Abstract

Background:

Duodenal adenocarcinoma (DA) is a rare malignancy, and the effectiveness of adjuvant chemotherapy on survival after surgical intervention remains ambiguous. This study investigates the impact of adjuvant chemotherapy on overall survival (OS) and recurrence-free survival (RFS) in patients with resectable DA.

Materials and methods:

Data from ninety-eight stage I-III DA patients who underwent surgical resection between January 1998 and December 2021 at two Chinese institutions were analyzed retrospectively. Survival outcomes were assessed using the Kaplan-Meier method, and multivariable Cox proportional hazards models identified significant prognostic factors.

Results:

Of the 98 patients, 45 received adjuvant chemotherapy and 53 did not. The median follow-up was 49 months. Patients who had adjuvant chemotherapy showed a longer median OS (52.9 vs. 25.1 months, p=0.003) and RFS (38.2 vs. 9.9 months, p<0.001) than those without. N stage was associated with reduced OS and RFS. Multivariate analysis identified adjuvant chemotherapy, N stage, and CA199 as independent predictors for RFS and OS. For patients with initial CA199 ≥ 27U/mL, adjuvant chemotherapy appeared to be associated with improvements in RFS and OS, alongside suggestive reductions in recurrence and mortality risks.

Conclusion:

Adjuvant chemotherapy correlates with an extended median OS and RFS, as well as a reduced risk of recurrence and mortality in patients who have undergone surgical resection for DA. Specifically, patients with DA who are in the N1 or N2 stage, or those with a CA199≥27U/mL at their initial diagnosis, may show potential clinical benefits from adjuvant chemotherapy as a preliminary exploratory finding.

1 Introduction

Small bowel adenocarcinoma (SBA), a rare gastrointestinal malignancy, accounts for less than 5% of all malignant digestive system tumors. In 2022, approximately 11,790 new SBA cases were reported in the United States, resulting in 1,960 deaths (1). Duodenal adenocarcinoma (DA) is the predominant subtype of SBA, comprising over 50% of SBA cases despite representing less than 0.5% of all gastrointestinal malignancies (2).

The main treatment for local (i.e., stages I–III) SBA is surgical resection, including en-bloc lymph node removal. Treatment for DA varies based on tumor location, typically involving pancreaticoduodenectomy for proximal tumors and segmental resection for distal tumors. In cases of unresectable SBA, neoadjuvant therapy incorporating fluoropyrimidine or platinum-based chemotherapy, sometimes in combination with radiotherapy, is recommended (3).

The prognosis is poor, with a 5-year overall survival(OS) rate of 14–33% for all patients and 40–60% for those who are curatively resectable (1, 4). A significant proportion of patients undergoing curative resection experience recurrence (5), underscoring the need for effective adjuvant strategies to enhance OS and reduce recurrence rates. Adjuvant chemotherapy, particularly regimens such as FOLFOX (folinic acid, fluorouracil, and oxaliplatin), CAPEOX (capecitabine and oxaliplatin), 5-FU/LV (5-fluorouracil and leucovorin), or capecitabine alone (3), in SBA remains controversial due to the mixed results from retrospective studies, meta-analyses, and the scarcity of randomized controlled trials (RCTs). While some studies indicate the benefits of adjuvant therapy (chemotherapy or chemoradiotherapy) (6, 7), others show no significant impact or equivocal results (8–11). Notably, most research has not specifically addressed DA, nor has it substantially focused on the response of the Chinese population to these adjuvant therapies.

Given this context, the principal aim of our study is to conduct a focused assessment of the effects of adjuvant chemotherapy on survival outcomes in Chinese patients with resectable DA. Additionally, this study seeks to identify the specific patient cohort that is most likely to benefit from adjuvant chemotherapy, thereby contributing valuable insights to current clinical practices.

