ABSTRACT
Background
Relationship between timing of initiating adjuvant chemotherapy (AC) and clinical outcomes after surgery for pancreatic cancer remains controversial.
Methods
In this ancillary analysis of the JASPAC 01 trial, 187 patients were classified according to the interval between surgery and chemotherapy initiation to Early (< 6 weeks, n = 45), Standard (6–8 weeks, n = 70), and Delayed (> 8 weeks, n = 72). Overall survival (OS) and relapse‐free survival (RFS) were analyzed.
Results
Baseline characteristics were comparable among three groups. The Standard group showed significantly longer OS than the Early group (median, 66 vs. 37 months; HR 0.61, 95% CI 0.38–0.99; p = 0.041), and relatively longer OS compared with the Delayed group (median, 45 months; HR 0.68, 95% CI 0.44–1.05; p = 0.077). RFS was longer in the Standard group (median, 46 months) compared with the Early group (20 months; HR 0.61, 95% CI 0.38–0.99; p = 0.040) and the Delayed group (20 months; HR 0.59, 95% CI 0.39–0.89; p = 0.011). Multivariate analysis identified operative procedure, R1 resection, lymph node metastasis, and nonstandard initiation (< 6 or > 8 weeks) as independent adverse prognostic factors.
Conclusions
Initiation of S‐1 AC at 6–8 weeks after pancreatectomy was associated with favorable survival outcomes.
Keywords: adjuvant chemotherapy, JASPAC 01, pancreatectomy, pancreatic cancer, S‐1 timing
Lee and colleagues found, in an exploratory ancillary analysis of the JASPAC 01 trial, that initiation of adjuvant S‐1 chemotherapy 6–8 weeks after pancreatectomy was associated with improved overall and relapse‐free survival. Nonstandard initiation before 6 or after 8 weeks remained an independent adverse prognostic factor for survival.
1. Introduction
Pancreatic ductal adenocarcinoma (PDAC) is one of the most aggressive and fatal malignancies. PDAC remains the third leading cause of cancer‐related deaths in the United States in 2025, with a 5‐year survival rate of 13% [1]. Complete surgical resection remains the only potentially curative treatment for localized PDAC; however, recurrence occurs in most patients even after R0 resection. Several large randomized controlled trials have demonstrated the efficacy of adjuvant chemotherapy (AC) for improving survival after curative pancreatectomy [2, 3, 4]. Among these, the Japan Adjuvant Study Group of Pancreatic Cancer (JASPAC) 01 trial [4] compared S‐1, an oral fluoropyrimidine, with gemcitabine in 385 patients (S‐1, n = 187; gemcitabine, n = 190) and demonstrated a clear survival advantage in the S‐1 group (median OS, 46.5 vs. 25.5 months; HR, 0.57; p < 0.0001). Consequently, adjuvant S‐1 chemotherapy has been recognized as the standard postoperative AC regimen for resected pancreatic cancer in Japan and other Asian countries.
However, the optimal timing for initiating AC after pancreatectomy remains controversial. It has been suggested that delayed initiation is associated with poorer outcomes, while some studies have reported that the timing of initiation has no significant impact on survival [5, 6, 7, 8, 9, 10]. In other gastrointestinal cancers, clinically relevant time thresholds for initiating AC have been proposed. In colon cancer, initiation within 8 weeks after surgery has been associated with improved survival and is generally recommended [11, 12]. In gastric cancer, the ACTS‐GC trial [13], which established the efficacy of adjuvant S‐1 therapy, defined initiation within 6 weeks after surgery as an inclusion criterion, and subsequent studies have reported improved outcomes with initiation within this time frame [14]. While delayed initiation has been linked to worse prognosis, recent large database analyses in pancreatic cancer have shown that excessively early initiation may also fail to improve survival, suggesting a U‐shaped relationship between timing of initiation and prognosis [6]. Similar findings have been reported in other malignancies, such as ovarian cancer, where initiating AC too early after surgery was associated with worse outcomes [15].
In this ancillary analysis of the JASPAC 01 trial, we investigated the relationship between the timing of initiating AC with S‐1 and survival outcomes, focusing on the survival impact of excessively early and delayed initiation.
