A 72-Year-Old Man With Duodenal Adenocarcinoma

CASE REPORTS

Dr. Tamraz: A 72-year-old male patient, previously healthy and had never smoked, presented with intermittent and crampy abdominal pain of about 1-month duration that was occasionally associated with nausea and vomiting. The patient had no fevers, night sweats, or weight loss. He also had no melena or change in bowel habits. His past medical history was significant for hypertension, which was well controlled on angiotensin receptor blockers.

Laboratory evaluation showed a mild microcytic anemia, with a Hb of 11 g/dL, Hct = 33%, MCV = 76 fl, normal WBCs and platelets. Occult blood was positive.

Dr. Lowery: With the presence of occult blood positivity, I assume an endoscopy was done.

Dr. el-Oubeidi: An upper endoscopy was undertaken and showed a tumor at the apex of the duodenal bulb, which was circumferential, ulcerated, and protruding into the lumen causing partial obstruction.

Dr. Zaatari: Biopsy of this tumor was conflicting between carcinoid and adenocarcinoma. Pathologists find it difficult sometimes to differentiate on limited biopsy material between a duodenal, pancreatic, biliary, and ampullary carcinoma due to the closeness of these structures and their overlapping histologic features of carcinomas arising at those sites. Immunohistochemical stains are also of limited value in this context. Molecular profiling may offer an opportunity to better categorize tumors that arise at that site. According to a recent presentation by Overmann et al¹ at the 2011 Gastrointestinal Cancers Symposium, gene expression analysis incorporates carcinomas of ampulla with that of the duodenum and classifies ampullary carcinomas into 2 groups, biliary like and intestinal like, each with a different prognosis and treatment.

Dr. Abou-Alfa: Dr. Shamseddine, how often do you see such presentation in Lebanon?

Dr. Shamseddine: Duodenal tumors whether carcinoid or adenocarcinoma in origin are very rare, with only a few cases reported in our registry.

Dr. O'Reilly: Adenocarcinoma of the small intestine is an uncommon neoplasm that represents up to 42% of all small bowel malignancies and fewer than 0.5% of all gastrointestinal malignancies.^2,3 The majority of patients present with locoregional disease, stage II or III disease present in 53% of patients.⁴ The diagnosis of these tumors is often difficult due to the rarity of these lesions and the nonspecific and variable nature of the presenting signs and symptoms. Thus, delay in diagnosis is common, which may result in discovery of the disease at a late stage and a poor treatment outcome. The etiology of small bowel cancers is unknown, although a number of risk factors and predisposing conditions have been described. The most important of these include hereditary nonpolyposis colorectal cancer, Peutz-Jeghers syndrome, familial adenomatous polyposis, chronic inflammatory disorders such as Crohn's disease, and autoimmune disorders such as celiac disease. The incidence of adenocarcinoma is highest in the duodenum of any site in the small bowel. In two large single-institution series of patients with small bowel adenocarcinoma 65% and 57% arose in the duodenum.^5,6

I am curious, though, if any further testing was obtained to help evaluate the extent of disease. Periampullary carcinomas include adenocarcinomas of the ampulla, pancreas, and duodenum as well as the common bile duct.

Dr. Al-Kutoubi: A computer tomography (CT) scan with oral and IV contrast was obtained that showed an ill-defined mass at the level of the duodenum measuring 6 × 5.4 cm, causing distortion of the head of the pancreas, partially obstructing the duodenum with no evidence of biliary ductal dilatation (Figures 1A and 1B).

Figure 1.

(A) Axial computed tomography images of the abdomen with intravenous and oral contrast showing a large heterogeneously enhancing lesion involving the proximal duodenum and distorting the pancreatic head. (B) Coronal reconstruction of computed tomography images of the upper abdomen showing the large duodenal lesion and surrounding structures (GB = gallbladder, * = common bile duct; long arrow = the portal vein; and short arrows = the pancreatic head).

Dr. Tamraz: In view of the resectable nature of the disease and the absence of metastatic disease, the patient underwent an exploratory laparotomy and a Whipple procedure was performed, which I will ask Dr. Faraj to comment on.

