- Email: [email protected]
Positron Emission Tomography with 18F-Fluorodeoxyglucose of Uterine Sarcoma: A Comparison with Magnetic Resonance Imaging and Power Doppler Imaging
Gynecologic Oncology 80, 372–377 (2001) doi:10.1006/gyno.2000.6081, available online at http://www.idealibrary.com on
Positron Emission Tomography with 18F-Fluorodeoxyglucose of Uterine Sarcoma: A Comparison with Magnetic Resonance Imaging and Power Doppler Imaging Naohiko Umesaki, M.D., Tetsuji Tanaka, M.D.,* Masato Miyama, M.D.,* Naoki Kawamura, M.D.,* Sachio Ogita, M.D.,* Jyoji Kawabe, M.D.,† Terue Okamura, M.D.,† Koichi Koyama, M.D.,† and Hironobu Ochi, M.D.† Department of Obstetrics and Gynecology, Wakayama Medical College, 811-1 Kimiidera, Wakayama 640-0012, Japan; and *Department of Obstetrics and Gynecology, and †Division of Nuclear Medicine, Osaka City University Medical School, 1-4-3 Asahi-machi, Abeno-ku, Osaka 545-8585, Japan Received August 11, 2000; published online January 31, 2001
Objective. The effectiveness of positron emission tomography with 18F-fluorodeoxyglucose (FDG-PET) for diagnosis of uterine sarcoma was evaluated in comparison to the effectiveness of magnetic resonance (MR) imaging and power Doppler imaging. Method. The cases of five Osaka City University Hospital patients diagnosed with uterine sarcoma based on histopathological examination, in whom FDG-PET, MR imaging, and power Doppler imaging studies had been performed preoperatively, were reviewed. A comparative study of the usefulness of these three imaging modalities for diagnosis of sarcoma was conducted. Tumors comprised three leiomyosarcomas, one endometrial stromal sarcoma, and one carcinosarcoma. Results. FDG-PET examinations were 100% positive for the five sarcomas; MR imagings were 80% positive (four of five cases), and US was 40% positive (two of five cases). The mean strandardized uptake value of the sarcomas was 4.5 ⴞ 1.3. Conclusion. The sarcoma lesions were clearly imaged by FDGPET. FDG-PET may be a most useful diagnostic method for uterine sarcoma. © 2001 Academic Press
INTRODUCTION Uterine sarcomas are generally classified as leiomyosarcoma (LMS), endometrial stromal sarcoma (ESS), or carcinosarcoma. Although uterine sarcoma occurs in less than 5% of all uterine corpus malignancy cases, these tumors are the most lethal. The clinical features of LMS resemble those of leiomyoma (LM), the most common benign neoplasm of the female genital tract. Thus, differentiation between the two is very important. Magnetic resonance (MR) imaging is at present considered the most useful method for diagnosing uterine sarcoma, but it is not applicable to atypical sarcoma cases [1]. Some patients require unnecessary surgery for a differential diagnosis. Moreover, the recent trend toward conservative treatment of LM with gonadotropin-releasing hormone agonist raises the possibilities for mis- or delayed diagnosis. Thus, a new diagnostic approach to LMS has been anticipated. 0090-8258/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.