2. Methods
2.1. Study Design and Patients
This study was conducted as an ancillary analysis of the JASPAC 01 trial, a phase 3 randomized controlled trial comparing AC with S‐1 versus gemcitabine for curatively resected pancreatic cancer. The details of the original study have been described previously [4]. Briefly, eligible patients aged ≥ 20 years with an ECOG performance status of 0 or 1 and adequate organ function, and no history of chemotherapy and radiotherapy within 3 years before surgery. All patients underwent macroscopically curative resection (R0 or R1) of histologically proven invasive ductal carcinoma of the pancreas. Pathological staging was defined according to the Union for International Cancer Control TNM classification, 6th edition, and patients with stage I–III disease were eligible [16]. This ancillary analysis included patients assigned to the S‐1 arm who actually received S‐1 (per‐protocol population, n = 187). According to the JASPAC 01 protocol, eligible patients were required to be enrolled within 10 weeks after surgery, and protocol treatment was to be initiated within 15 days after registration. The protocol of the original JASPAC 01 trial (UMIN000000655), which included use of the obtained data for the ancillary analysis, was approved by the institutional review boards of all participating centers, including Kobe University Graduate School of Medicine (Approval No. 190013), and was conducted in accordance with the Declaration of Helsinki. All patients provided written informed consent both for participating in the trial and for the future ancillary analysis before enrollment to the JASPAC 01 study. This ancillary analysis was approved by the study group and performed using de‐identified data.
2.2. Treatment
S‐1 was administered orally twice daily for 28 consecutive days followed by a 14‐day rest (one cycle). The dose was based on body surface area: 40 mg (< 1.25 m2), 50 mg (1.25–1.49 m2), or 60 mg (≥ 1.5 m2). Treatment was repeated every 6 weeks for up to four cycles. Dose modification and discontinuation criteria followed the JASPAC 01 protocol [4].
2.3. Grouping According to Timing of Initiation
Patients were categorized into three groups according to the interval between surgery and initiation of AC with S‐1: Early group: < 6 weeks, Standard group: 6–8 weeks, Delayed group: > 8 weeks. These cut‐off values were defined a priori based on clinically relevant time thresholds reported in other gastrointestinal malignancies. In gastric cancer, initiation of adjuvant S‐1 within 6 weeks after surgery was adopted as an inclusion criterion of the ACTS‐GC trial [13], and earlier initiation within this period has been associated with improved survival [14]. In colon cancer, initiation within approximately 8 weeks after surgery has been associated with better outcomes, whereas delays beyond this time frame have been linked to inferior prognosis [11, 12]. Based on these established thresholds, 6 and 8 weeks were selected as clinically meaningful cut‐off points in the present study.
2.4. Endpoints
The primary outcome of this ancillary analysis was overall survival (OS), defined as the interval from surgery to death from any cause. The secondary endpoints included relapse‐free survival (RFS), defined as the interval from surgery to the first recurrence or death.
2.5. Statistical Analysis
Baseline characteristics were compared among groups using the chi‐square test or Fisher's exact test for categorical variables as appropriate, and the Mann–Whitney U test for continuous variables. Survival curves were estimated using the Kaplan–Meier method and compared by log‐rank test. Hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated using the Cox proportional hazards model. Sensitive analyses were performed in patients who completed all four cycles of S‐1. Multivariate analysis was performed using the Cox proportional hazards model in which not only factors showing p value < 0.05 in univariate analysis but also clinically important factors such as operative procedure, nodal status, and residual tumor status regardless of their p values were included as variables. Statistical significance was defined as a two‐sided p value < 0.05. All statistical analyses were performed using JMP software (version 16; SAS Institute, Japan).
3. Results
3.1. Patient Characteristics
A total of 187 patients who received adjuvant S‐1 in the JASPAC 01 trial were included in this ancillary analysis. The median interval from surgery to initiation of adjuvant S‐1 chemotherapy was 50 days (around 7 weeks). According to the interval between surgery and initiation of S‐1, 45 patients (24%) were classified into the Early group (< 6 weeks), 70 (37%) into the Standard group (6–8 weeks), and 72 (39%) into the Delayed group (> 8 weeks). Baseline characteristics are summarized in Table 1. No significant differences were observed among the three groups in age, sex, ECOG performance status, preoperative CA19‐9 level, operative procedure, R0 resection rate, or pathological stage, while the proportion of patients with lymph node metastasis tended to be higher in the Standard group than in the Delayed group (71% vs. 56%, p = 0.050). Postoperative complications occurred less frequently in the Early group than in the Standard group (16% vs. 39%, p = 0.007), and the median postoperative hospital stay was correspondingly shorter (18 vs. 23 days, p = 0.007).
TABLE 1.
