Research ArticleCLINICAL INVESTIGATIONS

Imaging δ- and μ-Opioid Receptors by PET in Lung Carcinoma Patients

Igal Madar, Badredin Bencherif, John Lever, Richard F. Heitmiller, Stephen C. Yang, Malcolm Brock, Julie Brahmer, Hayden Ravert, Robert Dannals and James J. Frost
Journal of Nuclear Medicine February 2007, 48 (2) 207-213;

This article has a correction. Please see:

Abstract

In the present study, we measured the kinetics and distribution in vivo of the selective δ-opioid antagonist 11C-methylnaltrindole (11C-MeNTI) and the μ-opioid agonist 11C-carfentanil (11C-CFN) in patients with lung carcinoma using PET. Methods: Paired measurements of 11C-MeNTI and 11C-CFN binding were performed in biopsy-proven small-cell (n = 2), squamous (n = 2), and adenocarcinoma (n = 3) lung cancer patients. Dynamic PET scans of increasing duration (0.5–8 min) were acquired over 90 min after an intravenous bolus injection of 370 MBq of tracer. Time–activity curves for tumor and normal lung parenchyma binding were generated using the region-of-interest (ROI) method. The mean activity at equilibrium was measured, and the specific-to-nonspecific binding ratio (tumor − lung)/lung was calculated. Four of 7 patients underwent an additional static 18F-FDG PET scan for clinical indications. Three of 7 patients underwent surgery, and stained sections of tumor were inspected for inflammation, necrosis, and scar tissue. Results: Increased binding of 11C-MeNTI and 11C-CFN was detected in all tumor types studied. 11C-MeNTI binding in tumor and healthy lung tissue was significantly more intense than that of 11C-CFN. The average specific-to-nonspecific binding ratio across cell types for 11C-MeNTI (4.32 ± 1.31; mean ± SEM) was greater than that of 11C-CFN (2.42 ± 1.17) but lower than that of 18F-FDG (7.74 ± 0.53). Intravenous naloxone produced 50% and 44% decreases in the specific-to-nonspecific binding ratios of 11C-MeNTI and 11C-CFN, respectively. Conclusion: These data provide in vivo evidence for the presence of δ- and μ-opioid receptor types in the 3 major human lung carcinomas and suggest the suitability of 11C-MeNTI and 11C-CFN as investigational probes of lung carcinoma biology.

Lung carcinoma continues to be a major health problem of high incidence and high mortality rate, surpassing the mortality of colon, breast, and prostate cancers combined (1). Definitive surgery is the most effective treatment for patients with localized, resectable lung tumors. The extent of disease, as determined by tumor staging, is critical in determining patient treatment options and prognosis. Noninvasive imaging has a key role in preoperative evaluation of disease staging. In recent years, major efforts have been made to identify molecular targets in malignant cells and to develop selective probes as diagnostic tools for detection of solid tumors by PET.

In vitro binding assays have shown that small-cell lung carcinoma (SCLC) and non-SCLC cell lines express multiple membrane receptors of μ-, δ-, and κ-types (2–6). Investigation of a series of lung cancer cell lines of diverse histologic types suggests that δ-opioid receptors are more abundant than the μ-receptor (3). Other studies show a differential distribution of opioid receptor types in lung carcinoma cells. Reverse-transcription polymerase chain reaction of the GLC8 SCLC cell line has found expression of only the δ-opioid receptor (6). Reversible binding of the δ-antagonist [125I][H-Tyr(3′-125I)-Tic(ψ)[CH2NH]Phe-Phe-OH] (ITIPP) was demonstrated in tumor-bearing mice (7). Saturable, high-affinity binding sites for μ- and δ-agonists were found in the rat lung parenchyma (8). The opioid receptor density in the pulmonary tissue was <10% of that reported for lung carcinoma cell lines.

Over the last 2 decades, our laboratory has developed and characterized several positron-emitting opioid ligands, including the 2 high-affinity PET tracers 11C-carfentanil (11C-CFN), a μ-opioid receptor agonist, and 11C-methylnaltrindole (11C-MeNTI), a δ-opioid receptor antagonist (9–11). MeNTI is highly selective for the δ-site (inhibitory concentration = 0.02 nmol/L), it is 700-fold selective for δ- over μ-sites, and it is >3,000-fold selective for δ- over κ-sites (12). In mouse brain, 11C-MeNTI binding is inhibited selectively by δ-ligands, but not by μ- or κ-ligands, and the regional distribution correlated well with the established δ-receptor density in vitro (13). 11C-MeNTI selectivity for δ-sites was validated in human brain by using quantitative PET (11). 11C-CFN is a selective agonist with high affinity for μ-opioid receptors in animal and human brain (9,10). 11C-MeNTI and 11C-CFN have been used to localize and quantitate alterations in regional distribution of opioid receptor types associated with various brain disorders (14–18).

