Nuclear Medicine and Biology 40 (2013) 240–244

Contents lists available at SciVerse ScienceDirect

Nuclear Medicine and Biology

j ourna l homepage: www.e lsev ie r .com/ locate /nucmedb io

Longitudinal observation of [11C]4DST uptake in turpentine-inducedinflammatory tissue

Jun Toyohara ⁎, Muneyuki Sakata, Keiichi Oda, Kenji Ishii, Kiichi IshiwataResearch Team for Neuroimaging, Tokyo Metropolitan Institute of Gerontology, 1-1 Naka-cho, Itabashi-ku, Tokyo 173-0022, Japan

⁎ Corresponding author. Tel.: +81 3 3964 3241; fax:E-mail address: toyohara@pet.tmig.or.jp (J. Toyohara

0969-8051/$ – see front matter © 2013 Elsevier Inc. Alhttp://dx.doi.org/10.1016/j.nucmedbio.2012.10.008

a b s t r a c t

a r t i c l e i n f o

Article history:

Received 22 September 2012Accepted 9 October 2012

Keywords:4DSTInflammationCell proliferationKi-67Positron emission tomography

Introduction: Longitudinal changes of 4′-[methyl-11C]thiothymidine ([11C]4DST) uptake were evaluated inturpentine-induced inflammation.Methods: Turpentine (0.1 ml) was injected intramuscularly into the right hind leg of male Wistar rats.Longitudinal [11C]4DST uptake was evaluated by the tissue dissection method at 1, 2, 4, 7, and 14 days afterturpentine injection (n=5). The tumor selectivity index was calculated using the previously publishedbiodistribution data in C6 glioma-bearing rats. Dynamic PET scan was performed on day 4 when maximum[11C]4DST uptake was observed during the longitudinal study. Histopathological analysis and Ki-67immunostaining were also performed.Results: The uptake of [11C]4DST in inflammatory tissue was significantly increased on days 2–4 after

turpentine injection, and then decreased. On day 14, tracer uptake returned to the day 1 level. The maximumSUV of inflamed muscle was 0.6 and was 3 times higher than that of the contralateral healthy muscle on days2–4 after turpentine injection. However, tumor selectivity index remains very high (N10) because of the lowinflammation uptake. A dynamic PET scan showed that the radioactivity in inflammatory tissues peaked at5 min after [11C]4DST injection, and then washed out until 20 min. At intervals N20 min, radioactivity levelswere constant and double that of healthy muscle. The changes in Ki-67 index were paralleled with those of[11C]4DST uptake, indicating cell proliferation-dependent uptake of [11C]4DST in inflammatory tissues.Conclusion: In our animal model, low but significant levels of [11C]4DST uptake were observed insubacute inflammation.

© 2013 Elsevier Inc. All rights reserved.

1. Introduction

Differentiation between residual tumor and inflammatory re-sponse is the major premise for accurate therapeutic monitoring oftumor tissues with positron emission tomography (PET). Althoughfluorine-18 fluorodeoxyglucose ([18F]FDG) is currently the mostwidely used radiopharmaceutical for this purpose, various forms ofinflammatory lesion take up large amounts of [18F]FDG [1–4]. Severallines of evidence suggest that the accumulation of [18F]FDG is basedon enhanced glycolysis, which has often been associated with thegrowth rate and malignancy of tumor cells. However, all living cellsrequire glucose, and high levels of [18F]FDG uptake are observed inhighly metabolic inflammatory cells.

More specific information on the in vivo status of tumors may beacquired by the use of a radiopharmaceutical that is a proliferationmarker, as proliferative activity is one of the key factors of malignantdisease. For this reason, we proposed carbon-11-labeled 4′-thiothy-midine ([11C]4DST, originally designated as [11C]S-dThd) as a tumorimaging agent based on the mechanisms of incorporation into DNA

+81 3 3964 2188.).

l rights reserved.

