- Email: [email protected]
Prognostic Value of 2-Deoxy-2-[F-18]Fluoro-D-Glucose Positron Emission Tomography Imaging for Patients with Prostate Cancer
PII S1095-0397(01)00065-6
Molecular Imaging and Biology Vol. 4, No. 1, 99–104. 2002 Copyright © 2001 Elsevier Science Inc. Printed in the USA. All rights reserved. 1536-1632/02 $–see front matter
ORIGINAL ARTICLE
Prognostic Value of 2-Deoxy-2-[F-18]Fluoro-D-Glucose Positron Emission Tomography Imaging for Patients with Prostate Cancer Nobuyuki Oyama, MD1, Hironobu Akino, MD1, Yuji Suzuki, MD1, Hiroshi Kanamaru, MD1, Yoshiji Miwa, MD1, Harutoshi Tsuka, MD1, Norihiro Sadato, MD2, Yoshiharu Yonekura, MD2, Kenichiro Okada, MD1 1
Department of Urology and 2Biomedical Imaging Research Center, Fukui Medical University, Fukui, Japan Purpose: The purpose of this study was to investigate the prognostic value of measuring glucose metabolism of primary prostate cancer lesions, using 2-Deoxy-2-[F-18]Fluoro-D-Glucose positron emission tomography (FDG-PET). Procedures: Forty-two patients with prostate cancer were investigated with FDG-PET, and standardized uptake value (SUV) of the prostate was calculated. After PET study, radical prostatectomy was performed in 17 patients (RPT group), and endocrine therapy in 25 patients (ET group). Relapse-free survival curves were created by the Kaplan-Meier method. Results: In the RPT group, the patients with high SUV had a poorer prognosis compared to those with low SUV (P 0.033). In the ET group, the patients with high SUV were likely to have a poorer prognosis with low significance at a level of P 0.087. Conclusions: FDG-PET appeared to have a defined prognostic value for patients with prostate cancer undergoing radical prostatectomy, and more patients need to be studied for patients undergoing endocrine therapy. (Mol Imag Biol 2002;4:99–104) © 2001 Elsevier Science, Inc. All rights reserved. Key Words: Prostate Cancer; Glucose Metabolism; FDG-PET; Prognosis.
Introduction
S
ince serum prostate specific antigen (PSA) has been introduced to clinical use, prostate cancer has been the most common neoplasm in men in the United States, and its incidence has also been increasing in Japan. Today, its precise diagnosis and treatment are major medical problems remaining to be resolved in both the United States and Japan. It is important to predict whether therapy will be a benefit to patients with prostate cancer before therapy. In the current assessment, clinical or pathological stage, the Gleason sum for histological grading, and serum PSA levels have been used as parameters to predict clinical behavior of prostate cancer.1–3 However, these parameters do not always correlate well with clinical outcomes.4–6 Address correspondence to: Nobuyuki Oyama, MD, Mallinckrodt Institute of Radiology, Washington University School of Medicine, Campus Box 8225, 510 South Kingshighway Boulevard, St. Louis, MO 63110. E-mail: [email protected]
Therefore, a new predictor of clinical outcome for prostate cancer before therapy is needed. In the last decade, 2-Deoxy-2-[F-18]Fluoro-D-Glucose positron emission tomography (FDG-PET) has significantly changed the diagnosis of malignant tumors. FDGPET is useful for detection of many kinds of neoplasms such as brain tumor, head and neck tumors, lung cancer, and pancreatic cancer, because of their high glucose utilization.7–10 With regard to the prognostic value of FDGPET, some reports have been published for glioma, lung cancer, breast cancer, and pleural mesothelioma,11–14 demonstrating the usefulness of FDG-PET as a predictor of prognosis of the patients with these malignant tumors. On the other hand, the prognostic value of FDG-PET for prostate cancer has not been investigated because of PET’s relatively low sensitivity for prostate cancer. Our previous research indicated the future possibility of using FDG-PET in the qualitative diagnosis of prostate cancer, as the FDG uptake for prostate cancer correlated well with histological grade and clinical stage.15 Thus, in the present study, we present a more detailed 99
100
Molecular Imaging and Biology, Volume 4, Number 1
evaluation of the prognostic value of FDG-PET imaging for patients with prostate cancer.
Materials and Methods Forty-two patients with mean age of 68.9 years (ranging 52–85) were enrolled in this study. They were histologically diagnosed with adenocarcinoma of the prostate at Fukui Medical University between January 1996 and April 1999. Histological diagnosis was accomplished with specimens obtained by transrectal systematic sextant prostate biopsy. Clinical staging was done according to the fifth edition of TNM classification,16 and histological grading was evaluated with the Gleason grading system.17 PET studies were performed before starting any type of treatment. Clinical stages were T1–T2N0M0 in 15, T3–T4N0M0 in 12 and N()/M() in 15 patients. Seventeen of 42 patients underwent radical prostatectomy, including four patients who received endocrine therapy as neoadjuvant therapy. Among these patients, bladder invasion or lymph node metastases were found, after surgery, in five patients who had received post-surgical endocrine therapy with LH-RH agonist (3.6mg goserelin over a 28-day period). The other 25 received only endocrine therapy with LHRH agonist (3.6mg goserelin over a 28-day period) without surgery. Patients who received radiation therapy or were followed up without treatment were excluded from this study. The Ethics Committee of Fukui Medical University approved the protocol. All patients were informed of the purpose of this study, the method of scanning, the time required, and the necessary pre-treatment, and gave their informed consent.