Baseline characteristics of the patients.
| Values | Early (n = 45) | Standard (n = 70) | Delayed (n = 72) | Standard versus early p | Standard versus delayed p |
|---|---|---|---|---|---|
| Age, years | 66 (36–86) | 66 (45–84) | 67 (34–84) | 0.814 | 0.618 |
| Sex, male | 23 (51) | 43 (61) | 40 (56) | 0.275 | 0.478 |
| ECOG performance status, 0 | 28 (62) | 50 (71) | 53 (74) | 0.302 | 0.771 |
| Preoperative CA19‐9 levels, U/mL | 58 (0–4812) | 98 (1–8591) | 65 (0–14 148) | 0.643 | 0.408 |
| Operative procedure, PD/DP | 25/20 | 45/25 | 51/21 | 0.349 | 0.405 |
| Combined portal vein resection | 12 (27) | 24 (34) | 18 (25) | 0.390 | 0.226 |
| Combined celiac axis resection | 2 (4) | 4 (6) | 2 (3) | 0.765 | 0.385 |
| Operative time, min | 394 (93–765) | 372 (155–811) | 379 (158–824) | 0.540 | 0.620 |
| Intraoperative blood loss, ml | 703 (50–3636) | 675 (143–7130) | 789 (105–2834) | 0.770 | 0.297 |
| Residual tumor status, R0 | 40 (89) | 64 (91) | 60 (83) | 0.651 | 0.147 |
| Primary tumor size, TS1/TS2/TS3/TS4 | 7/33/5/0 | 18/43/6/3 | 17/46/8/1 | 0.261 | 0.709 |
| Primary tumor status, T1/T2/T3/T4 | 3/2/39/1 | 2/6/62/0 | 6/2/63/1 | 0.369 | 0.173 |
| Regional lymph node status, N1 | 30 (67) | 50 (71) | 40 (56) | 0.588 | 0.050 |
| Pathological Stage, IA/IB/IIA/IIB/III | 3/0/12/29/1 | 2/3/15/50/0 | 4/2/26/39/1 | 0.296 | 0.188 |
| Histological type, Tubular adenocarcinoma | 41 (91) | 61 (87) | 60 (83) | 0.512 | 0.523 |
| Postoperative complications | 7 (16) | 27 (39) | 33 (46) | 0.007 | 0.381 |
| Postoperative hospital stay, days | 18 (9–55) | 23 (9–92) | 24 (9–81) | 0.007 | 0.777 |
| Surgery to adjuvant chemotherapy, days | 35 (17–41) | 49 (42–56) | 66 (57–85) | < 0.001 | < 0.001 |
Note: Values are n (%) and median (range). Primary tumor size and status, regional lymph node status, and pathological stage were described according to the International Union Against Cancer (UICC) TNM Classification of malignant tumors, 6th edition.
Abbreviations: CA19‐9, carbohydrate antigen 19–9; DP, distal pancreatectomy; PD, pancreatoduodenectomy.
3.2. Treatment Feasibility and Compliance
Treatment feasibility and compliance according to the timing of AC with S‐1 are summarized in Table 2, and the reasons for treatment discontinuation are provided in Table S1. Completion rates of AC were comparable among the Early, Standard, and Delayed groups (64%, 76%, and 74%, respectively), with no significant differences between the Standard group and either the Early or Delayed groups. The proportion of patients who completed AC without dose reduction was slightly lower in the Early group (Early 51%, Standard 61%, and Delayed 63%). Median relative dose intensity was high, maintained, and comparable among the three groups (99.4%, 97.1%, and 96.3%, respectively). Treatment discontinuation due to adverse events, physician decision, and refusal occurred in 24% of patients in the Early group, 20% in the Standard group, and 21% in the Delayed group, while discontinuation due to disease recurrence was observed in 9%, 4%, and 3% of patients, respectively.
TABLE 2.
Treatment compliance according to the timing of adjuvant chemotherapy with S‐1.
| Values | Early (n = 45) | Standard (n = 70) | Delayed (n = 72) | Standard versus early p | Standard versus delayed p |
|---|---|---|---|---|---|
| Completion of adjuvant chemotherapy | 29 (64) | 53 (76) | 53 (74) | 0.192 | 0.773 |
| Completion of adjuvant chemotherapy without dose reduction | 23 (51) | 43 (61) | 45 (63) | 0.275 | 0.895 |
| Relative dose intensity | 99.4 | 97.1 | 96.3 | 0.282 | 0.731 |
| Discontinuation of adjuvant chemotherapy due to adverse events, physician decision, refusal | 11 (24) | 14 (20) | 15 (21) | 0.573 | 0.902 |
| Discontinuation of adjuvant chemotherapy due to recurrence | 4 (9) | 3 (4) | 2 (3) | 0.314 | 0.626 |
Note: Values are n (%) and median (range).