[5–8]. Recently, we evaluated the tissue kinetics and biodistribution of[11C]4DST in a rodent tumor and acute sterile inflammationmodel [9].In our animal model, [11C]4DST produced promising results. [11C]4DST showed the highest level of tumor uptake with high tumorselectivity. Furthermore, [11C]4DST showed completely differentkinetics in tumor and inflammatory tissues.

Although the usefulness of [11C]4DST for differentiation betweenmalignant tumor and inflammationwas indicated in our experimentalmodel, this previous study had some limitations in that we used theinflammation model in the acute phase without any contribution ofproliferative inflammatory cells. In contrast, chronic inflammatorygranulomatous lesions include a Ki-67-positive (6.3%) lymphocytefraction [10]. Inflammatory lung diseases are accompanied bylymphocyte infiltration and involve growth factors that enhance theproliferation of lymphocytes [11]. If [11C]4DST is a real proliferationmarker, it may accumulate in chronic granulomatous lesions withproliferative inflammation. To confirm [11C]4DST selectivity, furtherexperimental studies with chronic inflammation models are needed.Therefore, in this study, we evaluated the longitudinal changes of[11C]4DST uptake and cell proliferation status indicated by Ki-67immunostaining in turpentine-induced acute, subacute, and chronicphases of inflammatory tissues. A tumor selectivity index was also

241J. Toyohara et al. / Nuclear Medicine and Biology 40 (2013) 240–244

calculated using the previously published biodistribution data in theC6 glioma-bearing rat model [9].

2. Materials and methods

2.1. Tracer

[11C]4DST was produced by palladium-mediated Stile cross-coupling reaction of 5-tributylstannyl-4′-thio-2′-deoxyuridine with[11C]methyl iodide [7]. The specific activities and radiochemicalpurities were 103.3±64.9 GBq/μmol and 99.4%±0.2%, respectively,at the time of injection.

2.2. Animals and inflammation model

MaleWistar rats weighing 80–100 gwere obtained from Japan SLC(Hamamatsu, Japan). Animals were housed under constant environ-mental conditions with a 12–12-h light–dark cycle. Food and waterwere provided ad libitum. To induce inflammation, 0.1 ml ofturpentine was injected intramuscularly into the thigh of the righthind leg. Turpentine injection is an established model of sterileinflammation [12,13]. The animal studies were approved by theAnimal Care and Use Committee of the Tokyo Metropolitan Instituteof Gerontology.

2.3. Longitudinal study of [11C]4DST

[11C]4DST (30–65 MBq/0.2–0.7 nmol) was injected intravenouslyinto rats at 1, 2, 4, 7, and 14 days post-turpentine injection (n=5).Animals were killed at 1 h after tracer injection and tissue sampleswere excised and weighed, followed by counting 11C radioactivityusing a CompuGamma CS 1282 counter (LKB-Wallac, Turku,Finland) with decay correction. The results were expressed asdimensionless standardized uptake values (SUVs) (cpm measuredper gram of tissue/cpm injected per gram of body weight). Theinflamed-to-healthy muscle concentration ratio of radioactivity wasalso calculated.

2.4. Small animal PET imaging

Rats on day 4 after turpentine injection were selected as themaximum [11C]4DST uptakewas observed on the same day during thelongitudinal study. Four rats were scanned dynamically with theinflammatory tissues in the field-of-view using a semiconductor smallanimal PET scanner (MIP-100; Sumitomo Heavy Industries, Tokyo,Japan) [14]. Rats were anesthetized with a mixture of isoflurane/air(inhalation anesthesia, 5% ratio during induction, later reduced tob2%). Rats were confirmed to be under anesthesia before tracerinjection. [11C]4DST was injected via the tail vein (45–115 MBq/0.5–4.0 nmol). A list-mode protocol was used for 60 min. Scanning wasstarted at the injection of radioactivity into the rat.

List-mode data were reframed into a dynamic sequence of 8×30 s,3×60 s, 2×120 s, 2×180 s, 3×300 s, 2×540 s, and 1×600 s frames.The data were reconstructed per time frame using an iterativereconstruction algorithm (3-dimensional ordered-subset expectationmaximization, provided by Sumitomo Heavy Industries; 1 iteration,32 subsets). The final datasets consisted of 31 slices, with a slicethickness of 0.85 mm and an in-plane image matrix of 256×256pixels of size 0.3×0.3 mm. Datasets were fully corrected for randomcoincidences and scatter. Images were smoothed with a Gaussianfilter (1.2 mm in both transverse and axial directions).

