Ann Nucl Med (2010) 24:241–247 DOI 10.1007/s12149-010-0363-0

ORIGINAL ARTICLE

18

F-fluoride uptake in bone metastasis: morphologic and metabolic analysis on integrated PET/CT

Masashi Kawaguchi • Ukihide Tateishi Kazuya Shizukuishi • Akiko Suzuki • Tomio Inoue

•

Received: 24 July 2009 / Accepted: 14 December 2009 / Published online: 24 March 2010 Ó The Japanese Society of Nuclear Medicine 2010

Abstract Objective The aim of our study was to evaluate detectability of bone metastatic lesions and evaluate the correlation between 18F-fluoride uptake patterns on positron emission tomography (PET) and morphologic changes on CT using integrated PET/CT. Methods We performed whole-body 18F-fluoride PET/CT staging for 27 patients with known cancer. Tumor types comprised breast (n = 7), prostate (n = 7), and others (n = 13). A total of 154 uptake lesions were evaluated. Both tracer uptake patterns determined by 18F-fluoride PET and morphologic patterns based on CT findings such as morphologic changes, involved locations, and grades scored using five-point scale were compared with histologic tumor subtypes and clinical laboratory data. Results CT patterns of metastatic lesion were lytic or unclassified in 26 lesions, sclerotic in 53 lesions, and mixed in 75 lesions. Multiple linear regression analysis revealed that metastatic bone lesions with high maximum standardized uptake value (SUVmax) tended to show sclerotic or mixed changes on CT (P \ 0.0001), and were also distributed in bone cortex alone or both bone cortex and medulla (P \ 0.0001). Conclusion In patients with bone metastasis, the lesions with sclerotic or mixed changes or located in bone cortex alone or both bone cortex and medulla tend to show high SUVmax on 18F-fluoride PET/CT.

M. Kawaguchi (&)
U. Tateishi
K. Shizukuishi
A. Suzuki
T. Inoue Department of Radiology, Yokohama City University Hospital, Graduate School of Medicine, 3-9 Fukuura, Kanazawa-ku, Yokohama, Kanagawa 236-0004, Japan e-mail: [email protected]

Keywords 18F-fluoride
PET
PET/CT
Bone metastasis
Skeletal imaging

Introduction Bone scintigraphy (BS) and magnetic resonance imaging (MRI) play an important role in the diagnosis of bone metastasis. Hamaoka and colleagues [1] discussed the advantages and disadvantages of these modalities with regard to bone metastasis in patients with breast cancer. However, BS has lower specificity and higher false-positive rates because the findings of BS reflect the metabolic reaction [2–6]. In addition, MRI is less desirable for detecting destruction of bone structures, because cortical bone does not produce a signal [7–12]. 18F-fluorodeoxyglucose (FDG) positron emission tomography (PET) has been used to measure glucose metabolism in many types of cancer and can be useful for distinguishing benign from malignant bone lesions. FDG PET produced more false-negative findings for bone metastases than for non-osseous metastases because of the low uptake of FDG by bone [13, 14]. 18 F-fluoride PET has been used for evaluating bone metastasis in various malignancies [15]. Fluoride ions exchange with hydroxy groups in hydroxyapatite crystal bone to form fluoroapatite. Accumulation of 18F-fluoride in sclerotic and lytic lesions reflects increased regional blood flow and bone turnover [16]. Several studies have demonstrated the diagnostic properties of 18F-fluoride PET for bone metastasis especially in lesions with pathologic changes [17]. 18F-fluoride PET is more useful for detecting bone metastasis than BS [18–22]. Combined PET/CT scanner has been developed and provides detailed anatomic information regarding the sites of the abnormal PET findings. We can also use CT to

123

242

assess morphologic features of lesions that are suspected of being bone metastasis on PET. In the present study, we retrospectively assessed the morphologic and metabolic findings of 18F-fluoride PET/CT in patients with various malignancies. Our aim was to clarify the relationship between morphologic and metabolic changes of 18F-fluoride PET/CT for detecting bone metastasis.

