A 71-Year-Old Man with a Large Cystic Pancreatic Mass

CASE REPORT

Walid Faraj: A 71-year-old man presented with a 3-month history of abdominal pain radiating to the back, anorexia, and 10kg weight loss. Abdominal examination revealed a palpable, nontender epigastric mass. A computed tomography (CT) scan showed an 8 × 7 cm heterogeneous mass in the body of the pancreas with thrombosis of the splenic vein. Endoscopic ultrasound (EUS) confirmed a complex 10 × 9 cm mass with solid and cystic components located in the pancreatic body and tail, with two enlarged lymph nodes along the celiac axis. Serum tumor markers were elevated: CEA 10 ng/mL (0–4 ng/mL), CA 19-9 1,145 U/mL (0–37 U/mL). Liver and renal function tests were normal. A staging laparoscopy was performed, which showed no evidence of liver or peritoneal metastases, and the patient underwent a distal pancreatectomy and splenectomy.

Mohamed Khalife: Is there any evidence for neoadjuvant therapy for pancreatic cancer?

Eileen O'Reilly: There have been no large randomized controlled studies of neoadjuvant therapy and surgery compared to surgery and adjuvant therapy in resectable pancreatic cancer. A recent systematic review of preoperative/neoadjuvant therapy for pancreatic cancer found that for patients with initially resectable disease, resection frequencies and survival after neoadjuvant therapy were similar to those of patients undergoing primary resection and adjuvant therapy. Approximately one third of patients with initially nonresectable tumors were resected following neoadjuvant therapy, with survival rates comparable with those of patients with initially resectable disease.¹ In a single-arm phase II study of gemcitabine-based neoadjuvant chemoradiotherapy performed at MD Anderson Cancer Center, 74% of patients underwent successful resection. Median survival for those patients was 34 months, compared to only 7 months for patients who had unresectable disease.² We are currently undertaking a trial at Memorial Sloan-Kettering Cancer Center evaluating preoperative gemcitabine and oxaliplatin for operable pancreatic cancer.

Walid Faraj: Dr Zaatari, can you please comment on the pathology of the tumor?

Ghazi Zaatari: Histopathologic examination revealed a cystic pancreatic tumor measuring 11cm in greatest dimension. The infiltrating tumor was composed of sheets of malignant epithelial cells with pleomorphic nuclei and moderately abundant clear cytoplasm (negative mucin and PAS-D stains) arising from a central intraductal papillary mucinous neoplasm (IPMN). Some of the smaller ducts showed variable degrees of pancreatic intraepithelial neoplasia (PanIN) ranging from grade I to III. The tumor involved the posterior margin of resection at the site of cystic tumor rupture; the pancreatic margin was free of tumor. The poorly differentiated clear cell carcinoma showed evidence of angiolymphatic and perineural invasion. Metastatic carcinoma was seen in 39 of 41 peripancreatic lymph nodes.

Ali Shamseddine: Is there evidence that carcinomas arising from IPNMs have a better outcome?

Deborah Mukherji: IPMNs of the pancreas encompass a spectrum of intraductal mucin-producing neoplasms with cystic dilatation of the pancreatic ducts. IPNMs comprise less than 5% of exocrine pancreatic tumors; however, heightened awareness and advances in imaging have led to an increase in their detection. Invasive carcinomas associated with IPMNs occur in approximately 35% of cases, most arising in areas of high-grade dysplasia. Two distinct types of carcinoma commonly occur in association with IPMNs: mucinous adenocarcinoma and conventional ductal adenocarcinoma.³

Small case series have reported the postoperative 5-year survival rate for IPMN associated with invasive carcinoma to be from 31% to 58% which is higher than seen in conventional pancreatic adenocarcinoma.⁴ However, other reports have demonstrated that when lymph nodes are involved, survival is similar to conventional ductal adenocarcinoma.⁶ A new subclassification of IPMNs based on morphologic phenotypes and immunohistochemical variations has been reported categorizing IPMNs into four subtypes: intestinal type, gastric type, pancreatobiliary type, and an oncocytic type.⁵

Sadakari et al found that invasive carcinoma derived from the nonintestinal type of IPMN was characterized by lymphatic invasion with 5-year survival as poor as that of patients with conventional ductal adenocarcinoma. In contrast, patients with invasive carcinoma derived from the intestinal type of IPMN had a more favorable prognosis than patients with conventional ductal adenocarcinoma.⁴ Although the numbers in the study were small (30 patients), the importance of further work in determining the molecular basis of these differences has been demonstrated.

Nadim El-Majzoub: How common are clear cell carcinomas of the pancreas and what are the differential diagnoses of clear cell neoplasms in the pancreas?