2.5. Small animal PET data analysis

In inflamed muscle and contralateral healthy muscle, three-dimensional regions of interest (3D-ROIs) were carefully drawn

around the outside rim of the non-inflamed hind leg (in the coronalimages), starting in the plane closest to the animal bed andproceeding upward until the skeleton was approached. Care wastaken not to include the skeleton in the ROIs. As the bone marrow hadmuch higher uptake of [11C]4DST than any other part of the hind leg,this was achieved easily. The left hind leg had a smaller volume thanthe swollen right hind leg (inflamed thigh). The volumes (ml) for thehealthy muscle 3D-ROI were about 50% of the entire volume of theinflamed muscle.

Time–activity curves and volumes (ml) for the 3D-ROIs werecalculated using standard software (PMOD, version 3.307; PMODTechnologies Ltd., Zurich, Switzerland). PET SUVs were calculated,using measured body weights and injected doses and assuming aspecific gravity of 1 g/ml for tissue.

2.6. Histology

Histopathological analysis and Ki-67 immunostaining wereexamined (Morphotechnology, Hokkaido, Japan). Inflammatorytissue samples were fixed in 10% neutral-buffered formalin,dehydrated by passage through a graded alcohol series, and finallyembedded in paraffin. Another five rats without turpentine injectionwere used as external controls (Day 0). Thin sections were stainedwith hematoxylin and eosin (H&E) for histological examination.Immunohistochemical staining was also carried out using theadjacent sections. Antigen retrieval was performed by heating afterimmersion of slides into target retrieval solution (pH 9.0) (Dako,Glostrup, Denmark) at 95 °C for 20 min. Endogenous peroxideactivity was blocked with 0.3% hydrogen peroxide. The anti-rat Ki-67 mouse monoclonal antibody (IgG1, Clone No. MIB-5; Dako) wasapplied to the slides at a dilution of 1:50 for 60 min at roomtemperature. Slides were washed, then incubated with the secondaryantibody (Histofine Simple Stain Rat MAX-PO; Nichirei, Tokyo,Japan) at room temperature for 30 min. Color development wascarried out by 5-min incubation in diaminobenzidine, followed bycounterstaining with hematoxylin.

Ki-67 labeling index (L.I.) was estimated by counting thepercentage of Ki-67-positive cell nuclei per 2500–10000 cells in 5regions of the tissues with a ×200 objective. To exclude regionalvariations in Ki-67 staining due to cell death, adjacent sectionsstained with H&E were examined for necrosis. Necrotic areas with alarge number of small cells with abnormal nuclei or apoptotic bodieswere excluded.

2.7. Statistical analysis

Differences between the two types of lesion were tested forstatistical significance using the two-sided Students t test. One-way analysis of variance (ANOVA) with Dunnett's multiplecomparison tests was used for comparison of time-dependentchanges of [11C]4DST uptake to day 1. In all analyses, Pb0.05 wastaken to indicate statistical significance. A tumor selectivity indexwas calculated using the following formula: (tumor SUV−muscleSUV)/(inflammation SUV−muscle SUV). This figure represents thetumor-to-inflammation ratio corrected for background activity.The selectivity index was calculated from the average SUV ininflammation and muscle for each group of rats. The previouslypublished C6 glioma data were used as input values for averagetumor SUV [9].

3. Results

3.1. Longitudinal study of [11C]4DST

Fig. 1 shows the time-dependent changes of [11C]4DST uptakemeasured on several days after turpentine injection. The uptake of

Fig. 1. [11C]4DST uptake (SUV) in inflamed muscle, healthy muscle, blood, andinflamed-to-healthy muscle ratio several days after turpentine injection (n=5). Errorbars indicate standard deviation (S.D.). ○=blood; ●=healthy muscle; □=inflamedmuscle; ■=inflamed/healthy.