The objective of the present study was to characterize the biodistribution of 11C-MeNTI and 11C-CFN in the primary tumor and lung tissue in biopsy-proven lung cancer patients with SCLC and non-SCLC of the squamous and adenocarcinoma cell types.

MATERIALS AND METHODS

Patient Characteristics

Seven consecutive newly diagnosed patients (3 male, 4 female; mean age ± SD, 69.1 ± 6.6 y) with biopsy-proven tumor of SCLC or non-SCLC type, tumor size ≥ 2 cm on CT, and no opiate therapy 1 wk or more before the PET studies were entered into the study (Table 1). Informed consent, using guidelines established by the Johns Hopkins University School of Medicine Joint Committee on Clinical Investigation was obtained from each subject before the imaging studies.

View this table:
TABLE 1

Patient Characteristics

Among the 7 patients, there were 2 squamous cancers, 3 adenocarcinomas, and 2 SCLCs (Table 1). Two patients (patients 1 and 7) had a resectable tumor with an extensive chest wall extension. Two patients (patients 2 and 3) had a resectable tumor confined to the lung parenchyma. Two patients (patients 4 and 5) had a nonresectable tumor due to pleural invasion, metastatic disease, and clinical conditions. One patient (patient 6) had a nonresectable SCLC tumor with tumor mass extending from the aortic arch to the main pulmonary artery.

PET Acquisition

Two PET studies were performed on each patient. For 5 subjects, a 11C-MeNTI PET scan, followed by a 11C-CFN PET scan, was acquired. A 1-h rest period between scans was allowed. For 2 subjects, an inhibition study using the opioid antagonist naloxone was performed. The baseline study was followed by the inhibition study with a 1-h rest period between scans. One subject underwent the baseline 11C-MeNTI scan and administration of naloxone (1.0 mg/kg, intravenously) 5 min before commencement of the inhibition scan. The second subject underwent paired 11C-CFN PET scans with administration of 0.16 mg/kg naloxone before commencement of the inhibition scan.

Four of the 7 patients underwent a third PET study of tumor metabolic activity using 18F-FDG. Metabolic imaging was conducted within 3 d before or after opioid receptor studies. There was no treatment during the period between the opioid and metabolic studies. Subjects were positioned in the scanner using the PET laser with the tumor in the center of the scanner field of view (FOV). Positioning of subjects was aided by the anatomic information obtained by the chest CT, by measuring the distance between the sternal notch and the center of the tumor mass.

11C-MeNTI was synthesized via the reaction of 11C-methyliodide with O-benzyl-naltrindole (13) (injected dose, 640.1 ± 85.1 MBq; specific activity, 83.4 ± 31.4 GBq/μmol). 11C-CFN was synthesized via the reaction of 11C-methyliodide with the normethyl precursor (19) (injected dose, 712.6 ± 26.8 MBq; specific activity, 177.7 ± 52.7 GBq/μmol). The mean injected dose of 18F-FDG was 373.1 ± 56.6 MBq.

PET data were acquired on an Advance scanner (GE Healthcare; 35 slices, 4.25-mm thick, 14.4-cm FOV). The patients were placed on the scanner bed, and the PET laser was positioned on the sternal notch; then the bed was moved along the distance measured on the CT scan for centering the tumor in the scanner FOV. A 10-min transmission scan using a 68Ge/68Ga source was performed for subsequent attenuation correction. After intravenous administration of the radiotracer, a series of 25 serial PET images was acquired. The image duration increased progressively from 30 s to 8 min until 90 min after injection. The whole-body 18F-FDG PET scan, commencing 30 min after a bolus injection of the tracer, was performed in 6 bed positions, 8 min per bed position. Subjects were monitored continuously throughout the scanning period for correction of any deviation.

PET Scan Quantification

Scans were corrected for radioactivity decay and for attenuation by using the acquired transmission scan. Reconstruction of images was accomplished by ordered-subset expectation maximization using a 26 × 26 cm FOV, a 256 × 256 pixel matrix (2 × 2 × 4.5 mm voxel size), and a final cubic resolution of 7.8 mm.