Metabolic imaging by positron emission tomography (PET) with positron tracers such as L- [methyl- 11C] methionine (carbon-11-methionine) or 2-[ 18F] fluoro-2-deoxy-D-glucose (FDG) is now being used to evaluate suspected malignant tumors [2– 4]. PET imaging differs considerably from conventional imaging modalities such as MR, computerized tomography, and ultrasonography because it quantifies the functional activity of tissue but not its morphological abnormality. We reviewed FDG-PET images of five uterine sarcomas. Histopathological examinations, blood tests, and MR and power Doppler ultrasonography images were also reviewed. The effectiveness of FDG-PET for diagnosis of sarcoma was evaluated in comparison to that of these other examination methods. MATERIALS AND METHODS Patients The cases of five patients (age, 50 – 69 years; mean ⫾SD, 58.0 ⫾ 10.1) who were admitted to Osaka City University Hospital between April 1998 and March 2000 and diagnosed with uterine sarcoma by histopathological examination and in whom preoperative FDG-PET, MR, and power Doppler imaging studies had been performed were reviewed. The usefulness of these imaging modalities for diagnosis of uterine sarcoma was evaluated retrospectively. As points of reference, blood tests and endometrial cytology or histology findings were also evaluated. The sarcomas included three LMSs, one ESS, and one carcinosarcoma. Four patients had primary tumors and one patient a recurrent tumor. The histopathological diagnosis of sarcoma was made based on surgical specimens in four cases and needle biopsy specimen in the one recurrent case. PET Study Radiolabeled FDG was produced with an NKK-Oxford superconducting cyclotron and an NKK synthesis system (NKK Corp., Kawasaki, Japan). A Headtome IV SET-1400W-10
372
FDG-PET FOR UTERINE SARCOMA
373
FIG. 1. A 52-year-old woman with uterine endometrial stromal sarcoma. (A) MR T1-weighted image (TSE; TR/TE⫽631/15) of the pelvis (transaxial view). The ESS lesion is observed as a high signal intensity area (T) in the uterus (U). Leiomyomas (L) are located posteriorly and to the right of the ESS lesion. (B) MR T2-weighted image (TSE; TR/TE⫽4000/102) of the pelvis (transaxial view). The ESS lesion is observed as a heterogeneous area of medium or high intensity (T) in the uterus (U). (C) FDG-PET image (transaxial view). Accumulated FDG (SUV⫽5.5) is seen in the tumor (T). (D) Transvaginal power Doppler image (sagital section) of the uterine tumor. A rich mosaic-like vasculature (arrow) is observed in the tumor (T).
(Shimadzu Corp., Kyoto, Japan) was used for the PET study. All patients fasted for at least 4 h before PET scanning. Transmission scans were obtained using a 68Ge rod source for attenuation correction, followed by intravenous injection of 185 MBq FDG. Emission scanning was performed from 45 to 55 min after the injection of FDG. The urinary bladder was
irrigated just before and during imaging in all patients with 2 L of saline solution to reduce the accumulation of FDG in the bladder, since bladder FDG can interfere with proper evaluation of PET images of the pelvic cavity. Tracer accumulation in elliptical regions of interest (ROI, 5 mm in diameter) on PET images was measured based on the standardized uptake value
374
UMESAKI ET AL.
FIG. 1—Continued
(SUV), i.e., the radioactive concentration (Bq/ml) in a hot spot divided by the injected FDG dose (Bq) and the patient’s body weight (g).
Power Doppler Ultrasonographic Examination Ultrasonographic examination was performed with a LOCIQ 500 ultrasonic scanner (GE Yokogawa Medical Systems, Tokyo, Japan) in power Doppler modes.
MR Imaging RESULTS MR imaging was performed with a 1.5-T imaging system (Signa, General Electric Medical System, Milwaukee, WI). Sagittal and transverse T1-weighted images and T2-weighted images were obtained.
The results of histopathological examinations, blood tests, and imaging diagnoses of the five patients are summarized in Table 1. Cytological and histopathological examinations re-
375
FDG-PET FOR UTERINE SARCOMA
FIG. 2. A 50-year-old woman with uterine leiomyosarcoma. (A) MR T1-weighted image (TSE; TR/TE⫽500/15) of the pelvis (transaxial view). A high signal intensity lesion (arrow) in the uterus (U) is observed. (B) FDG-PET image (transaxial view). Accumulated FDG (arrow) (SUV ⫽ 3.0) is seen in the uterus (U). A high signal intensity lesion in (2A) is necrosis. (C) Transvaginal power Doppler image (sagital section) of the uterine tumor. No vasculature is seen in the tumor (T).
vealed malignancy in only one case; positive LDH level (ⱖ500 IU/L) was demonstrated in one case, and a positive CA-125 level (ⱖ35 U/ml) was demonstrated in one case. MR imaging showed a high intensity area on T1-weighted images in four cases, and moderate- to high-intensity areas were demonstrated on T2-weighted images in all cases (Figs. 1 and 2). Intra-tumoral blood vessel patterns on ultrasonic power Doppler images were mosaic in two cases.
The SUVs of the sarcomas varied from 3.0 to 6.3; the mean ⫾ SD SUV was 4.5 ⫾ 1.3. PET studies were positive in all five cases, lending a strong suspicion for malignancy. DISCUSSION For diagnosis of uterine malignancy, endometrial cytology examinations or histopathological examination of endometrial
376
UMESAKI ET AL.