Fig. 2. Small animal PET image of rat hind leg region acquired after injection of [11C]4DST (summed frames from 41 min to the end of the scan) into the rat on day 4 afterturpentine injection.

242 J. Toyohara et al. / Nuclear Medicine and Biology 40 (2013) 240–244

[11C]4DST in inflammatory tissue was significantly increased on days2–4 after turpentine injection, and then decreased thereafter. On day14, tracer uptake levels returned to those on day 1. Themaximum SUVlevel of inflamed muscle was 0.6, and was approximately 3 timeshigher than that of healthy muscle on days 2–4 after turpentineinjection. The radioactivity levels in healthy muscle and blood weresignificantly decreased on day 2 and remained at the low level duringthe 14-day observation period. Statistical analysis of uptake differ-ences between inflamed and healthy muscles, and selectivity indexare summarized in Table 1. Although [11C]4DST uptake in inflamedmuscle was significantly greater than that of the healthy muscle, theselectivity index remained N10 on days 2, 4, 7, and 14 afterturpentine injection.

3.2. Small animal PET images

Small animal PET image acquired after injection of [11C]4DST arepresented in Fig. 2. A high level of tracer uptake was observed in bonemarrow. Inflamed swollenmuscle showed significant but low levels ofpatchy and rim-like uptake. Healthy muscle showed the backgroundlevel of radioactivity.

3.3. Kinetics of radioactivity in tissues

Fig. 3 shows the kinetics of [11C]4DST-derived radioactivity (PETSUV in the tissues as a function of time). In healthy muscle (non-proliferating tissue), tracer uptake was low and constant between 20and 60 min. In inflamed muscle, radioactivity peaked at 5 min, andthen washed out by 20 min. On and after 20 min, radioactivity levelswere constant and double that of the healthy muscle.

Table 1Standardized uptake values in the inflamed and healthy muscle at 60 min afterinjection of [11C]4DST.

Day Inflamed Healthy Inflamed vs. healthy Selectivity indexa

1 0.36±0.03 0.31±0.03 N.S. 91.42 0.59±0.04 0.18±0.01 b0.001 10.64 0.60±0.03 0.20±0.01 b0.001 10.87 0.45±0.02 0.22±0.01 b0.001 19.514 0.41±0.02 0.22±0.01 b0.001 23.8

Data represent means and standard deviation (n=5). N.S.=not statistically significant.a Selectivity index: tumor-to-inflammation ratio corrected for background in normal

muscle. Tumor data were used as an input value for average tumor SUV [9]. No S.D. forselectivity index can be given because, in some animals, tracer uptake in inflamedmuscle was equivalent to that in the contralateral healthy muscle.

3.4. Histopathology

Typical turpentine-induced inflammatory tissues are shown inFig. 4. One day after turpentine injection, histolysis and edema wereobserved at the site of injection. Massive infiltration of granulocytes,consisting predominantly of neutrophils and a few scattered mono-cytes, was noted adjacent to the abscess cavity 2 days after turpentineinjection. In addition, endothelial cells of blood vessels and youngfibroblasts appeared. These cells included a Ki-67-positive cellsfraction. On day 4, massive inflammatory cell layers and denseyoung fibroblast zones appeared. The inflammatory cell layersincluded larger numbers of round and polymorphonuclear cells,macrophages, and dead cells. On day 7, the lesions of inflammatorycell layers were reduced in comparison to day 4, and granulationtissue appeared. On day 14, thick granulation tissues with maturedfibroblast were obvious.

Fig. 5A shows the Ki-67 L.I. measured on several days afterturpentine injection. The Ki-67 L.I. in inflammatory tissue wassignificantly increased on days 2–4 after turpentine injection, andthen decreased. On day 14, Ki-67 L.I. returned to the control level. Thechanges in Ki-67 L.I. were paralleledwith those of [11C]4DST uptake inthe inflammatory tissues. As a result, [11C]4DST uptake in theinflammatory tissues was significantly correlated with Ki-67 L.I.(r2=0.93; P=0.002) (Fig. 5B).

