Postoperative Adjuvant Chemotherapy Cancel Out the Negative Survival Impact on Stage II/III Gastric Cancer Patients With Postoperative Complications


 Purpose The potential additive influence of adjuvant chemotherapy (AC) on prognosis of patients with stage II/III gastric cancer (GC) who experienced complications after radical surgery is unclear.Methods The whole group was divided into a postoperative complication (PC) group and a postoperative non-complication (NPC) group, and the overall survival (OS) rate, recurrence-free survival (RFS) rate and recurrence rate were compared between the two groups of patients. Results A total of 1563 patients between January 2010 and December 2015 in our center were included in this analysis. There were 268 patients (17.14%) in the PC group and 1295 patients (82.86%) in the NPC group. The 5-year OS rate of the PC group was 55.2%, the NPC group was 63.3%; and the 5-year RFS rate of the PC group was 53.7%, the non-PC group was 58.8%. Recurrence patterns showed no significant difference between the two group (all p>0.05). Adjuvant chemotherapy (AC) significantly improved the OS and RFS rates of patients with and without PCs (both p<0.05), and it showed no significant difference between the PC group and the NPC group who received AC (both p> 0.05). Stratified analysis showed that AC only improve the OS or RFS rates of stage III patients (both p<0.05). Further stratified analysis of the time interval (TI) from operation to initiation of AC in the PC group showed that a TI after 6 weeks (≥6eeks) improved only the OS and RFS rates of stage III patients, while when a TI within 6 weeks (＜6weeks), a benefit was observed in stage II and III patients (both p<0.05).Conclusion AC can abolish the negative effect of PCs on the long-term survival of patients with stage III GC; for stage II patients, the above offset effect is affected by the TI. Delaying AC initiation after 6 weeks may not improve the survival of patients experienced stage II GC with complications.


Introduction
Gastric cancer (GC) is the fth most common malignant tumor in humans and ranks third in tumorrelated mortality [1]. Surgery is the main treatment for GC, but the incidence of complications after radical gastrectomy plus D2 lymph node dissection remains high, at approximately 15-50% [2][3][4][5]. Previous literature has shown that postoperative complications signi cantly reduce the overall survival rate of patients with malignant tumors and are closely related to the increased risk of disease recurrence [6][7][8][9][10][11][12].
Scholars such as Sierzega M, Bohle B and others believe that complications prolong the postoperative in ammation time of patients and then aggravate in ammatory factors to stimulate the proliferation and growth of circulating cancer cells, which may lead to a poor long-term prognosis of patients after surgery [10,13].
In East Asia, adjuvant chemotherapy (AC) for patients with stage II/III GC after D2 radical resection has become the standard treatment plan [11][12][13][14]. However, the occurrence of postoperative complications may prolong the time to chemotherapy initiation for patients with GC, and whether the delayed initiation of AC could affect the survival outcome of these patients remains unclear. In recent years, studies have shown that although some patients experience complications after colorectal cancer resection, the initiation of AC treatment "as soon as possible" reduces the increased risk of tumor recurrence in these patients due to complications and improves overall survival [15 ,16]. However, related studies in patients with GC have not been reported. This study aimed to evaluate whether AC can abolish the poor prognosis of patients who undergo radical GC surgery caused by PCs and whether the time interval(TI)from radical operation to initiation of postoperative adjuvant chemotherapy for gastric cancer patients has an impact on this abolishment. The aim of this study was to explore the effect of AC on the prognosis of GC patients with postoperative complications.