FIG. 2—Continued
curettage specimens are usually performed. Sarcomas derived from uterine muscle or mesenchymal tissue, however, cannot be found easily until malignant tissue invades the endometrium. Endometrial curettage is useful for diagnosis in only 30% of patients with LMS [5]. This is also true in patients with ESS. In the present five cases, three of four primary uterine
sarcomas were negative based on endometrial cytology and endometrial curettage specimens. LDH can be a useful marker of sarcoma in some cases [6]. Only one of our patients showed a positive LDH value. A positive CA-125 pattern was found in one case, the value of which in the diagnosis of sarcoma is considered low. As Peter
TABLE 1 Patient Characteristics and Results of Laboratory and Imaging Studies Tumor marker
Patient
Age (years)
Tumor site
1
69
Uterus
CS
IIIc
2
69
Uterus
LMS
3
52
Uterus
4
50
5
50
a
Clinical stage
Histology a
Cytology (EM) b
Histology
CA-125 (U/ml) c
LDH (IU/L) d
Carcinosarcoma
16
454
IV
Positive, adeno Negative
Benign
58
788
ESS
I
Negative
Benign
25
312
Back
LMS
Recurrent
h
h
10
340
Uterus
LMS
I
Negative
Benign
15
366
CS, carcinosarcoma; LMS, leiomyosarcoma; ESS, endometrial stromal sarcoma. EM, endometrium; adeno, adenocarcinoma. c Normal range ⬍35 U/ml. d Normal range ⬍500 IU/L. e Low: low signal intensity; High: high signal intensity. f (⫺), no vasculature; (⫹), rich vasculature. g Normal range ⬍2.5. h Not done. b
MRI findings e T1: T2: T1: T2: T1: T2: T1: T2: T1: T2:
low high high high high high high high high high
Intratumoral blood vessel by US power Doppler f
PET (SUV) g
(⫺)
6.3
(⫹)
3.9
(⫹)
5.5
(⫺)
4.0
(⫺)
3.0
377
FDG-PET FOR UTERINE SARCOMA
et al. reported that a positive CA-125 pattern was detected in five of six patients with carcinosarcoma [7], we should reevaluate the significance of CA-125 for diagnosing carcinosarcoma. These laboratory examinations were not useful for the diagnosis of sarcoma in our patients. In cases of suspected sarcoma, MR imaging is usually used. The typical MR features of sarcoma are high signal intensity on T1-weighted images and a heterogeneous appearance of medium or high signal intensity on T2-weighted images [1]. High signal intensity on T1-weighted images has been reported to be particularly important for diagnosis of sarcoma. However, these findings are not specific to the sarcoma lesion, as highintensity lesions on T1-weighted images are often found in relation to bleeding of the muscle or mesenchymal tissue. In our series, four patients (80%) showed high signal intensity on T1-weighted images. Because vasculature-rich lesions are often found in malignancies, intensive examinations by power Doppler or pulse Doppler ultrasonography were also performed. Kurjak et al. [8] reported the resistance index (RI) of the intratumoral artery in sarcoma to be significantly lower than that of the LM. Hata [9] reported no difference in RI and pulsative index (PI) between sarcoma and LM but found that peak systolic velocity (PSV) was significantly higher in sarcoma lesions than in LM. Sase et al. [10] reported no difference in RI, PI, or PSV between the two lesions but showed that rich vascular flow was found in sarcoma lesions. We found rich vascular mosaic patterns in only two of our five cases. Ultrasound was less than satisfactory in our series. FDG-PET, which is based on metabolic activity in tissue, can be a useful diagnostic tool, and its diagnostic usefulness for various malignant tissues has been reported [3, 4]. A few diagnoses of ovarian cancer [11, 12] and cervical cancer [13] by PET have been reported. However, diagnostic PET findings for uterine sarcoma lesions have not been reported. As the cutoff SUV value for pelvic malignancy is set at 2.5 in our institution based on preclinical study, all positive PET studies indicate lesions with increased