4. Discussion

This study demonstrated the longitudinal changes of [11C]4DSTuptake and cell proliferation status in inflammatory tissues inturpentine-induced acute, subacute, and chronic phases. In addition,the tumor selectivity index was evaluated using the previouslypublished biodistribution data in C6 glioma-bearing rats.

Although [11C]4DST uptake in subacute inflammation was 3 timeshigher than that of healthy muscle, tumor selectivity index remained

Fig. 3. Kinetics of [11C]4DST-derived radioactivity in inflamed and healthy muscle (n=5). Error bars indicate standard deviation (S.D.) ○=inflamed muscle; ●=healthymuscle.

Fig. 4. Microscopy images (×200) of hematoxylin and eosin staining (upper), and Ki-67 immunostaining (middle) of inflamed muscle on days 1, 2, 4, 7, and 14 after turpentineinjection and of external healthy muscle that had not been injected with turpentine as a control. Red insets indicate magnified areas in the bottom row (×400).

243J. Toyohara et al. / Nuclear Medicine and Biology 40 (2013) 240–244

very high (N10), due to the low uptake (SUV 0.6) in subacuteinflammatory tissues. The longitudinal time-dependent changes of[11C]4DST uptake in inflammatory tissue were paralleled withthose of Ki-67 L.I. These observations indicated that [11C]4DSTuptake reflects the cell proliferation status of inflammatory tissues.These findings provide useful information for the optimized scantiming and interpretation of [11C]4DST PET images in tumortreatment evaluation.

The uptake kinetics of [11C]4DST in subacute inflammatory tissuesdiffered significantly from those reported previously in tumors andacute inflammation [9]. [11C]4DST uptake in the tumor increased

Fig. 5. Time course of Ki-67 labeling index (L.I.) in the inflamed muscle after turpentine injecLinear regression analysis of [11C]4DST uptake (SUV) and proliferative activity (Ki-67 L.I.) inbars indicate standard deviation (S.D.).

steadily with time, while that in acute inflammatory tissue showedslow exponential clearance after the initial distribution. In subacuteinflammation, radioactivity decreased gradually after an initialincrease, and then reached a plateau. The initial peak of acute andsubacute inflammation may have been due to increased blood flow inthe inflamed tissue. These findings suggest that dynamic [11C]4DSTPET has additional value for differentiating not only tumor frominflammation but also activity of inflammatory tissues.

Previously, we compared the tumor selectivity indexes of varioustracers in the same established tumor and acute inflammation model[9]. The results indicated that [11C]4DST showed the highest tumor

tion (A). A control group without turpentine injection was designated as day 0 (n=5).inflamedmuscle (B). A significant correlation was observed (r2=0.93; P=0.002). Error

244 J. Toyohara et al. / Nuclear Medicine and Biology 40 (2013) 240–244

selectivity compared with four well-defined PET tracers: [18F]FLT,[18F]FDG, [11C]choline, and [11C]methionine. In this study, the lowesttumor selectivity index of [11C]4DST was observed in the subacutephase (10.6). However, this value was still much larger than thosereported previously in the acute phase for [18F]FLT (1.4), [18F]FDG(3.4), [11C]choline (0.9), and [11C]methionine (5.3). Taken together,these observations support the high tumor selectivity of [11C]4DST ina range of inflammation types.