Patients
The clinicopathological data of 2476 patients who underwent radical GC surgery between January 2010 and December 2015 in the Department of Gastric Surgery, Union Hospital of Fujian Medical University, were retrospectively analyzed. The exclusion criteria were as follows: 1. Age <20 years or >80 years (n=67); 2. Neoadjuvant chemotherapy (n=63); 3. Remnant stomach cancer (n=59); 4. The clinicopathological data were not complete (n=36); and 5. pStage I (n=688). Ultimately, a total of 1563 patients were enrolled (Table 1), of whom 268 were in the postoperative complication (PC) group, and 1295 were in the postoperative non-complication (NPC) group (see the ow chart in supplement material eFig. 1 for details). And our patients never experienced radiotherapy. Both preoperative clinical staging and postoperative pathological staging were carried out according to the 8th edition of the AJCC-TNM staging system [17]. All operations were performed by the same experienced surgical team. All patients signed an informed consent form. This study was approved by the FMUUH Ethics Committee.

Statistical Analysis
Continuous variables are reported as the means ± SDs. Categorical and continuous variables were compared using a χ2 test or Fisher's exact test and a t test, respectively. The cumulative survival rate was calculated using the Kaplan-Meier method and a log-rank test. A Cox proportional hazards regression model was used to determine the independent prognostic factors associated with OS and RFS. Variables with a p value <0.05 in the univariate analysis were subsequently entered into the multivariate Cox proportional hazards analysis. All statistical analyses were performed using SPSS v.26.0 for Windows (SPSS Inc., Chicago, IL, USA). P values less than 0.05 were considered statistically signi cant.

De nitions
Postoperative complication (PC) group: According to previous studies, we recorded all complications that occurred in all patients within one month after surgery and used the CD classi cation [18] to classify the severity of postoperative complications. Patients with grade II and above complications (≥ grade II CD) were de ned as the PC group. Patients without complications and complication events below grade II were de ned as the NPC group [19].
Adjuvant chemotherapy (AC): Adjuvant chemotherapy was recommended for patients with pathological stages II and III [20]. The patient has received at least one cycle of postoperative adjuvant chemotherapy. Based on 5-uorouracil treatment, Most patients received a chemotherapy regimen based on S-1 plus platinum 49.5% .Those who never received any adjuvant chemotherapy were de ned as nonchemotherapy (NAC) patients.
The time interval (TI)from radical operation to initiation of postoperative adjuvant chemotherapy for gastric cancer patients was de ned as the time from the date of surgery to the initiation of postoperative adjuvant chemotherapy. According to tumor-related diagnoses and treatment guidelines, the rst postoperative chemotherapy time for GC patients is recommended to be within 6 weeks [21]. Therefore, in this study, 6 weeks was used as the observation cutoff point. Patients were divided into a TI within 6 weeks (<6eeks) group, a TI after 6 weeks (≥6weeks) group and a never received postoperative adjuvant chemotherapy (NAC) group.
Overall survival (OS): OS was de ned as the time from the date of surgery to death or the last follow-up.
All patients were followed up until death or at least 60 months, with a median follow-up time of 60 months (4-124 months).
Recurrence: Recurrence was con rmed by an imaging examination or biopsy of suspicious lesions.
Recurrence-free survival (RFS) was de ned as the period from the rst day after surgery to the discovery of recurrence. For RFS, if the deceased patient did not have tumor recurrence, the last follow-up record was reviewed. The recurrence patterns were classi ed as local recurrence (LR) (anastomotic or remnant stomach), peritoneal metastasis (PM) (including positive ascites in tumor cells, progressive thickening of the peritoneum found on imaging, newly emerging peritoneal lymphadenopathy, or tumor invasion into the female uterus or ovaries) and distant metastasis (DM) (including lymph nodes outside the D2 range and liver, lung, bone and other organs) [22]. Table 1 shows a comparison of the general clinicopathological data of the PC group (268 patients) and the NPC group (1295 patients). Compared with the NPC group, the PC group had older patients, a longer operation time and more blood loss during the operation (p<0.05). However, there was no signi cant difference in pathological staging between the two groups (p>0.05). The median times to recurrence in the two groups were 17.6 months and 15.8 months. in the PC group, the median time to local recurrence was 55 months, that to peritoneal metastasis was 24 months, and that to distant metastasis was 14.7 months; in the NPC group, the median time to local recurrence was 50 months, that to peritoneal metastasis was 17.7 months, and that to distant metastasis was 16.6 months; There was no signi cant difference in the location of recurrence or time to recurrence between the two groups (p>0.05) (Supplement material eTable 2). According to the strati ed analysis, there was no signi cant difference in the location of recurrence or time to recurrence between stage II and stage III patients (p>0.05) (Fig. 1B, Fig. 1C; Supplement material eTable 3, eTable 4).