Serial images from 0 to 90 min were summed to produce an image for placement of regions of interest (ROIs). The summed images were resampled using linear interpolation to a cubic voxel of 2 mm3. Images were presented in the coronal plane, and a rectangular ROI (8 × 8 mm) was drawn at the center of the tumor activity profile. The slice representing peak activity in the tumor was selected for further analysis. In some tumors a central area of reduced binding was present. Its diameter was measured on PET from the activity profile across the longitudinal axis as the distance between voxels with value of 50% the difference between tumor maximum activity and mean activity of the surrounding tissue. In 1 subject (patient 6), having a large tumor (6 × 8 × 4 cm), the area of increased receptor binding was interrupted, and tracer tumor uptake was sampled by 3 ROIs. In the remaining subjects, a continuous distribution of radioligands throughout the tumor area was observed, and tumor tracer uptake was sampled by a single ROI. Nontumor radioactivity was quantified by placing a rectangular ROI on the healthy lung parenchyma above and below the tumor mass and on the contralateral lung in a position corresponding to the ROIs ipsilateral to the malignant lesion (total number of ROIs = 6). The ROI template was transferred to the serial images of the corresponding PET section, and time–activity curves of tracer in tumor and healthy lung tissue were plotted. Image-processing procedures were performed using Analyze software (Mayo Clinic, Rochester, MN).

The time–activity curves were used to determine time-to-plateau activity. In all subjects, 11C-MeNTI and 11C-CFN activity in tumor plateaued over the 50- to 90-min postinjection time interval (Fig. 1). For each ROI, the mean radioactivity over the 50- to 90-min time interval was calculated from the corresponding time−radioactivity curves. The ROIs' counts/cm3/min were normalized to the injected dose (Bq/cm3/MBq injected). In addition, counts were translated to standardized uptake value (SUV) per body weight ([kBq/cm3/MBq] × kg).

FIGURE 1. 
FIGURE 1. 

Kinetics of 11C-MeNTI and 11C-CFN tumor binding in non-SCLC patient. (A) Radioligand activity in tumor (MeNTI, ▴; CFN, •) and host lung tissue (MeNTI, ▵; CFN, ○). Activity in healthy lung represents mean of 5 ROIs. (B) Specific-to-nonspecific binding ratio. 11C-MeNTI activity (▴) in tumor is significantly greater than that in 11C-CFN (•). 11C-MeNTI and 11C-CFN binding reaches plateau in tumor but decreases gradually in nonmalignant lung tissue; consequently, specific-to-nonspecific binding ratios for both radioligands increase continuously.

In some tumors, a necrotic core was present. An assessment of the core diameter was obtained by generating the activity profile across the tumor long axis and surrounding healthy tissue. The core boundaries were determined as 50% of the difference between the minimum activity in the tumor area and the mean activity in the normal surrounding tissue. The Student t test was used to test the hypothesis of significant differences between 11C-MeNTI and 11C-CFN binding and between tumor and nontumor structures. Differences with P < 0.05 were judged as significant.

RESULTS

Tumor Characteristics

The extent of inflammation, necrosis, and scarring in the tumor area was evaluated on 7-μm hematoxylin- and eosin- stained histologic sections of surgical specimens of 3 patients, as summarized in Table 2. One patient (patient 3) demonstrated a focal central scar (0.5 cm) and desmoplasia with cystic changes accompanied with arterial occlusion by tumor thrombosis. One patient (patient 1) demonstrated intermediate chronic as well as intense acute inflammation in some regions with a focal abscess formation; a 2-cm scar in lung parenchyma surrounded by viable tumor cells and obliteration of large-caliber lobar arteries were observed in the scar region. One patient (patient 7) demonstrated extensive necrosis with zones of viable cells measuring up to 0.5 cm at the tumor margin.

View this table:
TABLE 2

Histopathology of Lung Carcinoma

11C-MeNTI and 11C-CFN Binding Kinetics

An example of a time–activity profile for 11C-MeNTI and 11C-CFN binding in tumor and healthy lung parenchyma obtained in SCLC tumor (patient 6) is shown in Figure 1A. In tumor, 11C-MeNTI and 11C-CFN binding reach a plateau within a few minutes, which is maintained throughout most of the scanning period (90 min). 11C-MeNTI plateaus at significantly greater values than does 11C-CFN. In some tumors, binding kinetics were slower than those observed in Figure 1 but, in all subjects, plateau activity was obtained during the 50- to 90-min period. In contrast, 11C-MeNTI and 11C-CFN activity in healthy lung parenchyma, after an early peak, gradually washed out throughout the scanning time (Fig. 1A). A washout of activity from the healthy pulmonary tissue was observed in all subjects for both 11C-MeNTI and 11C-CFN. Figure 1B depicts the specific-to-nonspecific activity ratio, calculated as (tumor − lung)/lung, which increased linearly with time for both 11C-MeNTI and 11C-CFN.