Patient Preparation Each patient underwent FDG-PET after fasting for at least four hours. During scanning, the bladder was irrigated continuously with 10 L of physiological saline through a 20 Fr. 3-way balloon catheter, indwelt to prevent retention of FDG in the bladder, enabling accurate evaluation of FDG accumulation in the prostate.
prostate gland were obtained for up to 60 minutes after the injection. The mode of dynamic data acquisition consisted of four 30-second frames, eight 60-second frames, and five 600-second frames. Subsequently, the lower abdominal region was scanned for 20 minutes. Plasma glucose concentrations were measured in all patients.
Data Analysis A circular region of interest (ROI) was placed on the prostate of transaxial PET images, which were correspondent with the anatomic location as shown on computed tomography (CT) scans. If there was an area with a high accumulation of FDG in the prostate, this area was selected as the ROI, and if the accumulation was low, the ROI was placed with anatomical reference to the magnetic resonance imaging (MRI) or CT. As an index of FDG uptake, the standardized uptake value (SUV) was calculated for each patient according to the following formula: SUV radioactivity in ROI (Bq/cm3)/ injected dose (Bq)/body weight (g) To minimize the effect of ROI size on SUV, we utilized the maximum value of SUV within an ROI to represent FDG uptake in that particular region. Follow-up evaluation of patients consisted of the physical examination, serum PSA level measurement (1/ month), pelvic CT (1/3months), bone x-rays, and bone scintigraphy (1/3months). Patients were deemed to have a clinical relapse when any of the following four clinical features was identified: (1) a continuous increase of the serum PSA level over consecutive twice measurements; (2) 25% increase of prostate size measured by CT; (3) new lesions; or (4) progression at lymph node or distant metastatic sites. PSA value was determined with a double monoclonal antibody radioimmunoassay (Tandem-R; Hybritech, Inc., San Diego, CA). Relapse-free survival time was defined as the time from the date of the PET study to relapse or the last follow-up.
FDG-PET Imaging Procedure FDG was produced with the method of Hamacher et al.,18 using an automated FDG synthesis system (NKK, Tokyo, Japan) with a small cyclotron (OSCAR3, Oxford Instruments, UK). PET scanning was performed with a GE Advance system (GE, Milwaukee, WI). The physical characteristics of this scanner have been described in detail by DeGrado et al.19 Two transmission scans covering the prostate and adjacent lower abdominal regions were obtained for 10 minutes each using a standard pin source of 68Germanium (Ge)/68Gallium (Ga) for attenuation correction of the emission images. A 350 MBq dose of FDG was administered via the cubital vein over 10 seconds. Dynamic scans covering the
Statistical Evaluation The equality of the age, serum PSA value, and Gleason sum of each subgroup were tested with the Mann-Whitney nonparametric 2-sample test. Due to the covariables, the Kaplan-Meier product limit estimator was used to estimate survival probabilities for the patients’ subgroups.
Results Among the 17 patients with radical prostatectomy (RPT group), cancer relapse was identified in seven patients. The median follow-up of the RPT group was 17.6
FDG PET in Prostate Cancer/ Oyama et al.
101
Table 1. Relapse-free rates according to clinical stage and Gleason sum in RPT group
Months after PET 6 12 18 24 30 36 a,b
Relapse-free rate (%) T1/T2N0M0 (n 10)a
T3/T4N0M0 (n 7)b
Gs 2–4 (n 2)
Gs 5–7 (n 1)
Gs 8–10 (n 14)
100 87.5 87.5 87.5 87.5 0
100 71.4 57.1 42.9 42.9 42.9
100 100 100
100 74.1 64.8 55.6 55.6 27.8
100 100 100 100
NS (P 0.068).
months (ranging from 5–41 months). Among the 25 patients with endocrine therapy (ET group), cancer relapse was identified in 11 patients. The median followup of these 25 patients was 8.5 months (ranging from 1–31 months).
showed relapse during the period. However, log rank test could not be calculated because of the small number of patients with well- and poorly differentiated tumor. In the ET group, the log rank test across tumor grade was not significant (P 0.222).
Clinical Stage and Relapse Free Survival
SUV and Relapse-Free Survival
According to clinical stage, patients were divided into three groups; T1–2N0M0, T3–4N0M0, and N/M group. The relapse-free survival rates according to clinical stage are shown in Table 1 (RPT group) and Table 2 (ET group). In our series of the RPT group, no significant difference was demonstrated between the T1–2 and T3–4 group. On the other hand, in the ET group, T3–4 and N/M group showed poorer prognosis (P 0.05) than the T1–2 group.