2 Materials and methods2.1 Data source and patient selection

Patients diagnosed with duodenal adenocarcinoma (DA) who underwent curative resection at the Department of Medical Oncology at Chaoyang Sanhuan Cancer Hospital and the National Clinical Research Center for Cancer were enrolled between January 1998 and October 2021.The main inclusion criteria were as follows: 1)Age > 18 years;2)Pathologically confirmed diagnosis of DA;3)No evidence of distant metastasis;4)Complete R0 resection with regional lymph node dissection;5)No evidence of other concurrent malignancies;6)No adjuvant chemotherapy or completion of at least four cycles of adjuvant chemotherapy;7)Adjuvant chemotherapy initiated within 3–4 weeks after surgery;8)No severe postoperative complications that delayed adjuvant chemotherapy beyond 4 weeks. Patients were excluded if they had incomplete prognostic data, unavailable medical records, R1 resection, or severe postoperative complications precluding timely adjuvant chemotherapy. The study was approved by the ethics committee of Beijing Chaoyang District SanHuan Cancer Hospital(SH-2023020, 2023.7.17), and informed consent was waived due to the retrospective nature of the study.

2.2 Data variables

A comprehensive review of patient medical records was conducted to gather demographic data, tumor characteristics, and survival outcomes. The analysis included demographic variables such as age at diagnosis, gender, and the ECOG performance status. In terms of tumor features, TNM stage(version8), the histological grade, the T stage, and the N stage and tumor location in the duodenum (D1: bulb, D2: descending, D3: horizontal, D4: ascending) were recorded. Surgical interventions were analyzed, detailing the number of lymph nodes examined. The presence of perineural, lymphatic, and vascular invasion was also recorded. The use of adjuvant chemotherapy was documented, noting the specific regimens used, predominantly oxaliplatin combined with fluoropyrimidine or other chemotherapeutic agents. Furthermore, CEA was measured in ng/L with normal values below 4.7 ng/L, and CA199 in U/mL with normal values below 27 U/mL at the time of initial diagnosis. This normal reference range for CA199 (≤27 U/L) is the standardized clinical testing criterion of the participating institutions, consistent with the chemiluminescent immunoassay platform used for digestive system malignancies during the study period.

2.3 Treatment and assessment

Patients in this study were categorized into two groups based on adjuvant chemotherapy status: those who underwent the treatment and those who did not. The primary adjuvant chemotherapy regimens administered were as follows: 1) modified FOLFOX6: fluorouracil 2400mg/m2 continuous infusion following a 400mg/m2 or 600mg/m2 IV bolus, leucovorin 400 mg/m2, and oxaliplatin 85 mg/m2 every two weeks. 2) capecitabine and oxaliplatin: capecitabine at 1,000 mg/m2 twice daily from day 1 to 14, repeated every three weeks, oxaliplatin 130 mg/m2 every three weeks. 3) S-1 and oxaliplatin: S-1S-1 at 80 mg/m2 from day 1 to 14, repeated every three weeks, oxaliplatin at 85 mg/m2 every two weeks. During adjuvant therapy, treatment efficacy was evaluated every four cycles. After the completion of therapy, assessments were conducted every 3 to 6 months for the first two years, and annually thereafter. Patients experiencing any discomfort could undergo additional evaluations at any time. These follow-up visits included a physical examination and routine imaging assessments, such as chest, abdominal, and pelvic CT scans to check for signs of health deterioration, recurrence or metastasis. RFS was defined as the time from surgery to recurrence (local/distant), death from any cause, or last imaging follow-up (whichever occurred first). OS was calculated from resection to death or last confirmed survival status. We applied standard right-censoring in Kaplan-Meier analyses. For RFS: Patients without recurrence/metastasis and alive were censored at the last imaging confirmation date. For OS: Patients alive at data cutoff were censored at the last follow-up contact (e.g., clinic visit or telephone confirmation).

