Gallium-68-DOTA-NOC PET/CT of Patients With Gastroenteropancreatic Neuroendocrine Tumors: A Prospective Single-Center Study

Nuclear Medicine and Molecular Imaging Original Research Naswa et al. PET/CT of Gastroenteropancreatic NETs Nuclear Medicine and Molecular Imaging Original Research Niraj Naswa 1 Punit Sharma 1 Abhishek Kumar 1 Aftab Hasan Nazar 1 Rakesh Kumar 1 Sunil Chumber 2 Chandrashekhar Bal 1 Naswa N, Sharma P, Kumar A, et al. Keywords: 68 Ga-DOTA-NOC, conventional imaging, gastroenteropancreatic neuroendocrine tumors, neuroendocrine tumors, PET/CT DOI:10.2214/AJR.11.7298 Received May 21, 2011; accepted after revision July 12, 2011. 1 Department of Nuclear Medicine, All India Institute of Medical Sciences, Ansari Nagar, New Delhi 110029, India. Address correspondence to C. Bal (csbal@hotmail.com). 2 Department of Surgical Disciplines, All India Institute of Medical Sciences, New Delhi, India. AJR 2011; 197:1221 1228 0361 803X/11/1975 1221 American Roentgen Ray Society Gallium-68-DOTA-NOC PET/CT of Patients With Gastroenteropancreatic Neuroendocrine Tumors: A Prospective Single-Center Study OBJECTIVE. The objective of this study was to evaluate the role of 68 Ga-labeled [1, 4, 7, 10-tetraazacyclododecane-1, 4, 7, 10-tetraacetic acid]-1-nai 3 -octreotide (DOTA-NOC) PET/CT in the diagnosis and management of gastroenteropancreatic neuroendocrine tumors (NETs). SUBJECTS AND METHODS. One hundred nine patients (median age, 50 years) with gastroenteropancreatic NETs underwent 68 Ga-DOTA-NOC PET/CT. PET/CT was performed after injection of 132 222 MBq (4 6 mci) of 68 Ga-DOTA-NOC. Images were evaluated by two experienced nuclear medicine physicians both qualitatively as well as quantitatively (maximum standardized uptake value [SUV max ]). Results of PET/CT were compared with the results of conventional imaging. Histopathology results, when available, and follow-up PET/CT or conventional imaging with biochemical markers were considered to be the reference standards. RESULTS. Gallium-68-DOTA-NOC PET/CT showed sensitivity and specificity of 78.3% and 92.5%, respectively, for primary tumor and 97.4% and 100% for metastases. It was better than a conventional imaging modality for the detection of both primary tumor (p < 0.001) and metastases (p < 0.0001). It changed the management strategy in 21 patients (19%) and supported management decisions in 32 patients (29%). CONCLUSION. Gallium-68-DOTA-NOC PET/CT appears to be a highly sensitive and specific modality for the detection of gastroenteropancreatic NET. It is better than conventional imaging for the evaluation of gastroenteropancreatic NETs and can have a significant impact on patient management. N euroendocrine tumors (NETs) are a rare and heterogeneous group of neoplasms originating from the neural crest. The main localization of NETs is in the gastroenteropancreatic tract; lung; and, rarely, ovary and thymus [1]. Gastroenteropancreatic NETs are generally slow-growing and may remain indolent for many years. A distinguishing feature of these tumors is the production of peptides that may cause symptoms characteristic of excessive hormonal production. Because many gastroenteropancreatic NETs are malignant, the initial clinical presentation may be dominated by symptoms associated with tumor bulk or metastatic localizations, seriously affecting patient prognosis [2 4]. The liver is the most common site for metastasis, which decreases the 5-year survival rate [5 7]. Complete surgical excision is the treatment of choice for patients with NET, but in many patients such treatment is not possible at diagnosis [8 12]. Palliative treatments such as surgical debulking, chemotherapy, interferon therapy, and hepatic artery embolization have resulted in symptomatic improvement and in partial tumor regression [12]. Somatostatin analogs offer an alternative therapeutic modality. Also, high-level somatostatin receptor expression makes NETs amenable to treatment using peptide receptor radionuclide therapy in surgically inoperable cases and in advanced cases [13]. Conventional imaging modalities, such as CT, ultrasound, and MRI, represent the most widely used investigations for the diagnosis of gastroenteropancreatic NETs. Of these modalities, MRI is considered to be the most sensitive imaging technique especially for the detection of liver metastases, which are sometimes very difficult to localize because of their small size [14]. AJR:197, November 2011 1221

