Case study 91.1

• What imaging studies are most appropriate in the initial workup of neuroendocrine tumors (NETs)?

A 56-year-old female with small bowel, well-differentiated NET with metastasis to the liver is coming into clinic for her 6-month follow-up visit. She had previously profuse diarrhea and flushing. She was found to have a positive somatostatin scintigraphy scan (Octreoscan) and was started on octreotide with resolution of her symptoms.

1. What imaging modality is most appropriate for following the patient?

Contrast-enhanced computed tomography (CT) scan of the abdomen and pelvis with liver triphasic
Somatostatin scintigraphy scan
Magnetic resonance imaging (MRI) of the abdomen and pelvis
Routine positron emission tomography (PET) scan
A or C
B or D

Scintigraphy with octreotide labeled with indium 111 (Octreoscan) was first used in the 1990s to localize previously undetected primary or metastatic NETs. At that time, radiopeptide functional imaging allowed for better localization and staging of neuroendocrine tumors as compared to CT or MRI scans. However, in the last decade, imaging with CT and MRIs has undergone tremendous technological advancement with improvement in resolution. A recent evaluation was performed to determine the utility of modern octreotide scans when used in conjunction with modern CT or MRI scans. The study found that multiphase contrast-enhancing CT or MRI scans detected more pathologic NET lesions than did single-photon emission computed tomography (SPECT) octreotide scanning. Particularly, octreotide scans did not identify additional primary tumors or soft tissue lesions that were not seen by CT or MRI scans, but did identify unsuspected bone metastasis that were not identified on CT or MRI. Given that cross-sectional imaging with modern CT and MRI can detect lesions down to 2–3 mm, it is no longer reasonable to expect that routine indium 111 octreotide scans would be able to detect lesions smaller than that. In general, current somatostatin scintigraphy is sensitive down to a size of approximately 7 mm. More sensitive and specific radiopharmaceuticals are under development, but have not yet been established as helpful and are not currently part of standard care.

Octreotide scans are appropriate to use as a baseline to determine the presence or absence of somatostatin receptors in vivo on the patient’s tumor. This is particularly important for those patients with hormonally nonfunctional tumors. Those patients who are found to have a negative somatostatin scintigraphy scan are not appropriate for treatment with somatostatin analogs, as there is no reason to believe that somatostatin would be effective in the absence of receptors on the tumor. In our opinion, once a baseline octreotide scan is performed, there is rarely a utility for routine use of octreotide scan for staging or surveillance. Octreotide scans do appear to be very sensitive for picking up asymptomatic bone lesions, and may be performed for the detection of bone lesions that are not visualized by CT or MRI if this would change management.

CT scans are potentially useful for following patients with NET. However, because neuroendocrine tumors can often be isodense with normal hepatic parenchyma, contrast enhancement is mandatory, and a triphasic view is more likely to give a complete assessment of liver involvement. MRI scans are also an excellent means of following patients with NET, as they are both sensitive and specific for neuroendocrine tumor involvement. Because well-differentiated neuroendocrine tumors tend to have a low metabolic rate, PET scans are not routinely recommended, as they have a relatively high false-negative rate.

Case study 91.2

• How does grade (low vs. intermediate vs. high grade) impact the prognosis and treatment of NETs?

A 45-year-old male has experienced increased fatigue, intermittent abdominal pain, and 30-lb. weight loss over the last 3 months. Colonoscopy and esophagogastroduodenoscopy (EGD) are negative. CT scan of the abdomen and pelvis shows a jejunal mass and innumerable liver lesions. A core biopsy of a liver lesion is performed and shows a grade 3, poorly differentiated neuroendocrine tumor with 30 mitotic figures/10 HPF and Ki67 index of 40%.

1. What treatment would be recommended?

Octreotide
Debulking surgery
Liver RFA
Platinum-based chemotherapy

NETs are epithelial neoplasms with predominant neuroendocrine differentiation and are classified histologically into two main groups, well differentiated and poorly differentiated. Well-differentiated tumors, which were traditionally referred to as carcinoid or pancreatic neuroendocrine (islet cell) tumors, are further separated into low grade (mitotic count of <2/10 HPF and Ki67 index of <3%) and intermediate grade (mitotic count of 2-20/10 HPF and Ki67 index of 3–20%). In poorly differentiated tumors, mitotic count usually exceeds 20/10 HPF and Ki67 index is over 20%. Assignment of grade based upon the proliferative rate correlates with patient survival independent of tumor stage in both primary and metastatic gastroenteropancreatic NETs. In one series of 425 patients with a pancreatic NET, 5-year survival rates for low-, intermediate-, and high-grade tumors were 75%, 62%, and 7%, respectively. Although patients with low-grade and intermediate-grade NETs of the digestive system are treated similarly at this time, as new treatment modalities become available, it is likely that the histological grade of a low versus intermediate NET will affect the selection of appropriate treatment.

Poorly differentiated or high-grade NETs, in contrast, more closely resemble small-cell neuroendocrine carcinomas of the lung. They rarely secrete hormones or express somatostatin receptors. For this reason, octreotide scans are usually negative, and somatostatin analogs, with their limited antiproliferative effect, would be unlikely to be effective treatment. Furthermore, poorly differentiated NETs are typically associated with a rapid clinical course and median survival is poor with localized, regional, or distal disease, having an overall survival of 34, 14, and 5 months, respectively. As these tumors behave like small-cell lung cancer, a platinum-based regimen with cisplatin or carboplatin plus etoposide or irinotecan is usually recommended as first-line therapy. There is currently no standard for second-line treatment, although topotecan, paclitaxel, docetaxel, vinorelbine, gemcitabine, and irinotecan have all been reported to show some activity. Furthermore, even if a patient presents with local, resectable, poorly differentiated neuroendocrine tumors, local therapy with surgery and/or radiation therapy is usually not curative, and systemic chemotherapy needs to strongly be considered.