2.4 Statistical analysis

Patients were stratified based on whether they received adjuvant chemotherapy, and baseline characteristics were presented using descriptive statistics, with categorical variables summarized by frequency counts. Missing data were non-randomly distributed due to incomplete retrospective clinical records. Regression imputation was used to handle missing values for key clinicopathological variables. Comparisons between categorical groups were conducted using the Chi−square test. Fisher’s exact test was used when the expected frequency was less than 1. Continuous variables were compared using the Wilcoxon rank-sum test where appropriate. Survival curves were visualized and estimated using the Kaplan-Meier method, with differences between the curves assessed using log-rank tests. Cox proportional hazards regression models were applied to assess the association between patient characteristics and time-to-event outcomes while controlling for covariates. A backward elimination technique was used to build multivariable models for certain outcomes in which factors that showed univariate associations at P < 0.1 were passed to a second stage for multivariable modeling. In the second stage, all variables were placed in the model and the least significant factors were removed until only factors with P <0.05 were left in the model. Stratified and subgroup analyses were carried out on the variables from the multivariate analysis to pinpoint the groups that benefited most from adjuvant chemotherapy. For all analyses, a P value of < 0.05 was considered statistically significant. All statistical evaluations were executed using SPSS 27.0 (SPSS Inc., Chicago, IL, USA).

3 Results3.1 Patient characteristics

From January 1998 to December 2021, we analyzed 98 DA patients: 61 males and 37 females, aged 35-75 years (median age 55). Tumor location was distributed as D1 (n=6), D2 (n=70), D3 (n=8), and D4 (n=14). A total of 61 patients underwent pancreaticoduodenectomy (PD), while 28 patients had segmental resection (SR). Additionally, there were 9 cases where the surgical approach was not clearly documented. 45 received surgical resection and adjuvant chemotherapy, while 53 had surgery alone. Staging (AJCC 8th edition) showed 8 at stage I, 5 at IIA, 38 at IIB, 18 at IIIA, and 17 at IIIB. Most (75) were T3-T4, and 10 were T1-2. Nodal involvement was N0 in 46, N1 in 18, and N2 in 17. Of 88 patients with histological grading, 10 had well-differentiated, 50 moderately differentiated, and 28 poorly differentiated tumors. Guidelines recommend examining at least 8 regional lymph nodes; 50 patients met this, and 29 did not. Elevated CEA was found in 34 patients, and CA199 in 42. Adjuvant chemotherapy included fluoropyrimidines plus oxaliplatin in 33 patients, with 12 receiving other agents. Baseline characteristics were comparable between groups (P > 0.05) (Table 1). Consistent with our study’s enrollment criteria (which excluded patients with severe postoperative complications precluding timely adjuvant chemotherapy), no in-hospital or 30-day postoperative mortality was observed in the cohort, and no significant postoperative morbidity was documented among all included patients.

CharacteristicsStratificationAll patientsAdjuvant chemotherapyNo chemotherapyp-valuen=98n=45n=53Age(year)Median(range)55(41-73)54(34-75)0.91<4515870.4445-65632934>6520812Gendermale6131300.21female371423ECOG02411130.521331815NA571625TNM stage18260.112A5232B3815233A181353B1789NA1257T stageT13120.66T2743T31486T4612734NA1358N stageN04619270.42N118126N21789NA17611HistologyWD103100.35MD502723PD281315NA1028Lymphatic/vascular invasionYES16880.77NO743440NA835Perineural invasionYES11560.93NO793742NA835Lymph nodes examined<82911180.17≥8502723NA19712CA199(U/mL)>274221210.92≤27432221NA13211CEA(ng/L)>4.73417170.93≤4.7532726NA11110Tomor locationD16330.01D2702446D3880D414104Surgery procedurewhipple6121400.01Segmental28199NA945

Baseline characteristics of DA patients - adjuvant chemotherapy vs. no chemotherapy.

3.2 Adjuvant chemotherapy for DA

The median follow-up was 49 months (range 15-80). Of the cohort, 45 patients (45.92%) received adjuvant chemotherapy. The most common regimens were modified FOLFOX6 (n=12), capecitabine plus oxaliplatin (n=11), and S-1 plus oxaliplatin (n=7). Other regimens included various combinations of fluorouracil, gemcitabine, oxaliplatin, and other agents (n=15). Kaplan-Meier analysis showed no significant differences in median OS (36.5 vs. 55.9 months, p=0.86) or RFS (38.2 vs. 39.4 months, p=0.60) between oxaliplatin and fluoropyrimidine -based regimens and other chemotherapy regimens.