SUV of the RPT group ranged from 2.279 to 8.060 mg/ ml, and median SUV was 4.067 mg/ml. According to SUV, the RPT group was subdivided into two groups; the high SUV group (including a patient with median SUV) and low SUV group (Table 3). There were no significant differences between preoperative backgrounds including serum PSA value and Gleason score or numbers of patients receiving adjuvant endocrine therapy of each group (data not shown), however, patients’ ages did vary. Lymph node metastases were found in three patients of the high SUV group after surgery. The survival curves for these two groups shown in Figure 1 indicate that patients of the high SUV group had a significantly poorer prognosis (P 0.033). SUV of the ET group ranged from 2.991 to 21.161 mg/ml, and median SUV was 5.586 mg/ml. This group was also subdivided into a high SUV group and low SUV group (Table 4). There were no significant differences between preoperative backgrounds including age, serum PSA value, and Gleason score between two
Gleason Sum and Relapse-Free Survival Utilizing the Gleason sum (Gs), patients were divided into three groups; the well- (Gs2–4), moderately (Gs5– 7), and poorly differentiated group (Gs8–10). The relapse-free survival rates according to tumor grades are shown in Table 1 (RPT group) and Table 2 (ET group). In the RPT group, two of well-differentiated tumor and one of poorly differentiated tumor had no relapse, while seven of moderately differentiated tumor
Table 2. Relapse-free rates according to clinical stage and Gleason sum in ET group Relapse-free rate (%) T1/T2N0M0 T3/T4N0M0 (n 5)a (n 5)b Months after PET 6 12 18 24 30 a,b
P 0.05, a,c P 0.05.
100 100 100 100 100
75.0 37.5 0
N/M (n 15)c
Gs 2–4 (n 4)
Gs 5–7 (n 17)
Gs 8–10 (n 4)
59.4 37.1 37.1 18.6
100 100 100 66.7 66.7
66.0 33.9 17.0
50.0 50.0 50.0
102
Molecular Imaging and Biology, Volume 4, Number 1
Table 4. Patients for endocrine therapy
Table 3. Patients for radical prostatectomy
Age (mean SD)a Serum PSA valueb Gleason sum (mean SD)c Clinical stage T1/2 T3/4N0M0 N and/or M Pathological stage T1/2 T3/4 N Adjuvant therapy
Low SUV group (n 8)
High SUV group (n 9)
70.3 4.71 26.3 21.2 4.9 1.8
63.4 6.41 42.9 42.4 6.2 0.9
4 4 0
6 3 0
4 4 0 2
5 4 3 3
Age (mean SD)a Serum PSA valueb Gleason sum (mean SD)c Clinical stage T1/2 T3/4N0M0 N and/or M a
Low SUV group (n 12)
High SUV group (n 13)
70.7 9.89 339.7 701.3 5.6 1.7
70.1 6.22 472.1 898.9 6.4 1.8
5 0 7
0 3 10
NS (P 0.935), b NS (P 0.142), c NS (P 0.141).
In general, cancer tissue consumes a large amount of glucose as an energy source, and has a high rate of glycolysis. Therefore, evaluation of glucose metabolism can be informative in detection of malignant lesions. With respect to prostate cancer, there are some published reports of clinical studies on FDG-PET. In the study by Effert et al.,20 FDG-PET could not differentiate prostate cancer from benign prostatic hyperplasia (BPH). Also, Laubenbacher and colleagues,21 using FDG-PET, could not distinguish local recurrence from scar tissue after rad-
ical prostatectomy. Concerning the detection of metastatic sites with FDG-PET, Shreve et al.22 reported a sensitivity of 65% for bone metastases. In the study by Yeh et al.,23 FDG-PET could detect bone metastasis in only 20% of the patients. From these clinical reports, FDG-PET has been regarded as not being sensitive enough to detect prostate cancer for wide clinical use. Our previous report reconfirmed these results, however, it was demonstrated that FDG uptake of prostate cancer was significantly increased in tumors of high histological grade or advanced clinical stage with metastases.15 This suggested a possibility that the malignant behavior of prostate cancer could be metabolically evaluated with FDG-PET. In current studies of FDG-PET for other malignant tumors, the prognostic value of FDG-PET has clearly been shown.12–14 These reports addressed the role of FDG-PET in providing prognostic information and its potential to serve as a guide for therapeutic options. This was one of the most expected roles for FDG-PET in cancer diagnosis; hence, we investigated the prognostic value of FDG-PET for prostate cancer. The present study showed that in our series of the
Figure 1. Kaplan-Meier curves show patients with high SUV have a significantly poorer prognosis in the RPT group (P 0.033).
Figure 2. Kaplan-Meier curves show patients with high SUV are likely to have a poorer prognosis, however, it is only significant at a level P 0.087.
a
Significant (P 0.029), b NS (P 0.149), c NS (P 0.077).
subgroups. The survival curves for these groups are shown in Figure 2, indicating a trend that patients with high SUV were likely to have a poorer prognosis, however, it was only significant at a level P 0.087.
