Abstract
The aim of this study was to evaluate the clinical value of tomographic 99mTc-octreotide acetate (hereafter, 99mTc-octreotide) scintigraphy in the detection of patients with suspected lung cancer in comparison with that of 18F-FDG dual-head coincidence imaging (DHC). Methods: Forty-four consecutive patients with suspected pulmonary neoplasms underwent tomographic 99mTc-octreotide scintigraphy and 18F-FDG coincidence imaging using the same gantry. The region of interest was drawn on the entire primary lesion. The tumor-to-normal tissue tracer values for both 99mTc-octreotide and 18F-FDG were determined using region of interests and expressed as T/Nr and T/Nm, respectively. Final diagnosis was confirmed by histopathologic analysis or clinical follow-up. Results: Thirty-one of the 44 patients had lung cancer—6 with small cell lung cancer (SCLC) and 25 with non–small cell lung cancer (NSCLC). Thirteen of the 44 patients had benign lung lesions. The sensitivity, specificity, positive predictive value, and negative predictive value of 99mTc-octreotide were 100%, 75.7%, 90.1%, and 100%, respectively, and of 18F-FDG DHC were 100%, 46.1%, 83.8%, and 100%, respectively. In the 31 patients with malignant tumors, all 38 abnormal lymph nodes in 20 patients showed abnormal high focal uptake of 18F-FDG; only 7 patients with 10 regional lymph adenopathies showed moderate uptake of 99mTc-octreotide. Thirteen patients with 39 distant sites of abnormal uptake visualized (imaging stage IV) with 99mTc-octreotide included 2 patients with brain metastases, 6 patients with pleural invasion and multiple bone metastasis, 2 patients with contralateral internal lung metastasis and pleural invasion, and 3 patients with only multiple bone metastasis. The final diagnosis was confirmed by histopathology or clinical follow-up. Conclusion: The sensitivity of 99mTc-octreotide for the detection of lung cancer at the primary lesion was comparable with that of 18F-FDG coincidence imaging. Tomographic 99mTc-octreotide scintigraphy had lower sensitivity for the detection of hilar and mediastinal lymph node metastasis compared with that of 18F-FDG coincidence PET, but it had high sensitivity for the detection of remote metastatic lesions. However, because of the small population, further investigation is necessary.
Lung cancer is the leading cause of cancer death in both developed and developing countries, including China (1,2). The overall 5-y survival rate in patients with lung cancers after treatment is about 15% (1,3). Two thirds of patients with non–small cell lung cancer (NSCLC), which accounts for 80% of lung cancers, were diagnosed at the unresectable advanced stage (4,5).
The conventional noninvasive diagnostic modalities currently used in evaluating lung tumors—chest x-ray, CT, and sputum cytology—have a high percentage of indeterminate diagnosis (6,7). 18F-FDG PET and PET/CT have been widely studied in the detection of lung cancers for diagnosis, staging, and restaging (8–16). However, in developing countries, the high cost of this equipment and the lack of general availability have limited their widespread clinical use.
The affinity of receptor analogs such as somatostatin (SST) in some malignant lung tumors has been demonstrated (17–20). The aim of this study was to determine the efficacy of the SST receptor ([SSTR] octreotide acetate; hereafter, octreotide) imaging in diagnosing patients with suspected lung cancer in comparison with that of 18F-FDG coincidence imaging in the same consecutive 44 patients.
MATERIALS AND METHODS
Patients
The Institutional Review Board of the Nanjing Medical University as well as the local ethics committee approved this investigation. Written consent was obtained from all patients. Forty-four consecutive patients with suspected lung cancer (13 women, 31 men; age range, 44–83 y; average age, 62 ± 10 y [mean ± SD]) were prospectively enrolled in this study. Conventional methods, such as plain chest x-ray and chest CT were performed on all patients, who either refused or could not endure an invasive technique such as bronchoscopy or transthoracic biopsy. All patients underwent 99mTc-octreotide scintigraphy and 18F-FDG coincidence imaging. The final diagnosis and stage were confirmed by histopathologic findings or clinical follow-up.
99mTc-Octreotide Image Acquisition and Interpretation
Ten micrograms of octreotide (Sandostatin; Novartis Pharma Schweiz AG) were labeled with 99mTc-pertechnetate in a 1-mL volume. The mean radiochemical purity ± SD was 95.1% ± 2.9%. 99mTc-Octreotide (1,110 MBq [30 mCi]) was administrated intravenously, and planar whole-body static images were acquired about 4 h later. Immediately thereafter, SPECT/CT images of the chest and the upper abdomen, including the adrenal glands (Millennium VG, Hawkeye; GE Healthcare), were obtained. SPECT data were acquired in a 128 × 128 matrix through 360° rotation with 64 projections. The acquisition time for each projection was 40 s. Transverse slices were reconstructed with an ordered-subset expectation maximization (OSEM) iterative algorithm and formatted as a 128 × 128 matrix. The CT images were used for attenuation correction of the SPECT image datasets and for better anatomic delineation of the SPECT images.