Inhibition Study

The opioid antagonist naloxone produced a significant decrease of 11C-MeNTI and 11C-CFN tumor binding. Figure 2 demonstrates the effect of intravenous administration of naloxone on the time–activity profiles of 11C-MeNTI in adenocarcinoma primary tumor and healthy pulmonary tissue (patient 4). In the absence of naloxone (Fig. 2A), 11C-MeNTI binding in tumor rapidly plateaued, whereas binding in healthy lung parenchyma gradually declined. In contrast, intravenous administration of naloxone produced a significant washout of radioligand from tumor at a rate similar to that observed in the healthy lung parenchyma. Naloxone did not affect 11C-MeNTI binding kinetics in the healthy lung tissue (Fig. 2B). A similar effect of naloxone on binding kinetics of 11C-CFN in tumor and healthy lung parenchyma was observed. Administration of naloxone at a dose of 1.0 mg/kg resulted in a 50% decrease of the specific-to-nonspecific binding ratio for 11C-MeNTI, whereas a lesser dose (0.16 mg/kg) was required to exert a comparable inhibition (44%) of 11C-CFN binding ratio.

FIGURE 2. 
FIGURE 2. 

Effect of intravenous naloxone on 11C-MeNTI kinetics in lung adenocarcinoma (patient 4). Traces represent activity profile of 11C-MeNTI in tumor (▴) and healthy lung tissue (○) in absence (A) and presence (B) of naloxone. Nontumor represents mean of 5 samplings, using rectangular (8 × 8 mm) ROI, of activity in healthy lung parenchyma. Naloxone produces washout of binding in tumor but does not affect pharmacokinetics in normal lung.

Tumor Uptake

Elevated tumor binding of 11C-MeNTI and 11C-CFN was detected in all but 1 subject (patient 7) with a large necrotic core in tumor (Table 3). The mean 11C-MeNTI tumor's SUV, across subjects, was 7.4 ± 0.85 (mean ± SEM) compared with 1.8 ± 0.44 in the healthy lung tissue. 11C-CFN binding in tumor was significantly (P = 0.003) lower than that of 11C-MeNTI. The average 11C-CFN SUVs in tumor and host lung were 2.7 ± 0.72 and 0.91 ± 0.25, respectively. Consequently, the mean specific-to-nonspecific binding ratio, was greater for 11C-MeNTI (4.32 ± 1.30) than that for 11C-CFN (2.42 ± 1.17). The mean 18F-FDG SUVs in tumor and host lung were 5.7 ± 1.29 and 0.66 ± 0.14, respectively (Table 3).

View this table:
TABLE 3

Tumor and Lung Binding of 11C-MeNTI, 11C-CFN, and 18F-FDG

The distribution of receptor binding across the tumor area was not uniform. In 3 patients (patients 1–3), intense opioid receptor binding was confined to the margin of the tumor, composing a ring of increased binding surrounding a central cold area. In 2 other patients (patients 4 and 6), regions of increased receptor binding were distributed in discrete patches throughout the tumor mass. Figure 3 demonstrates a coronal view of PET images acquired from a patient with a squamous lung carcinoma tumor (patient 1). The tumor mass was localized in the lateral upper left lobe invading the chest wall. 11C-MeNTI and 11C-CFN binding comprises a rim of increased activity in the medial aspect of the tumor with no binding in the tumor core or in the tumor lateral aspect of the tumor bordering the chest wall.

FIGURE 3. 
FIGURE 3. 

Coplanar coronal sections of 11C-MeNTI (A), 11C-CFN (B), and 18F-FDG (C) PET images acquired in a patient with squamous lung carcinoma (patient 1): an 8 × 3 × 3 cm tumor, localized in lateral aspect of upper left lobe, extending to chest wall. Increased 11C-MeNTI and 11C-CFN binding forms an arch surrounding a cold center. Quantitative values obtained at the peak activity profile of the tumors are presented in Table 3. All 3 tracers demonstrate reduced binding in lateral aspect of the tumor bordering the chest wall, but this decrease is more pronounced for the δ-opioid receptors. Likewise, all 3 tracers demonstrate reduced activity at the tumor center, but this area is larger for the opioid agents. 11C-MeNTI binding in tumor and healthy lung is more intense than that of 11C-CFN. L = lung; H = heart; T = tumor.