Naswa et al. The expression of somatostatin receptors by NETs has led to the use of peptide receptor scintigraphy with radiolabeled somatostatin analogs for imaging. Indium-111-labeled pentetreotide (OctreoScan, Covidien) is one such radiotracer that has proven to be an important tool for the diagnosis of gastroenteropancreatic NETs [15 20]. Somatostatin receptor scintigraphy (SRS) has been reported to be more sensitive than a conventional imaging modality [20]. However, SRS with 111 In-pentetreotide suffers from the drawbacks of the limited resolution of SPECT technology and uneven biodistribution of the radiotracer in the liver and spleen, obscuring small lesions. With the availability of gallium-68 a novel generator-derived, positron-emitting radionuclide that can be easily labeled to macrocyclic chelators such as DOTA (1, 4, 7, 10-tetraazacyclododecane-1, 4, 7, 10-tetraacetic acid) new PET radiotracers aimed at detecting somatostatin receptors with greater specificity have been devised [21]. Gallium- 68-DOTA-1-Nal 3 -octreotide (NOC) is one such tracer. The aim of the current study was to prospectively evaluate the role of 68 Ga-DOTA- NOC PET/CT in patients with gastroenteropancreatic NETs and see to what extent it can influence management decisions in such patients. Subjects and Methods This was a prospective study. A total of 109 patients with gastroenteropancreatic NETs were evaluated with 68 Ga-DOTA-NOC PET/CT between October 2007 and September 2010. Only the initial PET/CT studies were used for the analysis; follow-up studies were excluded. This study was conducted in accordance with our institute s ethics protocol and written informed consent was obtained from all patients. The results of 68 Ga-DOTA-NOC PET/CT were analyzed from the following perspectives: staging disease in patients with already-diagnosed NET, detecting sites of recurrence in patients with treated NET (restaging), diagnosing NET in patients suspected of having NET on the basis of clinical features or biochemical results, comparing PET/ CT results with the results of conventional imaging modalities, selecting potential candidates for peptide-based radioreceptor therapy, and establishing whether the results of 68 Ga-DOTA-NOC PET/CT have an impact from a management point of view. Gallium-68-DOTA-NOC PET/CT Acquisition Gallium-68-DOTA-NOC synthesis was performed as previously detailed [22]. Gallium-68-DOTA- NOC PET/CT was performed on a dedicated PET/ CT unit (Biograph 2, Siemens Healthcare). If the patient was receiving somatostatin analog therapy, it was stopped before PET/CT: short-acting analogs for 3 days before PET/CT and long-acting analogs for 4 6 weeks. Fasting was not mandatory. A dose of 132 222 MBq (4 6 mci) of 68 Ga-DOTA-NOC was injected IV. After a 45- to 60-minute uptake period, patients underwent PET/CT. An oral contrast agent was used. No IV contrast agent was used. In the PET/CT system, the CT acquisition was performed with a slice thickness of 4 mm and a pitch of 1 on a helical dual-slice CT unit. Images were acquired using a matrix of 512 512 pixels and a pixel size of 1 mm. After the CT acquisition, the table was moved toward the FOV of PET, and PET acquisition of the same axial range was started with the patient in the same position. The PET components of the PET/CT system are based on a full-ring lutetium oxyorthosilicate PET system. Three-dimensional PET acquisition was performed from the base of the skull, including the pituitary fossa, to the mid thighs. Additional spot views were obtained when necessary. PET data were acquired using a matrix of 128 128 pixels and a slice thickness of 1.5 mm. CT-based attenuation correction of the emission images was used. PET images were reconstructed by iterative method ordered subset expectation maximization (2 iterations and 8 subsets). After completion of PET acquisition, the reconstructed attenuation-corrected PET images, CT images, and fused images of matching pairs of PET and CT images were available for review in the axial, coronal, and sagittal planes, as well as in maximum intensity projections and in the 3D cine mode. Image Analysis Gallium-68-DOTA-NOC PET/CT studies were evaluated by two experienced nuclear medicine physicians in consensus. They were blinded to the findings of structural imaging. They evaluated the PET images both qualitatively and semiquantitatively. Any nonphysiologic focal area of increased 68 Ga-DOTA-NOC uptake was sought. For 68 Ga- DOTA-NOC PET/CT, any nonphysiologic uptake that was greater than uptake in the surrounding area was interpreted as a positive finding for NET. Positive findings on 68 Ga-DOTA-NOC PET were localized to anatomic images from the unenhanced CT study. The criteria for a correct detection of NET by PET/CT are both positive 68 Ga-DOTA- NOC uptakes and the correct anatomic localization of the