Dr. Faraj: At laparotomy and following exploration of the peritoneal cavity, the lesser sac was opened and the duodenum was mobilized. The common hepatic duct was divided after cholecystectomy, and a tunnel was created anterior to the portal vein extending to the superior mesenteric vein. The pancreas was divided, and then the specimen was excised. Reconstruction was done in one limb using duct-to-mucosa anastomosis for the pancreaticojejunostomy and about 15 cm distally, and a hepaticojejunostomy was performed in a retrocolic fashion. Then about 70 cm gastro-jejunostomy was done, and a Braun anastomosis was carried by side-to-side jejuno-jejunostomy (Figure 2). The patient tolerated the procedure well and was transferred to the recovery room in good condition.

Figure 2.

Schematic diagram of surgical procedure.

Dr. Zaatari: There was a large fungating mass at the duodenum in proximity to the head of pancreas, and actually short of looking at the histologic sections, it was very difficult to determine that this was indeed a primary from the duodenum or pancreas because of the proximity of the two structures. But looking at the slides, you can see that most of the pathology was consistent with adenocarcinoma, moderately differentiated with many well-formed glands. You can see as well that the bulk of the tumor was within the wall of the duodenum with extension to the pancreatic head. Based on that, the argument was more in favor of duodenal primary rather than pancreatic primary. The ampulla was also involved by the tumor. There were 5 negative peripancreatic lymph nodes. Duodenal, gastric, common bile duct, and pancreatic margins were free of tumor. Gerota's fascia was negative for malignancy as well as hilar lymph nodes (Figures 3, 4, and 5). Immunohistochemistry showed the following: CK7 focally positive, CK20 negative, neuron-specific enolase (NSE) negative, chromogranin negative, CD56 negative.

Figure 3.

Photomicrograph demonstrating invasive moderately to poorly differentiated adenocarcinoma involving the mucosa of the duodenum and extending into the submucosa.

Figure 4.

Photomicrograph showing well- to moderately differentiated adenocarcinoma of the duodenum.

Figure 5.

Photomicrograph revealing extension of the adenocarcinoma within the muscularis propria of the duodenal wall.

Dr. Abou-Alfa: This is staged as a T4N0M0, AJCC stage II cancer. Dr. Ang, what would you recommend for therapy?

Dr. Ang: No standard of care exists for the adjuvant treatment of resected duodenal malignancies. We usually extrapolate from adjacent areas in the GI tract after classifying such tumors as being of an intestinal phenotype or pancreaticobiliary phenotype. Thus, as per Dr. Zaatari, the tumor morphologically is of intestinal phenotype due to the presence of a mucinous component within the glands. Duodenal tumors represent a therapeutic challenge due to the scarcity of these tumors and hence the absence of prospective randomized trials addressing the issue of adjuvant therapy. A 2007 Cochrane analysis concluded that there were no suitable trials to analyze.⁷ However, the role of adjuvant chemotherapy and chemoradiotherapy has been investigated in several retrospective studies. MD Anderson Cancer Center investigators studied the effect of adjuvant chemotherapy following surgery in patients with small bowel adenocarcinoma. Records of 217 patients were reviewed retrospectively, and results showed no significant effect on survival (P = .49).⁵ In a study conducted by Indiana University, 30 patients with resectable duodenal adenocarcinoma underwent surgery alone (n = 23) or surgery combined with chemotherapy or chemoradiotherapy (n = 7). Results also showed no significant effect on survival (P = .355).⁸ In another retrospective review by MD Anderson Cancer Center, Overmann et al identified 54 patients with small bowel adenocarcinoma. Thirty patients received adjuvant therapy consisting of systemic chemotherapy with or without radiation in 28 and radiation alone in 2 patients. Results revealed no benefit of adjuvant therapy on overall survival (P = .23). In patients with a high risk of relapse (defined as a lymph node ratio >10%), adjuvant therapy appeared to improve overall survival, P = .04, but not disease-free survival, P = .15.⁹ Swartz et al compared survival results of patients treated with adjuvant chemoradiotherapy with historical controls. Results also showed a nonimproved survival effect of adjuvant chemoradiotherapy (43% vs. 44%).¹⁰

From these retrospective studies, there seems to be no role of adjuvant therapy in the treatment of duodenal adenocarcinoma. Prospective trials are needed to further assess the role of adjuvant therapy. Thus it is reasonable to extrapolate from trials of adjuvant therapy in colorectal cancer or from trials of adjuvant therapy in gastric cancer when managing patients with duodenal tumors.