Discussion
FDG PET in Prostate Cancer/ Oyama et al. 103
RPT group, clinical stage and tumor grade were not significant in predicting the tumor prognosis. On the other hand, the high SUV group according to FDG-PET studies had a significantly worse prognosis than the low SUV group. Although multivariate analysis could not be performed because of the small number of patients, our results suggested that SUV would provide prognostic information in patients with prostate cancer before surgery. Thus, patients with tumors that are more active metabolically, as demonstrated by FDG-PET, should be considered to be at high risk for cancer relapse after prostatectomy regardless of clinical stage or tumor grade. Moreover, this information could also be useful for patients in decision making about the acceptance or rejection of the invasive medical intervention. In contrast, there was only significance at a level of P 0.087 in relapse-free survival between the high and low SUV group among the ET group. The ET group (n 25) included 15 patients with lymph node or bone metastases. Regarding those patients with metastases, their prognoses were considered dependent upon not only metabolic activity of the primary tumors but also that of the tumors at distant metastatic sites. Therefore, in ET group significance in relapse-free survival might be weaker than that in RPT group in our study. More patients need to be studied to make it clear. The reason for correlation between tumor glycolysis and prognosis has been revealed by degrees recently. In the study using prostate cancer xenograft, Agus et al.24 demonstrated that glucose metabolism of prostate tumors correlated with cell proliferative activity expressed by Ki-67 index. This result suggested that the prostate cancer with higher glucose metabolism might show rapid progress with poorer prognosis. Until now, there has been no published report investigating the correlation between glucose metabolism and tumor growth in clinical series. There is a need to verify this correlation in clinical prostate cancer patients. In this study, we do not conclusively claim that SUV is the most important prognostic factor. The number of patients analyzed in the present study was relatively small, and the observation period was short. So further study on FDG-PET with large number of patients with multivariate analysis will be necessary to compare the clinical utility of glucose metabolism of prostate cancer measured on FDG-PET.
Conclusion This is the first study that demonstrates the prognostic significance of FDG-PET for prostate cancer in patients who underwent radical prostatectomy or endocrine therapy. Further study on FDG-PET with a large number of patients will be necessary to verify the correlation between the FDG-PET results and clinical outcome us-
ing multivariate analysis. Nevertheless, FDG-PET imaging promises to be a useful prognostic indicator, and may provide relevant information when determining treatment options for prostate cancer.
The authors thank K. Sugimoto and other staff members of the Biomedical Imaging Research Center of Fukui Medical University for their expert technical support. We also thank Prof. M.J. Welch, Prof. T.R. Miller, and Prof. B.A. Siegel for critical reading of the manuscript.
References 1. Oefelein, M.G.; Grayhack, J.T.; McVary, K.T. Survival after radical retropubic prostatectomy of men with clinically localized high grade carcinoma of the prostate. Cancer 76:2535–42; 1995. 2. Montie, J.E. Current prognostic factors for prostate carcinoma. Cancer 78:341–4; 1996. 3. Lipponen, P.; Vesalainen, S.; Kasurinen, J.; Ala-opas, M.; Syrjanen, K. A prognostic score for prostatic adenocarcinoma based on clinical, histological, biochemical and cytometric data from the primary tumor. Anticancer Res. 16:2095–100; 1996. 4. Scher, H.I.; Chung, L.W. Bone metastases: improving the therapeutic index. Semin. Oncol. 21:630–56; 1994. 5. Scher, H.I.; Mazumdar, M.; Kelly, W.K. Clinical trials in relapsed prostate cancer: defining the target. J. Natl. Cancer Inst. 88:1623–34; 1996. 6. Eisenberger, M.A.; Nelson, W.G.; How much can we rely on the level of prostate-specific antigen as an end point for evaluation of clinical trials? A word of caution! J. Natl. Cancer Inst. 88:779–81; 1996. 7. Pantronas, N.J.; Chiro, G.D.; Kufta, C.; et al. Prediction of survival in glioma patients by means of positron emission tomography. J. Neurosurg. 62:816–22; 1985. 8. Rege, S.; Maass, A.; Chaiken, L.; et al. Use of positron emission tomography with fluorodeoxyglucose in patient with extracranial head and neck cancers. Cancer 73: 3047–58; 1994. 9. Nolop, K.B.; Rhodes, C.G.; Brudin, L.H.; et al. Glucose utilization in vivo by human pulmonary neoplasms. Cancer 60:2682–9; 1987. 10. Kato, T.; Fukatsu, H.; Ito, K.; et al. F-18-FDG-PET in pancreatic cancer: an unsolved problem. Eur. J. Med. 22:32–9; 1995. 11. Barker, F.G.; Chang, S.M.; Valk, P.E.; Pounds, T.R.; Prados, M.D. 18-fluorodeoxyglucose uptake and survival of patients with suspected recurrent malignant glioma. Cancer 79:115–26; 1997. 12. Ahuja, V.; Coleman, R.E.; Herndon, J.; Patz, EF. The prognostic significance of fluorodeoxyglucose positron emission tomography imaging for patients with nonsmall cell lung carcinoma. Cancer 83:918–24; 1998. 13. Oshida, M.; Uno, K.; Suzuki, M.; et al. Predicting the prognoses of breast carcinoma patients with positron emission tomography using 2-deoxy-2- fluoro[18F]-d-glucose. Cancer 82:2227–34; 1998. 14. Benard, F.; Sterman, D.; Smith, R.J.; et al. Prognostic value of FDG PET imaging in malignant pleural mesothelioma. J. Nucl. Med. 40:1241–5; 1999. 15. Oyama, N.; Akino, H.; Suzuki, Y.; et al. The increase of ac-
104
16. 17.