3.3. Survival Outcomes
Among these 187 patients, 123 patients (66%) experienced disease recurrence during the follow‐up period. The initial sites of recurrence were the local region (n = 34), para‐aortic or peritoneal lymph nodes (n = 20), liver (n = 33), lung (n = 19), peritoneum (n = 20), and other sites (n = 23). There were no significant differences in the initial sites of recurrence among the three groups.
The Standard group showed longer OS (median, 66 months) than the Early group (median, 37 months; HR 0.61, 95% CI 0.38–0.99; p = 0.041) and a trend toward longer OS compared with the Delayed group (median, 45 months; HR 0.68, 95% CI 0.44–1.05; p = 0.077) (Figure 1a). Similarly, RFS was longer in the Standard group (median, 46 months) compared with the Early group (median, 20 months; HR 0.61, 95% CI 0.38–0.99; p = 0.040) and the Delayed group (median, 20 months; HR 0.59, 95% CI 0.39–0.89; p = 0.011) (Figure 1b).
FIGURE 1.
Kaplan–Meier curves of overall survival (OS) (a) and relapse‐free survival (RFS) (b) in all patients (n = 187) according to the timing of adjuvant S‐1 chemotherapy initiation: Early (< 6 weeks, n = 45), Standard (6–8 weeks, n = 70), and Delayed (> 8 weeks, n = 72). The Standard group showed longer OS (median, 66 months) than the Early group (median, 37 months; p = 0.041) and the Delayed group (median, 45 months; p = 0.077). RFS was longer in the Standard group (median, 46 months) compared with the Early group (median, 20 months; p = 0.040) and the Delayed group (median, 20 months; p = 0.011). *p < 0.05; NS = not significant. Numbers below the plots indicate the number of patients at risk at each time point.
3.4. Sensitive Analysis
As the sensitive analysis, patients who completed adjuvant S‐1 chemotherapy showed more favorable survival outcomes than those who did not complete treatment (Figure 2). The median OS was 59 months in the completion group compared with 32 months in the non‐completion group (p = 0.008; HR 0.59, 95% CI 0.40–0.88). Median RFS was significantly longer in the completion group (30 vs. 18 months; p = 0.045; HR 0.67, 95% CI 0.46–0.99). In the sensitive analysis limiting to patients who completed 4 courses of adjuvant S‐1 chemotherapy, this subset in the Standard group demonstrated longer OS (median, 88 months) than that in the Early group (median, 38 months; HR 0.49, 95% CI 0.27–0.90; p = 0.019) and that in the Delayed group (median, 47 months; HR 0.56, 95% CI 0.33–0.95; p = 0.028) (Figure 3a). Similarly, RFS was longer in this subset in the Standard group (median, 88 months) compared with that in the Early group (median, 23 months; HR 0.52, 95% CI 0.30–0.93; p = 0.023) and that in the Delayed group (median, 21 months; HR 0.53, 95% CI 0.32–0.87; p = 0.009) (Figure 3b).
FIGURE 2.
Kaplan–Meier curves of overall survival (OS) (a) and relapse‐free survival (RFS) (b) according to completion of adjuvant S‐1 chemotherapy. Patients who completed adjuvant S‐1 chemotherapy showed significantly longer OS and RFS than those who did not complete treatment. The median OS was 59 months in the completion group compared with 32 months in the non‐completion group (p = 0.008). Median RFS was significantly longer in the completion group (30 vs. 18 months; p = 0.045). *p < 0.05; **p < 0.01. Numbers below the plots indicate the number of patients at risk at each time point.
FIGURE 3.
Kaplan–Meier curves of overall survival (OS) (a) and relapse‐free survival (RFS) (b) among patients who completed adjuvant S‐1 chemotherapy (n = 105) according to the timing of chemotherapy initiation: Early (< 6 weeks, n = 29), Standard (6–8 weeks, n = 53), and Delayed (> 8 weeks, n = 23). The Standard group demonstrated longer OS (median, 88 months) than the Early group (median, 38 months; p = 0.019) and the Delayed group (median, 47 months; p = 0.028). RFS was longer in the Standard group (median, 88 months) compared with the Early group (median, 23 months; p = 0.023) and the Delayed group (median, 21 months; p = 0.009). *p < 0.05; **p < 0.01. Numbers below the plots indicate the number of patients at risk at each time point.