Ann Nucl Med (2010) 24:241–247

Emission scans were obtained 60 min after intravenous administration of approximately 185 MBq 18F-fluoride. The acquisition time was 2 min per bed position in the three-dimensional mode. Images were reconstructed with attenuation-weighted ordered-subset expectation maximization with attenuation correction. Image interpretation and morphologic analyses

Materials and methods Patients A retrospective search of our institutional 18F-fluoride PET/CT database revealed 27 patients (17 males and 10 females, mean age 66 years, age range 49–83 years) who were diagnosed with bone metastases by other imaging modalities such as BS, CT and MRI, or clinical follow-up, and referred for assessment of bone metastasis between September 2008 and February 2009 (Table 1). The patients had various cancers, including cancer of the breast (n = 7), prostate (n = 7), pancreas (n = 4), colon (n = 2), liver (n = 2), lung (n = 1), stomach (n = 1), kidney (n = 1), parotid gland (n = 1), and uterine cervix (n = 1). The mean serum levels of alkaline phosphatase (ALP), lactate dehydrogenase (LDH), and calcium were 627.7 ± 539.5 [standard deviation (SD)] IU/l, 254.2 ± 163.7 IU/l, and 9.3 ± 0.4 mg/dl, respectively. Some therapies such as surgery, chemotherapy, radiotherapy, or hormone therapy were administered for their primary diseases before 18F-fluoride PET/CT. Initial surgery was performed in 13 patients (48%). Adjuvant chemotherapy was performed in 22 patients (79%). Radiotherapy was performed in 8 patients (30%). Hormone therapy was performed in 8 patients (30%) with breast or prostate cancer. One patient received no therapy when 18Ffluoride PET/CT was performed. Sixteen patients are alive with disease. Five patients died with disease. Six patients were referred to other hospital. This study was approved by the local ethics committee, and all patients gave written consent to participate in it. Imaging PET/CT images were acquired with an integrated PET/CT device (Aquiduo; Toshiba Medical Systems, Tokyo, Japan). Before PET, unenhanced CT was performed from the base of the skull to the upper thigh according to a standardized protocol performed with the following settings: transverse 2- and 5-mm section thickness, 120 kV, automated tube current (the upper limit was 100 mA), 0.5 s per CT rotation, a pitch of 15, and with breath-holding instructions.

123

Two radiologists reviewed the medical records of all patients. The reference standard was based on the followup imaging studies including whole-body CT, BS, MRI, and the medical records. PET and CT images obtained in all standard planes were reviewed on a viewer (synapse, FUJIFILM Medical, Tokyo, Japan). Two reviewers visually and quantitatively analyzed the images and recorded their findings after they reached consensus. 18F-fluoride PET images were assessed for the presence of bone metastasis using a five-point grading system. PET images were scored from 0 to 4 (0 = definitely negative, 1 = probably negative, 2 = possibly positive, 3 = probably positive and 4 = definitely positive) for each individual lesion. Thus grade 3 or 4 lesions on PET were considered to be suspected bone metastases on PET alone through the consensus of the two readers. Evaluation of CT images was performed by two radiologists using CT planes in which abnormal uptakes were detected on the corresponding PET planes. The radiologists used a viewer to display the CT scans with bone and softtissue windows. Diagnostic certainty in the presence or absence of bone metastasis on CT was expressed using the same five-point grading system that had been used for PET. When definite morphologic changes within the bone that corresponded to the areas of abnormal uptake were identified, a grade of 4 was given on CT, while a lesion was given a grade of 0 when no abnormal findings were seen. When some morphologic changes were seen in the bone but the finding was indeterminate for bone metastasis, the lesion was scored as grade 2. When abnormal 18F-fluoride uptake was present in bone, the exact location of abnormal uptake was identified on CT images [location (cortex, medulla, or both)]. Patients were classified as having lytic, sclerotic, mixed, or unclassified metastatic bone disease at the time of the diagnosis of bone metastasis on the basis of findings on CT. Unclassified disease included lesions having minimal sclerotic or lytic change, and lesions without change on CT when compared with the adjacent bone. The quantitative analysis was done using the maximum standardized uptake value (SUV). We divided metastatic lesions into two groups including sclerotic or mixed lesions and lytic or unclassified lesions for statistical analysis because of the small number of patients.

Ann Nucl Med (2010) 24:241–247 Table 1 Patient demographic

M male, F female, CT chemotherapy, RT radiotherapy, HT hormone therapy

243

Patient no.