18.
19.
20.
Molecular Imaging and Biology, Volume 4, Number 1
cumulation of 18-fluorodeoxyglucose in untreated prostate cancer. Jpn. J. Clin. Oncol. 29:623–62; 1999. Sobin, L.H. TNM classification of malignant tumors. 5th ed. New York, NY: Wiley-Liss, 1997. Gleason, D.F.; the Veterans Administration Cooperative Urological Research Group. Histologic grading and clinical staging of carcinoma of the prostate in urologic pathology. In: Tannenbaum, M., ed. The Prostate. Philadelphia, PA: Lea & Febiger; 1997: 171–97. Hamacher, K.; Coenen, H.H.; Stocklin, G. Efficient stereospecific synthesis of no-carried-added 2-[18F]-fluoro2-deoxy-D-glucose using aminopolyether supported nucleophilic substitution. J. Nucl. Med. 27:235–8; 1986. DeGrado, T.R.; Turkington, T.G.; Williams, J.J.; et al. Performance characteristics of whole-body PET scanner. J. Nucl. Med. 35:1398–406; 1994. Effert, P.J.; Bares, R.; Handt, S.; et al. Metabolic imaging of
21.
22.
23.
24.
untreated prostate cancer by positron emission tomography with18fluorine-labeled deoxyglucose. J. Urol. 155: 994–8; 1996. Laubenbacher, C.; Hofer, C.; Avril, N.; et al. F-18 FDG PET for differentiation of local recurrent prostate cancer and scar [abstract]. J. Nucl. Med. 36 (suppl):198P; 1995. Shreve, P.D.; Grossman, H.B.; Gross, M.D.; Wahl, R.L. Metastatic prostate cancer: initial findings of PET with 2- deoxy2-[F-18] fluoro -D- glucose. Radiology 199:751–6; 1996. Yeh, S.D.J.; Imbriaco, M.; Garza, D.; et al. Twenty percent of bony metastases of hormone resistant prostate cancer are detected by FDG-PET whole body scanning [abstract]. J. Nucl. Med. 36(suppl ):198P; 1995. Agus, D.B.; Golde, D.W.; Sgouros, G.; et al. Positron emission tomography of a human prostate xenograft: association of changes in deoxyglucose accumulation with other Measures of outcome following androgen withdrawal. Cancer Res. 58:3009–14; 1998.
Molecular Imaging and Biology Vol. 4, No. 1, 99–104. 2002 Copyright © 2001 Elsevier Science Inc. Printed in the USA. All rights reserved. 1536-1632/02 $–see front matter
ORIGINAL ARTICLE
Prognostic Value of 2-Deoxy-2-[F-18]Fluoro-D-Glucose Positron Emission Tomography Imaging for Patients with Prostate Cancer Nobuyuki Oyama, MD1, Hironobu Akino, MD1, Yuji Suzuki, MD1, Hiroshi Kanamaru, MD1, Yoshiji Miwa, MD1, Harutoshi Tsuka, MD1, Norihiro Sadato, MD2, Yoshiharu Yonekura, MD2, Kenichiro Okada, MD1 1
Department of Urology and 2Biomedical Imaging Research Center, Fukui Medical University, Fukui, Japan Purpose: The purpose of this study was to investigate the prognostic value of measuring glucose metabolism of primary prostate cancer lesions, using 2-Deoxy-2-[F-18]Fluoro-D-Glucose positron emission tomography (FDG-PET). Procedures: Forty-two patients with prostate cancer were investigated with FDG-PET, and standardized uptake value (SUV) of the prostate was calculated. After PET study, radical prostatectomy was performed in 17 patients (RPT group), and endocrine therapy in 25 patients (ET group). Relapse-free survival curves were created by the Kaplan-Meier method. Results: In the RPT group, the patients with high SUV had a poorer prognosis compared to those with low SUV (P 0.033). In the ET group, the patients with high SUV were likely to have a poorer prognosis with low significance at a level of P 0.087. Conclusions: FDG-PET appeared to have a defined prognostic value for patients with prostate cancer undergoing radical prostatectomy, and more patients need to be studied for patients undergoing endocrine therapy. (Mol Imag Biol 2002;4:99–104) © 2001 Elsevier Science, Inc. All rights reserved. Key Words: Prostate Cancer; Glucose Metabolism; FDG-PET; Prognosis.
