Prognostic effect of postoperative duration until adjuvant chemotherapy and cumulative S-1 dose in gastric cancer

Background Adjuvant chemotherapy (AC) following curative gastrectomy for stage II/III gastric cancer (GC) is recommended in Japan. However, for various reasons, patients cannot always start AC at the appropriate time. This study was designed to investigate the effect of the postoperative duration until adjuvant chemotherapy (PDAC) and cumulative S-1 dose on prognosis.


Background
Gastric cancer is the fourth most commonly diagnosed cancer and the third most common cause of cancer-related death worldwide [1]. Treatment for gastric cancer has been enhanced by improvements in surgical procedures and perioperative management [2][3][4][5][6].
Although surgical resection mainly involves macroscopic tumor resection and lymphadenectomy, the oncologic effect of surgical treatment is often limited to the local control. Therefore, perioperative therapy has been recommended to clear remaining microscopic metastasis [7].
The ACTS-GC (Adjuvant Chemotherapy Trial of TS-1 for Gastric Cancer) study and its 5-year follow-up results demonstrated that adjuvant chemotherapy (AC) with S-1 monotherapy following curative gastrectomy with lymphadenectomy significantly improved relapse-free survival and overall survival rates. Therefore, AC following curative gastrectomy has been recommended as the standard treatment in Japan for stage II/III gastric cancer since 2014 (JGCTG 2014) [8,9].
The ACTS-GC study proved the beneficial prognostic effect of starting S-1 treatment within 6 weeks of undergoing curative surgery. However, patients cannot always start AC at the appropriate time following surgery due to various clinical issues that can result in delayed recovery. The prognostic effect of any delay with regard to AC remains controversial [10,11]. Even if the prognosis may be affected by a delay in starting AC, until now there has been no definite cut-off value for the postoperative duration following curative surgery with which to stratify the prognosis [12]. Moreover, it is not clear whether a sufficient total S-1 dose during AC could affect the prognosis or diminish the poor prognostic effect resulting from a delay in starting AC.
In this study, we investigated the clinical significance of the postoperative duration until adjuvant chemotherapy (PDAC) and the total cumulative S-1 dose, with particular reference to the prognosis. The results of our study suggest a crucial indicator starting AC. Our study also suggested that the cumulative S-1 dose might be key to diminishing any poor prognostic effect, even if there is a delay in starting AC.

Patients and surgical procedures
This study was approved by the Kyoto Prefectural University of Medicine, Japan, and was therefore performed in accordance with the ethical standards laid down in the Declaration of Helsinki. Written informed consent was obtained from all patients to participate in the research. A total of 214 consecutive patients underwent curative gastrectomy with lymphadenectomy for stage II or III GC at our institute between January 2008 and December 2014. Of these 214 patients, 97 patients received S-1 monotherapy as AC. We excluded 21 patients from the study because of neoadjuvant chemotherapy (n = 2), insufficient follow-up (n = 1), recurrence during AC (n = 8), remnant GC (n = 3), and multiple carcinomas (n = 7). As a result, we retrospectively investigated 76 consecutive patients who received S-1 AC (Fig. 1).
Patients with clinical stage IA (clinical T1 and clinical N0 GC) underwent D1+ lymphadenectomy, while those with clinical stage IIA or higher underwent D2 or D2+ lymphadenectomy. All patients underwent macroscopic curative resection (R0); resected specimens were examined by pathologists and evaluated in accordance with the Japanese classification of GC [13]. Dissected lymph nodes were fixed in buffered formalin and embedded in paraffin prior to pathological examination. Pathologists in our institution examined embedded lymph nodes by sectioning slices in the plane of the largest node dimension to confirm the presence of metastasis. Clinicopathological findings from these patients were determined retrospectively on the basis of their hospital records.

Follow-up after curative gastrectomy followed by AC
Postoperative follow-up was performed in the outpatient clinic every three months following surgery. Blood chemistry was also measured every three months. Endoscopic examinations were performed annually, and computed tomography (CT) examinations were performed every three-to-six months for five years after surgery. The average followup period was 52.7 months.
AC was begun in the outpatient clinic following discharge. The dose of S-1 was determined according to a patient's body surface area (BSA). Specifically, patients with BSA < 1.25 m 2 received 80 mg per day; patients with 1.25 m 2 ≤ BSA < 1.5 m 2 received 100 mg per day; and patients with 1.5 m 2 ≤ BSA received 120 mg per day. The cumulative S-1 dose received by each patient during AC was calculated from their hospital records.