3.5. Prognostic Factors
Univariate and multivariate analyses of prognostic factors for OS are summarized in Table 3. In multivariate analysis, operative procedure (pancreaticoduodenectomy) (HR 1.52, 95% CI 1.02–2.26, p = 0.039), R1 resection (HR 1.75, 95% CI 1.05–2.90, p = 0.031), positive lymph node metastasis (HR 2.31, 95% CI 1.50–3.57, p < 0.001), and nonstandard initiation of AC (< 6 weeks or > 8 weeks) (HR 1.77, 95% CI 1.18–2.65, p = 0.005) were independently associated with worse overall survival.
TABLE 3.
Prognostic factors for overall survival.
| Factors | Patients (n = 187) | Univariate p | Hazard ratio (95% CI) | Multivariate p |
|---|---|---|---|---|
| Age (≥ 70 years) | 67 (36) | 0.787 | — | — |
| Sex (male) | 106 (57) | 0.377 | — | — |
| Performance status, 1 | 38 (20) | 0.526 | — | — |
| Preoperative CA19‐9 level (≥ 37 U/mL) | 121 (65) | 0.170 | — | — |
| Operative procedure (PD) | 121 (65) | 0.054 | 1.52 (1.02–2.26) | 0.039 |
| Residual tumor status, R1 | 23 (12) | 0.013 | 1.75 (1.05–2.90) | 0.031 |
| Primary tumor size (≥ TS2) | 145 (78) | 0.683 | ||
| Regional lymph node status, N1 | 120 (64) | < 0.001 | 2.31 (1.50–3.57) | < 0.001 |
| Postoperative complications | 67 (36) | 0.282 | ||
| Nonstandard initiation of AC (< 6 or > 8 week) | 117 (63) | 0.032 | 1.77 (1.18–2.65) | 0.005 |
Note: Values are n (%). Primary tumor size and regional lymph node status were described according to the International Union Against Cancer (UICC) TNM Classification of malignant tumors, 6th edition.
Abbreviations: AC, adjuvant chemotherapy; CA19‐9, carbohydrate antigen 19‐9; PD, pancreatoduodenectomy.
4. Discussion
In this ancillary analysis of the JASPAC 01 trial, we demonstrated that the timing of adjuvant S‐1 chemotherapy initiation was significantly associated with survival outcomes after curative pancreatectomy for PDAC. Patients who initiated AC 6–8 weeks after surgery achieved the most favorable OS and RFS, compared with both earlier (< 6 weeks) and delayed (> 8 weeks). There was no significant difference in adherence of S‐1 among the three groups. In addition, this association between timing of initiating S‐1 and survival outcomes persisted in the sensitive analysis restricted to the patients who completed AC. These findings suggest that treatment timing may be associated with survival outcomes independently from treatment adherence. Furthermore, because the proportion of distal pancreatectomy was higher in the Early group, exploratory subgroup analyses stratified by operative procedure were conducted. Among patients receiving pancreaticoduodenectomy, the Standard group showed significantly longer OS than the Early group. Among patients receiving distal pancreatectomy, the Standard group showed significantly longer RFS than the Delayed group (Figures S1 and S2). Multivariate analysis further indicated that initiation at 6–8 weeks remained independently associated with favorable survival outcomes after adjustment for potential confounders as much as possible. However, given the limited sample size and the exploratory nature of these analyses, these findings should be interpreted cautiously.
Large‐scale clinical studies have consistently demonstrated that delayed initiation of AC worsens prognosis after curative pancreatectomy for PDAC [6, 9]. Ma et al. retrospectively analyzed 5453 patients in the National Cancer Database and showed that initiation of AC later than 8 weeks resulted in significantly shorter survival compared with initiation at 4–8 weeks (median OS, 22.4 vs. 20.4 months; p = 0.01) [6]. Similarly, a nationwide Japanese multicenter retrospective study of 3361 patients demonstrated improved survival when adjuvant S‐1 chemotherapy was initiated within 10 weeks compared to after 10 weeks (median OS, 57.0 vs. 43.8 months; p < 0.001) [9]. Consistently with these findings, delayed initiation (> 8 weeks) in our cohort was associated with significantly worse RFS and showed a trend toward inferior OS not only in the overall population but also in patients who completed AC. Unlike previous database analyses, because our study was well‐controlled to maintain high RDI, delayed initiation might not have an adverse impact on compliance of S‐1. Indeed, as shown in Table 2, no significant differences in treatment compliance were observed between the Delayed and Standard groups. It is speculated that prolonged postoperative intervals might lead to tumor regrowth or progression of occult micrometastases, which might diminish the efficacy of S‐1. In addition, an important finding of our analysis was that the Standard group had a higher proportion of patients with lymph node–metastatic disease (71%) than the Delayed group (56%, p = 0.050). While lymph node metastasis was one of the strongest independent adverse prognostic factors in our multivariate analysis, the Standard group—despite including a greater number of patients with biologically advanced disease—demonstrated the most favorable survival outcomes. This observation suggests that initiation at 6–8 weeks was associated with favorable survival outcomes even in a cohort including a higher proportion of node‐positive disease.