In 4 patients, the distribution of the opioid receptor binding was compared with the tumor 18F-FDG uptake. In 2 of 4 patients (patients 1 and 2), increased uptake of 18F-FDG was confined to the tumor margin, similar to the distribution pattern observed by opioid receptor studies (Fig. 3). However, the central core of decreased receptor binding was larger than the metabolic cold center. In the example shown in Figure 3, the cold center diameter in the long axis is 38 and 13 mm for 11C-MeNTI and 18F-FDG studies, respectively. The opioid receptor cold spot was also larger than the necrotic core, as measured on histologic sections and the cavity observed on the CT scans. In 2 patients, a necrotic core of 20 and 5 mm was measured on CT, whereas the opioid receptor cold spot was 38 and 10 mm in diameter. 11C-MeNTI and 11C-CFN binding sites demonstrated similar distribution across the malignant tissue, and the region of reduced binding was of a similar diameter.

In 3 patients, tumor histologic sections were inspected for the presence of tumor infiltrating macrophages and lymphocytes and visually compared with the distribution of 11C-MeNTI and 11C-CFN. Immune cells were detected in only 1 patient. Their distribution did not correspond with the distribution of 11C-MeNTI or 11C-CFN binding.

DISCUSSION

This study provides in vivo evidence for the presence of μ- and δ-opioid receptors in human lung carcinoma, using the receptor type–selective tracers 11C-MeNTI and 11C-CFN and PET. The results of this study suggest that opioid receptors are significantly more abundant in lung carcinoma than in the normal host tissue and that the relative increase in δ-opioid binding sites in malignant lesions is greater than that of the μ-opioid subtype.

Expression of Opioid Binding Sites in Lung Carcinoma

Several lines of evidence indicate the selectivity of 11C-MeNTI and 11C-CFN for their related δ- and μ-opioid receptor types in lung carcinomas tumor. First, the linear increase in the specific component of tracer binding in tumor, as well as the washout of activity from the normal lung parenchyma, is typical of kinetics observed for 11C-MeNTI and 11C-CFN in opioid receptor-rich and -poor regions in the human brain, respectively, (10,11,20). This finding is in line with the abundance of δ-opioid (800 fmol/mg protein) and μ-opioid (1,700 fmol/mg) receptors in lung carcinoma cell lines (3), compared with the paucity of the opioid receptor density in lung parenchyma cells, measured in rats (3 and 150 fmol/mg protein, respectively) (8).

Second, the mean tumor-to-nontumor uptake ratio was significantly greater for 11C-MeNTI than that for 11C-CFN. This finding confirms in vitro reports, suggesting that receptor binding in lung carcinoma cells, normalized to healthy parenchymal cells, is much greater for δ- than for μ-opioid–selective radioligands (3,8).

Third, administration of the opioid antagonist naloxone reduced the binding in tumor but did not affect the activity in the healthy lung tissue. This finding suggests that activity in healthy lung is primarily nonspecific. In addition, naloxone produced comparable inhibition of 11C-MeNTI and 11C-CFN binding. However, the naloxone dose required to exert comparable inhibition was 5 times greater for 11C-MeNTI than that for 11C-CFN. This finding suggests a relatively greater density of δ-opioid receptors. In the human brain, the inhibitory potency of the opioid antagonist naltrexone was directly correlated (R2 = 0.88) with the regional density of δ-sites (11).