3.3 The univariate and multivariate COX analysis for RFS

During a median follow-up of 49 months (range 15-80), 70.4% of patients (69 individuals) experienced recurrence or metastasis, with an average RFS of 13.0 months. Univariate Cox regression showed that patients receiving adjuvant chemotherapy had longer RFS than those who did not (38.2 vs. 9.9 months, p<0.001) (Figure 1). N stage (N0, N1, N2) was significantly associated with worse RFS (39.4 vs. 13.1vs. 7.1 months, p=0.01) (Figure 2), as were elevated preoperative CA19-9 levels (38.2 vs. 13.1 months, p=0.01) (Figure 3) and TNM stage (p=0.05). Other factors like age, gender, ECOG, histological grade, T stage, perineural invasion, and preoperative CEA were not significantly associated with RFS (p>0.5).

Kaplan-Meier curves for RFS and OS according to adjuvant chemotherapy status.

Comparison of RFS and OS by the status of N stage.

Comparison of RFS by the status of preoperative CA199 levels.

In the multivariate Cox model, adjuvant chemotherapy was linked to longer RFS (HR = 0.24, 95% CI: 0.12-0.46, p<0.001), while N stage (HR = 1.93, 95% CI: 1.33-2.79, p<0.001) and elevated CA19-9 levels (HR = 1.98, 95% CI: 1.06-3.71, p=0.03) were associated with worse RFS. Age had no significant impact on RFS (HR = 1.35, 95% CI: 0.82-2.23, p=0.24).

3.4 The univariate and multivariate COX analysis for OS

The median OS was 24.9 months. Univariate Cox analyses showed that patients receiving adjuvant chemotherapy had significantly longer OS than those who did not (52.9 vs. 25.1 months, p=0.003) (Figure 1). N stage (N0, N1, N2) was significantly associated with poorer OS (48.4 vs. 23.2vs. 20.4months, p=0.01) (Figure 2), and perineural invasion also correlated with reduced OS (27.1 vs. 39.7 months, p=0.01) (Figure 4). Other factors, such as age, gender, ECOG, histological grade, TNM stage, T stage, lymphatic/vascular invasion, number of lymph nodes examined, tumor location (D1–D4), surgical approach (Whipple vs. isolated duodenal resection), and elevated preoperative CA19-9 and CEA levels, were not significantly associated with OS (p>0.5) (Table 2).

Comparison of OS by the status of perineural invasion.

CharacteristicPFSOSmPFS95%CIp-valuemOS95%CIp-valueAge0.060.13 <4514.210.6-17.823.419.1-27.7 45-6522.26.3-38.140.027.5-52.4 >6511.23.6-18.827.113.7-40.5Gender0.510.50 male16.510.6-22.430.723.9-37.5 remale17.74.7-30.736.420.3-52.6ECOG0.730.18012.29.8-14.623.218.4-27.9112.15.1-19.131.428.6-34.2TNM stage0.0490.1 I18.914.3-23.531.830.9-32.6 IIA29.314.1-44.539.718.5-60.9 IIB28.30.0-61.236.519.1-53.8 IIIA17.78.9-26.531.423.0-39.7 IIIB7.10.0-15.019.316.5-22.0Histology grade0.310.68 Well differentiated12.20.0-27.325.99.4-42.4 Moderately differentiated19.20.7-37.731.416.3-46.5 Poorly differentiated9.66.3-12.928.712.1-45.3T Stage0.740.42 T117.0–60.1– T212.21.4-23.029.820.8-38.8 T317.73.1-32.345.224.9-65.4 T412.110.5-13.727.116.7-37.5N Stage0.010.01 N039.421.8-57.0