Walid Faraj: Clear cell carcinomas arising in the pancreas are rare and poorly defined. This is the first report of clear cell carcinoma arising from an intraductal papillary mucinous neoplasm of the pancreas. The differential diagnosis of clear cell neoplasms in the pancreas includes metastatic renal cell carcinoma, perivascular epithelioid cell tumor, variants of solid and pseudopapillary tumors, and pancreatic endocrine tumors.^7–9

Kim et al reviewed 84 cases of pancreatic ductal adenocarcinoma and found 20 (24%) to have a significant degree of involvement by a clear cell component. In 12 cases the clear cells made up greater than 75% of the invasive tumor. Hepatocyte nuclear factor 1B (HNF1B) has been identified as a specific biomarker of clear cell tumors of the female genital tract. HNF1B was overexpressed in all cases of clear cell pancreatic carcinoma and in the clear cell components of eight ductal carcinomas with clear cell features. In the cases of conventional ductal carcinoma reviewed, eight cases (15%) showed overexpression of HNF1B. Survival of patients with conventional, mixed, and clear cell carcinoma was not significantly different. However, survival analysis stratified by degree of HNF1B staining, regardless of morphologic subtype, showed significantly decreased survival in cases with high HNF1B staining (P < .01).¹⁰

Further studies are needed to determine the significance of this marker with respect to tumor biology. Kim et al excluded carcinomas arising from IPMN from their analysis; we have demonstrated that the spectrum of tumors associated with IPMN includes this rare case of aggressive clear cell carcinoma. Future studies should not exclude ductal tumors associated with IPMN from analysis. The patient made a good postoperative recovery and commenced adjuvant gemcitabine chemotherapy 3 weeks post surgery.

Manish Shah: Given the unusual histology, positive margins and extensive lymph node metastasis, was adjuvant gemcitabine chemotherapy appropriate?

Ali Shamseddine: The CONKO-001 trial demonstrated a significant prolongation of disease-free survival and overall survival in patients who received adjuvant gemcitabine compared to observation alone after resection of pancreatic cancer. However, it is not noted whether patients with carcinomas arising from IPMN were included.^11,12 Patients with large, node-positive tumors were shown to benefit from adjuvant chemotherapy with gemcitabine, so the management of this patient was in line with current evidence and practice guidelines.

Walid Faraj: After receiving two weekly doses of gemcitabine chemotherapy, the patient presented with abdominal pain and malaise. A CT scan performed 2 months postoperatively revealed within that short period of time a large heterogeneous multicystic mass measuring 6.6 × 5.5 × 6.1 cm involving the proximal jejunum with evidence of liver metastases, omental metastases and ascites. What are the treatment options for metastatic pancreatic cancer?

Eileen M. O'Reilly: Current treatment options for metastatic pancreatic cancer include chemotherapy either as one drug option or as a gemcitabine-based combination with either erlotinib, capecitabine, or a platinum agent. Gemcitabine chemotherapy remains a standard first-line treatment for advanced pancreatic cancer, offering modest improvements in tumor-related symptoms and survival.¹³ Many different cytotoxic agents and targeted therapies have been compared with or combined with gemcitabine in randomized phase III trials; however, progress remains disappointing.

The epidermal growth factor receptor (EGFR) inhibitor erlotinib has shown a statistically significant survival advantage in combination with gemcitabine in a randomized phase III trial vs. gemcitabine alone.¹⁴ Debate is ongoing in the clinical arena regarding the clinical significant of the increase in overall survival from 5.91 months to 6.24 months associated with erlotinib treatment. Meta-analyses have indicated a survival benefit associated with gemcitabine in combination with fluoropyrimidine or platinum doublet chemotherapy.^15–17 More recent data indicate that for fit patients with metastatic pancreas adenocarcinoma, FOLFIRINOX (5-fluoruracil[5-FU]/leucovorin/irinotecan/oxaliplatin) may be a reasonable option following demonstration of a significant 3.5 month survival advantage over gemcitabine along with a tripling of response rate and doubling of progression-free survival.¹⁸ One other important point regarding evaluation of patients for adjuvant therapy is that a baseline CT scan following surgery may demonstrate overt disease. The level of CA 19-9 postoperatively can also provide insight regarding the likelihood of residual disease.¹⁹

Options remain limited for patients with gemcitabine-refractory disease, and there is no current standard of care. Commonly used second-line chemotherapy regimens include oxaliplatin and 5-FU/leucovorin (OFF),²⁰ gemcitabine and oxaliplatin (GEMOX),²¹ oxaliplatin and capecitabine (XELOX),²² and irinotecan plus oxaliplatin,²³ all dependent on what the patient has received as initial therapy.

Walid Faraj: The patient's performance status deteriorated rapidly. His symptoms were palliated using opiates and symptomatic drainage of ascites. A celiac plexus nerve block was considered, but given the quick deterioration and the bulk of metastatic disease, the procedure was technically difficult and considered unlikely to have been successful; it was not undertaken. Addressing the expectations of the patient and the family, given the rapid relapse and progression of disease, was challenging, requiring the input of palliative care and psychosocial supports. The patient died within 3 weeks of the documentation of metastatic disease.

SUMMARY

Deborah Mukherji: In conclusion, this was a rare case of clear cell carcinoma arising from pancreatic IPNM that demonstrated exceptionally aggressive behaviour. The focus of care quickly changed from potentially curative adjuvant treatment to palliation of symptomatic metastatic disease that resulted in substantial psychological distress for the patient and his family. Further work to elucidate the pathways of pancreatic tumorigenesis and the molecular pathology involving the spectrum of IPNM may provide both prognostic information and potential targets for therapy.