Age

Gender

Histologic subtypes of cancers

Treatments

Dominant pattern of bone metastasis

1

65

M

Prostate

CT, HT

Sclerotic

2

74

M

Prostate

Surgery, CT, RT, HT

Sclerotic

3

68

M

Prostate

CT, HT

Sclerotic

4

70

M

Prostate

HT

Sclerotic

5

81

M

Prostate

CT, HT

Sclerotic

6

64

M

Prostate

CT, RT, HT

Mixed

7

63

M

Prostate

CT, HT

Mixed

8

57

F

Breast

Surgery, CT, HT

Sclerotic

9

54

F

Breast

Surgery, CT, RT

Sclerotic

10

53

F

Breast

Surgery, CT

Sclerotic

11 12

59 56

F F

Breast Breast

Surgery, CT, RT Surgery, CT, RT

Mixed Mixed

13

73

F

Breast

CT

Lytic

14

49

F

Breast

Surgery, CT, RT, HT

Lytic

15

76

M

Pancreas

CT

Sclerotic

16

83

M

Pancreas

CT

Sclerotic

17

66

M

Pancreas

CT

Mixed

18

75

M

Pancreas

None

Unclassified

19

74

M

Liver

CT

Lytic

20

51

M

Liver

Surgery, CT

Lytic

21

79

M

Colon

Surgery

Mixed

22

65

M

Colon

Surgery

Unclassified

23

65

F

Lung

CT

Sclerotic

24

61

F

Stomach

Surgery, CT

Mixed

25

71

F

Cervix of uterus

Surgery, CT, RT

Sclerotic

26 27

70 66

M M

Kidney Parotid gland

Surgery CT, RT

Mixed Mixed

Statistical analysis The maximum period of clinical follow-up was 1 year and 9 months. Comparison of mean values between groups was performed with the Student’s t test and Mann–Whitney U test. P \ 0.05 was considered to indicate a significant difference. Spearman rank correlation test was performed to investigate relationships between serum level of ALP, LDH, calcium, and CT or PET grade. Multiple linear regression analysis was performed to clarify associations between CT findings and SUVmax. Statistical analysis was performed with SPSS software (version 12; SPSS, Chicago).

Results All patients were positive on 18F-fluoride PET/CT (Figs. 1, 2). A total of 154 lesions were identified. The sites most frequently involved were vertebra (37%), ilium (14%) and ribs (14%) (Table 2). Mean SUVmax was

22.3 ± 9.8 (SD) ranging from 4.5 to 32.8. Frequency of bone lesions was similar regardless of histologic subsets. Metastatic lesions that showed the highest SUVmax included lumbar vertebra (55%), sacrum (55%), and thoracic vertebra (43%). CT patterns of lesions included lytic (n = 10, 6%), sclerotic (n = 53, 34%), mixed (n = 75, 49%), and unclassified (n = 16, 10%) (Table 3). The mean SUVmax of sclerotic, mixed, lytic and unclassified lesions was 24.1 ± 8.7, 22.6 ± 9.0, 13.7 ± 6.8, and 14.0 ± 8.0, respectively (Fig. 3). The SUVmax of sclerotic or mixed lesion was significantly greater than that of lytic or unclassified lesions (23.2 ± 8.9 vs. 13.9 ± 7.4, P \ 0.0001). When stratified by lesions with involvement of medulla, the SUVmax was 18.0 ± 8.8 (Fig. 4). A significant difference was found in SUVmax between lesions involving the medulla only and those involving both the cortex and medulla (P \ 0.0001). The mean SUVmax of lesions in patients with each histologic subtype of cancer is listed in Table 4. CT grades expressed by the five-point grading system were 0 (n = 16, 10%), 1 (n = 16, 10%), 2 (n = 22,

123

244

Ann Nucl Med (2010) 24:241–247

Fig. 1 A 74-year-old man with hepatocellular carcinoma. Transverse CT image (a) shows lytic lesion in lumbar vertebra (L1). Corresponding transverse PET (b) and PET/CT (c) images show increased uptake of 18 F-fluoride (SUVmax, 20.5)

Fig. 2 A 68-year-old man with prostate cancer. Transverse CT image (a) shows sclerotic lesion in left iliac wing and sacrum. Corresponding transverse PET (b) and PET/CT (c) images show increased uptake of 18 F-fluoride (SUVmax, 32.8 and 26.0, respectively)

14%), 3 (n = 25, 16%), and 4 (n = 75, 49%). Similarly, PET grades were 0 (n = 0), 1 (n = 2, 1%), 2 (n = 7, 5%), 3 (n = 30, 19%), and 4 (n = 115, 75%). When stratified by sclerotic or mixed lesions, the percentage of CT and PET grades was 0 and 0% (grade 0), 12 and 0% (grade 1), 16 and 1% (grade 2), 16 and 18% (grade 3), and 56 and 81% (grade 4), respectively. When stratified by lytic or unclassified lesions, the percentage of CT and PET grades was 16 and 0% (grade 0), 1 and 2% (grade 1), 2 and 6% (grade 2), 4 and 7% (grade 3), and 3 and 11% (grade 4), respectively.

123

Both the CT and PET grades in the sclerotic or mixed lesions were significantly greater than those of the lytic or unclassified lesions (P \ 0.0001). The serum level of ALP correlated weakly with CT grade (r = 0.259, P = 0.001) and PET grade (r = 0.165, P = 0.042). The coefficients of the serum level of LDH and that of calcium did not reach a significant difference. When the SUVmax of metastatic lesions divided into two groups (sclerotic and mixed vs. lytic and unclassified) was analyzed, multiple linear regression analysis revealed that factors independently associated with SUVmax were

Ann Nucl Med (2010) 24:241–247

245

Table 2 Anatomical sites of metastatic lesions Location

No. of lesions

Skull

10 (7)

Ribs, sternum, clavicle and scapula

22 (14)

Cervical vertebrae

14 (9)

Thoracic vertebrae

21 (14)

Lumbar vertebrae

22 (14)

Sacrum

11 (7)

Ilium

21 (14)

Ischium

11 (7)

Pubis

11 (7)

Femur

10 (7)

Tibia

1 (1)

Data in parentheses are percentages

Table 3 CT findings and five-point grade by CT and PET

Fig. 3 Relation between SUVmax and CT findings (mixed, sclerotic, lytic, unclassified). The SUVmax of sclerotic or mixed lesions tends to be greater than that of lytic or unclassified lesions

CT findings Mixed

75 (49%)

Sclerotic

53 (34%)

Nonspecific

16 (10%)

Lytic

10 (6%)

Involved area of bone Medulla

77 (50%)

Both cortex and medulla

70 (45%)

Cortex

3 (2%)

None

4 (3%)

Five-point grade

CT

PET

0

16 (10%)

0

1 2

16 (10%) 22 (14%)

2 (1%) 7 (5%)

3

25 (16%)

30 (19%)

4

75 (49%)

115 (75%)

sclerotic or mixed CT patterns and locations of cortex and medulla (Table 5).

Discussion Our results mirrored those of several studies evaluating the detectability of bone metastasis with 18F-fluoride PET or PET/CT [17–22]. Our results also showed that the SUVmax of metastatic lesion was independently associated with CT findings: sclerotic or mixed pattern and location of cortex or cortex and medulla. 18F-fluoride PET/CT seems to be a useful modality for the detection of metastatic

Fig. 4 Relation between SUVmax and involved areas (medulla, cortex, medulla and cortex). The SUVmax of lesions with involvement of both medulla and cortex tends to be greater than that of medulla or cortex only

lesions especially those with such morphologic changes on CT. In our study, frequent bone metastasis detected by fluoride PET was located in vertebra and ileum, whereas skull was an uncommon location of bone metastasis. These results are in agreement with those of other studies [17–19] and might be related to the relatively numerous small metastatic deposits in vertebra and ileum detected by 18 F-fluoride PET/CT. With 18F-fluoride PET/CT the majority of bone metastases showed a sclerotic or mixed CT pattern. The detection of lesions with lytic CT pattern was only 6.5% (12/153). These results might be attributable to selection

123

246

Ann Nucl Med (2010) 24:241–247

Table 4 SUVmax for each cancer Histologic subtypes of cancers

Number of lesions

SUVmax ± SD

Prostate

50 (32%)

25.5 ± 1.2

Breast

44 (29%)

23.2 ± 8.9

Pancreas

19 (12%)

17.9 ± 2.4

Colon

13 (8%)

23.2 ± 2.6

Stomach

10 (6%)

17.8 ± 1.5

Liver

6 (4%)

15.1 ± 3.2

Parotid gland Lung

6 (4%) 3 (2%)

19.7 ± 3.4 21.1 ± 5.8

Cervix of uterus

2 (1%)

32.0

Kidney

1 (1%)

32.0

Table 5 Factors associated with SUVmax on NaF PET/CT Variable

B

HR

95% CI

P value

Sclerotic or mixed CT pattern 1.759 5.804 1.767–19.601 \0.0001 Location of cortex only or cortex and medulla

1.759 5.809 2.817–11.979 \0.0001

HR hazard ratio, CI confidence interval

bias related to the retrospective nature of the data collection. However, these results suggest that 18F-fluoride PET/ CT was more sensitive to detect lesions with sclerotic osseous reaction than those with lytic destruction. Even-Sapir and colleagues [18] described that 18 F-fluoride PET/CT showed the highest sensitivity and specificity to detect bone metastasis in patients with prostate cancer compared to conventional imaging. In contrast, Schirrmeister et al. [21] described that 18F-fluoride PET showed the highest area under ROC curve in patients with lung cancer among three different modalities. In our study, we evaluated various histologic subtypes of cancers, including breast, prostate, pancreas, colon, liver, lung, kidney, parotid gland, and uterine cervix. Our results suggest the utility of 18F-fluoride PET/CT in the detection of bone metastasis in patients with various cancers. 18 F-fluoride PET has been shown to be substantially more accurate than BS in the assessment of bone metastasis in patients with malignant tumors. Although PET has been proved to be an effective tool in the management of patients with malignant tumors, it provides limited information on the morphologic abnormalities in bone. Co-registered functional and morphologic data sets are generated with integrated PET/CT systems, and previous results for the diagnosis of bone metastasis with this system have been promising [18]. The use of PET/CT system improves the specificity of 18F-fluoride PET by determining the morphology of uptake lesions on the CT data [17]. Thus, 18F-fluoride PET/CT can provide precise information

123

regarding both the morphologic and metabolic changes occurring in bone metastasis. Patients with bone metastasis have often been excluded from clinical trials because bone metastases are considered immeasurable. This difficulty is relevant when using an imaging modality that detects bone metastasis based on the morphologic findings alone. In contrast, 18F-fluoride PET/CT enables semi-quantitative assessment of uptake, which makes it convenient to use when evaluating tumor viability during treatment in addition to morphologic monitoring by the CT portion. Therefore, we propose that sequential 18F-fluoride PET/CT performed at the baseline and follow-up has the possibility to predict the therapeutic effect of bone metastasis. Our study had some potential limitations. Since data collection was performed retrospectively, selection bias may have affected our results. The enrolled patients showed a heterogeneous histologic diagnosis. In addition, all patients underwent various kinds of treatment for bone metastasis. We started whole-body CT, MRI findings and clinical follow-up as the reference standard. However, another difficulty in interpreting the results of our study is the lack of histologic proof of lesions detected with 18 F-fluoride PET/CT. Obtaining histologic proof of all skeletal lesions is impractical and is unethical when there is no impact on the clinical management. It was unclear whether 18F-fluoride PET/CT proved to be more costeffective than conventional strategies. However, to our knowledge, the costs and health outcomes associated with 18 F-fluoride PET/CT in addition to those of conventional studies for bone metastasis in clinical practice have not been assessed in the clinical or economic context. In conclusion, our study results provide evidence that the degree of uptake of bone metastasis on 18F-fluoride PET/CT is independently associated with a sclerotic or mixed CT pattern and locations of cortex and medulla. Our results need to be validated in a prospective study to further clarify this issue.

References 1. Hamaoka T, Madewell JE, Podoloff DA, Hortobagyi GN, Ueno NT. Bone imaging in metastatic breast cancer. J Clin Oncol. 2004;22:2942–53. 2. Rybak LD, Rosenthal DI. Radiological imaging for the diagnosis of bone metastases. Q J Nucl Med. 2001;45:53–64. 3. Citrin DL, Bessent RG, Greig WR. A comparison of the sensitivity and accuracy of the 99TCm-phosphate bone scan and skeletal radiograph in the diagnosis of bone metastases. Clin Radiol. 1977;28:107–17. 4. Rosenthal DI. Radiologic diagnosis of bone metastases. Cancer. 1997;80:1595–607. 5. Galasko CS, Doyle FH. The detection of skeletal metastases from mammary cancer: a regional comparison between radiology and scintigraphy. Clin Radiol. 1972;23:295–7.