Statistical analysis
Statistical analyses were conducted using JMP version 10 (SAS Institute Inc., Cary, NC).
The Mann-Whitney U test for unpaired data comprising continuous variables was used to compare clinicopathological variables. For the analysis of survival, Kaplan-Meier survival curves were constructed for groups based on univariate predictors, and differences between the groups were tested using a generalized Wilcoxon test. The Cox proportional hazards model was used for further evaluations of multivariate survival analysis. A p-value < 0.05 was considered statistically significant. Table 1 shows the clinical characteristics of GC patients who received curative gastrectomy followed by S-1 AC. The mean age of the patients was 62.9 years. Of 76 patients, 47 (61.8%) were male and 29 (38.2%) were female; 30 patients were at pStage IIA, 17 patients were at pStage IIB, 17 patients were at pStage IIIA, and 12 patients were at pStage IIIB. Total gastrectomy was performed in 30 patients (39.5%), distal gastrectomy in 44 patients (57.9%), and proximal gastrectomy was performed in two patients (2.6%) for curative resection dependent on the location of the tumor. A total of 19 patients (25.0%) underwent D1+ lymphadenectomy based on their clinical stage, while the remaining patients performed D2 or D2+ lymphadenectomy [13].

Cut-off value of PDAC to stratify the prognosis and correlation between PDAC and clinicopathological factors
We performed a minimum p-value analysis for relapse-free survival (RFS) using various cut-off values for PDAC, as shown in Fig. 2. The cut-off value of 7 weeks post-surgery was confirmed to be the upper-limit to stratify the prognosis (p = 0.017; 5-year RFS: PDAC ≥ 7 weeks vs. PDAC < 7 weeks; 48.5 % vs. 77.0 %). Although there were no significant differences in 5-year overall survival (OS) rates, patients in the PDAC ≥ 7 weeks group had a poorer prognosis than those in the PDAC < 7 weeks group (p = 0.49; 5-year OS: PDAC ≥ 7 weeks vs. PDAC < 7 weeks; 74.3 % vs. 80.7 %). (Fig. 3).

Cut-off value of cumulative S-1 dose to stratify the prognosis and combined survival curves using PDAC and cumulative S-1 dose factors
To clarify the clinical effect of cumulative S-1 dose, we performed a minimum p-value analysis for RFS using various cut-off values of cumulative S-1 dose (data not shown). A cut-off value of 12,000 mg was confirmed to stratify the prognosis most (p = 0.006; 5-year RFS: cumulative S-1 dose ≥ 12,000 mg vs. < 12,000 mg; 84.7 % vs. 52.4 %).

Comparison of PDAC with clinicopathological factors
Next, we evaluated correlations between PDAC and clinicopathological factors using the Mann-Whitney U test. As shown in Table 2, a high peak in the C-reactive protein (CRP) cut-off value of 8 mg/dl or above was significantly associated with a delay in starting S-1 AC. In addition, the incidence of postoperative complications (Clavien-Dindo classification ≥ II) tended to be linked to longer PDAC (p = 0.065). There were no other significant differences between the groups with regard to other clinicopathological factors.

Univariate and multivariate analysis using Cox's proportional hazard model
To elucidate the prognostic factors for recurrence-free survival, univariate and multivariate analysis using Cox's proportional hazard model were performed. As shown in