Introduction
S
ince serum prostate specific antigen (PSA) has been introduced to clinical use, prostate cancer has been the most common neoplasm in men in the United States, and its incidence has also been increasing in Japan. Today, its precise diagnosis and treatment are major medical problems remaining to be resolved in both the United States and Japan. It is important to predict whether therapy will be a benefit to patients with prostate cancer before therapy. In the current assessment, clinical or pathological stage, the Gleason sum for histological grading, and serum PSA levels have been used as parameters to predict clinical behavior of prostate cancer.1–3 However, these parameters do not always correlate well with clinical outcomes.4–6 Address correspondence to: Nobuyuki Oyama, MD, Mallinckrodt Institute of Radiology, Washington University School of Medicine, Campus Box 8225, 510 South Kingshighway Boulevard, St. Louis, MO 63110. E-mail: [email protected]
Therefore, a new predictor of clinical outcome for prostate cancer before therapy is needed. In the last decade, 2-Deoxy-2-[F-18]Fluoro-D-Glucose positron emission tomography (FDG-PET) has significantly changed the diagnosis of malignant tumors. FDGPET is useful for detection of many kinds of neoplasms such as brain tumor, head and neck tumors, lung cancer, and pancreatic cancer, because of their high glucose utilization.7–10 With regard to the prognostic value of FDGPET, some reports have been published for glioma, lung cancer, breast cancer, and pleural mesothelioma,11–14 demonstrating the usefulness of FDG-PET as a predictor of prognosis of the patients with these malignant tumors. On the other hand, the prognostic value of FDG-PET for prostate cancer has not been investigated because of PET’s relatively low sensitivity for prostate cancer. Our previous research indicated the future possibility of using FDG-PET in the qualitative diagnosis of prostate cancer, as the FDG uptake for prostate cancer correlated well with histological grade and clinical stage.15 Thus, in the present study, we present a more detailed 99
100
Molecular Imaging and Biology, Volume 4, Number 1
evaluation of the prognostic value of FDG-PET imaging for patients with prostate cancer.
Materials and Methods Forty-two patients with mean age of 68.9 years (ranging 52–85) were enrolled in this study. They were histologically diagnosed with adenocarcinoma of the prostate at Fukui Medical University between January 1996 and April 1999. Histological diagnosis was accomplished with specimens obtained by transrectal systematic sextant prostate biopsy. Clinical staging was done according to the fifth edition of TNM classification,16 and histological grading was evaluated with the Gleason grading system.17 PET studies were performed before starting any type of treatment. Clinical stages were T1–T2N0M0 in 15, T3–T4N0M0 in 12 and N()/M() in 15 patients. Seventeen of 42 patients underwent radical prostatectomy, including four patients who received endocrine therapy as neoadjuvant therapy. Among these patients, bladder invasion or lymph node metastases were found, after surgery, in five patients who had received post-surgical endocrine therapy with LH-RH agonist (3.6mg goserelin over a 28-day period). The other 25 received only endocrine therapy with LHRH agonist (3.6mg goserelin over a 28-day period) without surgery. Patients who received radiation therapy or were followed up without treatment were excluded from this study. The Ethics Committee of Fukui Medical University approved the protocol. All patients were informed of the purpose of this study, the method of scanning, the time required, and the necessary pre-treatment, and gave their informed consent.
Patient Preparation Each patient underwent FDG-PET after fasting for at least four hours. During scanning, the bladder was irrigated continuously with 10 L of physiological saline through a 20 Fr. 3-way balloon catheter, indwelt to prevent retention of FDG in the bladder, enabling accurate evaluation of FDG accumulation in the prostate.
prostate gland were obtained for up to 60 minutes after the injection. The mode of dynamic data acquisition consisted of four 30-second frames, eight 60-second frames, and five 600-second frames. Subsequently, the lower abdominal region was scanned for 20 minutes. Plasma glucose concentrations were measured in all patients.
Data Analysis A circular region of interest (ROI) was placed on the prostate of transaxial PET images, which were correspondent with the anatomic location as shown on computed tomography (CT) scans. If there was an area with a high accumulation of FDG in the prostate, this area was selected as the ROI, and if the accumulation was low, the ROI was placed with anatomical reference to the magnetic resonance imaging (MRI) or CT. As an index of FDG uptake, the standardized uptake value (SUV) was calculated for each patient according to the following formula: SUV radioactivity in ROI (Bq/cm3)/ injected dose (Bq)/body weight (g) To minimize the effect of ROI size on SUV, we utilized the maximum value of SUV within an ROI to represent FDG uptake in that particular region. Follow-up evaluation of patients consisted of the physical examination, serum PSA level measurement (1/ month), pelvic CT (1/3months), bone x-rays, and bone scintigraphy (1/3months). Patients were deemed to have a clinical relapse when any of the following four clinical features was identified: (1) a continuous increase of the serum PSA level over consecutive twice measurements; (2) 25% increase of prostate size measured by CT; (3) new lesions; or (4) progression at lymph node or distant metastatic sites. PSA value was determined with a double monoclonal antibody radioimmunoassay (Tandem-R; Hybritech, Inc., San Diego, CA). Relapse-free survival time was defined as the time from the date of the PET study to relapse or the last follow-up.
FDG-PET Imaging Procedure FDG was produced with the method of Hamacher et al.,18 using an automated FDG synthesis system (NKK, Tokyo, Japan) with a small cyclotron (OSCAR3, Oxford Instruments, UK). PET scanning was performed with a GE Advance system (GE, Milwaukee, WI). The physical characteristics of this scanner have been described in detail by DeGrado et al.19 Two transmission scans covering the prostate and adjacent lower abdominal regions were obtained for 10 minutes each using a standard pin source of 68Germanium (Ge)/68Gallium (Ga) for attenuation correction of the emission images. A 350 MBq dose of FDG was administered via the cubital vein over 10 seconds. Dynamic scans covering the
Statistical Evaluation The equality of the age, serum PSA value, and Gleason sum of each subgroup were tested with the Mann-Whitney nonparametric 2-sample test. Due to the covariables, the Kaplan-Meier product limit estimator was used to estimate survival probabilities for the patients’ subgroups.