Notably, excessively early initiation of AC (< 6 weeks) was also associated with inferior survival compared with initiation at 6–8 weeks, despite similar RDI and fewer postoperative complications. This was consistent with a large retrospective National Cancer Database study of PDAC patients that showed early (< 4 weeks) initiation of AC was associated with worse outcomes, although that study included heterogeneous regimens and lacked detailed RDI data [6]. Similar findings have been reported in ovarian cancer, where excessively early chemotherapy initiation within 21 days after surgery was associated with poorer survival [15]. Several hypotheses may explain this association. The immediate postoperative phase is characterized by systemic inflammation, transient immunosuppression, and activation of wound‐healing pathways, which have been proposed to influence tumor progression and treatment response in several reports [17, 18, 19]. In addition, nutritional depletion and incomplete hepatic or renal recovery after major pancreatic surgery might affect S‐1 metabolism [20, 21]. However, these mechanisms were not directly evaluated in the present study and remain speculative. Furthermore, although baseline tumor‐related prognostic factors such as tumor size, nodal involvement, margin positivity, or elevated preoperative CA19‐9 levels were comparable among groups, unmeasured clinical factors—such as physician judgment, intraoperative impressions, or perceived pathological risk—may have influenced the timing of chemotherapy initiation. Thus, confounding by indication cannot be completely excluded. Moreover, although multivariate adjustment was performed, residual confounding related to postoperative recovery status and other unmeasured factors may persist. Importantly, in this cohort, no deaths or recurrences occurred within the first 4 months after surgery. Because adjuvant S‐1 was initiated within 85 days in all patients, no early postoperative events occurred during the timing classification period (< 6 or 6–8 weeks). Accordingly, survival conditioning bias due to early mortality or recurrence is likely limited in the present analysis. Nevertheless, as treatment timing was not randomized, residual structural bias cannot be entirely excluded. Collectively, our analysis of S‐1 monotherapy within a uniform clinical trial suggests that excessively early initiation was associated with less favorable survival outcomes.
This study has several limitations. First, as this was an ancillary analysis, inherent biases and unmeasured confounding factors—such as postoperative nutritional status and inflammatory markers—were not captured in the original JASPAC 01 dataset. Although no deaths or recurrences occurred before initiation of S‐1 AC in this prospective cohort, survival conditioning and immortal time bias cannot be excluded completely because treatment timing was not randomized. A schematic representation of the potential bias structure underlying this observational analysis is provided in Figure S3 to facilitate interpretation of these findings. Therefore, our findings should be interpreted as an observational association rather than causal evidence of an optimal therapeutic window. Second, the relatively small sample size of the Early group, together with the lack of detailed information regarding the clinical reasons for early or delayed initiation of AC, limits the interpretation of our findings and precludes further exploration of causality. Third, the time windows for initiating AC were defined based on evidence from other gastrointestinal cancers, and an optimal initiation window specific to this population may be identified in future studies. Finally, because all patients received S‐1 monotherapy, generalization of these results to modern multi‐agent regimens such as FOLFIRINOX or gemcitabine plus nab‐paclitaxel should be made with caution. Nonetheless, the prespecified treatment and prospective data collection of the original clinical trial strengthen the internal validity of our findings.
In conclusion, initiation of S‐1 AC at 6–8 weeks after pancreatectomy was associated with favorable survival outcomes in this exploratory ancillary analysis. These findings should be interpreted cautiously and warrant validation in future prospective studies.
Funding
This work was supported by Pharma‐Valley Center, Shizuoka Industrial Foundation (currently The Mt. Fuji Foundation for Healthcare Innovation and Cluster Development), and Taiho Pharmaceutical.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Figure S1: Kaplan–Meier curves of overall survival (OS) (a) and relapse‐free survival (RFS) (b) among patients who underwent pancreaticoduodenectomy (PD) according to the timing of adjuvant S‐1 chemotherapy initiation: Early (< 6 weeks, n = 25), Standard (6–8 weeks, n = 45), and Delayed (> 8 weeks, n = 51).