metabolic activity that are strongly suspect for malignancy. FDG-PET was very useful for the diagnosis of sarcoma, even though the SUV values were a little low. Patient 2 had diabetes, which was controlled by oral medicine. However, this patient’s blood sugar value was 170 mg/dl at PET study. The high glucose level in this patient could have been the cause of the low SUV [13]. In patients 3 and 4, large degenerative lesions were found, degenerative changes that might have influenced the SUV values. Even if the SUV value is a little low, sarcoma should be suspected when the value is positive. In our institution, FEG-PET, MR, and power Doppler imaging studies had been performed for five cases of LM, preoperatively. The SUV values of these cases were 2.0, 1.8. 1.8, 1.6, and 1.0, respectively, and the mean ⫾SD SUV was 1.6 ⫾
0.4. This result revealed that SUV values of LM cases were all negative. MR imaging showed a high-intensity area on T1weighted image in one case of LM. Mosaic patterns of intratumoral blood vessel on ultrasonic power Doppler images were not seen in all cases of LM. PET study was positive in 100% of our cases of uterine sarcoma. PET study can be a most useful diagnostic method for uterine sarcomas in comparison to other diagnostic methods. Based on the positive imaging of results with PET, a prospective evaluation of this modality in patients with uterine tumors is warranted. REFERENCES 1. Fujii S. In: Fujii S, editor. An atlas of MRI with histopathology on smooth muscle tumors of the uterus. Tokyo: Medical View, 1999. 2. Lapela M, Leskinen-Kallio S, Varpula M, Grenman S, Alanen K, Nagren K, Lehikoinen P, Ruotsalainen U, Teras M, Joensuu H. Imaging of uterine carcinoma by carbon-11-methionine aand PET. J Nucl Med 1994;35: 1618 –23. 3. Lamki L. Positron emission tomography in oncology. General acceptance of its roles is overdue. Cancer 1996;78:2039 – 42. 4. Rigo P, Paulus P, Kaschten B, Hustinx R, Bury T, Jerusalem G, Benoit T, J F-W. Oncological applications positron emission tomography with fluorine-18 fluorodeoxyglucoase. Eur J Nucl Med 1996;23:1641–74. 5. Schwartz L, Diamond M, Schwarz P. Leiomyosarcomas: clinical presentation. Am J Obstet Gynecol 1993;168:180. 6. Seki K, Hoshihara T, Nagata I. Leiomyosarcoma of the uterus: ultrasonography and serum lactate dehydrogenase level. Gynecol Obstet Invest 1992;33:114. 7. Peters WA III, Bagley CM, Smith R. CA-125. Use as tumor marker with mixed medodermal tumors of the female genital tract. Cancer 1986;58: 2625–7. 8. Kurjak A, Kupesic S, Shalan H, Jukic S, Kosuta D, Ilijas M. Uterine sarcoma: a report of 10 cases studied by transvaginal color and pulsed Doppler sonography. Gynecol Oncol 1995;59:342– 6. 9. Hata K. Uterine sarcoma: can it be differentiated from uterine leiomyoma with Doppler ultrassonography? A preliminary report. Ultrasound Obstet Gynecol 1997;9:101. 10. Sase M, Ogata T, Numa F, Ogata H, Nakamura Y, Suminami Y, Kato H. Findings of uterine sarcoma and uterine myoma by color and power Doppler. Acta Obst Gynaecol Jpn 1996;48:1159 – 60. 11. Hoh CK, Hawkins RA, Glaspy JA, Dahlbom M, Tse NY, Hoffman EJ, Schiepers C, Choi Y, Rege S, Nitzsche E, Maddashi J, Phelps ME: Cancer detection with whole-body PET using 2-[ 18F]fluoro-2-deoxy-D-glucose. J Comput Assist Tomogr 1993;17:582–9. 12. Hubner K, Mcdonald T, Niethammer J, Smith G, Gould H, Buonocore E. Assessment of primary and metastatic ovarian cancer by positron emission tomography (PET) using 2-[ 18F]deoxyglucose (2-[ 18F]FDG). Gynecol Oncol 1993;51:192– 4. 13. Sugawara Y, Eisbruch A, Kosuda S, Recker B, Kison P, Wahl R. Evaluation of FDG-PET in patients with cervical cancer. J Nucl Med 1999; 40:1125–31. 14. Langen KH, Braun U, Kops ER. The influence of plasma glucose levels on fluorine-18-fluorodeoxyglucose uptake in bronchial carcinoma. J Nucl Med 1993;34:355.