The SPECT/CT images of the thorax and upper abdomen were displayed in the 3 orthogonal planes on the camera workstation and were interpreted separately and independently by 3 nuclear medicine physicians who were unaware of the clinical diagnosis and the results of other imaging tests. Focal areas of increased tracer uptake compared with adjacent lung tissue activity were considered positive for malignancy. Lesions with radiotracer uptake less than or equal to adjacent or surrounding lung tissue activity were interpreted as benign. Hilar and mediastinal foci with activity greater than the adjacent mediastinal tissue were interpreted as abnormal lymph adenopathy. Regions of interest (ROIs) were drawn on the entire primary lesion and the contralateral normal lung tissue; tumor uptake of 99mTc-octreotide activity was estimated by using ROIs that were drawn around the primary lesions, and tumor-to-normal tissue activity (T/Nr) ratios were generated.
18F-FDG Coincidence Image Acquisition and Image Interpretation
One day after tomographic 99mTc-octreotide imaging, 18F-FDG dual-head coincidence imaging (DHC) was performed 60 min after intravenous injection of 18F-FDG (296–370 MBq). Each patient had fasted for at least 6 h prior to the injection. Emission data were acquired for 10 rotations of 360° in 1.8 min per revolution using hybrid SPECT/CT with a coincidence system (Millennium VG, Hawkeye; GE Healthcare). Images from the base of the skull to the pelvic floor were acquired with 2 separate tomographic examinations. Brain images were acquired separately. Low-dose CT (2.5 mA) was acquired for 40 revolutions of 360° in 15 s per revolution for attenuation correction and image fusion. Data were reconstructed with an ordered-subset expectation maximization (OSEM) iterative algorithm. Functional, anatomic, and fusion images were obtained; all images were formatted to a 128 × 128 matrix. The 18F-FDG images were displayed in the 3 orthogonal planes on the camera workstation.
Images were analyzed visually by the same 3 experienced nuclear medicine physicians who had analyzed the 99mTc-octreotide images and were unaware of the final clinical diagnosis. Interpretation was based on attenuation-corrected images. Any focal activity in a known lesion with an intensity higher than the normal mediastinal activity was considered malignant. Any hilar or mediastinal focal increased activity greater than that of the surrounding mediastinal activity was regarded as a metastatic lymph node. Any focal distant increased activity that did not correspond to a normal physiologic structure and had increased intensity was considered as a metastatic lesion. Tumor-to-normal tissue uptake ratios of 18F-FDG were also calculated for the primary lesion (T/Nm).
Statistical Analysis
The data were analyzed and presented as the mean ± SD. The sensitivity, specificity, negative predictive value (NPV), and positive predictive value (PPV) were estimated for 99mTc-octreotide and 18F-FDG images using SPSS software, version 10.0. The McNemar test was used to assess the 18F-FDG coincidence PET images. The efficacy of tomographic 99mTc-octreotide for differentiation of lung cancer from benign lesions was determined with the Student t test.
RESULTS
The demographics, histopathology, and staging of the patients are provided in Table 1.
Clinical Information and Histopathologic Database Used in Study
Primary Lung Lesion
Of the consecutive 44 patients, 31 (70.5%) had malignant neoplasms identified by histopathologic findings—6 patients with small-cell lung cancer (SCLC) and 25 patients with NSCLC. Thirteen of the 44 patients had benign lesions (Table 1). Tissues for histopathology were obtained by thoracic surgery in 18 patients, with fine-needle aspiration in 12 patients, and during bronchoscopy in 8 patients. The benign nature of the lesions was confirmed on clinical follow-up in 6 patients. One patient with a benign clear cell tumor showed a high uptake of 18F-FDG but no obvious uptake of 99mTc-octreotide (Fig. 1). Pleural invasions in 2 patients were clearly visualized on 99mTc-octreotide and 18F-FDG images (Figs. 2 and 3).
A 46-y-old male patient with benign clear cell tumor. (A) CT scans showed abnormality in right upper lobe. (B) 18F-FDG coincidence images had intense uptake in lesion. (C) 99mTc-Octreotide images were negative in lesion.
A 75-y-old male patient with adenosquamous lung cancer in left upper lobe. (A) CT scans showed abnormality in left upper lobe. (B) 18F-FDG coincidence images had intense uptake in lesion (arrows). (C) 99mTc-Octreotide images showed intense uptake in primary tumor and pleura (arrows). (D) Adenosquamous lung cancer and pleural invasion were verified on histology.
