Primary adenocarcinoma of the duodenum is an uncommon disease, representing less than 0.5% of gastrointestinal malignancies and approximately 45% of small intestine adenocarcinomas.1 Although nodal metastases, margins of resection, and depth of invasion are of prognostic importance in many malignant neoplasms, these factors have not consistently been shown to be of prognostic significance for patients with duodenal adenocarcinoma. The prognosis for patients with curatively resected duodenal adenocarcinoma is relatively better (5-year survival, 60%)2 than for patients with ampullary carcinoma (46%),3 distal bile duct (27%),4 or cancer in the head of pancreas (10%).5 The development of radiologic and endoscopic examination has led to an increased identification of earlier stages of duodenal cancer, but the oncogenic characteristics of primary duodenal cancer remain poorly understood. We herein demonstrate the outcomes of 21 patients with primary duodenal adenocarcinoma who were analyzed to clarify the predictive factors of patient survival.

Methods and patients

Results

Comparison of clinicopathologic features

The mean age of the patients at the time of admission was 61.19 years of age, and 11 male subjects and 10 female subjects were evaluated in the present study. The initial patient symptoms included epigastralgia, appetite loss, anemia, jaundice, hematemesis, and vomiting, as well as others (Table 1).

Table 1 Patients and treatment characteristics of 21 patients with duodenal adenocarcinoma

View Fullscreen

Treatment procedures for duodenal adenocarcinoma

Surgical procedures were used to examine the invasiveness or metastasis of duodenal adenocarcinoma. Pancreaticoduodenectomy (PD) and pylorus-preserving PD (PpPD) were performed for patients with advanced carcinoma, and local resection was carried out in patients with less invasive cancers. Endoscopic mucosal resection (EMR) was performed for early-stage cancers that remained within the mucosal layer without lymph node (LN) metastasis. Twelve (57.1%) patients underwent curative resection (R0): 4 PDs, 4 PpPDs, 2 local resections of the duodenum, and 2 EMRs. A curative resection included T3/T4 (50%) and N2-positive tumors (16.7%). One portal vein reconstruction and one local colectomy were included in PD or PpPD. Palliative surgery was performed in 9 patients (42.9%), including a gastro-intestinal bypass for advanced cancers with infiltration and distant metastasis. From the pathologic diagnoses, all curative resections were found to be R0 (Table 1).

From the aspect of curative or palliative surgery, the results of the clinicopathologic features between the 2 groups are shown in Table 2. The tumor size, depth, LN metastasis and distant metastasis were found to be statistically significant between the curative resection group and palliation surgery group (P  =  0.024, P  =  0.019, P  =  0.0186, P < 0.0001, respectively, using Fisher exact probability test). The 12 curative resections indicated 8 well-to-moderately differentiated adenocarcinomas, but 4 palliative surgery patients had tumors that could not be identified according to the tissue differentiation type.

Table 2 The comparison of tumor characteristics between patients who underwent curative resection and those who had palliative surgery

View Fullscreen

Cause-specific survival curves

The survival curves for the curative and palliative surgery groups are shown in Fig. 1. The median cause-specific survival times calculated using the Kaplan-Meier method were 1784 days (range 160–8650 days) in the 12 patients of the curative surgery group and 261 days (range 27–857 days) in the 9 patients of the palliative surgery group. The curative surgery group exhibited significantly better survival rates compared with the palliative surgery group (P  =  0.0003, log-rank test).

View Figure

• Download as Powerpoint
• Download Figure

The factors predictive of survival in 21 patients with primary duodenal adenocarcinoma

The tumor size (P  =  0.007), tumor depth (P  =  0.007), tumor differentiation (P  =  0.010), and LN metastasis (P  =  0.009) were found to significantly and negatively impact survival according to the univariate analysis (Table 3).

Table 3 Univariate analysis of the factors affecting the survival of 21 patients with primary duodenal adenocarcinoma

View Fullscreen

Factors affecting survival of 12 patients with curative surgery

PD, PpPD, partial resection of the duodenum, and EMR were performed in 4, 4, 2, and 2 patients, respectively. A total of 4 male subjects and 8 female subjects were examined. The factors affecting survival of curative surgery are indicated in Table 4. In this group, the gender, tumor location, tumor size, macroscopic appearance, tumor depth, pancreas invasion, lymphatic vessel invasion, blood vessel invasion, LN metastasis, and surgical procedure did not significantly impact survival. The age and tumor differentiation factors only impacted overall survival (P < 0.05, log-rank test).

