Kaohsiung J Med Sci December 2007 • Vol 23 • No 12 639 The “seed-and-soil hypothesis” of the mechanism of bone metastasis (via marrow) was first described by Stephan Paget in 1889 and is supported by findings from animal models of bone metastasis [1]. Bone metas-tases are a frequent complication of cancer, occurring Received: January 19, 2007 Accepted: April 25, 2007
Address correspondence and reprint requests to: Dr Yu-Wen Chen, Department of Nuclear Medicine, Kaohsiung Medical University Hospital, 100 Tzyou 1stRoad, Kaohsiung 807, Taiwan. E-mail: [email protected]
D
ISCORDANT
F
INDINGS OF
S
KELETAL
M
ETASTASIS
B
ETWEEN
T
C
99
M
MDP B
ONE
S
CANS AND
F
18
FDG PET/CT I
MAGING FOR
A
DVANCED
B
REAST AND
L
UNG
C
ANCERS
—T
WO
C
ASE
R
EPORTS AND
L
ITERATURE
R
EVIEW
Yu-Wen Chen,1,6Ming-Yii Huang,2,6Jan-Sing Hsieh,3,6Ming-Fung Hou,3,5,6
Shah-Hwa Chou,4,6and Chih-Liang Lin7
Departments of 1Nuclear Medicine, 2Radiation Oncology, 3Gastroenteric Surgery, 4Thoracic Surgery,
and5Cancer Center, Kaohsiung Medical University Hospital, 6College of Medicine, Kaohsiung
Medical University, and 7Department of Gastroenteric Surgery, Kaohsiung City
United Hospital, Da-Ton Division, Kaohsiung, Taiwan.
Traditionally, Tc99mmethyl diphosphate (MDP) bone scintigraphy provides high-sensitivity
detec-tion of skeletal metastasis from breast and lung cancers in regular follow-up. Fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT), based on the glucose metabolism of malignant cells, plays a role in describing tumor growth, proliferation of neoplasm and the extent of metastasis. In general, concordant findings of skeletal metastasis are seen on both types of image, especially in cases of breast and lung cancer. However, there were extremely discordant findings of skeletal metastasis between bone scans and F18FDG PET/CT imaging in
two cases among 300 consecutive F18FDG PET/CT follow-up exams of patients with
malignan-cies, during the past year, in our center. Both cases, one of breast cancer and one of lung cancer, had negative bone scintigraphic findings, but a diffusely high grade of F18FDG avid marrow
infiltration in the axial spine, leading to the diagnosis of stage IV disease in both cases. Owing to variant genetic aberrance of malignance, F18FDG PET/CT reveals direct evidence of diffuse,
rapid neoplasm metabolism in the bone marrow of the spine, but not of secondary osteoblastic reactions in vivo. F18FDG PET/CT should always be employed in the follow-up of patients with
malignancies.
Key Words:F18FDG PET/CT, skeletal metastasis, Tc99mMDP bone scintigraphy
(Kaohsiung J Med Sci 2007;23:639–46)
The material in this article first appeared in poster form at the 9thWorld Federation of Nuclear
in up to 70% of patients with advanced breast or prostate cancer [2], and in approximately 15–30% of patients with carcinomas of the lung, colon, stomach, bladder, uterus, rectum, thyroid or kidney. Bone metas-tases have been characterized as osteolytic, osteoblastic or mixed lesions containing both elements [3]. Most patients with breast cancer have predominantly oste-olytic lesions, although at least 15–20% of them have predominantly osteoblastic lesions [4]. Once tumors metastasize to bone, they are usually incurable: only 20% of patients with breast cancer are still alive 5 years after the discovery of bone metastasis [5].
Tc99mmethyl diphosphate (MDP) bone
scintigra-phy, which is a cost-effective and useful tool in wide-spread disease, is the most commonly used means of detecting bone metastasis and has variable diagnostic sensitivity with comparatively low specificity. How-ever, F18 fluorodeoxyglucose (FDG) positron
emis-sion tomography/computed tomography (PET/CT) based on the glucose metabolism of malignant cells, may describe tumor growth, proliferation of neoplasm and the extent of metastasis, including distal skeletal involvement. In general, concordant findings of skele-tal metastasis are seen on both types of image, espe-cially in those of breast and lung cancers. However, there were extremely discordant findings of skeletal metastasis between bone scans and F18FDG PET/CT in two cases among 300 consecutive F18FDG PET/
CT follow-up exams during the past year in our PET center.
C
ASEP
RESENTATIONSCase 1
A 51-year-old male patient with chronic cough for more than 3 weeks in February (2006) had received clinic medication to no avail, and was admitted for advanced diagnosis. Tracing his history, only duode-nal ulcer and acute appendicitis post appendectomy, 20 years ago, were revealed. He was, however, a heavy smoker and habitually chewed betel nut. Initially, a large soft mass in the right lower lobe (RLL) of the lung was seen on chest X-ray, and thoracic CT sug-gested lung cancer with mediastinal nodal involve-ment. Based on imaging, the stage was considered to be IIIb, T3N2Mx. Pathology demonstrated adenocar-cinoma, which was determined to be grade II after CT-guided biopsy. Tc99m MDP skeletal scintigraphy
revealed only a hot spot in the right anterior sixth rib, heterogeneous lower grade of Tc99mMDP activity in
the spine, and extraosseously faint Tc99mMDP activity in the RLL lung mass, as shown in Figure 1. Less con-vincing evidence of distal bony metastases was found by a bone scan in March. The patient then received his first course of chemotherapy. F18FDG PET/CT was arranged for treatment follow-up due to an uncontrol-lable lung mass on chest X-ray after 1 month.
Extremely discordant imaging patterns between F18FDG PET/CT and bone scans were seen, as shown in Figure 2. Diffusely extensive spotty high-grade F18
FDG avid marrow activity was distributed in the spine, pelvis and proximal long bone, but there was no obvious lytic or sclerotic bony destruction on CT. A higher grade of RLL lung mass with central necro-sis and mediastinal nodal activity was also noted, but the patient was free of hepatic metastases. Under F18 FDG PET/CT, stage IV disease was diagnosed. Radiotherapy and gefitinib treatment were planned for aggressive control in subsequent months. Another Tc99m MDP bone scan was arranged for July, and a
lower grade of Tc99m MDP with heterogeneous pat-tern in the lumbar spine, bilateral proximal long bones and the old hot spot in the right anterior sixth rib were detected. However, the patient died in late July due to poor hemodynamic function, hepatic failure from severe hepatic metastases and pancytopenia.
Case 2
A 53-year-old female with left breast cancer (infil-trative ductal carcinoma with negative immuno-chemistry of ER/PR/Her2neu) had received modified radical mastectomy in October 2003. Chemotherapy and radiation therapy were also arranged. She had her follow-up in the city hospital. Tc99mMDP skeletal
scintigraphy was arranged to be administered annu-ally. On serial Tc99mMDP scans, there was a hot spot
in the left hemivertebral body of L2 with mildly progressive activity. Therefore, possible degenerative change was considered first, as shown in Figure 3. However, progressive elevation of the tumor marker CA153 to over 130 IU/mL was noted. Under an im-pression of distal metastases, F18FDG PET/CT was
arranged. Typically, multiple high-grade F18FDG avid
osteolytic bony metastases in the spine, pelvis and proximal long bones were seen, as shown in Figure 4. Magnetic resonance imaging also demonstrated distal bony metastases. Due to severe pathologic fractures,
Ant. Post. Ant. Post.
March 2006 July 2006
Figure 1.Two MDP skeletal scintigraphies were performed in early March and July. Visualized activity in both soft tissue and kidneys did not support the diagnosis of the super scan. Initial skeletal imaging in March revealed tiny spot activity in the right anterior aspect of the sixth and fourth ribs, heterogeneous spinal activity and extraosseous tumor activity in the right lower lobe of the lung. The second skeletal imaging in July showed diffusely heterogeneous increased MDP activity in bilateral proximal humerus, femur and sacroiliac joint. Although a progressive change in the MDP scintigraphic pattern was seen, less convincing evidence of typical multiple bony metastases was demonstrated.
Figure 2.FDG PET/CT was performed later in March due to restaging of lung adenocarcinoma. Unfortunately, a large high-grade FDG avid pulmonary mass with central necrosis, conglomerated bilateral mediastinal/supraclavicle nodal activity, right pleural effu-sion and extremely disseminated heterogeneous nest-like high-grade FDG avid neoplasm metastatic foci within the marrow cavity of the spine, pelvis, rib cage and bilateral proximal long bones were seen on FDG PET/CT. Stage IV disease and poor prognosis were indi-cated. However, no evidence of hepatic metastasis was demonstrated by FDG PET/CT at this time.
Ant. Post. Ant. Post.
A B
Figure 3.Anterior and posterior views of skeletal scintigraphy in: (A) 2005; (B) 2006. Progressively heterogeneous MDP activity in the posterior aspect of L2 (arrow) and faint activity in L3 were seen between serial annual skeletal scintigraphies. However, no overt evidence of abnormal MDP activity in the rest of the skeletal system was noted. Based on one spinal hot lesion and elevation of serum CA153, FDG PET/CT was arranged for further evaluation.
A B
Figure 4.(A) Magnetic resonance imaging (MRI). (B) FDG PET/CT. An L2 pathologic compression fracture (thin arrows) is demon-strated on MRI and FDG PET/CT. Multiple vertebral bony metastases (thick arrows) in the T/L spine and sacrum are also seen on MRI and FDG PET/CT, but are discordant with the findings of serial skeletal scintigraphy. Diagnosis of osteolytic bone metastases is favored.
she received surgical fixation. Final pathology showed bony metastasis of breast cancer (Figure 5).
D
ISCUSSIONHigh blood flow, certain adhesive molecules of tumors and a large repository of immobilized growth factors, including transforming growth factor-β, insulin-like growth factor I and II, fibroblast growth factors, platelet-derived growth factors, bone morphogenetic proteins and calcium, together, result in bone becom-ing the preferred site of metastases [3]. These growth factors, which are released and activated during bone resorption, provide a fertile ground in which tumor cells can grow. Stephan Paget first proposed the “seed-and-soil hypothesis” to explain the mechanism of bone metastasis in 1889, and animal models of bone metas-tasis support this hypothesis [1]. During normal bone turnover, there is a balanced remodeling sequence: osteoclasts resorb bone first and then osteoblasts
form bone at the same site. Osteoclasts arise from precursor cells in the monocyte-macrophage lineage [6], and these differentiate into inactive osteoclasts. Expression of receptor activator of nuclear factor-κB ligand (RANKL) and macrophage colony stimulating factor by stroma cells or osteoblasts plays a critical role in the maturation of osteoclasts.
Most osteotropic factors, such as parathyroid hor-mone, 1,25-dihydroxyvitamin D3 and prostaglandins, induce the formation of osteoclasts by increasing the expression of RANKL on marrow stroma cells and osteoblasts, rather than by acting directly on clast precursors [7,8]. The ratio of RANKL to osteo-protegerin regulates the formation and activity of osteoclasts. Osteoclasts resorb bone by secreting pro-teases that dissolve the matrix and produce acid that releases bone mineral into the extracellular space under the ruffled border of the plasma membrane of osteoclasts, which faces bone and is the resorbing organelle of the cell. Osteoblasts, the bone-forming cells, arise from mesenchymal stem cells, which form
A B
C
Figure 5.Pathology shows: (A) infiltrative ductal carcinoma of the breast; (B, C) malignant bony metastasis from bony fragment after fixation of pathologic fracture. [Figures provided by United City Hospital, Kaohsiung, Taiwan.]
osteoblasts, adipocytes and muscle cells. CBFA1 (core-binding factor α1) is a critical factor for the differen-tiation of osteoblasts, and many growth factors can enhance the maturation of these cells. However, the differentiation of osteoblasts is less well understood than that of osteoclasts.
We are familiar with purely osteolytic bone destruc-tive metastasis in multiple myelomas, due to osteoblas-tic dysfunction with unknown etiology. However, overproduction of urokinase-type plasminogen acti-vator (u-PA) by prostate-cancer cells increases bone metastasis [9]. Prostate cancer cells also release prostate-specific antigen and kallikrein, a serine pro-tease, which can cleave parathyroid hormone (PTH)-related peptide, resulting in block of tumor-induced bone resorption [3]. Kallikrein may also activate osteo-blast growth factors. Bone metastases from prostate cancer are predominantly osteoblastic, with increased numbers of irregular bone trabeculae [10].
In osteolytic metastases, the destruction of bone is mediated by osteoclasts rather than tumor cells [11,12]. However, the factors responsible for the activation of osteoclasts vary depending on the tumor. Several osteoclastogenic factors have been implicated in the in-creased activity of osteoclasts, including interleukin-1, interleukin-6, macrophage inflammatory protein 1α and RANKL. In the vicious cycle of osteolytic metas-tasis, especially in breast cancer, tumor cells secrete PTH-related peptide as the primary stimulator of osteo-clastogenesis. In addition, tumor cells produce other factors that increase the formation of osteoclasts, including interleukin-6, prostaglandin E2, tumor necrosis factor and macrophage colony-stimulating factor. These factors increase the expression of receptor activator of RANKL, which acts directly on osteoclast precursors to induce the formation of osteoclasts and bone resorption.