Comparison of Expression of Opioid Binding Sites in Lung Tumor and Brain

Previously, 11C-MeNTI and 11C-CFN were used for the quantification of opioid binding sites in the human brain by PET (9–11,20,21). Comparison between these studies and the present work reveals interesting differences in the distribution of opioid binding sites. In the human brain, μ-opioid receptors are predominant compared with δ-receptors, whereas in lung carcinoma a reversed relationship was found. The absolute amount of 11C-MeNTI binding to δ-opioid receptors in lung carcinoma (98.2 Bq/cm3/MBq) was 3 times greater than that measured in brain regions known to contain high δ-receptor density (30 Bq/cm3/MBq) (11). In contrast, 11C-CFN binding in lung carcinoma tumor was significantly less than that in the human brain (27.5 Bq/cm3/MBq in tumor vs. 50 Bq/cm3/MBq in thalamus). Moreover, this difference is even more accentuated when the tracer's binding is expressed as the specific-to-nonspecific activity ratio, a measure of opioid receptors' binding potential (dissociation constant/maximum number of binding sites) (21). In lung carcinoma tumor, the δ-receptor binding potential is 6 times greater than that in receptor-rich brain regions (4.3 vs. 0.73), whereas this ratio for 11C-CFN is about half that in the brain (2.4 vs. 5.1) (11). This finding is in line with in vitro reports. δ-Receptor concentration in human brain, as determined by quantitative autoradiography, is in the range of a few picomoles per gram of protein (22), whereas receptor density in lung carcinoma cell lines is in the range of a several hundreds of picomoles per gram (3).

Potential Clinical Implications

In addition to their analgesic effect, opioids appear to be important in the growth regulation of normal and neoplastic tissue (23). Opioid alkaloids, endogenous opioid analogs, and food-derived opioid peptides inhibit in a dose-dependent manner the proliferation of a wide diversity of human tumors (24), including lung carcinoma cells (3,25–27). Expression of opioid binding sites correlated inversely with tumor growth in vivo, and administration of opioid agonists resulted in increased survival rates of tumor-bearing mice (28). In lung, colon, and breast carcinoma cells, opioid-induced growth suppression was associated with elevated expression of apoptosis markers (29). Opioid agonists potentiated the antitumor activity of paclitaxel (30). The well-documented negative growth effect suggests that 11C-MeNTI and 11C-CFN PET may afford an effective diagnostic tool of important prognostic value. Currently, opioid-based anticancer drugs are in early clinical trials (31). Opioid receptor imaging using PET may facilitate the development of future anticancer drugs and assist in selecting proper candidates for treatment using opioid peptides.

Animal studies have shown that opioids may not only suppress proliferation but also affect migration and invasive activity of neoplastic cells (32). Migration of immune and neoplastic cells are modulated by μ- and δ-opioid peptides acting as chemoattractants (33). Moreover, opioid peptides were implicated in the invasive activity of neoplastic cells either by remodeling actin cytoskeleton associated with phosphorylation of δ-opioid receptors or by enzymatic degradation of the extracellular matrix (34,35). These data may attribute functional significance to the localization of opioid binding sites to the tumor margin as observed in most patients in the present study. The finding that the region of reduced receptor binding was much larger than the histopathologic or radiographic core or the hypometabolic region suggests that the unique distribution of opioid receptors in lung tumors is not dictated solely by viability factors. Neoplastic cells involved in metastatic spreading or attraction of immune cells will most likely be localized in the tumor margin. However, further PET studies in a larger group of patients are needed to explore the relationship between opioid receptors and tumor proliferation and metastatic activity. Elucidation of the role of opioids in paracrine interactions between tumor cells and the surrounding tissue may propose a new noninvasive diagnostic tool of important prognostic significance in lung cancer and other types of solid tumors.

CONCLUSION

PET of specific opioid receptor-type probes provides evidence that SCLC and non-SCLC in humans express high levels of both δ- and μ-opioid types. δ-Receptors are more abundant than the μ-type across the 3 major lung carcinoma types. Both receptor types appear mainly in the tumor margin, suggesting a functional role of the nonbrain opioid system. Quantitative imaging of opioid receptor subtypes in patients may further the understanding of lung cancer biology and extend the biochemical characterization of individual tumors—thereby, improving prognosis and treatment of solid carcinomas.

Acknowledgments

The authors thank Drs. William Mathews and John Musachio for assisting in the radiosyntheses, David Clough and Karen Edmonds for their assistance in the PET imaging procedures, and Martin Stumpf for managing the data systems. This research was supported, in part, by National Cancer Institute grant CA 32845.