Results Among the 17 patients with radical prostatectomy (RPT group), cancer relapse was identified in seven patients. The median follow-up of the RPT group was 17.6
FDG PET in Prostate Cancer/ Oyama et al.
101
Table 1. Relapse-free rates according to clinical stage and Gleason sum in RPT group
Months after PET 6 12 18 24 30 36 a,b
Relapse-free rate (%) T1/T2N0M0 (n 10)a
T3/T4N0M0 (n 7)b
Gs 2–4 (n 2)
Gs 5–7 (n 1)
Gs 8–10 (n 14)
100 87.5 87.5 87.5 87.5 0
100 71.4 57.1 42.9 42.9 42.9
100 100 100
100 74.1 64.8 55.6 55.6 27.8
100 100 100 100
NS (P 0.068).
months (ranging from 5–41 months). Among the 25 patients with endocrine therapy (ET group), cancer relapse was identified in 11 patients. The median followup of these 25 patients was 8.5 months (ranging from 1–31 months).
showed relapse during the period. However, log rank test could not be calculated because of the small number of patients with well- and poorly differentiated tumor. In the ET group, the log rank test across tumor grade was not significant (P 0.222).
Clinical Stage and Relapse Free Survival
SUV and Relapse-Free Survival
According to clinical stage, patients were divided into three groups; T1–2N0M0, T3–4N0M0, and N/M group. The relapse-free survival rates according to clinical stage are shown in Table 1 (RPT group) and Table 2 (ET group). In our series of the RPT group, no significant difference was demonstrated between the T1–2 and T3–4 group. On the other hand, in the ET group, T3–4 and N/M group showed poorer prognosis (P 0.05) than the T1–2 group.
SUV of the RPT group ranged from 2.279 to 8.060 mg/ ml, and median SUV was 4.067 mg/ml. According to SUV, the RPT group was subdivided into two groups; the high SUV group (including a patient with median SUV) and low SUV group (Table 3). There were no significant differences between preoperative backgrounds including serum PSA value and Gleason score or numbers of patients receiving adjuvant endocrine therapy of each group (data not shown), however, patients’ ages did vary. Lymph node metastases were found in three patients of the high SUV group after surgery. The survival curves for these two groups shown in Figure 1 indicate that patients of the high SUV group had a significantly poorer prognosis (P 0.033). SUV of the ET group ranged from 2.991 to 21.161 mg/ml, and median SUV was 5.586 mg/ml. This group was also subdivided into a high SUV group and low SUV group (Table 4). There were no significant differences between preoperative backgrounds including age, serum PSA value, and Gleason score between two
Gleason Sum and Relapse-Free Survival Utilizing the Gleason sum (Gs), patients were divided into three groups; the well- (Gs2–4), moderately (Gs5– 7), and poorly differentiated group (Gs8–10). The relapse-free survival rates according to tumor grades are shown in Table 1 (RPT group) and Table 2 (ET group). In the RPT group, two of well-differentiated tumor and one of poorly differentiated tumor had no relapse, while seven of moderately differentiated tumor
Table 2. Relapse-free rates according to clinical stage and Gleason sum in ET group Relapse-free rate (%) T1/T2N0M0 T3/T4N0M0 (n 5)a (n 5)b Months after PET 6 12 18 24 30 a,b
P 0.05, a,c P 0.05.
100 100 100 100 100
75.0 37.5 0
N/M (n 15)c
Gs 2–4 (n 4)
Gs 5–7 (n 17)
Gs 8–10 (n 4)
59.4 37.1 37.1 18.6
100 100 100 66.7 66.7
66.0 33.9 17.0
50.0 50.0 50.0
102
Molecular Imaging and Biology, Volume 4, Number 1
Table 4. Patients for endocrine therapy
Table 3. Patients for radical prostatectomy
Age (mean SD)a Serum PSA valueb Gleason sum (mean SD)c Clinical stage T1/2 T3/4N0M0 N and/or M Pathological stage T1/2 T3/4 N Adjuvant therapy
Low SUV group (n 8)
High SUV group (n 9)
70.3 4.71 26.3 21.2 4.9 1.8
63.4 6.41 42.9 42.4 6.2 0.9
4 4 0
6 3 0
4 4 0 2
5 4 3 3
Age (mean SD)a Serum PSA valueb Gleason sum (mean SD)c Clinical stage T1/2 T3/4N0M0 N and/or M a
Low SUV group (n 12)
High SUV group (n 13)
70.7 9.89 339.7 701.3 5.6 1.7
70.1 6.22 472.1 898.9 6.4 1.8
5 0 7
0 3 10
NS (P 0.935), b NS (P 0.142), c NS (P 0.141).