tumor. The PET/CT findings were grouped as primary or metastatic disease. The maximum standardized uptake values (SUVs max ) of primary lesions and metastatic lesions were calculated. For final analysis, the lesion with the highest pathologic tracer accumulation within each region in each patient was recorded. Reference Standard Histopathology results and the results of conventional imaging and follow-up imaging in combination with biochemical markers were used as the reference standard. Histopathologic confirmation was not possible in all of the cases because of technical or ethical reasons. Hence, most of the lesions detected on 68 Ga-DOTA-NOC PET/ CT were confirmed with follow-up, if applicable (minimum duration = 6 months), with 68 Ga- DOTA-NOC PET/CT; with ultrasound; or, in the case of pancreatic lesions, with endoscopic ultrasound and biochemical markers. For comparison, conventional imaging modalities (contrastenhanced CT, MRI, ultrasound) reports were retrieved from the hospital database and reviewed. Statistical Analysis Continuous variables were expressed as medians and ranges. Categoric data were expressed as numbers and percentages. The sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV) of 68 Ga-DOTA- NOC PET/CT and conventional imaging modalities for detecting primary and metastatic lesions were calculated. The McNemar test was used to compare the diagnostic accuracy of 68 Ga-DOTA-NOC PET/ CT with conventional imaging modalities. The Mann-Whitney test (with two-tailed probability) was used to compare continuous variables. A p value of < 0.05 was considered to be statistically significant. All the data analyses were performed using a statistical software package (SPSS, version 11.5, SPSS). Results Patient Characteristics A total of 109 patients with known or suspected gastroenteropancreatic NET were included in the current study. Patient characteristics including age, sex, serum chromogranin A level, histology result, treatment received, and indication for PET/CT are detailed in Table 1. The most common histopathologic variant was carcinoid tumor followed by gastrinoma. Reference Standard The reference standard for considering a lesion detected on 68 Ga-DOTA-NOC PET/ CT as either positive or negative for NET was based on a combination of the following: histopathology results in 24 patients (biopsy, n = 18; fine-needle aspiration, n = 6 patients), follow-up 68 Ga-DOTA-NOC PET/ CT in 11 patients, correlation with conventional imaging (contrast-enhanced CT or MRI) in 71 patients, biochemical marker in 22 patients, endoscopic ultrasound in one patient, and upper gastrointestinal endoscopy in 1222 AJR:197, November 2011

PET/CT of Gastroenteropancreatic NETs TABLE 1: Patient and Disease Characteristics Variable Value % of Patients Age (y) Median 50 Range 21 76 Sex (no. of patients) Male 58 53.2 Female 51 46.8 Chromogranin A (ng/ml) a Median 243 Range 14 1000 Diagnosis Carcinoid 47 43.1 Gastrinoma 21 19.3 Insulinoma 9 8.3 Glucagonoma 1 0.9 Neuroendocrine tumor NOS 31 28.4 Treatment Surgery 30 27.5 Octreotide 5 4.6 Surgery and octreotide 11 10.0 Surgery and chemotherapy, radiotherapy, or both 2 1.8 Octreotide and chemotherapy 1 0.9 None 60 55.0 Indication for PET/CT Staging 60 55.0 Restaging 49 45.0 Note Dash ( ) indicates not applicable. NOS = not otherwise specified. a Chromogranin A values were available for 51 patients. four patients. Hence, in some patients, more than one method was used to confirm the diagnosis. Based on the reference standard, 72 patients had primary tumor and 77 patients had metastases. Gallium-68-DOTA-NOC PET/CT for Primary Tumor Gallium-68-DOTA-NOC PET/CT detected primary tumor in 57 of 72 patients who had primary tumor according to the reference standard. A total of 67 primary tumors were localized in these 57 patients. The most common site of primary tumor was the pancreas (Figs. 1 and 2) followed by duodenum (Table 2). In the remaining 15 patients, primary tumor was not localized by PET/CT. The overall sensitivity, specificity, PPV, NPV, and accuracy of 68 Ga-DOTA-NOC PET/CT for primary tumor are detailed in Table 3. Gallium-68-DOTA-NOC PET/CT findings were true-positive for primary tumor in 54 patients, true-negative in 37, falsepositive in three, and false-negative in 15. It localized primary tumor in 18 patients with carcinoid, 15 patients with gastrinoma, four patients with insulinoma, and 17 patients with NET not otherwise specified (NOS). Gallium-68-DOTA-NOC PET/CT for Metastases Gallium-68-DOTA-NOC PET/CT detected metastases in 75 of 77 patients who had one or more sites of metastases according to the reference standard. A total of 106 