Thorax 1996;51:727-732

Reduced adrenoceptor density in vivo inhuman lung tumours: a preliminary study withpositron emission tomography

F Qing, M J Hayes, C G Rhodes, T Krausz, S W Fountain, M M Burke, T Jones,J M B Hughes

Department ofMedicine (RespiratoryDivision)F QingM J HayesJ M B Hughes

Department ofHistopathologyT Krausz.

Royal PostgraduateMedical School,HammersmithHospital, Du CaneRoad, LondonW12 ONN, UK

MRC Clinical SciencesCentre, CyclotronUnit, HammersmithHospital, London, UKC G RhodesT Jones

Thoracic SurgicalUnit, HarefieldHospital, Harefield,MiddlesexUB9 6JH, UKS W Fountain

Department ofHistopathology, MountVernon Hospital,Middlesex, UKM M Burke

Correspondence to:Professor J M B Hughes.Received 21 December 1994Returned to authors7 March 1995Revised version received14 November 1995Accepted for publication9 February 1996

AbstractBackground - Reduced P adrenergic re-ceptor density in tumours has been re-ported in previous in vitro studies. Theaim of the present study was to assesswhether this occurs in vivo.Methods - Pulmonary P adrenoceptorswere imaged and quantified in vivo usingpositron emission tomography (PET)and the P antagonist radioligand (S)-["CJCGP-12177 in five men with lungtumours of mean age 58 years (range 42-68). The histology of the tumours wassquamous cell carcinoma in two cases,adenocarcinoma in one, carcinoid tumourin one, and large cell carcinoma in one.The regional blood volume and extra-vascular tissue density were also meas-ured using PET. Regions of interest weredrawn for both non-tumour and tumourlung tissue.Results - The mean (SD) blood volumewas 0-142 (0.025) mn/ml in tumour regionsand 0-108 (0-010) ml/ml in normal lung re-gions - a difference of 31%. Mean (SD)extravascular tissue density was 0-653(0.133) g/ml in tumour regions, sub-stantially higher than in normal lung re-gions (0 157 (0-021) gIml). On the contrary,D receptor density was 5-1 (1.8) pmoUg intumour regions, lower than the value of9*9 (1.6) pmoUg found in adjacent normallung - a difference of 48%.Conclusions - In vivo P adrenoceptor dens-ity is reduced in human lung tumours.(Thorax 1996;51:727-732)

Keywords: P adrenergic receptors, neoplasms, positronemission tomography.

The 1 adrenergic system has been shown toplay a modulatory role in cell growth' anddifferentiation.2 In previous in vitro studiesactivation of the P adrenergic system has beenreported to promote the differentiation of thehuman promonocytic cell line into monocytic-like cells3 and to induce the differentiationof promyelocytic cells.4 In addition, in vivoadministration of the 1 agonist isoprenaline inrat resulted in a suppression of the growth ofthe C-6 glioma tumour,' while ablation of thesympathetic nerve system augmented tumourgrowth.5 Furthermore, this suppressive re-sponse of isoprenaline was proportional to the1 adrenergic receptor density on the tumourcell surface.6 Reduced density of 13 adrenergic

receptors in tumour tissue has been previouslyreported in in vitro studies. Mouse lymphocyticleukaemia cells have been shown to have fewer1 adrenoceptors per cell and a smaller fractionof these receptors are functionally active cellsurface receptors.7 In lung, mouse neoplasticcells contained fewer 1 adrenoceptors and pro-duced less cAMP in response to isoprenaline.8In human lung in vitro 13 receptors were alsodecreased in cancer tissue.9

Positron emission tomography (PET) is aspecialised branch of medical imaging whoseaccuracy and specificity merit the term "in vivoautoradiography". It has opened up a new fieldof study in oncology by providing informationon glucose metabolism, oxygen utilisation,tumour hypoxia, haemodynamics, and proteinand DNA synthesis.'0 When used with suitableradiolabelled ligands, PET can also be usedto determine receptor density in vivo. Withfluorine-18 labelled fluoroestradiol, PET hasbeen used to study oestrogen receptors inpatients with breast cancer to assist in evalu-ation of metastatic or recurrent lesions and,more important, to predict the response ofindividual patients to anti-oestrogen therapy."In view of the potential therapeutic possibilityof modulating the sympathetic nervous systemto control tumour growth we have used PETand carbon-il labelled (S)-CGP-12177 tomeasure pulmonary P adrenoceptor density invivo in patients with lung cancer to assess thedegree to which the in vitro differences indensity of 13 receptors between tumour andnormal lung occur in vivo.