Dr. Geara: The patient then received adjuvant chemoradiation extrapolating from MacDonald's regimen for the treatment of adjuvant gastric cancer.¹¹ The patient received one cycle of 5 FU (425 mg/m²/day) and leucovorin calcium (20 mg/m²/day) daily for 5 days followed 1 month later by 45 Gy (1.8 Gy/day) radiotherapy given with 5 FU (400 mg/m²) and leucovorin (20 mg/m²) on days 1–4 and on the last 3 days of radiotherapy. One month after completion of radiotherapy, 2 more 5-day cycles of chemotherapy with 5FU (425 mg/m²/day) and leucovorin (20 mg/m²/day) were given at monthly intervals. Treatment ended on September 2008. His course was complicated by grade 2 oral mucositis, but he was able to tolerate the full adjuvant therapy. Subsequent to this the patient was followed expectantly.

Dr. Tamraz: One year later the patient presented with pain at the incision site. Physical examination identified a small mass, and thus the patient underwent excision of abdominal wall tumor extra and intra-abdominally and repair of abdominal wall defect with prolene mesh. Pathology showed metastatic adenocarcinoma consistent with duodenal primary and extending to the margins of resection.

Dr. Kelsen: This is now clear evidence of metastatic disease. Was there any disease left, and what did complete staging show?

Dr. Tamraz: One month later, a PET-CT scan showed a hypermetabolic lesion at the left lower anterior chest wall (SUVmax = 4.7) and a small hypermetabolic lesion with SUVmax = 3.8 at the left parasternal region.

Dr. Faraj: A decision was made to proceed with surgical resection with wider margins together with excision of abdominal muscles around the mass and extending to the xyphoid medially. Actually, it was a bulge at the lateral aspect of left subcostal incision and on examination; it was typical of an incisional hernia thus the additional justification for surgical intervention. An incision was performed over the previously extended right subcostal incision until the mass was encountered. Hence, upon surgery, it was not a hernia, it was a hard mass. So dissection around the tumor was done reaching the anterior abdominal fascia. The fascia was incised, and the external oblique muscle was incised, obtaining good margins from the tumor and followed down to the posterior fascia. The tumor was found to be extending up to the 8th and 7th rib cartilages. So the dissection was carried up as high as possible, and then the mass was excised and sent for pathology. Because there was a defect at the wound site, a prolene mesh was placed.

Dr. Shamseddine: I was not in favor of a surgical approach and would have preferred to have initiated systemic therapy. Nonetheless the patient subsequently received 6 cycles of capecitabine. However, 3 months later, a CT chest of chest, abdomen, and pelvis showed suspicion of metastasis to the soft tissues of the abdominal wall. A new PET-CT scan showed interval appearance of multiple hypermetabolic lesions in the left anterolateral abdominal wall at the level of mid-left kidney. An fine needle aspiration of the abdominal wall mass was negative for malignancy.

Dr. Sibai: Was there a reason for PET-CT and rebiopsy?

Dr. O'Reilly: I do not see the need for the PET-CT and biopsy because imaging clearly showed these soft tissue lesions in the abdominal wall, and after the prior surgery, which had documented recurrence, the need to do so again was not indicated.

Dr. Lowery: Is there a role for biologic therapy in this case?

Dr. Tamraz: Due to the rarity of small bowel tumors, the role of carcinogenesis is not well understood, unlike in colorectal and pancreatic carcinogenesis. In an attempt to analyze the role of KRAS and BRAF mutations in small bowel tumors, Schönleben¹² concluded that the oncogenic properties of KRAS and BRAF contribute to the tumorigenesis of periampullary and ampullary tumors and that BRAF mutations occur more frequently in periampullary than ampullary neoplasms. In another report by Arai et al¹³ they found that the mutational frequency of the K-ras gene in small bowel adenocarcinoma is similar to that of colorectal adenocarcinoma, while the frequency of the p53 gene was slightly lower than that of colorectal adenocarcinoma.

Dr. Abou-Alfa: Thank you, all. To summarize: Duodenal adenocarcinomas are quite rare. Their diagnosis is therefore often difficult due to the nonspecific and variable nature of the presenting signs and symptoms and the difficulty in ascertaining the exact site of origin of some periampullary malignancies. The lack of biliary dilatation is more in favor of a duodenal origin. There is no standard of care, and there is no consensus on adjuvant treatment because of a paucity of data addressing the benefit of adjuvant therapy, whether chemotherapy, radiation therapy, or both. Hence, its role remains undefined. Actually, a Cochrane meta-analysis in 2007 deduced that there were no suitable trials to analyze the place of adjuvant therapy.⁷ The few retrospective studies that addressed this issue revealed conflicting results with no survival benefit from any approach.