The effect of adjuvant chemotherapy (AC)
According to whether adjuvant chemotherapy (AC) was received, a strati ed analysis of the PC group and NPC group showed that AC signi cantly improved the 5-year OS and RFS rates of patients with and without PCs (both p<0.05). In addition, the 5-year OS and RFS rates of patients in the PC group who received AC were signi cantly higher than those of patients without adjuvant chemotherapy(NAC) group (both p<0.05), and there was no signi cant difference in the 5-year OS and RFS rates between the PC group and NPC group when they were given AC. (both p >0.05) (Fig. 3A, Fig. 3B).
Further strati ed analysis based on pathological staging showed that in stage II patients with PCs (Fig.   3C, Fig. 3D), AC did not signi cantly improve 5-year OS or RFS (both p>0.05). In stage III patients (Fig. 3E,  Fig. 3F), the 5-year OS and RFS rates of the PC group who received AC were signi cantly higher than NAC group (both p<0.05).
6. The effect of the time interval (TI) from radical operation to initiation of adjuvant chemotherapy eFig. 2A(Supplement material eFig. 2A) shows the distribution of the TI. The median TI of patients with PCs was 36 days. The 5-year OS and RFS rates of patients in the TI≥6weeks group and TI 6 weeks group were signi cantly better than those of patients in the NAC group (both p<0.05) (Fig. 3G, Fig. 3H).
The strati ed analysis of pathological stage showed that the median TI of stage II and stage III patients with PCs were 35 days and 36 days, respectively (Supplement material eFig. 2B, eFig. 2C). Among the stage II patients, only the 5-year OS (p=0.038) and RFS (p=0.005) rates of patients with PCs in the TI 6 weeks group were signi cantly higher than NAC group, while a TI≥6 weeks did not signi cantly improve the 5-year OS and RFS rates of stage II PC group (Fig. 3I, Fig. 3J). Among stage III patients, the 5-year OS and RFS rates of those with PCs in a TI≥6 weeks group and a TI 6 weeks group were signi cantly higher than NAC group (both p<0.05) (Fig. 3K, Fig. 3L).