References

  1. 1.↵
    Parkin DM, Bray FI, Devesa SS. Cancer burden in the year 2000: the global picture. Eur J Cancer. 2001;37(suppl 8):4–66.
    OpenUrl
  2. 2.↵
    Campa MJ, Schreiber G, Bepler G, et al. Characterization of delta opioid receptors in lung cancer using a novel nonpeptidic ligand. Cancer Res. 1996;56:1695–1701.
    OpenUrlAbstract/FREE Full Text
  3. 3.↵
    Maneckjee R, Minna JD. Opioid and nicotine receptors affect growth regulation of human lung cancer cell lines. Proc Natl Acad Sci U S A. 1990;87:3294–3298.
    OpenUrlAbstract/FREE Full Text
  4. 4.
    Roth KA, Barchas JD. Small cell carcinoma cell lines contain opioid peptides and receptors. Cancer. 1986;57:769–773.
    OpenUrlCrossRefPubMed
  5. 5.
    Schreiber G, Campa MJ, Prabhakar S, et al. Molecular characterization of the human delta opioid receptor in lung cancer. Anticancer Res. 1998;18:1787–1792.
    OpenUrlPubMed
  6. 6.↵
    Sher E, Codignola A, Passafaro M, et al. Nicotinic receptors and calcium channels in small cell lung carcinoma: functional role, modulation, and autoimmunity. Ann N Y Acad Sci. 1998;841:606–624.
    OpenUrlCrossRefPubMed
  7. 7.↵
    Collier TL, Schiller PW, Waterhouse RN. Radiosynthesis and in vivo evaluation of the pseudopeptide delta-opioid antagonist [125I]ITIPP(psi). Nucl Med Biol. 2001;28:375–381.
    OpenUrlPubMed
  8. 8.↵
    Bhargava HN, Villar VM, Cortijo J, Morcillo EJ. Binding of [3H][D-Ala2,MePhe4,Gly-ol5]enkephalin, [3H][D-Pen2 D-Pen5]enkephalin, and [3H]U-69,593 to airway and pulmonary tissues of normal and sensitized rats. Peptides. 1997;18:1603–1608.
    OpenUrlCrossRefPubMed
  9. 9.↵
    Frost JJ, Dannals RF, Duelfer T, et al. In vivo studies of opiate receptors. Ann Neurol. 1984;15(suppl):S85–S92.
  10. 10.↵
    Frost JJ, Douglass KH, Mayberg HS, et al. Multicompartmental analysis of 11C-carfentanil binding to opiate receptors in humans measured by positron emission tomography. J Cereb Blood Flow Metab. 1989;9:398–409.
    OpenUrlPubMed
  11. 11.↵
    Madar I, Lever JR, Kinter CM, et al. Imaging of delta opioid receptors in human brain by N1′-(11C-methyl)naltrindole and PET. Synapse. 1996;24:19–28.
    OpenUrlCrossRefPubMed
  12. 12.↵
    Portoghese PS, Sultana M, Takemori AE. Design of peptidomimetic delta opioid receptor antagonists using the message-address concept. J Med Chem. 1990;33:1714–1720.
    OpenUrlCrossRefPubMed
  13. 13.↵
    Lever JR, Kinter CM, Ravert HT, Musachio JL, Mathews WB, Dannals RF. Synthesis of N1′-(11C-methyl)naltrindole (11C-MeNTI): a radioligand for positron emission tomographic studies of delta opioid receptors. J Labelled Compds Radiopharm. 1995;36:137–145.
    OpenUrl
  14. 14.↵
    Bencherif B, Stumpf M, Wang G, McCaul M, Madar I, Frost J. Brain mu opioid receptor binding changes in chronic alcoholics before and after partial volume correction (PVC) [abstract]. J Nucl Med. 2001;42(suppl):107P.
    OpenUrl
  15. 15.
    Madar I, Bencherif B, Brandt J, Rabins PV, Zubeita JK, Frost JJ. Mu opioid receptors in Alzheimer's disease: statistical parametric mapping analysis [abstract]. J Nucl Med. 2001;42(suppl):12P.
    OpenUrl
  16. 16.
    Madar I, Lesser RP, Krauss G, et al. Imaging of delta- and mu-opioid receptors in temporal lobe epilepsy by positron emission tomography. Ann Neurol. 1997;41:358–367.
    OpenUrlCrossRefPubMed
  17. 17.
    Frost JJ, Mayberg HS, Fisher RS, et al. Mu-opiate receptors measured by positron emission tomography are increased in temporal lobe epilepsy. Ann Neurol. 1988;23:231–237.
    OpenUrlCrossRefPubMed
  18. 18.↵
    Zubieta JK, Gorelick DA, Stauffer R, Ravert HT, Dannals RF, Frost JJ. Increased mu opioid receptor binding detected by PET in cocaine-dependent men is associated with cocaine craving. Nat Med. 1996;2:1225–1229.