regions were noted in these 75 patients. The most common site of metastases on 68 Ga-DOTA-NOC PET/CT was the liver, followed by the lymph nodes (Table 2). Gallium-68-DOTA-NOC PET/CT findings were false-negative for liver metastases in two patients. The overall sensitivity, specificity, PPV, NPV, and accuracy of 68 Ga-DOTA-NOC PET/CT for detecting metastases are detailed in Table 4. Gallium-68-DOTA-NOC PET/CT was true-negative in 32 patients and there were no false-positive lesions. Semiquantitative Analysis The results of a semiquantitative analysis (i.e., SUV max ) of primary tumors and metastases are detailed in Table 5. Among the primary tumors, histologic subtype did not influence the uptake of 68 Ga-DOTA-NOC. No significant difference was found in the SUV max of carcinoid compared with gastrinoma (p = 0.571) and NET NOS (p = 0.641). Similarly, no significant difference was found between gastrinoma and NET NOS (p = 0.533). Because of the small number of insulinoma patients with positive 68 Ga-DOTA-NOC PET/ CT findings for primary tumor, insulinomas were not compared with the other histologic subtypes. However, for metastatic disease, the SUV max showed a significant difference: The SUV max of liver metastases was significantly higher than the SUV max of lymph node (p = 0.008) and bone (p < 0.001) metastases. However, no significant difference was found in SUV max between lymph node and bone metastases (p = 0.164). Overall, no significant difference existed in SUV max of primary tumor and of metastases (p = 0.695). Conventional Imaging for Primary Tumor Conventional imaging detected primary tumor in 44 of 72 patients who had primary tumor according to the reference standard. In the remaining 28 patients, primary tumor was not localized on conventional imaging. The overall sensitivity, specificity, PPV, NPV, and accuracy of conventional imaging for primary tumor are detailed in Table 3. Conventional imaging findings were truepositive for primary tumor in 44 patients, true-negative in 37, false-positive in three, and false-negative in 25. Conventional Imaging for Metastases Conventional imaging detected metastases in 63 of 77 patients who had one or more sites of metastasis according to the reference standard. A total of 77 lesions were noted in these 63 patients. The most common site of metastases on conventional imaging was the liver (n = 60), followed by the lymph nodes (n = 14) and bone (n = 3). The overall sensitivity, specificity, PPV, NPV, and accuracy of conventional imaging for detecting metastases are detailed in Table 3. Conventional imaging findings were true-negative in 32 patients and false-negative in 14. There were no falsepositive results. AJR:197, November 2011 1223

Naswa et al. Comparison of 68 Ga-DOTA-NOC PET/CT and Conventional Imaging for Primary Tumor Gallium-68-DOTA-NOC PET/CT detected primary lesions in 57 patients compared with 44 patients on conventional imaging (p < 0.001). In nine patients, additional primary sites were detected with 68 Ga-DOTA- NOC PET/CT. The total number of primary sites seen with 68 Ga-DOTA-NOC PET/CT was 67, whereas 57 primary sites were seen with conventional imaging. Fig. 1 48-year-old man with pancreatic carcinoid. DOTA-NOC = [1, 4, 7, 10-tetraazacyclododecane-1, 4, 7, 10-tetraacetic acid]-1-nai 3 -octreotide. A, Whole-body 68 Ga-DOTA-NOC PET maximumintensity-projection image shows multiple areas of increased tracer uptake throughout body. B and C, PET/CT (B) and CT (C) images show multinodular pancreas with areas of increased tracer uptake consistent with primary lesions. D and E, PET/CT (D) and CT (E) images show multiple hypodense lesions in liver with intense tracer uptake. F and G, PET/CT (F) and CT (G) images show skeletal metastasis (arrow) in sacrum that was not seen on conventional modalities. Although detection of bony metastasis by 68 Ga-DOTA-NOC PET/CT does not change management in this case, its presence is important in terms of prognosis. A Comparison of 68 Ga-DOTA-NOC PET/CT and Conventional Imaging for Metastases Gallium-68-DOTA-NOC PET/CT detected metastases in 75 patients compared with 63 patients on conventional imaging (p < 0.0001). A total of 106 metastatic regions were seen on 68 Ga-DOTA-NOC PET/ CT compared with 77 lesions on conventional imaging. In 26 patients, one or more additional metastatic sites were detected with 68 Ga-DOTA-NOC PET/CT. On a per-site B D F analysis, 68 Ga-DOTA-NOC PET/CT was superior to conventional imaging for the detection of lymph node (p < 0.0001) and bone (p = 0.002) metastases. However, no significant difference was found for the detection of liver metastases (p = 1.000). Impact of Gallium-68-DOTA-NOC PET/CT on Patient Management Of the 109 patients who underwent 68 Ga- DOTA-NOC PET/CT, findings from the study C E G 1224 AJR:197, November 2011

