence of lymphatic (p<0.0001) and venous invasion (p<0.0001), greater extent of lymph node metastasis (p<0.0001), and recurrence (p<0.0001) were significantly more frequent in patients with TI≥180 than in those with TI<180. A multivariate analysis using the Cox's proportional hazard model identified TI≥180 as an independent poor prognostic factor (hazard ratio=2.38, 95% confidence interval=1.47-3.92, p=0.0004) (Table III). We then compared the incidence of various recurrence patterns in 131 patients with recurrence according to TI. The incidence of peritoneal recurrence was significantly higher in patients with TI≥180 than in those with TI<180 (p=0.0394); no other relationship was observed between other types of recurrence and the incidence according to TI (Table IV).
Survival rates for different pathological tumor stages according to TI group. Table V shows the survival rates for each pT stage according to the TI cut-off value. TI had the ability to stratify the prognosis of patients with pT3 gastric cancer using all cut-off values (p=0.0141) and the cut-off value of 270 (p=0.0032). In pStage II gastric cancer, TI also had the ability to stratify the prognosis of patients using all cut-off values (p=0.0052) and the cut-off value of 270 (p=0.0011). About pStage III gastric cancer, prognosis of patients tend to be stratified using all cut-off value of 270 (p=0.1723) (Table VI). These results indicated that TI can contribute to decision-making on treatment strategies for patients with pT3 and pStage II-III gastric cancer by stratifying the prognosis of these patients.
Discussion
Although tumor depth has been identified as one of the most crucial factors for staging in gastric cancer (4, 15), tumor size as an indication of horizontal tumor invasion is also known to be an important prognostic factor in gastric cancer as with other solid cancers (5-12). The combined indicator of both tumor depth and size has not been examined before to our knowledge. Therefore, we herein developed a novel combined indicator, the TI. TI was associated with poor clinical outcomes and may be an independent poor prognostic factor. Furthermore, TI had the ability to stratify the prognosis of patients with pT3 and pStage II disease, and therefore may contribute to decision-making regarding treatment strategies for gastric cancer.
Previous studies reported the significance of tumor size in predicting the prognosis of patients with gastric cancer. Kunisaki et al. examined 1,215 patients with gastric cancer and classified them into two groups using a cut-off value of 100 mm as the maximal tumor diameter (9). This value may stratify the prognosis of stage II-III patients. Saito et al. also evaluated 1,473 patients with gastric cancer and divided them into two groups using a cut-off value of 80 mm for tumor size (8). Giuliani et al. categorized their patients with gastric cancer using a cut-off value of 25 or 50 mm (7). These findings suggested the clinical significance of various cut-off tumor sizes as an independent prognostic factor, but also showed that tumor size is not an alternative to the conventional factor of tumor depth because tumor depth has the ability to stratify the prognosis associated with larger or smaller tumors and is a prognostic factor itself independently of tumor size, as reported by Kunisaki et al. (9). This clinical issue prompted us to develop a combined indicator of tumor status, which included the benefits of both tumor depth and size.
In our study, gastric cancer tumors with high TI presented aggressive malignant clinical behavior, such as a higher incidence of the advanced type of macroscopic appearance, undifferentiated type, lymphatic invasion, venous invasion, lymph node metastasis, and recurrence. A tumor size of 8 cm or greater has already been included in the criteria of a recent clinical trial on preoperative chemotherapy for clinically resectable type 4 or large type 3 tumors in JCOG0210 (16) and JCOG0501 trials (17). TI may be a more sensitive criterion for such trials because it includes the clinical feature of tumor depth in addition to tumor size. Concerning preoperative or postoperative adjuvant chemotherapy, patients with stage III disease are considered to require further treatment strategies because the adjuvant chemotherapy trial of S-1 for gastric cancer trial did not demonstrate any survival benefit of postoperative adjuvant chemotherapy using the oral anticancer drug S-1 in these patients (18). Therefore, TI may also be a more useful indicator for selecting patients with stage III patients with poor outcomes in order to perform more intensive chemotherapy. Thus, TI may contribute to decision-making in future clinical trials.
Regarding recurrence, TI≥180 correlated with peritoneal recurrence (p=0.0394), among the various types of recurrence, in patients undergoing curative gastrectomy (Table IV). The rate of peritoneal recurrence in patients with 180< TI and TI <180 were 0% (0/2) and 0.7% (4/573) in pStage I, 14.3% (8/56) and 3.5% (4/115) in pStage II, and 29.6% (40/134) and 19.0% (11/58) in pStage IIII, respectively (Table VII). These results strongly suggested that a high TI may be a pivotal predictor of peritoneal recurrence; high TI may be an indicator in a meticulous follow-up for the early detection of peritoneal recurrence and for intensive adjuvant chemotherapy to prevent peritoneal recurrence.
Nevertheless, there are still limitations to this study. This was a retrospective single-center study. Furthermore, our identified TI cut-off value is still controversial. Preoperative TI, which was estimated by endoscopy or other modalities before surgery for the purpose of preoperative chemotherapy, may also need further investigation because preoperative tumor size may differ from pathological tumor size after gastrectomy. Therefore, a prospective observational study using several large cohorts may be needed to validate the significance of TI.
Footnotes
* These Authors contributed equally to this work.
- Received January 25, 2016.
- Revision received March 7, 2016.
- Accepted March 15, 2016.
- Copyright© 2016 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved