ORIGINAL ARTICLE 
Annals of Nuclear Medicine Vol. 14, No. 2, 103-109, 2000 
The usefulness of 99mTc-HMPAO-labeled leukocyte scintigraphy in the diagnosis of skeletal metastases of cancers 
Manabu KINOSHITA,* Yukinori IMAI,* Kenji FUJIWARA* and Tatsuya MIYAMAE** 
*Third Department of Internal Medicine, Saitama Medical School 
**Department of Radiology, Saitama Medical School 
The usefulness of bone marrow scintigraphy with 99mTc-HMPAO-labeled leukocytes (leukocyte bone marrow scintigraphy) in the diagnosis of skeletal metastases of cancers was investigated in 70 lesions in 27 patients with various types of cancer. The final diagnosis of skeletal metastases was based on one or more criteria consisting of histological confirmation, typical findings of metastases by bone radiograph, CT and MRI, or progressive swellings of the lesions with severe pain due to nerve compression. Of the 70 lesions, 55 were finally diagnosed as metastases, and 15 as benign lesions. Leukocyte bone marrow scintigraphy showed photopenic defects in 52 of the 55 metastatic lesions (sensitivity 95%), and the remaining 3 negative lesions were found positive for metastases by MRI. In contrast, MRI could evaluate only 39 of the 55 lesions because 16 lesions in the ribs, scapula and sternum were not visualized. Of these 39 lesions, MRI showed positive findings for metastases in 33 (sensitivity 85%), and negative findings in 6 with photopenic defects found by leukocyte bone marrow scintigraphy. Of the 15 benign lesions, 3 were false positive for metastases on leukocyte bone marrow scintigraphy (specificity 80%). We conclude that 99mTc-HMPAO-labeled leukocyte bone marrow scintigraphy may be useful in the diagnosis of skeletal metastases of cancers, particularly when MRI fails to evaluate the lesions. 
Key words: metastasis 99mTc-HMPAO-labeled leukocyte scintigraphy, bone marrow scintigraphy, skeletal 
INTRODUCTION 
BONE SCINTIGRAPHY is usually used as a screening test to detect skeletal metastases of cancers, but this technique alone cannot differentiate the metastases from benign skeletal lesions.1-5 Although MRI and CT are helpful in this differentiation,6,7 systemic examination with them is not practical because of the excessive radiation dose, time consumption or cost benefit.8 
Skeletal metastases of cancers occur via the blood-stream. Cancer cells are assumed to reach the bone marrow first, and then the bone,9-11 suggesting that the early 

Received August 25, 1999, revision accepted December 27, 1999. 
For reprint contact: Manabu Kinoshita, M.D., Third Department of Internal Medicine, Saitama Medical School, 38 morohongou, Moroyama-machi Iruma-gun, Saitama 350-0451, JAPAN. 

diagnosis of skeletal metastases of cancers may be made by evaluating bone marrow abnormalities on images. 111In-chloride and 99mTc-colloid are commonly employed for bone marrow scintigraphy,12 but such scintigraphy cannot always provide clear images to precisely analyze bone marrow lesions. 99mTc-HMPAO-labeled leukocyte scintigraphy is widely used in the detection of pathological conditions such as abdominal and skeletal inflammatory diseases.13-15 
In the present study we evaluated the utility of 99mTc-HMPAO-labeled leukocyte scintigraphy (leukocyte bone marrow scintigraphy) in the diagnosis of skeletal metastases of cancers. 
SUBJECTS 
From among patients who were admitted to our hospital between 1996 and 1998 with histological or clinical diagnosis of cancers, 27 patients who had one to ten suspicious skeletal metastases of cancers were selected as subjects. The primary cancers were hepatocellular carcinoma in 12 patients, gastric cancer in 7, gallbladder cancer in 2, and prostate cancer, lung cancer, breast cancer, colorectal cancer, esophageal cancer, and unknown primary tumor each in 1. Of these patients, 24 showed signs of abnormal accumulations on bone scintigraphy, and 3 had clinical symptoms characteristic of skeletal metastases of cancers despite having no abnormal accumulations on bone scintigraphy. In these 27 patients, altogether 73 skeletal lesions were evaluated. All the lesions consisted of those whose diagnosis of metastases were not definite on bone scintigraphy. 
METHODS 
Forty milliliter of venous blood was collected from each patient, and leukocytes were harvested according to the method of McAfee et al.16 The leukocytes were labeled with 99mTc-HMPAO (550 MBq) by the modified method of Aburano et al.,13 and administered intravenously back to the patient. At 4 hours after the intravenous administration, this leukocyte bone marrow scintigraphy could provide delineated clear planar anterior and posterior images, by using a dual-head gamma camera (Prism 2000 Picker) and a low-energy parallel-hole collimator. Spinal SPECT was performed under the following conditions: 360deg., 72 steps, 128 x 128 matrix, 30 second intervals. Images were prepared with a ramp filter and Butterworth filter (order: 8, cut-off frequency: 0.25 cycle per pixel). 
Focal photopenic defects found by leukocyte bone marrow scintigraphy were diagnosed as positive for skeletal metastases of cancers. 
MRI was performed with a 1.5 T machine (Magnetom H15SP, Siemens, Erlangen, Germany). T1-weighted images were captured with a spin echo sequence (TR/TE = 500/12), and T2-weighted with a spin echo sequence (TR/TE = 4500/120). STIR images were obtained with a Short-Turn-Inversion Recovering (TR/TE/TI = 4000/30/150). 
The final diagnosis of skeletal metastases of cancers in each patient was made when at least one of the following criteria was fulfilled. 
1) Cancer tissue in the skeletal lesion was confirmed by histological examination. 
2) Low signal T1-weighted and high signal T2-weighted images on MRI were detected in the skeletal lesions. 
3) Bone destruction, osteogenic change, and tumor formation detected by CT or bone radiograph were detected in the skeletal lesions. 
4) Patients were diagnosed as having skeletal metastasis when the results of follow-up tests confirmed skeletal metastasis even if the results of the initial tests were negative. 
5) Even if the above four criteria were negative, progressive swellings of the lesions were seen with severe pain due to nerve compression during the follow-up observation period. 
Verbal or written consent to participate in the study was obtained from each patient or members of his/her family. 
RESULTS 
Figure 1 shows a leukocyte bone marrow scintigram of a healthy man. On a planar image (posterior view) the entire bone marrow was clearly delineated, and on a SPECT image the spinal bone marrow was detectable without any interference by the liver and spleen. 
In 27 patients both planar and spinal SPECT images by leukocyte bone marrow scintigraphy were obtained in 70 of the 73 lesions. Two lesions in the thoracic vertebrae and one lesion in the cervical vertebra were too remote from other lesions to be examined at the same time. 
Table 1 shows the results of leukocyte bone marrow scintigraphy in 55 skeletal lesions in 16 patients who had a final diagnosis of skeletal metastases of cancers. Of the 55 lesions, 52 were positive for metastases on leukocyte bone marrow scintigraphy (sensitivity 95%), and the remaining 3 negative lesions were positive for metastases on MRI. These 3 lesions were less than 0.5 cm diameter. In contrast, MRI evaluated only 40 lesions, because 13 in the ribs, 1 in the scapula and 2 in the sternum could not be visualized. Of the 39 metastatic lesions, 6 lesions positive for metastases on leukocyte bone marrow scintigraphy were negative on MRI (sensitivity 85%). 
Table 2 demonstrates the results of leukocyte bone marrow scintigraphy in 14 skeletal lesions in 11 patients who were finally diagnosed as having no skeletal metastases of cancers. Of these 14 lesions, 3 were positive for metastases on leukocyte bone marrow scintigraphy, but negative on MRI. Two of these 3 lesions were found in older patients with markedly deformed vertebrae. But MRI was positive for metastases in one lesion found negative on leukocyte bone marrow scintigraphy. 
The specificity for the diagnosis of skeletal metastases of cancers by leukocyte bone marrow scintigraphy was 80%. 
CASES 
Patient No. 1 
A 26-year-old man with hepatocellular carcinoma complained of severe back pain. 99mTc-MDP bone scintigraphy (posterior view) showed no abnormal accumulation anywhere in the body. Nevertheless, leukocyte bone marrow scintigraphy (posterior view) disclosed focal photopenic defects in the second and eleventh thoracic vertebrae, first, third and fifth lumbar vertebrae and first sacral vertebra. MRI detected several low signal intensities on T1-weighted images of the same regions. Figure 2 shows three of these lesions. 
Patient No. 3 
A 62-year-old man with hepatocellular carcinoma developed severe pain in the chest, back and both legs 99mTc-MDP bone scintigraphy disclosed several accumulations in the ribs, sternum and thoracic and lumbar vertebrae. Leukocyte bone marrow scintigraphy also revealed focal photopenic defects in the same regions. Figure 3 shows such defects in the ribs and sternum. 
Patient No. 15 
A 71-year-old man with gastric cancer experienced severe back and lumbar pain. A bone radiograph showed compression fractures in the ninth thoracic vertebra and first lumbar vertebra 99mTc-MDP bone scintigraphy revealed abnormal accumulations in the same regions. A T1-weighted MRI image disclosed low signal intensity in the ninth thoracic vertebra. This site was enhanced by contrast imaging with Gd-DTPA. But in the first lumbar vertebra Gd-DTPA contrast imaging was negative despite a positive low signal intensity on MRI, characteristic of benign compression fracture. Leukocyte bone marrow scintigraphy (SPECT) showed focal photopenic defects in the ninth thoracic vertebra and first lumbar vertebra (Fig. 4). 
DISCUSSION 
Based on the assumption that the early diagnosis of skeletal metastases of cancers may be made by analyzing bone marrow abnormalities on images, we studied the utility of 99mTc-HMPAO-labeled leukocyte scintigraphy in the present study.
In order to improve unsatisfactory bone marrow scintigraphy images obtained with 111In-Cl3, 99mTc-colloid preparations, etc.,12 we employed 99mTc-HMPAO-labeled leukocyte scintigraphy, because granulocytes in the circulation derived from bone marrow are supposed to reach the reticuloendothelial system and accumulate in the bone marrow.10 To prepare leukocytes and to label with 99mTc-HMPAO, we used standard methods established in the detection of inflammatory abdominal, bone and joint diseases.13-17 Scans were performed 4 hours after intravenous administration of labeled leukocytes, as this leukocyte bone marrow scintigraphy provided delineated clear planar anterior and posterior images, by using a dual-head gamma camera (Prism 2000 Picker) and a low-energy parallel-hole collimator in our preliminary study. The subjects were selected from among patients with cancers who revealed abnormal accumulations on 99mTc-MDP bone scitigraphy routinely performed for cancerous patients and those who complained of severe pain despite having no such accumulations. The final diagnosis of skeletal metastases was made by fulfillment of one or more criteria consisting of histological confirmation, typical findings of metastases by bone radiograph, CT and MRI, or progressive swellings of the lesions accompanied by severe pain due to nerve compression. 
As shown in Table 1, 52 of 55 skeletal metastatic lesions showed signs of focal photopenic defects on this bone marrow scintigraphy (sensitivity 95%). The present results were noteworthy, because 16 of these 55 lesions including 13 in the ribs, 1 in the scapula and 2 in the sternum which could not be visualized by MRI. Moreover, leukocyte bone marrow scintigraphy showed positive findings for skeletal metastases in 6 lesions which were negative for metastases on MRI. MRI is capable of detecting the replacement of adipose cells by tumor cells in bone marrow in the presence of skeletal metastases of cancers.7 Similar replacement also occurs when bleeding exists in the bone marrow, suggesting that metastatic lesions cannot be detected clearly in cases of coexisting bleeding due to bone fractures.18-21 This may explain the higher sensitivity of leukocyte bone marrow scintigraphy than that of MRI. 
Of 14 skeletal lesions in 11 patients who were finally diagnosed as having no skeletal metastases of cancers, 3 revealed focal photopenic defects in the bone marrow on leukocyte bone marrow scintigraphy, as shown in Table 2. These false positive findings were not seen in MRI images. The exact reason for this false positive findings should be investigated in the future. Even if positive findings for metastases are obtained by leukocyte bone marrow scintigraphy, evaluation by MRI is advisable in cases of suspected benign compressions. 
Bone marrow scintigraphy with 99mTc-labeled antigranulocyte monoclonal antibody has been used to provide better images than those obtained with 111In-chloride scanning in the diagnosis of skeletal metastases as well.8,22,23 But this scintigraphy involves such problems as image deterioration caused by repeated use and allergic reaction caused by high serum human antimouse antibody.22 The use of 99mTc-HMPAO-labeled leukocyte scintigraphy in the evaluation of bone marrow was reported by Watanabe et al.24 This is the only study that utilized the present scintigraphy in the diagnosis of skeletal metastases in a case of malignant lymphoma. The present study is the first attempt to investigate the utility of the method in a large number of cases. 
For leukocyte bone marrow scintigraphy, as much venous blood as 40 ml is necessary, and separation and labeling of leukocytes take 2 hours. It would therefore be advisable to use this scintigraphy in the case of suspected skeletal metastases of cancers the diagnosis of which is undetermined by bone scintigraphy, along with conventional bone radiograph, particularly when skeletal metastases are present in multiple lesions including the ribs, sternum, scapula, and there are coexisting compression fractures. 
CONCLUSION 
99mTc-HMPAO-labeled leukocyte scintigraphy may be useful in the diagnosis of skeletal metastases of cancers. Its use is preferable when the diagnosis is undetermined by bone scintigraphy along with conventional bone radiograph, particularly in cases of multiple skeletal metastases existing in lesions undetectable by MRI and coexisting compression fractures. 
REFERENCES 
1. Kamby C, Vejborg I, Daugaard S, Guldhammer B, Dirksen H, Rossiing N. Clinical and radiological characteristics of bone metastases in breast cancer. Cancer 60: 2524-2531, 1987. 
2. Michel P, Soler M, Imhof E, Perrunchoud AP. Initial staging of non-small lung cancer value of routine radioisotope bone scanning. Thorax 146: 469-473, 1991. 
3. Piffer S, Amichetti M, Valentini A. Skeletal scintigraphy and physical examination in the staging of early breast cancer. Acta Oncol 27: 21-24, 1988. 
4. Brown B, Laorr A, Greenspan A. Negative bone scintigraphy with diffuse osteobrastic breast carcinoma metastases. Clin Nucl Med 19: 194-196, 1994. 
5. Clitrin DL, Bessent RG, Greig WR. A comparison of sensitivity and accuracy of 99mTc-phosphate bone scan and skeletal radiograph in the diagnosis of bone metastasis. Clin Radiol 28: 107-117, 1977. 
6. Mehta RC, Wilson MA, Perlman SB. False-negative bone scan in extensive metastatic disease: CT and MRI finding. J Compute Assist Tomogra 13: 717-719, 1989. 
7. Zimmer WD, Bergquist TH, McLeod RA. Bone tumors. Magnetic resonance imaging versus computed tomography. Radiology 155: 709-718, 1985. 
8. Lee KH, Chung JK, Choi CW, Jeong JM, Lee DS, Bang YJ, et al. Technetium-99m-Labeled Antigranulocyte Antibody Bone Marrow Scintigraphy. J Nucl Med 36: 1800-1805, 1995. 
9. Beth A, Berettoni BA, Carter JR. Mechanisms of cancer metastasis to bone. J Bone Joint Surg 68: 308-312, 1986. 
10. Distefano A, Tashima CK, Yap Y, Hortobagyi GN. Bone marrow metastasis without cortical bone involvement in breast cancer patients. Cancer 44: 196-198, 1979. 
11. Kamby C, Guldhammer B, Vejborg I, Rossing N, Dirksen H, Daugaard S. The presence of tumor cells in bone marrow at the time of first recurrence of breast cancer. Cancer 60: 1306-1312, 1987. 
12. Reske SN. Recent advances in bone marrow scanning. Eur J Nucl Med 18: 203-221, 1991. 
13. Aburano T, Taki J, Kinuya S, Shuke N, Yokoyama K, Takayama T, et al. Clinical application of Tc-99m HMPAO labeled leukocyte imaging in inflammatory disease. KAKU IGAKU (Jpn J Nucl Med) 28: 1143-1150, 1991. 
14. Frederick L, Datz FL, James E, Seabold E, Manuel L, Brownl L, et al. Guideline for Technetium-99m-leukocyte Scintigraphy. J Nucl Med 38: 986-990, 1997. 
15. Peters AM, Danpure HJ, Osman S, Hawker RJ, Henderson BL, Hodgson KH, et al. Clinical experience with 99mTc-hexamethyl-propylene-amine-oxime for labelling leucocytes and imaging inflammation. Lancet 2: 946-949, 1986. 
16. McAfee JG, Subramanian G, Gagne G. 99mTc-HMPAO for leucocyte labeling experimental comparison with 111In oxine in dog. Eur J Nucl Med 13: 353-357, 1987. 
17. Edelstyn GA, Gillespite PJ, Grebbell FS. The radiological demonstration of osseous metastasis. Experimental observation Clin Radiol 18: 158-162, 1967. 
18 Li KC, Poon PY. Sensitivity and specificity of MRI in detecting malignant spinal cord compression and distinguishing malignant from benign compression fracture of vertebrae. Magn Reson Imaging 6: 547, 1993 
19. Yuh WT, Zachar CK, Barloon T. Vertebral compression fractures distinction between benign and malignant causes with MR imaging. Radiology 172: 215-218, 1989 
20. Baker LL, Goodman SB, Perkesh I. Benign versus pathologic compression fracture of vertebral bodies assessment with conventional spin-echo, chemical-shift, and STIR MR imaging. Radiology 174: 495-502, 1990. 
21. Porter BA, Sields AF, Olson DO. Magnetic resonance imaging of bone marrow disorder. Radiol Clin North Am 24: 269, 1986. 
22. Joseph K, Hoffken H, Bosslet K. In vivo labelling of granulocytes with 99mTc anti-NCA monoclonal antibodies for imaging inflammation. Eur J Nucl Med 14: 367-373, 1988. 
23. Higuchi T, Inoue T, Sarwar M, Origuchi N, Karasawa M, Naruse T, et al. 99mTc-labelled chimeric human/mouse antigranurocyte antibody bone marrow scintigraphy: A preliminary clinical study. Nucl Med Commun 19: 463-474, 1998. 
24. Watanabe N, Nomura K, Tsuji S, Futatsuya R, Kamei T, Seto H, et al. A case of bone marrow metastasis of malignant lymphoma imaging with 99mTc-HMPAO-labeled leukocytes. Kaku Igaku Gazoushindan 6: 40-42, 1991.