PET/CT of Gastroenteropancreatic NETs A C Fig. 2 39-year-old man with pancreatic gastrinoma. DOTA-NOC = [1, 4, 7, 10-tetraazacyclododecane-1, 4, 7, 10-tetraacetic acid]-1-nai 3 -octreotide. A and B, PET/CT (A) and CT (B) images show lesion in head of pancreas with liver metastasis. C and D, PET/CT (C) and CT (D) images reveal second primary lesion in jejunum. This lesion was confirmed on histopathology to be well-differentiated neuroendocrine tumor. Gallium-68-DOTA-NOC PET/CT findings changed surgical strategy management in this case by showing additional primary lesion. led to a change in the management strategy for 21 patients (19%). On the basis of the results, six patients (5.5%) underwent surgery for primary lesions that were not detected on other imaging modalities. In another eight patients (6.4%), 68 Ga-DOTA-NOC PET/CT (Fig. 3) helped in surgical planning by depicting additional surgically resectable sites. Four patients (3.6%) were spared unnecessary surgery because evidence of the advanced stage of disease was seen. It changed the treatment regimen in two patients with liver metastases and ruled out liver metastases in one patient. In addition, the detection of the expression of somatostatin receptors in these patients led to the continuation of octreotide-based treatments (32 patients), including 11 patients with progressive disease undergoing peptide-based radionuclide therapy with 177 Lu-DOTA-d-Phe(1)- Tyr(3)-Octreotate (TATE). Although continuation of treatment cannot be considered to be a management change, 68 Ga-DOTA-NOC PET/CT definitely helped in supporting the treatment decisions in these patients (29%). Discussion In the past decade, SRS using 111 In-pentetreotide (OctreoScan) has played a central role B D TABLE 2: PET/CT Findings Finding Frequency Primary Localized a,b Stomach 4 Duodenum 19 Pancreas 26 Jejunum 9 Ileum 4 Rectum or colon 3 Others 2 Not localized a 15 Metastases c Liver 59 Lymph node 33 Bone 12 Others 2 a Of total 72 patients who had primary tumor (nontreated, recurrent, residual). b In nine patients, additional primary sites were localized. c Many patients had more than one site of metastasis, whereas some had no metastasis. in the functional assessment of NETs with a high degree of sensitivity and acceptable specificity for both primary as well as metastatic gastroenteropancreatic NETs. However, problems with SPECT emanate from the low resolution of the system that hamper imaging lesions that are small and that have low receptor densities [23 25]. More recently, the development of novel PET tracers ( 68 Ga-DO- TA peptides) that bind specifically to somatostatin receptors expressed on the surface of NET cells allows the visualization of NET on 68 Ga-DOTA-peptide PET/CT scans. Several different DOTA peptides (DOTA-d-Phe(1)- Tyr(3)-Octreotide [TOC], DOTA-NOC, and DOTA-TATE) have been used in this clinical setting for either NET diagnosis or peptide receptor radionuclide therapy. The utility of 68 Ga-DOTA-NOC PET/CT in this regard ranges from diagnosis of disease extent to a preliminary procedure for evaluation of somatostatin receptor expression before the start of peptide receptor radionuclide therapy or cold somatostatin analog treatment. Several studies in the past have established the efficacy of this tracer in detecting NETs with varying results [26 29]. The current study confirms that SRS using 68 Ga-DOTA-NOC PET/CT is an effective method for the evaluation of patients with AJR:197, November 2011 1225

Naswa et al. TABLE 3: Patient-Based Diagnostic Accuracy of Conventional Imaging and PET/CT for the Detection of Primary Tumor Performance Value Conventional Imaging (%) PET/CT (%) Sensitivity 63.8 (44/69) [51.3 75] 78.3 (54/69) [66.6 87.2] Specificity 92.5 (37/40) [79.5 98.3] 92.5 (37/40) [79.5 98.3] Positive predictive value 93.6 (44/47) [82.4 98.5] 94.7 (54/57) [85.3 98.8] Negative predictive value 59.7 (37/62) [46.4 71.9] 71.1 (37/52) [56.9 82.8] Accuracy 74.3 (81/109) 83.5 (91/109) Note Data in parentheses were used to calculate performance values; data in brackets are 95% CIs. TABLE 4: Patient-Based Diagnostic Accuracy of Conventional Imaging and PET/CT for Metastases Performance Value Conventional Imaging (%) PET/CT (%) Sensitivity 81.8 (63/77) [71.3 89.6] 97.4 (75/77) [90.9 99.6] Specificity 100 (32/32) [89 100] 100 (32/32) [89 100] Positive predictive value 100 (63/63) [94.2 100] 100 (75/75) [95.1 100] Negative predictive value 69.6 (32/46) [54.2 82.2] 94.1 (32/34) [80.2 