Figure S2: Kaplan–Meier curves of overall survival (OS) (a) and relapse‐free survival (RFS) (b) among patients who underwent distal pancreatectomy (DP) according to the timing of adjuvant S‐1 chemotherapy initiation: Early (< 6 weeks, n = 20), Standard (6–8 weeks, n = 25), and Delayed (> 8 weeks, n = 21).
Figure S3: Conceptual diagram illustrating the potential sources of bias in the analysis of the timing of S‐1 initiation and overall survival.
Table S1: Reasons for discontinuation of S‐1 chemotherapy among patients who discontinued treatment according to timing group.
Acknowledgments
The authors have nothing to report.
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
References
- 1. Siegel R. L., Kratzer T. B., Giaquinto A. N., Sung H., and Jemal A., “Cancer Statistics, 2025,” CA: A Cancer Journal for Clinicians 75, no. 1 (2025): 10–45, 10.3322/caac.21871. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Neoptolemos J. P., Stocken D. D., Bassi C., et al., “Adjuvant Chemotherapy With Fluorouracil Plus Folinic Acid vs Gemcitabine Following Pancreatic Cancer Resection: A Randomized Controlled Trial,” JAMA 304, no. 10 (2010): 1073–1081, 10.1001/jama.2010.1275. [DOI] [PubMed] [Google Scholar]
- 3. Oettle H., Neuhaus P., Hochhaus A., et al., “Adjuvant Chemotherapy With Gemcitabine and Long‐Term Outcomes Among Patients With Resected Pancreatic Cancer: The CONKO‐001 Randomized Trial,” JAMA 310, no. 14 (2013): 1473–1481, 10.1001/jama.2013.279201. [DOI] [PubMed] [Google Scholar]
- 4. Uesaka K., Boku N., Fukutomi A., et al., “Adjuvant Chemotherapy of S‐1 Versus Gemcitabine for Resected Pancreatic Cancer: A Phase 3, Open‐Label, Randomised, Non‐Inferiority Trial (JASPAC 01),” Lancet 388, no. 10041 (2016): 248–257, 10.1016/s0140-6736(16)30583-9. [DOI] [PubMed] [Google Scholar]
- 5. Abe K., Furukawa K., Fukuda M., et al., “Timing of TS1 Adjuvant Chemotherapy as a Prognostic Factor in Recurrent Pancreatic Cancer After Surgery,” Surgical Oncology 58 (2025): 102179, 10.1016/j.suronc.2024.102179. [DOI] [PubMed] [Google Scholar]
- 6. Ma S. J., Oladeru O. T., Miccio J. A., Iovoli A. J., Hermann G. M., and Singh A. K., “Association of Timing of Adjuvant Therapy With Survival in Patients With Resected Stage I to II Pancreatic Cancer,” JAMA Network Open 2, no. 8 (2019): e199126, 10.1001/jamanetworkopen.2019.9126. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Mirkin K. A., Greenleaf E. K., Hollenbeak C. S., and Wong J., “Time to the Initiation of Adjuvant Chemotherapy Does Not Impact Survival in Patients With Resected Pancreatic Cancer,” Cancer 122, no. 19 (2016): 2979–2987, 10.1002/cncr.30163. [DOI] [PubMed] [Google Scholar]
- 8. Murakami Y., Uemura K., Sudo T., et al., “Early Initiation of Adjuvant Chemotherapy Improves Survival of Patients With Pancreatic Carcinoma After Surgical Resection,” Cancer Chemotherapy and Pharmacology 71, no. 2 (2013): 419–429, 10.1007/s00280-012-2029-1. [DOI] [PubMed] [Google Scholar]
- 9. Tomimaru Y., Eguchi H., Shimomura Y., et al., “Standard Versus Delayed Initiation of S‐1 Adjuvant Chemotherapy After Surgery for Pancreatic Cancer: A Secondary Analysis of a Nationwide Cohort by the Japan Pancreas Society,” Journal of Gastroenterology 58, no. 8 (2023): 790–799, 10.1007/s00535-023-01988-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Xia B. T., Ahmad S. A., Al Humaidi A. H., et al., “Time to Initiation of Adjuvant Chemotherapy in Pancreas Cancer: A Multi‐Institutional Experience,” Annals of Surgical Oncology 24, no. 9 (2017): 2770–2776, 10.1245/s10434-017-5918-z. [DOI] [PubMed] [Google Scholar]