MethodsPATIENTSFive men of mean age 58 years (range 42-68)with lung tumours were investigated. Theirdetails are summarised in the table. None ofthepatients was taking 13 agonists, 13 antagonists, orany other drugs that might interfere with thesympathetic nervous system. All patients gavewritten informed consent to the protocol whichwas approved by the Hammersmith HospitalResearch ethics committee and the UnitedKingdom Administration of Radioactive Sub-stances advisory committee.

PATHOLOGICAL ASSESSMENTResected tumours were available for ex-amination in cases 1, 3, and 4 and biopsy tissueonly was available in cases 2 and 5.At bronchoscopy a polypoid endobronchial

727

on April 15, 2022 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.51.7.727 on 1 July 1996. D

ownloaded from

Qing, Hayes, Rhodes, Krausz, Fountain, Burke, Jones, Hughes

Histopathological findings

Patient no. Age (years) Diagnosis Differentiation Inflammation Necrosis Fibrosis

1 60 Squamous cell carcinoma Moderate + + + + +2 64 Squamous cell carcinoma Poor NA, Bx only* NA, Bx only NA, Bx only3 58 Adenocarcinoma Moderate + + +4 42 Carcinoid tumour NA5 68 Large cell carcinomat Poor NA, Bx only NA, Bx only NA, Bx only

NA = not applicable; Bx = biopsy;-= none; + = occasional/small foci; + + = several/large foci; + + + = extensive.* Bronchial and thyroid biopsy specimens only. No thoracotomy followed due to the metastasis in the thyroid.t Bronchial biopsies and brushing and adrenal biopsy specimens only. No thoracotomy performed due to the metastasis.

tumour was seen in case 5 and, although brush-ings showed large cell carcinoma, a biopsy wasnon-diagnostic. Subsequently, a needle biopsyof an adrenal mass showed carcinoma. Thesefindings were felt sufficient to justify a diagnosisof primary lung carcinoma rather than pul-monary metastases.The grade of differentiation of the malignant

tumours was given as mild, moderate, or poor.The carcinoid tumour in case 4 was not graded.Histological slides of the three resectedtumours were assessed on the basis of thenumber and size of the foci for necrosis, in-flammation, and fibrosis. Occasional/small fociwere scored as +, several foci/large foci as + +,and extensive as + + +. The small volumes oftumour in the biopsy specimens from cases 2and 5 were not considered representative andwere therefore not quantified.

PET SCANNINGPET scans were performed using an ECAT931-08/12, 15-plane positron camera (Siemens/

CTI Inc, Knoxville, Tennessee, USA) as pre-viously described. 2 Each plane has a slice thick-ness of 6-6 mm full width at halfmaximum; thetotal thickness of lung imaged in the transaxialdirection is 10-8 cm. PET scanning consistedof (1) transmission, (2) Cl5O emission and (3)(S)-["C]CGP-12177 dynamic emission scans.The patients lay supine on the scanning table.

A venous cannula was inserted into a forearmvein for blood sampling and a second venouscannula was inserted into the other arm for thetracer infusion. Arterial blood pressure andelectrocardiography were recorded every 15minutes throughout the study.

Transmission scanThe scan information was recorded for a 20minute period during the exposure of a ringsource of positron emitting 68Ge/68Ga that en-circles the subject. These data were used forattenuation correction of all subsequent emis-sion data and also provided images of the lungdensity distribution, total density (vascular plus

Figure 1 (A) PET image of total lung density from case 3. Regions of interest were drawn on this image for the normallung regions as well as for the tumour, identified as a high density area in the right peripheral field. (B) CT scan of aplane close to that shown in (A). (C) PET image of regional pulmonary blood volume. (D) PET image of extravasculartissue density.

728

on April 15, 2022 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.51.7.727 on 1 July 1996. D

ownloaded from

In vivo ,B adrenoceptors in neoplasms

extravascular, g/ml [g lung/ml thoracic vol-ume]) (fig 1A). A computed tomographic (CT)scan (fig IB) is also provided for comparison.