99.1] Accuracy 87.2 (95/109) 98.2 (107/109) Note Data in parentheses were used to calculate performance values; data in brackets are 95% CIs. gastroenteropancreatic NETs and can influence patient management. In the current study, 68 Ga-DOTA-NOC PET/CT showed a sensitivity of 78.3% and specificity of 92.5% for primary tumor. These values are similar to those previously published in the literature [29]. The median SUV max of primary tumor lesions was 13, and it was 14.5 for metastatic lesions. The SUV max for primary tumor in the pancreas was not significantly different from that for primary tumor in other sites (p = 0.404). Traditionally pancreatic NETs are usually well differentiated with a higher expression of somatostatin receptor 2A receptor [30, 31]. Because most primary tumors in the present study were well differentiated, this might explain the discrepancy. The sensitivity and specificity of 68 Ga- DOTA-NOC PET/CT for metastatic disease was 97.4% and 100%, respectively. The high rate of positivity obtained with in vivo 68 Ga- DOTA-NOC PET/CT is not surprising because high-affinity somatostatin-binding sites have been found in vitro on most gastroenteropancreatic endocrine tumors [32, 33]. The majority of these tumors contain high numbers of receptors homogeneously distributed throughout the tumor and expressed on both the primary and metastatic sites [32]. There were three false-positive studies for primary tumor that were later confirmed to be physiologically high uptake in the uncinate process of the pancreas. Recent reviews routinely mention this finding of increased uptake in the head of pancreas as physiologic [34]; however, it can sometimes be confusing and TABLE 5: Results of Semiquantitative Analysis on 68 Ga-DOTA-NOC PET/CT Site Median Range Primary (all) 13 1 125 Carcinoid 13 3.3 125 Gastrinoma 11 4.5 58.2 Insulinoma 29.3 23.5 48.3 Neuroendocrine tumor NOS 11.3 1 69 Metastases (all) 14.5 1.3 145 Liver 16.8 4.3 145 Lymph node 9.6 2.3 50 Bone 5.5 1.3 53 Note DOTA-NOC = [1, 4, 7, 10-tetraazacyclododecane-1, 4, 7, 10-tetraacetic acid]-1-nai 3 -octreotide, SUV max = maximum standardized uptake value, NOS = not otherwise specified. SUV max may be interpreted as positive if the uptake pattern is nontypical, as was seen in three patients in our study. Compared with conventional imaging, 68 Ga-DOTA-NOC PET/CT showed higher accuracy for both primary (83.4% vs 74.3%) and metastatic (98.2% vs 87.2%) gastroenteropancreatic NETs. Conventional imaging modalities alone were inferior to 68 Ga- DOTA-NOC PET/CT in the detection of both primary (p < 0.001) and metastatic (p < 0.0001) disease. In nine patients, additional primary sites were detected with 68 Ga- DOTA-NOC PET/CT. The total number of primary sites detected with 68 Ga-DOTA-NOC PET/CT was 67 and with conventional imaging, 57. One or more additional metastatic sites were detected with 68 Ga-DOTA-NOC PET/ CT in 26 patients. On a per-site analysis, 68 Ga- DOTA-NOC PET/CT was superior to conventional imaging for the detection of lymph node (p < 0.0001) and bone (p = 0.002) metastases. However, no significant difference was found for liver metastases (p = 1.000). In two patients with documented liver metastases on conventional imaging modalities, 68 Ga-DOTA- NOC PET/CT showed no uptake. This lack of uptake could be the result of poor differentiation of the lesions, as was shown in one of the patients in this study on fine-needle aspiration. Negative 68 Ga-DOTA-NOC PET/ CT findings are also crucial because negative 1226 AJR:197, November 2011

PET/CT of Gastroenteropancreatic NETs A Fig. 3 49-year-old woman with carcinoma of unknown primary of neuroendocrine origin. DOTA-NOC = [1, 4, 7, 10-tetraazacyclododecane-1, 4, 7, 10-tetraacetic acid]-1- NaI 3 -octreotide. A, Whole-body 68 Ga-DOTA-NOC PET maximum-intensity-projection image shows multiple areas of increased tracer uptake in liver and focal area of uptake in left side of pelvis. B E, CT (B and C) and PET/CT (D and E) images show multiple hypodense lesions in liver with intense uptake. Fine-needle aspiration from one liver lesion revealed features of neuroendocrine tumors. C and E show soft-tissue lesion (arrow) in ileal bowel loop in pelvis with focal uptake suggestive of primary lesion. This lesion was confirmed to be primary carcinoid on histopathology. findings help to select the appropriate therapy given that these lesions are known to respond to combined chemotherapy using etoposide and cisplatin [32, 35] rather than to the conventional octreotide-based treatment. In the current study, a significant overall impact of 68 Ga-DOTA-NOC PET/CT was seen on the management of patients with