    OpenUrlCrossRefPubMed
  19. 19.↵
    Dannals RF, Ravert HT, Frost JJ, Wilson AA, Burns HD, Wagner HN Jr. Radiosynthesis of an opiate receptor binding radiotracer: 11C-carfentanil. Int J Appl Radiat Isot. 1985;36:303–306.
    OpenUrlCrossRefPubMed
  20. 20.↵
    Smith JS, Zubieta JK, Price JC, et al. Quantification of delta-opioid receptors in human brain with N1′-(11C-methyl)naltrindole and positron emission tomography. J Cereb Blood Flow Metab. 1999;19:956–966.
    OpenUrlPubMed
  21. 21.↵
    Endres CJ, Bencherif B, Hilton J, Madar I, Frost JJ. Quantification of brain mu-opioid receptors with [11C]carfentanil: reference-tissue methods. Nucl Med Biol. 2003;30:177–186.
    OpenUrlCrossRefPubMed
  22. 22.↵
    Pfeiffer A, Pasi A, Mehraein P, Herz A. Opiate receptor binding sites in human brain. Brain Res. 1982;248:87–96.
    OpenUrlCrossRefPubMed
  23. 23.↵
    Zagon IS, McLaughlin PJ. Opioids and differentiation in human cancer cells. Neuropeptides. 2005;39:495–505.
    OpenUrlCrossRefPubMed
  24. 24.↵
    Hatzoglou A, Kampa M, Castanas E. Opioid-somatostatin interactions in regulating cancer cell growth. Front Biosci. 2005;10:244–256.
    OpenUrlPubMed
  25. 25.↵
    Munjal I, Minna J, Maneckjee R, Bieck P, Spector S. Possible role of endogenous morphine and codeine on growth regulation of lung tissue. Life Sci. 1995;57:517–521.
    OpenUrlPubMed
  26. 26.
    Heusch WL, Maneckjee R. Signalling pathways involved in nicotine regulation of apoptosis of human lung cancer cells. Carcinogenesis. 1998;19:551–556.
    OpenUrlAbstract/FREE Full Text
  27. 27.↵
    Maneckjee R, Minna JD. Opioids induce while nicotine suppresses apoptosis in human lung cancer cells. Cell Growth Differ. 1994;5:1033–1040.
    OpenUrlAbstract
  28. 28.↵
    Gomez-Flores R, Caballero-Hernandez D, Tamez-Guerra R, et al. Increased survival of tumor-bearing mice by the delta opioid SNC 80. Anticancer Res. 2005;25:4563–4567.
    OpenUrlAbstract/FREE Full Text
  29. 29.↵
    Hatsukari I, Hitosugi N, Matsumoto I, Nagasaka H, Sakagami H. Induction of early apoptosis marker by morphine in human lung and breast carcinoma cell lines. Anticancer Res. 2003;23:2413–2417.
    OpenUrlPubMed
  30. 30.↵
    Jaglowski JR, Zagon IS, Stack BC Jr. Opioid growth factor enhances tumor growth inhibition and increases the survival of paclitaxel-treated mice with squamous cell carcinoma of the head and neck. Cancer Chemother Pharmacol. 2005;56:97–104.
    OpenUrlCrossRefPubMed
  31. 31.↵
    Smith JP, Conter RL, Bingaman SI. Treatment of advanced pancreatic cancer with opioid growth factor: phase I. Anticancer Drugs. 2004;15:203–209.
    OpenUrlCrossRefPubMed
  32. 32.↵
    McLaughlin PJ, Zagon IS. Progression of squamous cell carcinoma of the head and neck is associated with down-regulation of the opioid growth factor receptor. Int J Oncol. 2006;28:1577–1583.
    OpenUrlPubMed
  33. 33.↵
    Fichna J, Janecka A. Opioid peptides in cancer. Cancer Metastasis Rev. 2004;23:351–366.
    OpenUrlPubMed
  34. 34.↵
    Entschladen F, Bastian P, Niggemann B, Zaenker KS, Lang K. Inhibition of cell migration via G protein-coupled receptors to opioid peptides and angiotensin. Ann N Y Acad Sci. 2004;1028:320–328.
    OpenUrlCrossRefPubMed
  35. 35.↵
    Harimaya Y, Koizumi K, Andoh T, Nojima H, Kuraishi Y, Saiki I. Potential ability of morphine to inhibit the adhesion, invasion and metastasis of metastatic colon 26-L5 carcinoma cells. Cancer Lett. 2002;187:121–127.
    OpenUrlCrossRefPubMed
  • Received for publication August 22, 2006.
  • Accepted for publication October 19, 2006.
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Bookmark this article

Jump to section