A 57-y-old male patient with squamous lung cancer. (A) CT scans showed neoplasm high in upper lobe of right lung. (B) 18F-FDG coincidence images had focal uptake in lesion. (C) 99mTc-Octreotide images had focal high uptake in lesion. (D) Squamous lung cancer and pleura and rib invasion were verified on histology.
The median size of the lesions (average of the 2 longest dimensions on the chest CT) was 4.8 cm (range, 1.0–11.0 cm).
PPVs and NPVs of 99mTc-octreotide scintigraphy and 18F FDG coincidence imaging for the diagnosis of primary lesions are shown in Table 2. The sensitivity and specificity of the detection of primary lesions were not statistically significant between these 2 imaging modalities (χ2 = 0.72; P = 0.395).
Efficacy of 99mTc-Octreotide for Detection of Primary Lesion and Lymph Node Involvement Compared with 18F-FDG
The T/Nr for octreotide was 3.03 ± 1.09, and the T/Nm for FDG was 5.51 ± 2.13. The 99mTc-octreotide tumor uptake correlated with that of 18F-FDG uptake (r = 0.66; P < 0.0001). The uptake of 99mTc-octreotide in benign lesions was 2.03 ± 1.02 and was significantly less than that for malignancy (3.45 ± 0.73; t = 4.82; P < 0.001).
Metastatic Lymph Adenopathy
The staging status and clinical information in the 31 patients with histopathologically documented lung cancer are shown in Table 3. In 20 patients, all of the 38 regional lymph node metastatic lesions showed a high degree of 18F-FDG uptake. However, only 7 of those 20 patients with metastatic lymph node adenopathy were detected on 99mTc-octreotide scintigraphy. The sensitivity of 99mTc-octreotide for the detection of lymph node metastases was low—35% compared with 18F-FDG coincidence imaging (100%).
Demographic and Clinical Staging of Study Population
Distant Metastatic Lesions
There were 39 distant metastatic sites in 13 patients, which were visualized by both 99mTc-octreotide tomography and 18F-FDG coincidence imaging. Two patients had brain metastases, 6 patients had pleural invasion and multiple bone metastases, 2 patients had contralateral internal lung metastases and pleural invasion, and 3 patients had only multiple bone metastases. The 2 patients with brain metastases had intense uptake of 99mTc-octreotide: 1 patient showed decreased uptake of 18F-FDG, as demonstrated in Fig. 4, whereas the other patient showed intense focal uptake of 18F-FDG. Both patients had a high accumulation of 99mTc-octreotide (Fig. 4). All of these metastatic lesions were confirmed by histopathology or other noninvasive imaging modalities such as bone scan, CT, and MRI.
A 65-y-old woman with lung adenocarcinoma. (A) CT scans revealed brain metastasis and peripheral edema in left temporal lobe. (B) 18F-FDG images showed low uptake in brain lesion. (C) 99mTc-Octreotide images showed focal uptake in brain lesion.
DISCUSSION
Conventional noninvasive imaging modalities, such as plain chest x-ray and CT, continue to play an important role in the detection of lung cancer but often fail to distinguish between malignant and benign lung tumors (21). Definitive diagnosis and staging have traditionally depended on more invasive techniques. However, even these invasive procedures—such as bronchoscopy and transbronchial or transthoracic biopsy—may fail to provide a conclusive diagnosis (22,23), despite their known potential complications (24).
SST and synthesized longer-acting analogs such as octreotide and lanteotide have many predominantly inhibitory hormonal functions on the gastrointestinal tract and nervous system as well as direct and indirect antineoplastic effects. SSTRs are integral membrane glycoproteins that are distributed in a variety of tissues. Five subtypes have been identified and cloned. Alteration of SSTR expression during disease states—such as their overexpression in many neoplasia—can be exploited by imaging techniques (18,25,26). SSTR scintigraphy with 111In-labeled SST analogs has been well documented, especially in neuroendocrine tumors (26–28).
The fact that SSTRs are overexpressed in many malignant tumors provides the basis for differentiating malignant tissues from other tissues by nuclear imaging using SSTR-binding radiotracers. Several reports have confirmed the presence of SSTRs on NSCLC tumors and cell lines (29–31). Other studies have shown that SST analog scintigraphy labeled with 111In- and 99mTc-pertechnetate can be used for the detection of solitary pulmonary nodules (19,20,32). 99mTc-Depreotide (NeoTect; Berlex Laboratories) is one example.
Octreotide is also a synthetic analog of SST, which is available in most hospitals. A 1-step labeling reaction, without additional purification, results in a high, stable radiochemical yield with 99mTc-pertechnetate (99mTc-octreotide); the radiochemical purity was >98%. No complications or side effects were observed, and the labeling method is very simple.