Ann Nucl Med (2010) 24:241–247 6. Lee YT. Bone scanning in patients with early breast carcinoma: should it be a routine staging procedure? Cancer. 1981;47:486– 95. 7. Sanders TG, Parsons TW 3rd. Radiographic imaging of musculoskeletal neoplasia. Cancer Control. 2001;8:221–31. 8. Tryciecky EW, Gottschalk A, Ludema K. Oncologic imaging: interactions of nuclear medicine with CT and MRI using the bone scan as a model. Semin Nucl Med. 1997;27:142–51. 9. Coleman RE. Monitoring of bone metastases. Eur J Cancer. 1998;34:252–9. 10. Zimmer WD, Berquist TH, McLeod RA, Sim FH, Pritchard DJ, Shives TC, et al. Bone tumors: magnetic resonance imaging versus computed tomography. Radiology. 1985;155:709–18. 11. Moon KL Jr, Genant HK, Helms CA, Chafetz NI, Crooks LE, Kaufman L. Musculoskeletal applications of nuclear magnetic resonance. Radiology. 1983;147:161–71. 12. Vogler JB 3rd, Murphy WA. Bone marrow imaging. Radiology. 1988;168:679–93. 13. Cook GJ, Fogelman I. The role of positron emission tomography in the management of bone metastases. Cancer. 2000;88:2927–33. 14. Moon DH, Maddahi J, Silverman DH, Glaspy JA, Phelps ME, Hoh CK. Accuracy of whole-body fluorine-18-FDG PET for the detection of recurrent or metastatic breast carcinoma. J Nucl Med. 1998;39:431–5. 15. Grant FD, Fahey FH, Packard AB, Davis RT, Alavi A, Treves ST. Skeletal PET with 18F-fluoride: applying new technology to an old tracer. J Nucl Med. 2008;49:68–78. 16. Hawkins RA, Choi Y, Huang SC, Hoh CK, Dahlbom M, Schiepers C, et al. Evaluation of skeletal kinetics of fluorine 18-fluoride ion and PET. J Nucl Med. 1992;33:633–42.

247 17. Even-Sapir E, Metser U, Flusser G, Zuriel L, Kollender Y, Lerman H, et al. Assessment of malignant skeletal disease: initial experience with 18F-fluoride PET/CT and comparison between 18F-fluoride PET and 18F-fluoride PET/CT. J Nucl Med. 2004;45:272–8. 18. Even-Sapir E, Metser U, Mihsani E, Mishani E, Lievshitz G, Lerman H, et al. The detection of bone metastases in patients with high-risk prostate cancer: 99mTc-MDP planar bone scintigraphy, single- and multi-field-of-view SPECT, 18F-fluoride PET, and 18F-fluoride PET/CT. J Nucl Med. 2006;47:287–97. 19. Schirrmeister H, Guhlmann A, Elsner K, Kotzerke J, Glatting G, Rentschler M, et al. Sensitivity in detecting osseous lesions depends on anatomic localization: planar bone scintigraphy versus 18F PET. J Nucl Med. 1999;40:1623–9. 20. Schirrmeister H, Guhlmann A, Kotzerke J, Santjohanser C, Ku¨hn T, Kreienberg R, et al. Early detection and accurate description of extent of metastatic bone disease in breast cancer with fluoride ion and positron emission tomography. J Clin Oncol. 1999; 17:2381–9. 21. Schirrmeister H, Glatting G, Hetzel J, Nu¨ssle K, Arslandemir C, Buck AK, et al. Prospective evaluation of the clinical value of planar bone scans, SPECT, and 18F-labeled NaF PET in newly diagnosed lung cancer. J Nucl Med. 2001;42:1800–4. 22. Schirrmeister H, Buck A, Guhlmann A, Reske SN. Anatomical distribution and sclerotic activity of bone metastases from thyroid cancer assessed with F-18 sodium fluoride positron emission tomography. Thyroid. 2001;11:677–83.

123