C'5O emission scanTo obtain regional blood volume (ml/ml [mlblood/ml thoracic volume]) (fig 1 C) a six min-ute emission scan was performed five minutesafter the start of a four minute inhalation ofoxygen-15 labelled carbon monoxide (C'50)in air. C'50 was administered at a concentrationof 3 MBq/ml and a flow rate of 500 ml/min.C'50 combines with haemoglobin to form 150-carboxyhaemoglobin in the red blood cells ofthe lung capillaries. Blood samples were taken0, 2, 4, and 6 minutes after the start of theC150 scan to relate vascular radioactivity to theequilibrium images of the C150 distribution,thereby allowing the calculation of regionalpulmonary blood volume, to be used later inthe calculation of receptor density. Values ofregional blood density (g blood/ml thorax) wereobtained by multiplying the regional blood vol-ume by 1 06 (whole blood density). Quan-titative images of pulmonary extravasculartissue density (g/ml) (fig 1D) were then cal-culated by subtracting blood density from thenormalised transmission scan as previously de-scribed. 13

S-["C] CGP-12177 dynamic emission scan(S)-CGP-12177 [(3'-tert-butylamino-2'-hydroxy-propoxy)-benzimidazol-2-one] was asym-metrically synthesised and labelled with theshort lived positron emitting radionuclide car-bon- 11 (half life = 20A4 minutes) on site. 14 (S)-["C]CGP-12 177 produced by this method ex-ceeded 99% chemical and radiochemical pur-ity. Measurement ofpulmonary 1B adrenoceptordensity was performed using a modification ofthe double injection method of Delforge et al. 15A high specific activity (S)-["C]CGP-12177preparation (- 185 MBq (S)-["1C]CGP-12177in 31tg cold (S)-CGP-12177) was given intra-venously over two minutes followed 30 minuteslater by a second injection of (S)-["C]CGP-12177 with a lower specific activity(- 370MBq (S)-["C]CGP-12177 in 25 igcold (S)-CGP-12177). Dynamic emissionscanning comprising 55 frames started at thetime of the-first injection and continued for 75minutes. A representative image of the (S)-["C]CGP-12177 uptake is shown in fig 2.

CALCULATION OF P ADRENOCEPTOR DENSITYImages were analysed on SUN work stationsby use of Analyze image analysis'6 and theMatlab (The MathWorks Inc, Natick, Mas-sachusetts, USA) mathematical software pack-age. Regions of interest (ROIs) for the normallung regions and for the tumour, the latteridentified as high density areas, were drawnon the transmission images. In the 15 planesscanned the most caudal plane was selectedfrom the second plane above the diaphragm. Togenerate pulmonary tissue tracer time/activitycurves ROIs were projected onto the dynamic

Figure 2 Quantitative image of # adrenoceptor binding(pmollml thorax) obtained by adding the dynamic timeframe images recorded between 10 and 30 minutesfollowing the first (S)-['C]CGP 12177 injection. Thesignal from the lung is low relative to the heart (top right)because of the low tissue occupancy of the lung (see fig1 (D)). Visual discrimination between tumour and normallung in this image is poor because of the similar level ofuptake (the tumour uptake is only twice that of lung) andregional statistical fluctuation.

(S)-["1C]CGP-12177 images. The mean traceractivity in serial lung planes (craniocaudal) wascalculated and plotted against time. The ex-travascular tissue tracer time/activity curve wasobtained by subtracting the pulmonary vascular(S)-[" C] CGP-1 2177 time/activity curve (cal-culated from the C'50 blood volume data and(S)-["C]CGP-12177 activity in the venousblood samples) from the regional (S)-["C]CGP-12177 time/activity curve. A graph-ical approach based on that originally proposedby Delforge et al for use in the heart"' wasused to calculate the density of pulmonary 1adrenoceptors in each ROI, taking into accountthe total amount of cold ligand in both in-jections. This technique was further modifiedto express 13 adrenoceptor density as pmol/g(pmol per gram of tissue) by normalising the1 adrenoceptor density to the local value ofextravascular tissue density. The P adreno-ceptor density in lung'7 and heart'8 esti-mated using this PET approach has beenshown to be correlated with the P receptordensity determined with classic in vitromeasurement (radioligand binding assay).

STATISTICAL ANALYSISData are presented as means (SD). Paired ttests were used to compare the values betweentumour and normal regions. A value ofp<0 05was considered statistically significant.