Case study 91.3

• Is there a role for adjuvant octreotide after resection of NET primary?

A 56-year-old female undergoing an elective cholecystectomy for management of gallstones is found incidentally at operation to have a small bowel tumor. A resection of 20 cm of small bowel is accomplished, and final pathology reveals a 2.5 cm well-differentiated, low-grade carcinoid tumor with three of six lymph nodes positive for tumor. A postoperative contrast-enhanced CT scan of the abdomen and pelvis with liver triphasic shows no evidence of disease, and she is currently asymptomatic.

1. What should her follow-up plan be?

Adjuvant octreotide
Adjuvant interferon alpha
Adjuvant streptomycin and fluorouracil
Observation

To date, no study has been done to investigate the use of any antineoplastic agent in the adjuvant treatment of resected carcinoid or pancreatic NET. Thus far, antitumor activity from all known agents would appear to be too modest to suggest that a meaningful survival benefit could be expected in the adjuvant setting. Given that the median survival of patients who have had resection of low-grade, local or low-grade, local-regional NETs is very long (often measured in years to decades), it would be extremely unlikely that a study could show a benefit in the absence of an agent with outstanding clinical activity.

The PROMID trial investigated the use of octreotide to treat patients with nonfunctional metastatic carcinoid tumors. The trial, which was not powered to assess a survival difference, randomized patients to octreotide versus observation at the time of diagnosis of metastatic disease. The results showed that octreotide was able to improve time to tumor progression, or progression-free survival, by approximately 7 months compared to placebo in patients with active carcinoid tumors of the midgut. However, given that the median survival exceeded 5 years in each arm, it is highly unlikely that early initiation of octreotide, as was done in this trial, is necessary in order to achieve a beneficial result. The relatively modest benefit demonstrated makes it exceedingly unlikely that the cure rate would be increased by subjecting patients to octreotide in the adjuvant setting. Furthermore, although somatostatin analogs are quite well tolerated, they do have multiple potential side effects, including abdominal pain, bloating, loose stools, fat malabsorption, mild glucose intolerance, and increased risk of gallstones.

Alpha interferon is an agent with considerable toxicity and minimal evidence of substantial antitumor activity in NETs. Early studies of interferon alpha in NETs have confused the literature by mixing “biologic” responses and “objective” responses together in their reporting, thus creating the misconception that substantial tumor regression was a common outcome from interferon therapy. In fact, regression to interferon is extremely rare. A later trial in the Mayo Clinic with interferon alpha showed a high degree of toxicity and minimal evidence of activity in carcinoid tumors. A review by Plockinger et al. (2007) estimated that approximately 10% of patients achieve some degree of actual tumor regression, and major objective responses are incredibly rare. Furthermore, interferon has not been studied in the adjuvant setting and has a high degree of toxicity, including flulike symptoms, fatigue, depression, myelosuppresion, alteration in thyroid function, and anorexia. Therefore, it should not be used in the adjuvant setting, and it has an extremely limited role in the treatment of NETs even in the metastatic setting.

The role of conventional chemotherapy in the metastatic setting for well-differentiated carcinoid and pancreatic neuroendocrine tumors is debated, as response rates vary considerably between different studies secondary to a wide range of assessment criteria and patient populations. In general, response to cytotoxic chemotherapy is rare in patients with advanced, well-differentiated carcinoid tumors. Cytotoxic chemotherapy for pancreatic NETs appears to be a bit more responsive than carcinoid tumors; however, older trials have been reported in a manner that may overstate the degree of actual activity. For instance, in the phase II–III Eastern Cooperative Oncology Group (ECOG) trial, 249 patients with carcinoid tumors were randomized to receive doxorubicin with fluorouracil or streptozocin with fluorouracil. The response rates were relatively low at approximately 16% for both arms, and substantial side effects were seen, with approximately one-third of patients who received streptozocin-based therapy developing mild to moderate renal toxicity. One of the earlier studies presented by Moertel et al. (1992) with streptozocin and doxorubicin reported a 69% response rate in patients with advanced pancreatic NETs. Modern criteria for response assessment were not applied in this trial, however, and it is likely that the objective response rates as they would be defined by today’s criteria were, in fact, substantially lower than reported. A retrospective evaluation of 16 patients treated with this combination at Memorial Sloan Kettering Cancer Center found an objective response in only 1 of 16 patients, and identified many methodological flaws in response reporting in the Moertel trial. The high response rate reported is now felt to be a gross overestimate of the degree of activity with streptozocin-based chemotherapy. More recently, a retrospective review from MD Anderson Cancer Center showed an overall response rate of 39% with fluorouracil, doxorubicin, and streptozocin in patients with pancreatic NETs. Based on the only modest response rates and high rate of toxicities, systemic chemotherapy is not recommend in the adjuvant setting and rarely plays a role in early management of metastatic low grade NETs”. It is usually reserved for patients with symptoms secondary to tumor bulk or uncontrolled hormonal excess once they have failed octreotide treatment. More recently, everolimus and sunitinib have been shown to have modest activity in metastatic pancreatic NETs (but not in carcinoid tumors). Use of these agents in the adjuvant setting has not been studied and is not recommended at this time.

< div class='tao-gold-member'>

Only gold members can continue reading. Log In or Register to continue