Discussion
There have been few reports of prognostic effects with regard to PDAC and the extent of S-1 treatment in AC [11,12,14]. In this study, we clearly demonstrated that a PDAC ≥ 7 weeks and a cumulative S-1 dose of more than 12,000 mg were independent prognostic factors in GC patients undergoing AC following curative gastrectomy. Moreover, in patients who received a cumulative S-1 dose of more than 12,000 mg, we showed that there were no prognostic differences between patients who had a PDAC of more than or less than 7 weeks. PDAC ≥ 7 weeks. Our results strongly suggested that a cumulative S-1 dose of more than 12,000 mg is a crucial indicator and might be the key dose for diminishing the poor prognostic effects arising from a delay in AC.
Concerning PDAC, we clearly demonstrated that 7 weeks was the best cut-off value to stratify the prognosis of patients with pStage II/III gastric cancer (p = 0.017; 5-year RFS: PDAC ≥ 7 weeks vs. PDAC < 7 weeks; 48.5 % vs. 77.0 %). From an oncological perspective, to eliminate microscopic metastasis it appears that AC should be started immediately following curative gastrectomy. PDAC has been reported to be an independent prognostic factor in other types of cancer, with various cut-off values suggested, including 12 weeks [10] and 8 weeks in colon cancer [15][16][17] and 13 weeks (91 days) in breast cancer [18]. With regard to gastric cancer, various cut-off values of PDAC have been reported as prognostic factors, ranging from 4 to 8 weeks [12,16,[19][20][21]. Our PDAC cut-off value of 7 weeks could also be a candidate indicator. Future studies are warranted, and big data analysis might be needed to determine the best cut-off value for clinical settings.
Regarding the cumulative S-1 dose, we also demonstrated that a cumulative S-1 dose of more than 12,000 mg was an independent prognostic factor for RFS [p = 0.033; HR 0.38 (95% CI: 0.14-0.93)]. The Japanese phase 3 randomized trial (JCOG1104 [OPAS-1]) revealed that four courses of S-1 AC were inferior to eight courses of S-1 AC for achieving RFS of pStage II gastric cancer, highlighting the importance of the duration of AC [22].
Fujitani et al. also reported that S-1 AC with a duration of more than 6 months could have a prognostic impact in patients with gastric cancer [11]. There results could be a crucial indicator of the importance of S-1 AC duration. However, the S-1 dose intensity in each patient is considerably different in clinical practice, and is indeed not as high as shown in previous reports [8,23,24]. Therefore, we suggest that the cumulative S-1 dose is also a pivotal factor, in addition to the PDAC of S-1.
The most striking finding in the present study was that there were no prognostic differences for RFS between patients with PDAC ≥ 7 weeks and those with PDAC < 7 weeks (p = 0.64; 5-year RFS: PDAC ≥ 7 weeks vs. PDAC < 7 weeks; 85.7 % vs. 84.4 %).
Specifically, these results suggested that a cumulative S-1 dose of more than 12,000 mg might diminish the poor prognostic effect associated with a delay in AC. Similar results were also reported that S-1 AC of more than 6 months [11] and S-1 AC with a relative dose intensity of more than 64.6 % [12] have more prognostic impact than PDAC. Although which factors have the optimal impact on prognosis is unclear, we enhance on the This study had some limitations: the results were obtained from a retrospective evaluation of a small number of patients at a single institute. A large-scale, multicenter cohort study is necessary to confirm the significance of PDAC and cumulative S-1 dose.

Conclusions
Patients with early initiation of AC had better prognosis than those with delayed initiation. 7 weeks could be a crucial indicator starting AC, and a cumulative S-1 dose of more than 12,000 mg might be a key dose for diminishing the poor prognostic effects of delaying AC.

Acknowledgements
Not applicable.

Ethical approval and consent to participate
This study was designed in accordance with the Declaration of Helsinki and was approved by the Institutional Review Board of Kyoto Prefectural University of Medicine. All patients received sufficient explanation of the study, and written informed consent was obtained.

Consent for publication
Not applicable.

Availability of data and materials
The datasets used during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.

Funding
There is no funding to be declared. Tables Table 1 Clinicopathological characteristics of patients who received S-1 monotherapy as adjuvant chemotherapy.      The cut-off value of PDAC to stratify relapse-free survival (RFS) rates the most effectively in pStage II/III gastric cancer was 7 weeks post-surgery (p = 0.017).  patients who received a cumulative S-1 dose ≥ 12,000 mg or S-1 dose < 12,000 mg, there was no difference between PDAC ≥ 7 weeks and PDAC < 7 weeks (p =