In this study, longitudinal time-dependent changes of [11C]4DSTuptake in inflammatory tissues were paralleledwith those of Ki-67 L.I.This was in good accordance with a recent report suggesting asignificant correlation between SUVs of [11C]4DST and proliferativeactivity of lung tumor lesions [8]. Although we have not identified theresponsible Ki-67-positive [11C]4DST-accumulating cells, activatedlymphocytes, fibroblasts, and endothelial cells are reasonable candi-dates. For example, bronchoalveolar lavage preparations of activesarcoidosis patients contained Ki-67-positive T lymphocytes (6.2%)and macrophages (6.5%) [15]. In the croton oil-induced rat inflam-mationmodel, the number of fibroblasts peaked on days 3–5 and thendeclined thereafter [16]. Sholley et al. reported the time sequence andmagnitude of endothelial DNA synthesis in sites of acute inflamma-tion induced by thermal injury to the skin in rats [17]. They foundincreased endothelial (10%–12%) and perivascular (9%) [3H]thymi-dine L.I. on days 2 and 3 after injury. The endothelial [3H]thymidineL.I. diminished progressively 10 days after injury (1.5%–5%). Further-more, Ezaki et al. evaluated the time course of endothelial cellproliferation and microvascular remodeling in a Mycoplasma pulmo-nis-infected mouse chronic inflammation model [18]. Endothelial cellproliferation, as determined by bromodeoxyuridine labeling, peakedon day 5 (19.4%), declined sharply until day 9 (5.9%), but remained at3 times the pathogen-free values until at least day 28 (2.8%) afterinfection. Taken together, the tendencies of cell proliferation timecourse were not different, despite the different cell types andinflammation models. Therefore, a transient increase in [11C]4DSTuptake may generally occur in the subacute phase of inflammation.These considerations also indicated that an interval of over 7 days isrequired after therapeutic treatment for evaluation of tumor responseby [11C]4DST PET.

Depression of Ki-67 L.I. was observed on day 1 after turpentineinjection. This initial depressed cell proliferation is a common featureof the epithelial response to various injuries [19,20]. The radioactivitylevels in healthy muscle and blood were significantly decreased onday 2 and continued during the 14-day observation period. This mayreflect the purging of primed lymphocytes from healthy muscle andblood due to the massive infiltration of lymphocytes in inflamedmuscle. Furthermore, various types of stress have been shown todepress cell proliferation as a result of rising blood levels of adrenalineand glucocorticoid hormones [21]. In contrast, epithelium with highmitotic activity in response to wounding is much less sensitive to theinhibitory effect of stress [22], so this could explain the significantuptake of [11C]4DST in inflamedmuscle. These findings emphasize theproblems inherent in the common practice of relying solely onapparently normal tissue from experimental animals as controlmaterial in longitudinal studies of this type.

5. Conclusion

In our experimental animal models, [11C]4DST uptake was low butincreased significantly in subacute inflammation and then decreasedin the chronic phase. This time-dependent changes of [11C]4DSTuptake corresponded well to the Ki-67 L.I. Therefore, [11C]4DSTuptake reflects the cell proliferation status of inflammatory tissues.

These data will aid in the interpretation of [11C]4DST PET data foranalysis of tumor treatment response.

Acknowledgments

This work was supported by a Grant-in Aid for Scientific Research(B) No. 22390241 from the Japan Society for the Promotion of Science(to Jun Toyohara) and a Grant from the National Center for GlobalHealth and Medicine (to Jun Toyohara and Kiichi Ishiwata). Theauthors thank Mr. Kunpei Hayashi and Mr. Takashi Yamaguchi fortechnical assistance.

References

[1] Yamada S, Kubota K, Kubota R, Ido T, Tamahashi N. High accumulation of fluorine-18-fluorodeoxyglucose in turpentine-induced inflammatory tissue. J Nucl Med1995;36:1301-6.

[2] Kaim AH, Weber B, Kurrer MO, Gottschalk J, von Schulthess GK, Buck A.Autoradiographic quantification of 18F-FDG uptake in experimental soft-tissueabscesses in rats. Radiology 2002;223:446-51.

[3] Kubota R, Kubota K, Yamada S, TadaM, Ido T, Tamahashi N.Microautoradiographicstudy for the differentiation of intratumoral macrophages, granulation tissues andcancer cells by the dynamics of fluorine-18-fluorodeoxyglucose uptake. J NuclMed 1994;35:104-12.

[4] Kubota R, Yamada S, Kubota K, Ishiwata K, Tamahashi N, Ido T. Intratumoraldistribution of fluorine-18-fluorodeoxyglucose in vivo: high accumulation inmacrophages and granulation tissues studied by microautoradiography. J NuclMed 1992;33:1972-80.