Table 4 Univariate analysis of the factors affecting the survival of 12 patients with curative resections

View Fullscreen

Discussion

The duodenum is anatomically located in the retroperitoneum. This anatomic characteristic complicates the treatment for duodenal cancer and leads to infiltrate in the pancreatic parenchyma directly and through retroperitoneal spaces.7,8

The primary patient symptoms are vague and not specific (epigastralgia, 38.1%; appetite loss, 14.3%), and the upper gastro-intestinal (GI) examinations by radiologic and endoscopic procedures may be useful for the prompt examination and earlier diagnosis of early-stage duodenal adenocarcinoma.9

Complete resection remains the only option for curability, but it is not always feasible in patients with locally advanced tumors or when tumors have invaded the retroperitoneal spaces. PD is the only appropriate operation to achieve complete resection when a tumor is located in the second portion of the duodenum. In the present study, curative resection was performed for 12 patients with duodenal cancer, and 4 PD and 4 PpPD resections were performed without mortality.

Several authors have suggested that duodenal adenocarcinomas should be treated by PD1,10,11 because regional lymph nodes are most thoroughly extirpated by this method. Although it is difficult to perform a well-defined regional lymphadenectomy in the area owing to reduced duodenal mesentery, the removal of all periduodenal soft tissues removes an adequate number of lymph nodes in the majority of patients. For tumors located in the first, third, or fourth portions of the duodenum, a curative resection may be accomplished by removing the affected regional tissue. Several authors have reported excellent survival and considerably lowered postoperative mortality and morbidity following segmental resection.2,7 However, it has been argued that a segmental resection results in inadequate resection margins and incomplete regional lymphadenectomy.12 In the present study, 2 patients (9.5%) successfully underwent a local resection of the third portion of the duodenum, and 8 patients (38.1%) underwent PD/PpPD. In light of this potential limitation, PD/PpPD has been suggested as a standard approach to primary duodenal cancer, regardless of the primary site of the tumor.13

However, primary duodenal adenocarcinoma may be diagnosed at an advanced tumor stage, which results in a low local excision rate. Surgical palliation continues to play an important role in the management of periampullary carcinoma.14 In the present study, palliative surgery was performed in 9 patients (42.9%). Surgical palliation can be accomplished with an acceptable perioperative mortality and morbidity with excellent long-term results.

There is general agreement that patients with resectable duodenal adenocarcinomas achieve better survival times than those with unresectable tumors.1,15–18 In the present study, the resectability rate was 57.1%; other reported resectability rates ranged from 43% to 87%.1,2,8,13,15,17,19 The survival figures for patients who undergo resection also vary widely. Several recent series have reported 5-year survival rates ranging from 25% to 44.1% in patients who have undergone curative resection.2,8,15,17,19–21 In our study, the survival after curative surgery (R0) was good, with 3- and 5-year survival rates of 72.9% and 65.7%, respectively, which are consistent with the 5-year survival rates of published series even though our rate was in the range of the other studies. Moreover, the survival was significantly higher in patients who underwent curative surgery (median survival 1784 days, range 160–8650 days) than for those who underwent palliative resection (median survival 261 days, range 27–857 days) (P  =  0.0003, log-rank test).

It has been reported that palliative surgery is not compatible with survival at 5 years.2,20,22,23 The 1- and 3-year survival rates of the present 9 patients who underwent palliative surgery were 24.9% and 1.0%, respectively. In the present study, the resectability was found to be associated with a better survival rate in patients with duodenal adenocarcinoma, thus indicating that an aggressive surgical approach that achieves complete tumor resection should be pursued whenever possible. Distant metastasis and local invasion are the main contraindications to curative surgery.

Regional LN metastases have been reported in 43% to 64.3%8,24 of tumors, which is consistent with the present findings (52.3%). Rose et al2 performed wide-ranging lymphadenectomies and achieved a 5-year survival rate of 43% in patients with a positive node. Therefore, a combination of PD/PpPD with wide-ranging lymphadenectomy may be appropriate for the detection of LN metastasis. Sarela et al24 demonstrated that improved survival can be achieved for patients diagnosed early with no nodal involvement. In our study, the presence of LN involvement did not contribute to survival for curative resection (P  =  0.129, Table 4). However, distant LN metastasis was more significant with palliative surgery than curative surgery (P  =  0.009, Table 3). A randomized prospective trial is therefore warranted in the future to evaluate the ability of extended lymphadenectomy to increase patient survival rates. We were unable to demonstrate that this treatment confers a survival benefit and emphasize that the statistical analysis suffers from many limitations (nonrandomized, potential selection bias, and inadequate numbers), thus precluding a statistically powerful and reliable analysis.

The present study indicates that the 5-year survival rates of patients after a curative resection are favorable, but the respectability (R0) of patients with primary duodenal adenocarcinoma was not efficient. An increased resectability of primary duodenal adenocarcinoma, the ability to diagnose tumors at earlier stages, and radical resection are therefore considered to be essential predictors for the overall patient survival.