Positron Emission Tomography with 18F-Fluorodeoxyglucose of Uterine Sarcoma: A Comparison with Magnetic Resonance Imaging and Power Doppler Imaging Naohiko Umesaki, M.D., Tetsuji Tanaka, M.D.,* Masato Miyama, M.D.,* Naoki Kawamura, M.D.,* Sachio Ogita, M.D.,* Jyoji Kawabe, M.D.,† Terue Okamura, M.D.,† Koichi Koyama, M.D.,† and Hironobu Ochi, M.D.† Department of Obstetrics and Gynecology, Wakayama Medical College, 811-1 Kimiidera, Wakayama 640-0012, Japan; and *Department of Obstetrics and Gynecology, and †Division of Nuclear Medicine, Osaka City University Medical School, 1-4-3 Asahi-machi, Abeno-ku, Osaka 545-8585, Japan Received August 11, 2000; published online January 31, 2001
Objective. The effectiveness of positron emission tomography with 18F-fluorodeoxyglucose (FDG-PET) for diagnosis of uterine sarcoma was evaluated in comparison to the effectiveness of magnetic resonance (MR) imaging and power Doppler imaging. Method. The cases of five Osaka City University Hospital patients diagnosed with uterine sarcoma based on histopathological examination, in whom FDG-PET, MR imaging, and power Doppler imaging studies had been performed preoperatively, were reviewed. A comparative study of the usefulness of these three imaging modalities for diagnosis of sarcoma was conducted. Tumors comprised three leiomyosarcomas, one endometrial stromal sarcoma, and one carcinosarcoma. Results. FDG-PET examinations were 100% positive for the five sarcomas; MR imagings were 80% positive (four of five cases), and US was 40% positive (two of five cases). The mean strandardized uptake value of the sarcomas was 4.5 ⴞ 1.3. Conclusion. The sarcoma lesions were clearly imaged by FDGPET. FDG-PET may be a most useful diagnostic method for uterine sarcoma. © 2001 Academic Press
INTRODUCTION Uterine sarcomas are generally classified as leiomyosarcoma (LMS), endometrial stromal sarcoma (ESS), or carcinosarcoma. Although uterine sarcoma occurs in less than 5% of all uterine corpus malignancy cases, these tumors are the most lethal. The clinical features of LMS resemble those of leiomyoma (LM), the most common benign neoplasm of the female genital tract. Thus, differentiation between the two is very important. Magnetic resonance (MR) imaging is at present considered the most useful method for diagnosing uterine sarcoma, but it is not applicable to atypical sarcoma cases [1]. Some patients require unnecessary surgery for a differential diagnosis. Moreover, the recent trend toward conservative treatment of LM with gonadotropin-releasing hormone agonist raises the possibilities for mis- or delayed diagnosis. Thus, a new diagnostic approach to LMS has been anticipated. 0090-8258/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.