Discussion
In recent years, an increasing number of scholars have reported that postoperative complications (PCs) after surgical operations are closely related to the poor prognosis of patients with malignant tumors [23][24][25][26][27]. This retrospective study analyzed this adverse effect and explored whether adjuvant chemotherapy (AC) and the time interval (TI) from radical operation to initiation of adjuvant chemotherapy could abolish the poor prognosis caused by PCs in patients with stage II/III GC. We found that compared with the NPC group, the long-term survival rate of the PC group was lower; AC signi cantly improved the OS and RFS rates of the NPC group and the PC group. Further strati ed analysis of the TI in patients with PCs showed that in stage III patients with PCs, a TI 6 weeks and a TI≥6 weeks could reduce the risk of disease recurrence and improve overall survival; however, in PC-II patients, only a TI 6 weeks could signi cantly improve 5-year OS (p=0.038) and RFS (p=0.005).
At present, surgical resection remains the primary method for the treatment of GC, but the occurrence of complications after gastrectomy cannot be completely avoided. Previous reports on the mechanism of postoperative complications negatively affecting the prognosis of patients with malignant tumors have been inconsistent. JH Saunders et al. found that postoperative anastomotic leakage did not affect the long-term survival of patients with esophageal cancer, but other serious complications (except for anastomotic leakage) were identi ed as independent prognostic factors for overall survival and diseasefree survival [25]. In the eld of GC, Masanori et al. found in the JCOG1001 study that, regardless of the type or degree of complications after radical GC, postoperative complications were associated with poor OS [26]. Kubota et al. analyzed 1395 patients with GC who underwent radical resection and found that compared with patients without complications, the 5-year survival rate of patients with postoperative complications was signi cantly lower, and they believed that this nding may have been related to complications. Complications are related to a prolonged postoperative in ammation time [12]. In addition, studies by Katai H, Okholm C and others [28][29][30] have shown that postoperative complications may actually alter the progression of the malignant tumor itself. In ammatory mediators and cellular effects are important parts of the local tumor environment. PC-induced systemic in ammatory responses through in ammatory mediators [31,32], such as interleukin-6 [33,34], nuclear factor-kappa B [35] and Creactive protein[36], stimulate residual microbes after surgery and the proliferation of metastatic tumor cells.
AC is usually administered to remove residual cancer cells and micro-metastasis after surgery, and it has been proven that AC after GC resection can bene t patient survival [37][38][39]. However, there is no conclusive conclusion about the time at which patients with PCs should start postoperative adjuvant chemotherapy. In recent studies on PCs in patients undergoing GC resection, it has been shown that patients with PCs and delayed AC initiation have reduced survival rates [39][40][41][42]. It has also been reported that the delayed initiation of AC is associated with poor survival outcomes for patients with colorectal cancer, ovarian cancer, and breast cancer [43][44][45]. In the eld of GC, a meta-analysis of 6017 patients showed that every 4-week delay in postoperative adjuvant chemotherapy initiation reduced the relative OS rate by 14% [46]. In this study, we found that in stage II patients with PCs, when the TI was advanced to less than 6 weeks, the 5-year OS and RFS rates were signi cantly improved. For stage III patients with PCs, chemotherapy can be started within 6 weeks or after 6 weeks and signi cantly improve 5-year OS and RFS. This may be because in ammatory mediators and cell effect factors are an important part of the local tumor environment, and the in ammatory state caused by PCs increases the risk of blood recurrence in advanced GC [47], which is related to poor long-term survival. The tumor burden of stage II patients is low, and the number of circulating cancer cells may be small. If the body can tolerate the condition, AC treatment should be given as soon as possible in the early stage of micro-metastasis ( 6 weeks), as it helps to inhibit tumor cell proliferation, reduce the risk of recurrence and improve overall survival.
This study has the following shortcomings: 1. This was a single-center retrospective study that lacked external veri cation, and the sample size was small. A large-sample multicenter study is needed to further con rm the conclusions of this study. 2. In this study, we did not evaluate whether the course of chemotherapy affects the long-term prognosis of patients. 3. The effect of chemotherapy on prognosis was not studied. Nevertheless, to the best of our knowledge, this study was the rst to investigate the effect of AC on the long-term prognosis of stage II/III GC patients with postoperative complications, and it was the rst to report the prognostic effect of the TI on GC patients with postoperative complications. This study also provides a preliminary basis for future multicenter or prospective studies.

Conclusion
This study showed that PCs can signi cantly reduce the 5-year OS and RFS rates of patients with GC.
Postoperative adjuvant chemotherapy can offset the negative impacts of PCs on the long-term survival of patients with stage III GC; for stage II patients, the above offset effect is affected by the TI, and AC may not improve the survival of stage II patients with PCs in whom postoperative adjuvant chemotherapy is initiated ≥6 weeks. However, the ndings of this study still need to be further veri ed via external data, especially Western data.

Declarations Acknowledgments
Human rights statement and informed consent All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1964 and later versions. Informed consent or substitute for it was obtained from all patients for being included in the study.

Consent for publication
This article doesn't report an individual participant's data in any form.

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Author Contributions: Drs Shen LL and Lu J contributed equally to this work and should be considered rst coauthors.
Drs Zheng CH and Huang CM had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.