In this study, 44 clinically indeterminate patients with lung neoplasms were enrolled, and the clinical value of 99mTc-octreotide scintigraphy in the detection of lung cancer was evaluated and compared with that of 18F-FDG DHC.
Our results demonstrated that 99mTc-octreotide scintigraphy is a sensitive test for the detection of primary lung cancer, as reported previously (32,33). The sensitivity of tomographic 99mTc-octreotide scintigraphy for the detection of primary lung cancer is similar to that of 18F-FDG coincidence imaging. However, the specificity and the PPVs of tomographic 99mTc-octreotide scintigraphy are higher than those of 18F-FDG coincidence PET. Despite the high sensitivity of 18F-FDG PET, false-positive findings—such as in patients with inflammatory processes—might hamper proper patient management (34). 99mTc-Octreotide scintigraphy does not seem to be affected as much by inflammatory processes. In our study, 1 patient with a benign clear cell tumor, 2 patients with granulomas, and another patient with tuberculosis showed no focal accumulation of 99mTc-octreotide, despite the presence of high focal 18F-FDG activity. Only half of the patients with tuberculosis (3/6) showed increased activity on 99mTc-octreotide scans. One might consider the combined use of these 2 radionuclides to decrease the false-positive rate of 18F-FDG scans. However, the discrepancy between these 2 diagnostic techniques could not be evaluated further because no patient with a carcinoid tumor was found in this relatively small group.
With respect to metastatic hilar and mediastinal nodes, a large meta-analysis indicated that 18F-FDG PET is a relatively accurate method for assessing the status of the mediastinal lymph node regions (35). Our results revealed that 18F-FDG coincidence imaging was more accurate than 99mTc-octreotide scintigraphy for the detection of mediastinal and hilar lymph node involvement. For example, all 20 patients with a total of 38 hilar and mediastinal metastatic lymph nodes had increased 18F-FDG activity in their lesions on 18F-FDG coincidence images. However, only 7 patients with 10 of the hilar and mediastinal nodes had mild-to-moderate 99mTc-octreotide uptake; the sensitivity for the detection of hilar and mediastinal lymph node metastasis was 35%, significantly lower than that for 18F-FDG coincidence imaging. This low sensitivity was due to the relatively high background of 99mTc-octreotide in the mediastinum and hilum of lung compared with that of 18F-FDG.
On the other hand, the sensitivity of 99mTc-octreotide scintigraphy for the detection of remote metastatic lesions was high. Thirty-nine distant metastatic sites in 13 patients were visualized by both 99mTc-octreotide and 18F-FDG imaging. These 13 patients included 2 with brain metastasis, 6 with pleural invasion and multiple bone metastasis, 2 with contralateral internal lung metastasis and pleural invasion, and 3 with only multiple bone metastasis. Other trials have reported that 18F-FDG PET could detect unexpected distant metastasis in approximately 13% of patients, which had a large impact on patient management (36,37). Our study suggested that 99mTc-octreotide scintigraphy was also valuable for the detection of distant metastatic disease, which led to further diagnostic investigation.
In the present study, we also noted that the resolution of the instrument used was limited compared with that of dedicated PET/CT. However, most patients were in advanced stages of disease, and the size of the primary lesion was relatively large in this study, which might have resulted in a high sensitivity of 99mTc-octreotide and 18F-FDG in the detection of the primary lesion.
This prospective study confirmed that 99mTc-octreotide tomography had a high sensitivity for the detection of the primary lesion and distant metastasis in patients with suspected lung cancer, whereas it had a low sensitivity for the detection of mediastinal and hilar lymph node metastasis. However, 99mTc-octreotide tomography might be an alternative method for the diagnosis of lung cancer, especially in developing countries, where the availability of PET/CT or dedicated PET is limited because of the high cost.
CONCLUSION
Tomographic 99mTc-octreotide scintigraphy is an effective noninvasive technique for diagnosing lung cancer, especially for the detection of primary lesions. Its sensitivity and specificity for the primary lesions are comparable to those of 18F FDG coincidence imaging. Although 99mTc-octreotide has a high sensitivity for the detection of remote metastatic lesions, it has a lower sensitivity for the detection of hilar and mediastinal lymph node metastasis. Further investigation is needed to confirm the results of this relatively small population study.
Acknowledgments
We are grateful to Dr. Xiu-Jie Liu (Cardiovascular Institute and Fu Wai Hospital, Beijing, China) for valuable advice and assistance in revision of this article. This research was supported by Jiangsu Government Science grant BS2004507 and Nanjing Health Bureau grant zx0214, China.
Footnotes
-
COPYRIGHT © 2007 by the Society of Nuclear Medicine, Inc.
References
- Received for publication February 12, 2007.
- Accepted for publication May 31, 2007.