Metabolic imaging by positron emission tomography (PET) with positron tracers such as L- [methyl- 11C] methionine (carbon-11-methionine) or 2-[ 18F] fluoro-2-deoxy-D-glucose (FDG) is now being used to evaluate suspected malignant tumors [2– 4]. PET imaging differs considerably from conventional imaging modalities such as MR, computerized tomography, and ultrasonography because it quantifies the functional activity of tissue but not its morphological abnormality. We reviewed FDG-PET images of five uterine sarcomas. Histopathological examinations, blood tests, and MR and power Doppler ultrasonography images were also reviewed. The effectiveness of FDG-PET for diagnosis of sarcoma was evaluated in comparison to that of these other examination methods. MATERIALS AND METHODS Patients The cases of five patients (age, 50 – 69 years; mean ⫾SD, 58.0 ⫾ 10.1) who were admitted to Osaka City University Hospital between April 1998 and March 2000 and diagnosed with uterine sarcoma by histopathological examination and in whom preoperative FDG-PET, MR, and power Doppler imaging studies had been performed were reviewed. The usefulness of these imaging modalities for diagnosis of uterine sarcoma was evaluated retrospectively. As points of reference, blood tests and endometrial cytology or histology findings were also evaluated. The sarcomas included three LMSs, one ESS, and one carcinosarcoma. Four patients had primary tumors and one patient a recurrent tumor. The histopathological diagnosis of sarcoma was made based on surgical specimens in four cases and needle biopsy specimen in the one recurrent case. PET Study Radiolabeled FDG was produced with an NKK-Oxford superconducting cyclotron and an NKK synthesis system (NKK Corp., Kawasaki, Japan). A Headtome IV SET-1400W-10
372
FDG-PET FOR UTERINE SARCOMA
373
FIG. 1. A 52-year-old woman with uterine endometrial stromal sarcoma. (A) MR T1-weighted image (TSE; TR/TE⫽631/15) of the pelvis (transaxial view). The ESS lesion is observed as a high signal intensity area (T) in the uterus (U). Leiomyomas (L) are located posteriorly and to the right of the ESS lesion. (B) MR T2-weighted image (TSE; TR/TE⫽4000/102) of the pelvis (transaxial view). The ESS lesion is observed as a heterogeneous area of medium or high intensity (T) in the uterus (U). (C) FDG-PET image (transaxial view). Accumulated FDG (SUV⫽5.5) is seen in the tumor (T). (D) Transvaginal power Doppler image (sagital section) of the uterine tumor. A rich mosaic-like vasculature (arrow) is observed in the tumor (T).
(Shimadzu Corp., Kyoto, Japan) was used for the PET study. All patients fasted for at least 4 h before PET scanning. Transmission scans were obtained using a 68Ge rod source for attenuation correction, followed by intravenous injection of 185 MBq FDG. Emission scanning was performed from 45 to 55 min after the injection of FDG. The urinary bladder was
irrigated just before and during imaging in all patients with 2 L of saline solution to reduce the accumulation of FDG in the bladder, since bladder FDG can interfere with proper evaluation of PET images of the pelvic cavity. Tracer accumulation in elliptical regions of interest (ROI, 5 mm in diameter) on PET images was measured based on the standardized uptake value
374
UMESAKI ET AL.
FIG. 1—Continued
(SUV), i.e., the radioactive concentration (Bq/ml) in a hot spot divided by the injected FDG dose (Bq) and the patient’s body weight (g).
Power Doppler Ultrasonographic Examination Ultrasonographic examination was performed with a LOCIQ 500 ultrasonic scanner (GE Yokogawa Medical Systems, Tokyo, Japan) in power Doppler modes.
MR Imaging RESULTS MR imaging was performed with a 1.5-T imaging system (Signa, General Electric Medical System, Milwaukee, WI). Sagittal and transverse T1-weighted images and T2-weighted images were obtained.
The results of histopathological examinations, blood tests, and imaging diagnoses of the five patients are summarized in Table 1. Cytological and histopathological examinations re-
375
FDG-PET FOR UTERINE SARCOMA
FIG. 2. A 50-year-old woman with uterine leiomyosarcoma. (A) MR T1-weighted image (TSE; TR/TE⫽500/15) of the pelvis (transaxial view). A high signal intensity lesion (arrow) in the uterus (U) is observed. (B) FDG-PET image (transaxial view). Accumulated FDG (arrow) (SUV ⫽ 3.0) is seen in the uterus (U). A high signal intensity lesion in (2A) is necrosis. (C) Transvaginal power Doppler image (sagital section) of the uterine tumor. No vasculature is seen in the tumor (T).
vealed malignancy in only one case; positive LDH level (ⱖ500 IU/L) was demonstrated in one case, and a positive CA-125 level (ⱖ35 U/ml) was demonstrated in one case. MR imaging showed a high intensity area on T1-weighted images in four cases, and moderate- to high-intensity areas were demonstrated on T2-weighted images in all cases (Figs. 1 and 2). Intra-tumoral blood vessel patterns on ultrasonic power Doppler images were mosaic in two cases.
The SUVs of the sarcomas varied from 3.0 to 6.3; the mean ⫾ SD SUV was 4.5 ⫾ 1.3. PET studies were positive in all five cases, lending a strong suspicion for malignancy. DISCUSSION For diagnosis of uterine malignancy, endometrial cytology examinations or histopathological examination of endometrial
376
UMESAKI ET AL.