gastroenteropancreatic NETs. In a study that evaluated the impact of 68 Ga-DOTA-NOC PET/CT on the management of patients with NETs, Ambrosini et al. [27] found that it either affected staging or caused a therapy modification in more than half of the patients (55.5%). In our study group, there was a substantial change in the management protocol for 21 patients (19%). Six patients (5.5%) underwent surgery for primary lesions that were not detected on other imaging modalities including confirmation of a primary tumor in the duodenum suspected on conventional imaging; resection of primary tumor in two patients with unknown primary presentation; and resection of an insulinoma, a glucagonoma, and a gastrinoma that were undetected on other modalities. In eight patients, 68 Ga-DOTA-NOC PET/CT helped in proper surgical planning with regard to regional extension in view of curative surgery. In five of these patients, additional primary foci that needed to be removed for complete surgical cure were detected. In two patients, additional nodal disease was detected, which led to complete excision. Four patients (3.7%) were spared unnecessary surgery because of the detection of advanced stage of disease. In two patients with biopsy-proven liver metastases, there was no uptake of the B D radiotracer, suggesting of poorly differentiated tumor. A change in therapy from a somatostatin analog to cisplatin-based chemotherapy resulted in a significant response in these patients. In one patient with a primary pancreatic gastrinoma, surgery was deferred because liver metastases were suspected. Gallium-68-DOTA-NOC PET/CT detected the primary tumor with no evidence of liver metastases. In another 21 patients who were on a somatostatin analog, therapy was allowed to continue after the detection of somatostatin receptor expression. Eleven patients who had gradual progressive disease were selected for peptide receptor radionuclide therapy with 177 Lu-DOTA-TATE. Thus, there was an overall influence in almost half of the patients (48%) regarding management decisions (change in 19% and support in 29%). C E AJR:197, November 2011 1227

Naswa et al. A few limitations of the current study should be kept in mind. First, histopathologic confirmation of all lesions was not available. This was not technically or ethically feasible. Second, the number of insulinoma patients in this study was small, precluding any subgroup analysis. Third, we did not attempt to reevaluate the conventional imaging modalities studies. Hence, we cannot comment on retrospective visibility of any lesion that was initially missed. Conclusion Gallium-68-DOTA-NOC PET/CT appears to be a highly sensitive and specific modality in the detection of gastroenteropancreatic NET. It is better than conventional imaging for this patient population and can have significant impact on patient management. The detection of a high degree of somatostatin receptor expression can be exploited to treat patients with octreotide therapy and peptide receptor radionuclide therapy and, conversely, a negative finding on 68 Ga-DOTA-NOC PET/ CT can guide the treating physician to choose an alternate form of treatment. References 1. Modlin IM, Lye KD, Kidd M. A 5-decade analysis of 13,715 carcinoid tumors. Cancer 2003; 97:934 959 2. Sutliff VE, Doppman JL, Gibril F, et al. Growth of newly diagnosed, untreated metastatic gastrinomas and predictors of growth patterns. J Clin Oncol 1997; 15:2420 2431 3. Fraker DL, Jensen RT. Pancreatic endocrine tumors. In: DeVita VT, Hellman S, Rosemberg SA, eds. Cancer: principles and practice of oncology, 5th ed. Philadelphia, PA: Lippincott-Raven, 1997:1678 1704 4. Eriksson B, Oberg K, Skogseid B. Neuroendocrine pancreatic endocrine tumors: clinical findings in a prospective study of 84 patients. Acta Oncol 1989; 28:373 377 5. Schillaci O, Spanu A, Scopinaro F, et al. Somatostatin receptor scintigraphy in liver metastasis detection from gastroenteropancreatic neuroendocrine tumors. J Nucl Med 2003; 44:359 368 6. Madeira I, Terris B, Voss M, et al. Prognostic factors in patients with endocrine tumours of the duodenopancreatic area. Gut 1998; 43:422 427 7. La Rosa S, Sessa F, Capella C, et al. Prognostic criteria in nonfunctioning pancreatic endocrine tumours. Virchows Arch 1996; 429:323 333 8. Malagelada JR, Edis AJ, Adson MA, van Heerden JA, Go VLW. Medical and surgical options in the management of patients with gastrinoma. Gastroenterology 1983; 84:1524 1532 9. Andersen DK. Current diagnosis and management of Zollinger-Ellison syndrome. Ann Surg 1989; 210:685 703 10. Norton JA, Collen MJ, Gardner JD. Prospective study of gastrinoma localization