ResultsThe histopathological findings are given in thetable.

PULMONARY BLOOD VOLUMEIn each subject the regional blood volume washigher in tumour regions than in normal lungregions. For the group as a whole the bloodvolume was 0-142 (0 025) ml/ml in tumourregions and 0- 108 (0010) ml/ml in normal lung

729

on April 15, 2022 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.51.7.727 on 1 July 1996. D

ownloaded from

Qing, Hayes, Rhodes, Krausz, Fountain, Burke, Jones, Hughes

0.18

E

-

E0

0nCU

C0E0-

0.16 K0.14 _

12 r.

0

0

0

0.12 H

0.10 _

.

0

00006

0E

._a)

C0Q0.'a)

0)0)

10

8

6

4

2

0

Normal Tumourregions regions

Figure 3 Pulmonary blood volume in nornal lung andtumour regions. Horizontal bars represent mean values.The blood volume was higher in the tumour regions thanin the corresponding normal lung regions (p<OO5).

o

000

0

00

0

-3-

0

Normal Tumourregions regions

Figure 5 ,B adrenoceptor density expressed as pmollgtissue in normal lung and tumour regions. Horizontal barsrepresent mean values (p<O-O1).

regions (fig 3) - a difference of 31% (p<0-05).The mean (95% confidence interval (CI)) forthe differences between tumour and normalregions was 0-034 (0-005 to 0-063) ml/ml.

EXTRAVASCULAR TISSUE DENSITYExtravascular tissue density was always higherin tumour regions than in normal lung regionsin every subject. The mean value of the extra-vascular tissue density for the group was 0-653(0-133) g/ml in tumour regions which wasmuch higher than the 0-157 (0-021) g/ml innormal lung regions (p<0-001) (fig 4). Themean (95% CI) for the differences betweentumour and normal regions was 0-496 (0-346to 0-646) g/ml.

1 ADRENOCEPTOR DENSITYThe density of 1 receptors was lower in tumourregions than in normal lung regions in everysubject, irrespective of the histopathologicaltype. The mean 13 receptor density for the groupwas 5-1 (1-8) pmol/g in tumour regions and 9-9

1.0 7

0.8 0

0.6

(1-6) pmol/g in normal lung regions (fig 5) - adifference of 48% (p<0-01). The mean (95%CI) for the differences between tumour andnormal regions was 4-8 (2-2 to 7-4) pmol/g.

DiscussionRECEPTOR DENSITY IN TUMOURSIn vivo studies on ,B receptors in patients withlung cancer have not previously been reporteddue to the lack of suitable methods. We reporthere the first in vivo study which shows fewer13 receptors in human lung neoplastic tissue.The reduced numbers of 13 receptors in neo-plastic tissue found in the present study areconsistent with previously reported in vitrostudies.The earlier studies relied mainly on in vitro

measurements on either cultured neoplasticcell lines or resected lung tissue obtained atthoracotomy. Lange-Carter and colleagues8 ex-amined 13 receptor density in cultured neo-plastic and normal mouse lung epithelial cells.They reported fewer 13 adrenergic receptors inneoplastic E9 and PCC4 cell membranes (6-5and 12 fmol/mg protein, respectively) than innormal C10 cell membranes (37 fmol/mg pro-tein). Kondratenko and colleagues9 measured1 receptor densities in normal lung tissue (ob-tained at segmental resection of patients withtuberculoma) and neoplastic lung tissue (ob-tained from patients with pulmonary adeno-carcinoma). The 1 receptor density was only84 fmol/mg protein in the neoplastic tissuecompared with 456 fmol/mg protein in normallung tissue.

I

0.4

0.2

0.0

0

Normal Tumourregions regions

Figure 4 Extravascular lung tissue density in normallung and tumour regions. Horizontal bars represent meanvalues (p<O-OO1).