FIG. 2—Continued
curettage specimens are usually performed. Sarcomas derived from uterine muscle or mesenchymal tissue, however, cannot be found easily until malignant tissue invades the endometrium. Endometrial curettage is useful for diagnosis in only 30% of patients with LMS [5]. This is also true in patients with ESS. In the present five cases, three of four primary uterine
sarcomas were negative based on endometrial cytology and endometrial curettage specimens. LDH can be a useful marker of sarcoma in some cases [6]. Only one of our patients showed a positive LDH value. A positive CA-125 pattern was found in one case, the value of which in the diagnosis of sarcoma is considered low. As Peter
TABLE 1 Patient Characteristics and Results of Laboratory and Imaging Studies Tumor marker
Patient
Age (years)
Tumor site
1
69
Uterus
CS
IIIc
2
69
Uterus
LMS
3
52
Uterus
4
50
5
50
a
Clinical stage
Histology a
Cytology (EM) b
Histology
CA-125 (U/ml) c
LDH (IU/L) d
Carcinosarcoma
16
454
IV
Positive, adeno Negative
Benign
58
788
ESS
I
Negative
Benign
25
312
Back
LMS
Recurrent
h
h
10
340
Uterus
LMS
I
Negative
Benign
15
366
CS, carcinosarcoma; LMS, leiomyosarcoma; ESS, endometrial stromal sarcoma. EM, endometrium; adeno, adenocarcinoma. c Normal range ⬍35 U/ml. d Normal range ⬍500 IU/L. e Low: low signal intensity; High: high signal intensity. f (⫺), no vasculature; (⫹), rich vasculature. g Normal range ⬍2.5. h Not done. b
MRI findings e T1: T2: T1: T2: T1: T2: T1: T2: T1: T2:
low high high high high high high high high high
Intratumoral blood vessel by US power Doppler f
PET (SUV) g
(⫺)
6.3
(⫹)
3.9
(⫹)
5.5
(⫺)
4.0
(⫺)
3.0
377
FDG-PET FOR UTERINE SARCOMA
et al. reported that a positive CA-125 pattern was detected in five of six patients with carcinosarcoma [7], we should reevaluate the significance of CA-125 for diagnosing carcinosarcoma. These laboratory examinations were not useful for the diagnosis of sarcoma in our patients. In cases of suspected sarcoma, MR imaging is usually used. The typical MR features of sarcoma are high signal intensity on T1-weighted images and a heterogeneous appearance of medium or high signal intensity on T2-weighted images [1]. High signal intensity on T1-weighted images has been reported to be particularly important for diagnosis of sarcoma. However, these findings are not specific to the sarcoma lesion, as highintensity lesions on T1-weighted images are often found in relation to bleeding of the muscle or mesenchymal tissue. In our series, four patients (80%) showed high signal intensity on T1-weighted images. Because vasculature-rich lesions are often found in malignancies, intensive examinations by power Doppler or pulse Doppler ultrasonography were also performed. Kurjak et al. [8] reported the resistance index (RI) of the intratumoral artery in sarcoma to be significantly lower than that of the LM. Hata [9] reported no difference in RI and pulsative index (PI) between sarcoma and LM but found that peak systolic velocity (PSV) was significantly higher in sarcoma lesions than in LM. Sase et al. [10] reported no difference in RI, PI, or PSV between the two lesions but showed that rich vascular flow was found in sarcoma lesions. We found rich vascular mosaic patterns in only two of our five cases. Ultrasound was less than satisfactory in our series. FDG-PET, which is based on metabolic activity in tissue, can be a useful diagnostic tool, and its diagnostic usefulness for various malignant tissues has been reported [3, 4]. A few diagnoses of ovarian cancer [11, 12] and cervical cancer [13] by PET have been reported. However, diagnostic PET findings for uterine sarcoma lesions have not been reported. As the cutoff SUV value for pelvic malignancy is set at 2.5 in our institution based on preclinical study, all positive PET studies indicate lesions with increased