and resection in patients with Zollinger-Ellison syndrome. Ann Surg 1986; 204:468 479 11. Norton JA, Shawker TH, Doppman JL. Localization and surgical treatment of occult insulinomas. Ann Surg 1990; 212:615 620 12. Howard TJ, Zinner MJ, Stabile BE, Passaro E Jr. Gastrinoma excision for cure: a prospective analysis. Ann Surg 1990; 211:9 14 13. Kwekkeboom DJ, Bakker WH, Kam BL, et al. Treatment of patients with gastro-entero-pancreatic (GEP) tumours with the novel radiolabelled somatostatin analogue [ 177 Lu-DOTA 0,Tyr 3 ] octreotate. Eur J Nucl Med Mol Imaging 2003; 30:417 422 14. Dromain C, de Baere T, Lumbroso J, et al. Detection of liver metastases from endocrine tumors: a prospective comparison of somatostatin receptor scintigraphy, computed tomography, and magnetic resonance imaging. J Clin Oncol 2005; 23:70 78 15. Lamberts SWJ, Hofland LJ, van Koetsueld PM, et al. Parallel in vivo and in vitro detection of functional somatostatin receptors in human endocrine pancreatic tumors: consequences with regard to diagnosis, localization and therapy. J Clin Endocrinol Metab 1990; 71:566 574 16. Krausz Y, Bar-Ziv J, de Jong R, et al. Somatostatinreceptor scintigraphy in the management of gastroenteropancreatic tumors. Am J Gastroenterol 1998; 93:66 70 17. Lamberts SWJ, Krenning EP, Reubi JC. The role of somatostatin and its analogs in the diagnosis and treatment of tumors. Endocr Rev 1991; 12:450 482 18. Schillaci O, Annibale B, Scopinaro F, delle Fave GF, Colella AC. Somatostatin receptor scintigraphy of malignant somatostatinoma with indium- 111-pentetreotide. J Nucl Med 1997; 38:886 887 19. Schillaci O, Massa R, Scopinaro F. Indium-111-pentetreotide scintigraphy in the detection of insulinomas: importance of SPECT imaging. J Nucl Med 2000; 41:459 462 20. Krenning EP, Kwekkeboom DJ, Bakker WH, et al. Somatostatin receptor scintigraphy with [ 111 In-DT- PA-D-Phe1]- and [ 123 I-Tyr3]-octreotide: the Rotterdam experience with more than 1000 patients. Eur J Nucl Med 1993; 20:716 731 21. Antunes P, Ginj M, Zhang H, et al. Are radiogallium-labelled DOTA-conjugated somatostatin analogues superior to those labelled with other radiometals? Eur J Nucl Med Mol Imaging 2007; 34: 982 993 22. Zhernosekov KP, Filosofov DV, Baum RP, et al. Processing of generator-produced 68 Ga for medical application. J Nucl Med 2007; 48:1741 1748 23. Krenning EP, Kwekkeboom DJ, de Jong M, et al. Essentials of peptide receptor scintigraphy with emphasis on somatostatin analog octreotide. Semin Oncol 1994; 21:6 14 24. Lamberts SWJ, Reubi JC, Krenning EP. Somatostatin and the concept of peptide receptor scintigraphy in oncology. Semin Oncol 1994; 21:1 5 25. Kwekkeboom D, Krenning EP, de Jong M. Peptide receptor imaging and therapy. J Nucl Med 2000; 41:1704 1713 26. Srirajaskanthan R, Kayani I, Quigley AM, Soh J, Caplin ME, Bomanji J. The role of 68 Ga- DOTATATE PET in patients with neuroendocrine tumors and negative or equivocal findings on 111 In- DTPA-octreotide scintigraphy. J Nucl Med 2010; 51:875 882 27. Ambrosini V, Campana D, Bodei L, et al. 68 Ga- DOTA-NOC PET/CT clinical impact in patients with neuroendocrine tumors. J Nucl Med 2010; 51:669 673 28. Maecke HR, Hofmann M, Haberkorn U. 68 Ga-labeled peptides in tumor imaging. J Nucl Med 2005; 46(suppl 1):172S 178S 29. Gabriel M, Decristoforo C, Kendler D, et al. 68 Ga- DOTA-Tyr 3 -octreotide PET in neuroendocrine tumors: comparison with somatostatin receptor scintigraphy and CT. J Nucl Med 2007; 48:508 518 30. Prasad V, Ambrosini V, Hommann M, Hoersch D, Fanti S, Baum RP. Detection of unknown primary neuroendocrine tumors (CUP-NET) using 68 Ga- DOTA-NOC receptor PET/CT. Eur J Nucl Med Mol Imaging 2010; 37:67 77 31. Campana D, Ambrosini V, Pezzilli R, et al. Standardized uptake values of 68 Ga-DOTA-NOC PET: a promising prognostic tool in neuroendocrine tumors. J Nucl Med 2010; 51:353 359 32. Reubi JC, Kvols LK, Waser B, et al. Detection of somatostatin receptors in surgical and percutaneous needle biopsy samples of carcinoids and islet cell carcinomas. Cancer Res 1990; 50:5969 5977 33. Reubi JC, Kvols L, Krenning E, Lamberts SWJ. In vitro and in vivo detection of somatostatin receptors in human malignant tissues. Acta Oncol 1991; 30: 463 468 34. Ruf J, Schiefer J, Furth C, et al. 68 Ga-DOTATOC PET/CT of neuroendocrine tumors: spotlight on the CT phases of a triple-phase protocol. J Nucl Med 2011; 52:697 704 35. Moertel CG, Kvols LK, O Connell MJ, Rubin J. Treatment of neuroendocrine carcinomas with combined etoposide and cisplatin: evidence of major therapeutic activity in the anaplastic variants of these neoplasms. Cancer 1991; 68:227 232 1228 AJR:197, November 2011