MEASURED IN VIVO VALUES

Pulmonary blood volume (ml blood/ml thorax)was 31% higher in the tumour regions thanin surrounding normal lung tissue, but whennormalised for the tissue occupancy (bloodvolume/g extravascular tissue) the blood vol-ume was 0-23 ml blood/g tissue in tumour re-gions and 0-65 ml blood/g tissue in normal lungregions. Although an interrelationship betweenregional pulmonary blood volume and bloodflow has been reported,"9 the interpretation of

E0

:LI0n

~0'a)0):3

-)cU

CU

xw3

n n.9,

730

on April 15, 2022 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.51.7.727 on 1 July 1996. D

ownloaded from

In vivo ,B adrenoceptors in neoplasms

these differences in blood volume between thetumour and normal lung regions is still difficult,since the major role of pulmonary blood flowis not nutritional but for gas exchange and thevascular structure in tumour tissue is usuallydistorted.The higher value of tissue density in tumour

regions reflects the fact that the tissue oc-cupancy in lung cancer tissue is higher thanin normal lung tissue which contains a largeproportion of air. We recently studied a groupof 17 normal subjects to assess the effect ofsalbutamol on pulmonary 1 adrenoceptors.20In that group, which was younger than thegroup in the present study (37 (9) years com-pared with 58 (10) years), the values of bloodvolume, extravascular tissue density, and P ad-renoceptor density were 0p151 (0.020) ml/ml,0- 182 (0 044) g/ml, and 10-7 (2 0) pmol/g, re-spectively, compared with the correspondingvalues of 0 108 (0 010) ml/ml, 0'157 (0 021) g/ml, and 9 9 (1 6) pmol/g obtained for the "nor-mal" regions in the present study. The valuesfor extravascular tissue density and density of13 adrenoceptors were close between the twogroups, whereas the value for the blood volumeis rather lower in the older group of patients.Whether this difference is due to the ageingeffect or due to the tumour itself (neoplastictissue might release some substances that couldaffect the blood volume in the adjacent normallung tissue) is unclear.

1 RECEPTORS AND MALIGNANCY,B receptors and cAMPThe P adrenergic receptor forms the frontierunit in the 13 receptor, G protein, adenylatecyclase, cyclic 3', 5'-adenosine monophosphate(cAMP) and protein kinase chain which formsone of the most important second messengersystems. Besides P receptor agonists, numerousother reagents such as cholera toxin, forskolinand prostaglandin E can also increase the cellu-lar cAMP level.2' A number of growth factorsutilise the same signal transduction pathwaysas P receptors - for example, platelet derivedgrowth factor (PDGF) has been reported toelicit cAMP accumulation in Swiss 3T3 cells.22cAMP is an important modulator of cellulargrowth and differentiation.2' The intracellularcAMP level fluctuates at different stages of thecell cycle. The most common finding is thatthe cAMP level is at a minimum during mitosis,gradually increases during GI (Gap 1, periodbetween mitosis and the start of DNA syn-thesis), reaches a peak near the G,-S border,and declines again during the S phase (DNAsynthesis).23 Addition of exogenous cAMP tovarious cultured cell types has demonstratedboth an inhibiting and a stimulating effect onproliferation.24 cAMP might modulate the ac-tivity of various oncogenes and growth factors.The ultimate response to cAMP would thendepend upon the cell types, the oncogene driv-ing its growth, the dose of cAMP, and theenvironment of the cells.24 Alterations in anystep in the cAMP synthesis pathway mighttherefore result in a malfunction of the re-gulation of cell proliferation and differentiation.

Potential therapeutic implicationsChanges in 13 adrenoceptors could affect intra-cellular cAMP levels which might, in turn,affect the growth and differentiation or qui-escence of a cell. The finding of a reduced :adrenoceptor density in neoplastic tissue pro-vides an opportunity for potential therapeuticintervention. cAMP has been reported to in-hibit the growth of small cell lung cancer celllines.25 A low receptor density might thereforebe indicative of a deficient cAMP productionwith a resulting diminution in the regulationof cell growth and differentiation. In certaintypes ofcancer in which cAMP exerts a negativeeffect on growth, this might contribute to fur-ther neoplastic progression. As stated above, anumber of agents can increase the cAMP level.The possibility therefore exists of employingthese agents to restore cAMP levels. This hasalready been tested in cultured cell lines andin animal preparations. An early study byChelmika-Schorr and coworkers demonstratedthat growth of a C-6 glioma was suppressed inrats treated with the P receptor agonist iso-prenaline. Addition of papaverine, a cAMPphosphodiesterase inhibitor, to the treatmentaugmented this effect. Because ofthe extremelycomplicated role of cAMP in modulating thegrowth and differentiation of cells, it is pre-mature to say if this kind of intervention isapplicable to humans.The lower P receptor density could itself be