[5] Toyohara J, Kumata K, Fukushi K, Irie T, Suzuki K. Evaluation of [methyl-14C]4′-thiothymidine for in vivo DNA synthesis imaging. J Nucl Med 2006;47:1717-22.

[6] Toyohara J, Okada M, Toramatsu C, Suzuki K, Irie T. Feasibility studies of 4′-[methyl-11C]thiothymidine as a tumor proliferation imaging agent in mice. NuclMed Biol 2008;35:67-74.

[7] Toyohara J, Nariai T, Sakata M, Oda K, Ishii K, Kawabe T, et al. Whole-bodydistribution and brain tumor imaging with 11C-4DST: a pilot study. J Nucl Med2011;52:1322-8.

[8] Minamimoto R, Toyohara J, Seike A, Ito H, Endo H, Morooka M, et al. 11C-4DSTPET/CT for proliferation imaging in non-small-cell lung cancer. J Nucl Med2012;53:199-206.

[9] Toyohara J, Elsinga P, Ishiwata K, Sijbesma J, Dierckx R, van Waarde A. Evaluationof 4′-[methyl-11C]thiothymidine (11C-4DST) in a rodent tumor and inflammationmodel. J Nucl Med 2012;53:488-94.

[10] Zhao S, Kuge Y, Kohanawa M, Takahashi T, Zhao Y, Yi M, et al. Usefulness of 11C-methionine for differentiating tumors from granulomas in experimental ratmodels: a comparison with 18F-FDG and 18F-FLT. J Nucl Med 2008;49:135-41.

[11] Drent M, du Bois RM, Poletti V. Recent advances in the diagnosis and managementof nonspecific interstitial pneumonia. Curr Opin Pulm Med 2003;9:411-7.

[12] VanWaarde A, Cobben DCP, Suurmeijer AJH, Maas B, VaalburgW, de Vries EF, et al.Selectivity of 18F-FLT and 18F-FDG for differentiating tumor from inflammation ina rodent model. J Nucl Med 2004;45:695-700.

[13] Van Waarde A, Jager PL, Ishiwata K, Dierckx RA, Elsinga PH. Comparison of sigma-ligands and metabolic PET tracers for differentiating tumor from inflammation. JNucl Med 2006;47:150-4.

[14] Ishii K, Kikuchi Y, Matsuyama S, Kanai Y, Kotani K, Ito T, et al. First achievement ofless than 1 mm FWHM resolution in practical semiconductor animal PET scanner.Nucl Instrum Meth A 2007;576:435-40.

[15] Chilosi M, Menestrina F, Capelli P, Montagna L, Lestani M, Pizzolo G, et al.Immunohistochemical analysis of sarcoid granulomas. Evaluation of Ki67+ andinterleukin-1+ cells. Am J Pathol 1988;131:191-8.

[16] Sholley MM, Cavallo T, Cotran RS. Endothelial proliferation in inflammation I.Autoradiographic studies following thermal injury to the skin of normal rats. Am JPathol 1977;89:277-96.

[17] Feher J, Jennings EH, Rannie I. Histochemical alterations and cell proliferation inexperimental inflammation. Br J Exp Path 1971;52:615-20.

[18] Ezaki T, Baluk P, Thurston G, La Barbara A, Woo C, McDonald DM. Time course ofendothelial cell proliferation and microvascular remodeling in chronic inflamma-tion. Am J Pathol 2001;158:2043-55.

[19] Willoughby SG, Hopps RM, Johnson NW. Changes in the rate of epithelialproliferation of rat oral mucosa in response to acute inflammation induced byturpentine. Arch Oral Biol 1986;31:193-9.

[20] Mackenzie IC. The effects of frictional stimulation on mouse ear epidermis. I. Cellproliferation. J Invest Derm 1974;62:80-5.

[21] Bullough WS. Mitotic and functional homeostasis: a perspective review. CancerRes 1965;25:1683-727.

[22] Bullough WS, Laurence EB. Stress and adrenalin in relation to the diurnal cycle ofepidermal mitotic activity in adult male mice. Proc R Soc Lond B 1961;154:540-56.