metabolic activity that are strongly suspect for malignancy. FDG-PET was very useful for the diagnosis of sarcoma, even though the SUV values were a little low. Patient 2 had diabetes, which was controlled by oral medicine. However, this patient’s blood sugar value was 170 mg/dl at PET study. The high glucose level in this patient could have been the cause of the low SUV [13]. In patients 3 and 4, large degenerative lesions were found, degenerative changes that might have influenced the SUV values. Even if the SUV value is a little low, sarcoma should be suspected when the value is positive. In our institution, FEG-PET, MR, and power Doppler imaging studies had been performed for five cases of LM, preoperatively. The SUV values of these cases were 2.0, 1.8. 1.8, 1.6, and 1.0, respectively, and the mean ⫾SD SUV was 1.6 ⫾
0.4. This result revealed that SUV values of LM cases were all negative. MR imaging showed a high-intensity area on T1weighted image in one case of LM. Mosaic patterns of intratumoral blood vessel on ultrasonic power Doppler images were not seen in all cases of LM. PET study was positive in 100% of our cases of uterine sarcoma. PET study can be a most useful diagnostic method for uterine sarcomas in comparison to other diagnostic methods. Based on the positive imaging of results with PET, a prospective evaluation of this modality in patients with uterine tumors is warranted. REFERENCES 1. Fujii S. In: Fujii S, editor. An atlas of MRI with histopathology on smooth muscle tumors of the uterus. Tokyo: Medical View, 1999. 2. Lapela M, Leskinen-Kallio S, Varpula M, Grenman S, Alanen K, Nagren K, Lehikoinen P, Ruotsalainen U, Teras M, Joensuu H. Imaging of uterine carcinoma by carbon-11-methionine aand PET. J Nucl Med 1994;35: 1618 –23. 3. Lamki L. Positron emission tomography in oncology. General acceptance of its roles is overdue. Cancer 1996;78:2039 – 42. 4. Rigo P, Paulus P, Kaschten B, Hustinx R, Bury T, Jerusalem G, Benoit T, J F-W. Oncological applications positron emission tomography with fluorine-18 fluorodeoxyglucoase. Eur J Nucl Med 1996;23:1641–74. 5. Schwartz L, Diamond M, Schwarz P. Leiomyosarcomas: clinical presentation. Am J Obstet Gynecol 1993;168:180. 6. Seki K, Hoshihara T, Nagata I. Leiomyosarcoma of the uterus: ultrasonography and serum lactate dehydrogenase level. Gynecol Obstet Invest 1992;33:114. 7. Peters WA III, Bagley CM, Smith R. CA-125. Use as tumor marker with mixed medodermal tumors of the female genital tract. Cancer 1986;58: 2625–7. 8. Kurjak A, Kupesic S, Shalan H, Jukic S, Kosuta D, Ilijas M. Uterine sarcoma: a report of 10 cases studied by transvaginal color and pulsed Doppler sonography. Gynecol Oncol 1995;59:342– 6. 9. Hata K. Uterine sarcoma: can it be differentiated from uterine leiomyoma with Doppler ultrassonography? A preliminary report. Ultrasound Obstet Gynecol 1997;9:101. 10. Sase M, Ogata T, Numa F, Ogata H, Nakamura Y, Suminami Y, Kato H. Findings of uterine sarcoma and uterine myoma by color and power Doppler. Acta Obst Gynaecol Jpn 1996;48:1159 – 60. 11. Hoh CK, Hawkins RA, Glaspy JA, Dahlbom M, Tse NY, Hoffman EJ, Schiepers C, Choi Y, Rege S, Nitzsche E, Maddashi J, Phelps ME: Cancer detection with whole-body PET using 2-[ 18F]fluoro-2-deoxy-D-glucose. J Comput Assist Tomogr 1993;17:582–9. 12. Hubner K, Mcdonald T, Niethammer J, Smith G, Gould H, Buonocore E. Assessment of primary and metastatic ovarian cancer by positron emission tomography (PET) using 2-[ 18F]deoxyglucose (2-[ 18F]FDG). Gynecol Oncol 1993;51:192– 4. 13. Sugawara Y, Eisbruch A, Kosuda S, Recker B, Kison P, Wahl R. Evaluation of FDG-PET in patients with cervical cancer. J Nucl Med 1999; 40:1125–31. 14. Langen KH, Braun U, Kops ER. The influence of plasma glucose levels on fluorine-18-fluorodeoxyglucose uptake in bronchial carcinoma. J Nucl Med 1993;34:355.