In Case 2, there was progressive Tc99mMDP
accu-mulation in L2, probably related to osteoblastic remod-eling after pathologic fracture. However, the other multiple high-grade F18FDG avid osteolytic
metas-tases in the T/L spines and pelvis, observed by mag-netic resonance imaging and F18 FDG PET/CT, are not consistent with Tc99mMDP skeletal scintigraphy
for osteoblastic activity. Distal marrow/bone metas-tases of solid malignant neoplasms, such as those of breast cancer/lung cancer, have variable patterns of osteoblastic/osteolytic activity. Therefore, moni-toring bony metastases should be based on multiple
imaging modalities, in addition to Tc99mMDP
skele-tal scintigraphy (which was the only modality used traditionally).
In Case 1, the patient with lung adenocarcinoma, extremely discordant imaging results were found among the osteoblastic activity detected by Tc99mMDP
skeletal scintigraphy, osteolytic activity detected by skeletal CT and marrow activity detected by F18FDG
PET. Based on this phenomenon, we considered the case to be one of diffusely high-grade FDG avid neo-plasm occupying the marrow, but not evoking an effect on osteocytes. Homogeneously reactive marrow activ-ity post chemotherapy seems less likely. Unfortunately, this finding is consistent with poor prognosis [13], despite shrinkage of tumor size after radiation therapy. F18FDG PET/CT should always be employed to detect
marrow/bony metastases in detail, following screen-ing studies usscreen-ing Tc99mMDP skeletal scintigraphy.
REFERENCES
1. Paget S. The distribution of secondary growths in can-cer of the breast. Lancet 1889;1:571–3.
2. Coleman RE, Rubens RD. The clinical course of bone metastases from breast cancer. Br J Cancer 1987;55:61–6. 3. Roodman GD. Mechanisms of bone metastasis. N Engl
J Med 2004;350:1655–64.
4. Coleman RE, Seaman JJ. The role of zoledronic acid in cancer: clinical studies in the treatment and preven-tion of bone metastases. Semin Oncol 2001;28(Suppl 6): 11–6.
5. Coleman RE. Metastatic bone disease: clinical features, pathophysiology and treatment strategies. Cancer Treat
Rev 2001;27:165–76.
6. Roodman GD. Cell biology of the osteoclast. Exp
Hematol 1999;227:1229–41.
7. Yasuda H, Shima N, Nakagawa N, et al. Osteoclast differentiation factor is a ligand for osteoprotegerin/ osteoclastogenesis-inhibitory factor and is identical to TRANCE/RANKL. Proc Natl Acad Sci USA 1998;95: 3597–602.
8. Hofbauer LC, Heufelder AE. Osteoprotegerin and its cognate ligand: a new paradigm of osteoclastogenesis.
Eur J Endocrinol 1998:139:152–4.
9. Achbarou A, Kaiser S, Tremblay G, et al. Urokinase overproduction results in increased skeletal metastasis by prostate cancer cells in vivo. Cancer Res 1994;54: 2372–7.
10. Roudier MP, Vesselle H, True LD, et al. Bone histology at autopsy and matched bone scintigraphy findings in patients with hormone refractory prostate cancer: the effect of bisphosphonate therapy on bone scintigraphy results. Clin Exp Metastasis 2003;20:171–80.
11. Taube T, Elomaa I, Blomquvist C, et al. Histomorpho-metric evidence for osteoclast-mediated bone resorp-tion in metastatic breast cancer. Bone 1994;15:161–6. 12. Boyde A, Maconnachie E, Reid SA, et al. Scanning
elec-tron microscopy in bone pathology: review of methods,
potential and applications. Scan Electron Microsc 1986; 4:1537–54.
13. Prevost S, Boucher L, Larivee P, et al. Bone marrow hypermetabolism on FDG PET as survival prognostic factor in non-small-cell lung cancer. JNM 2006;47:559–65.
!"VS==N==NV= !"VS==Q==OR= !"#$%&' !"!#$ %&' !( UMT !"#$NMM
!"#$%&'()*+,-.
!"#$%&'()*+,-.
!"#$= = !"
!
NIS== OIS== PIS== PIRIS
QIS== T !"!#$ %= =N !"= =O !"= =P !"#= =Q != =R ! S !"!= = ! T !"#$%&'(%= = !"# !=VVã= !"#$%&'()*+,-./$0123456789 !"#$%&'()*+,-./0123 456789:;<-./= !"#$%&'(!)*+,*-./0123456789:;<=> !"#$%&'()*+,-./0123456789*+:;<=>? !"#$%&'()*+,-./0123)456789:0;<3=,- !"#$%&'()*+,-./0123456/78-9:;%<=> !"#$%&'()*+,-./01"23456789:5;<=>?@ !"#$%&'()#$*+,-./012345$6789:2;<= !"#$%&#'()!*+,-./0123!456789:;<=>? !"#$%&'()*+,-./"01234 !"#$%&'()*+,-./0-123456.7+, E !=OMMTXOPWSPVQSF