- 11. Des Guetz G., Nicolas P., Perret G. Y., Morere J. F., and Uzzan B., “Does Delaying Adjuvant Chemotherapy After Curative Surgery for Colorectal Cancer Impair Survival? A Meta‐Analysis,” European Journal of Cancer 46, no. 6 (2010): 1049–1055, 10.1016/j.ejca.2010.01.020. [DOI] [PubMed] [Google Scholar]
- 12. Yang Y., Lu Y., Tan H., et al., “The Optimal Time of Starting Adjuvant Chemotherapy After Curative Surgery in Patients With Colorectal Cancer,” BMC Cancer 23, no. 1 (2023): 422, 10.1186/s12885-023-10863-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Sakuramoto S., Sasako M., Yamaguchi T., et al., “Adjuvant Chemotherapy for Gastric Cancer With S‐1, an Oral Fluoropyrimidine,” New England Journal of Medicine 357, no. 18 (2007): 1810–1820, 10.1056/NEJMoa072252. [DOI] [PubMed] [Google Scholar]
- 14. Yamamoto M., Sakaguchi Y., Kinjo N., et al., “S‐1 Adjuvant Chemotherapy Earlier After Surgery Clinically Correlates With Prognostic Factors for Advanced Gastric Cancer,” Annals of Surgical Oncology 23, no. 2 (2016): 546–551, 10.1245/s10434-015-4868-6. [DOI] [PubMed] [Google Scholar]
- 15. Meyer‐Wilmes P., Kennes L. N., Ignatov A., et al., “Earlier Is Not Always Better: Optimal Time to Initiate Adjuvant Chemotherapy After Surgery for Ovarian Cancer,” Archives of Gynecology and Obstetrics 312, no. 4 (2025): 1163–1173, 10.1007/s00404-025-08095-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Greene F. L., “TNM Staging for Malignancies of the Digestive Tract: 2003 Changes and Beyond,” Seminars in Surgical Oncology 21, no. 1 (2003): 23–29, 10.1002/ssu.10018. [DOI] [PubMed] [Google Scholar]
- 17. Desborough J. P., “The Stress Response to Trauma and Surgery,” British Journal of Anaesthesia 85, no. 1 (2000): 109–117, 10.1093/bja/85.1.109. [DOI] [PubMed] [Google Scholar]
- 18. Tohme S., Simmons R. L., and Tsung A., “Surgery for Cancer: A Trigger for Metastases,” Cancer Research 77, no. 7 (2017): 1548–1552, 10.1158/0008-5472.Can-16-1536. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. van der Bij G. J., Oosterling S. J., Beelen R. H., Meijer S., Coffey J. C., and van Egmond M., “The Perioperative Period Is an Underutilized Window of Therapeutic Opportunity in Patients With Colorectal Cancer,” Annals of Surgery 249, no. 5 (2009): 727–734, 10.1097/SLA.0b013e3181a3ddbd. [DOI] [PubMed] [Google Scholar]
- 20. Bozzetti F., “Nutritional Support of the Oncology Patient,” Critical Reviews in Oncology/Hematology 87, no. 2 (2013): 172–200, 10.1016/j.critrevonc.2013.03.006. [DOI] [PubMed] [Google Scholar]
- 21. Weimann A., Braga M., Carli F., et al., “ESPEN Practical Guideline: Clinical Nutrition in Surgery,” Clinical Nutrition 40, no. 7 (2021): 4745–4761, 10.1016/j.clnu.2021.03.031. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1: Kaplan–Meier curves of overall survival (OS) (a) and relapse‐free survival (RFS) (b) among patients who underwent pancreaticoduodenectomy (PD) according to the timing of adjuvant S‐1 chemotherapy initiation: Early (< 6 weeks, n = 25), Standard (6–8 weeks, n = 45), and Delayed (> 8 weeks, n = 51).
Figure S2: Kaplan–Meier curves of overall survival (OS) (a) and relapse‐free survival (RFS) (b) among patients who underwent distal pancreatectomy (DP) according to the timing of adjuvant S‐1 chemotherapy initiation: Early (< 6 weeks, n = 20), Standard (6–8 weeks, n = 25), and Delayed (> 8 weeks, n = 21).
Figure S3: Conceptual diagram illustrating the potential sources of bias in the analysis of the timing of S‐1 initiation and overall survival.
Table S1: Reasons for discontinuation of S‐1 chemotherapy among patients who discontinued treatment according to timing group.
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.