the result of the neoplastic process. This maybe due to a number of reasons. Firstly, pro-liferation of connective tissue (related to tissuerepair or inflammation) may increase the re-ceptor-deficient or receptor-lacking cell types,which together with oedema might contributeto the lower numbers of receptors per gramof tissue. However, this is unlikely to be thecomplete explanation of our results since mostof the tissue assessed histologically was neo-plastic. Definite fibrosis and necrosis werefound only in one of three patients and in-flammation was found only in two of threepatients (table), whereas the lower receptordensity was observed in every subject. Sec-ondly, neoplastic cells are genetically defectiveand poorly differentiated. Thus, the expressionand synthesis of 13 receptors in these "pre-mature" cells are poor and might result inlower receptor numbers per cell. Gope andcolleagues26 have reported alterations in the P2adrenergic receptor gene in some colorectalcancers. Hughes and colleagues27 reported thatthe 12 adrenergic receptor deficient S49 lym-phoma cells expressed reduced quantities of 12receptor specific mRNA although they con-tained the same amount of the 12 receptor geneas the control cells. In their studies the amountOf 12 adrenergic receptor specific mRNA cor-related very well with the reduction in receptorexpression in these cells. They believe that therelative paucity of gene transcripts is re-sponsible for the diminution of the receptors.

In summary, the present studies using PETas a new tool have found that 1 adrenergicreceptor density in vivo was considerably re-duced in five lung tumours. This is unlikelyto be solely explained by fibrosis, necrosis or

731

on April 15, 2022 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.51.7.727 on 1 July 1996. D

ownloaded from

Qing, Hayes, Rhodes, Krausz, Fountain, Burke, 3ones, Hughes

inflammation, and suggests a lower number ofa receptors per neoplastic cell.

The authors thank A D Williams, A R K Blyth, and the personnelof the blood counting laboratory and the radiochemistry andcyclotron operations groups for their assistance in performingthe PET scans. This work was supported by the NationalAsthma Campaign, Allen and Hanburys Ltd, and the BritishLung Foundation.

1 Chelmicka-Schorr E, Arnason BGW, Holshouser SJ. C-6glioma growth in rat: suppression with a [-adrenergicagonist and a phosphodiesterase inhibitor. Ann Neurol1980;8:447-9.

2 Cho-Chung YS, Clair T, Tagliaferri P, Ally S, Katsaros D,Tortora G, et al. Site-selective cyclic AMP analogs as new

biological tools in growth control, differentiation, andproto-oncogene regulation. Cancer Invest 1989;7: 161 -77.

3 Paul-Eugene N, Dugas B, Gordon J, Kolb JP, Caims JA,Paubert-Braquet JM, et al. [32-adrenoceptor stimulationaugments the IL-4-induced CD23 expression and releaseand the expression of differentiation markers (CD 14,CD18) by the human monocytic cell line, U937. Clin ExpAllergy 1993;23:317-25.

4 Bang BE, Aarbakke J, Sager G. Epinephrine induces [3-adrenergic desensitization and differentiation of HL-60cells. Scand 7 Clinl Lab Invest 1993;53:331-5.

5 Kinnard RL, Chelmicka-Schorr E, Checinski ME, JonesKH, Arnason BGW. Sympathetic nervous system andglioma growth. Eur 7 Cancer Clin Oncol 1986;22:501-3.

6 Chelmicka-Schorr E, Sportiello MG, Atweh SF, ArnasonBGW. Response of C-6 glioma to isoproterenol and pa-paverine in vivo depends on [-adrenergic receptor density.Ann Neurol 1984;15:96-9.

7 D'Amico C, Crescimanno M, Armata MG, Leonardi V,Palazzoadriano M, D'Alessandro N. Beta-adrenergic in-fluences on doxorubicin-sensitive or -resistant p388 leuk-emia cells. Anticancer Res 1992;12:2253-6.

8 Lange-Carter CA, Droms KA, Vuillequez JJ, MalkinsonAM. Differential responsiveness to agents which stimulatecAMP production in normal versus neoplastic mouse lungepithelial cells. Cancer Lett 1992;67:139-44.

9 Kondratenko TY, Zacharova IV, Kuzina NV, Katukov VY,Severin ES, Komilova ZC, et al. Alterations in humanlung adrenergic receptors in cancer. Biochenii Mol Biol IJut1993;29:123-30.

10 Koh WJ, Griffin TW, Rasey JS, Laramore GE. Positronemission tomography. A new tool for characterization ofmalignant disease and selection of therapy. Acta Onicol1994;33:323-7.

11 Mintun MA, Welch MJ, Siegel BA, Mathias CJ, BrodackJW, McGuire AH, et al. Breast cancer: PET imaging ofestrogen receptors. Radiology 1988;169:45-8.

12 Ueki J, Rhodes CG, Hughes JMB, Silva RD, Lefroy DC,Ind PW, et al. In vivo quantification of pulmonarv 1j-

adrenoceptor density in humans with (S)-["C] CGP12177 and PET.

_Appl Physiol 1993;75:559-65.

13 Rhodes CG, Wollmer P, Fazio F, Jones T. Quantitativemeasurement of regional extravascular lung density usingpositron emission and transmission tomography. 7 ConipitAssist Toniogr 1981;5:783-91.

14 Brady F, Luthra SK, Tochon-Danguy H-J, Steel CJ, WaterSL, Kensett MJ, et al. Asymmetric synthesis of a precursorfor the automated radiosynthesis of S-["C]CGP12177 asa preferred radioligand for [3-receptors. Appl Radiat Isot1991;42:621-8.

15 Delforge J, Syrota A, Lancon J-P, Nakajima K, Janier CLM,Vallois J-M, et al. Cardiac beta-adrenergic receptor densitymeasured in vivo using PET, CGP 12177 and a newgraphical method. 7 Nuicl Med 199 1;32:739-48.

16 Robb RA, Hanson DP. A software system for interactive andquantitative visualization of multidimensional biomedicalimages. Anistralas Phvs Eng Sci Med 1991;14:9-30.

17 Qing F, Rhodes CG, Hayes MJ, Ind PW, Krausz T, FountainSW, et al. In vivo quantification of human pulmonary -

adrenoceptor density using PET: comparison with in vitroradioligand binding. _7 Nucl Med 1996 (in press).

18 Merlet P, Delforge J, Syrota A, Angevin E, Maziere B,Crouzel C, et al. Positron emission tomography with "C-CGP- 12177 to assess 1-adrenergic receptor concentrationin idiopathic dilated cardiomyopathy. Circuilatiosn 1993;87:1169-78.

19 Brudin LH, Rhodes CG, Valind SO, Jones T, Hughes- JMB. Interrelationship between regional blood flow, blood

volume, and ventilation in supine humans. _7 Appl P/issiol1994;76:1205- 10.

20 Hayes MJ, Qing F, Rhodes CG, Ind PW, Sriskandan S,Jones T, et al. In vivo quantification of pulmonary 3i-adrenoceptor: effects of 3-agonist therapy. Ani _7 RespirCnit Care Med 1996 (in press).

21 Rozengurt E. Early signals in the mitogenic response. Scienice1986;234:161-6.

22 Rozengurt E, Stroobant P, Waterfield MD, Deuel TF,Keehan M. Platelet-derived growth factor elicits cyclicAMP accumulation in Swiss 3T3 cells: role of prosta-glandin production. Cell 1983;34:265-72.

23 Beebe SJ, Corbin JD. Cyclic nucleotide-dependent proteinkinases. In: Boyer PD, Krebs EG, eds. The enzvnies.Orlando: Academic Press, 1986;17:44-114.

24 Gottesman MM, Fleischmann RD. The role of cAMPin regulating tumor cell growth. Cancer Sin-v 1986;5:291-308.

25 Khan MZ, Freshnev RI, McNicol AM, Murray AM. In-duction of phenotypic changes in SCLC cell lines in vitroby hexamethylene bisacetamide, sodium butyrate, andcAMP. Atnn Onicol 1993;4:499-507.

26 Gope R, Gope ML, Thorson A, Christensen M, Smyrk T,Chun M, et al. Genetic changes at the beta-2-adrenergicreceptor locus on chromosome 5 in human colorectalcarcinomas. Anticancer Res 1991 ;11 :2047-50.

27 Hughes RJ, Howard MJ, Allen JM, Insel PA. Decreased 3i,-adrenergic receptor mRNA expression in receptor-deficient S49 lymphoma cells. Mol Phanniacol 1991;40:974-9.

732

on April 15, 2022 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.51.7.727 on 1 July 1996. D

ownloaded from