In general, cancer tissue consumes a large amount of glucose as an energy source, and has a high rate of glycolysis. Therefore, evaluation of glucose metabolism can be informative in detection of malignant lesions. With respect to prostate cancer, there are some published reports of clinical studies on FDG-PET. In the study by Effert et al.,20 FDG-PET could not differentiate prostate cancer from benign prostatic hyperplasia (BPH). Also, Laubenbacher and colleagues,21 using FDG-PET, could not distinguish local recurrence from scar tissue after rad-
ical prostatectomy. Concerning the detection of metastatic sites with FDG-PET, Shreve et al.22 reported a sensitivity of 65% for bone metastases. In the study by Yeh et al.,23 FDG-PET could detect bone metastasis in only 20% of the patients. From these clinical reports, FDG-PET has been regarded as not being sensitive enough to detect prostate cancer for wide clinical use. Our previous report reconfirmed these results, however, it was demonstrated that FDG uptake of prostate cancer was significantly increased in tumors of high histological grade or advanced clinical stage with metastases.15 This suggested a possibility that the malignant behavior of prostate cancer could be metabolically evaluated with FDG-PET. In current studies of FDG-PET for other malignant tumors, the prognostic value of FDG-PET has clearly been shown.12–14 These reports addressed the role of FDG-PET in providing prognostic information and its potential to serve as a guide for therapeutic options. This was one of the most expected roles for FDG-PET in cancer diagnosis; hence, we investigated the prognostic value of FDG-PET for prostate cancer. The present study showed that in our series of the
Figure 1. Kaplan-Meier curves show patients with high SUV have a significantly poorer prognosis in the RPT group (P 0.033).
Figure 2. Kaplan-Meier curves show patients with high SUV are likely to have a poorer prognosis, however, it is only significant at a level P 0.087.
a
Significant (P 0.029), b NS (P 0.149), c NS (P 0.077).
subgroups. The survival curves for these groups are shown in Figure 2, indicating a trend that patients with high SUV were likely to have a poorer prognosis, however, it was only significant at a level P 0.087.
Discussion
FDG PET in Prostate Cancer/ Oyama et al. 103
RPT group, clinical stage and tumor grade were not significant in predicting the tumor prognosis. On the other hand, the high SUV group according to FDG-PET studies had a significantly worse prognosis than the low SUV group. Although multivariate analysis could not be performed because of the small number of patients, our results suggested that SUV would provide prognostic information in patients with prostate cancer before surgery. Thus, patients with tumors that are more active metabolically, as demonstrated by FDG-PET, should be considered to be at high risk for cancer relapse after prostatectomy regardless of clinical stage or tumor grade. Moreover, this information could also be useful for patients in decision making about the acceptance or rejection of the invasive medical intervention. In contrast, there was only significance at a level of P 0.087 in relapse-free survival between the high and low SUV group among the ET group. The ET group (n 25) included 15 patients with lymph node or bone metastases. Regarding those patients with metastases, their prognoses were considered dependent upon not only metabolic activity of the primary tumors but also that of the tumors at distant metastatic sites. Therefore, in ET group significance in relapse-free survival might be weaker than that in RPT group in our study. More patients need to be studied to make it clear. The reason for correlation between tumor glycolysis and prognosis has been revealed by degrees recently. In the study using prostate cancer xenograft, Agus et al.24 demonstrated that glucose metabolism of prostate tumors correlated with cell proliferative activity expressed by Ki-67 index. This result suggested that the prostate cancer with higher glucose metabolism might show rapid progress with poorer prognosis. Until now, there has been no published report investigating the correlation between glucose metabolism and tumor growth in clinical series. There is a need to verify this correlation in clinical prostate cancer patients. In this study, we do not conclusively claim that SUV is the most important prognostic factor. The number of patients analyzed in the present study was relatively small, and the observation period was short. So further study on FDG-PET with large number of patients with multivariate analysis will be necessary to compare the clinical utility of glucose metabolism of prostate cancer measured on FDG-PET.
Conclusion This is the first study that demonstrates the prognostic significance of FDG-PET for prostate cancer in patients who underwent radical prostatectomy or endocrine therapy. Further study on FDG-PET with a large number of patients will be necessary to verify the correlation between the FDG-PET results and clinical outcome us-
ing multivariate analysis. Nevertheless, FDG-PET imaging promises to be a useful prognostic indicator, and may provide relevant information when determining treatment options for prostate cancer.
The authors thank K. Sugimoto and other staff members of the Biomedical Imaging Research Center of Fukui Medical University for their expert technical support. We also thank Prof. M.J. Welch, Prof. T.R. Miller, and Prof. B.A. Siegel for critical reading of the manuscript.
References 1. Oefelein, M.G.; Grayhack, J.T.; McVary, K.T. Survival after radical retropubic prostatectomy of men with clinically localized high grade carcinoma of the prostate. Cancer 76:2535–42; 1995. 2. Montie, J.E. Current prognostic factors for prostate carcinoma. Cancer 78:341–4; 1996. 3. Lipponen, P.; Vesalainen, S.; Kasurinen, J.; Ala-opas, M.; Syrjanen, K. A prognostic score for prostatic adenocarcinoma based on clinical, histological, biochemical and cytometric data from the primary tumor. Anticancer Res. 16:2095–100; 1996. 4. Scher, H.I.; Chung, L.W. Bone metastases: improving the therapeutic index. Semin. Oncol. 21:630–56; 1994. 5. Scher, H.I.; Mazumdar, M.; Kelly, W.K. Clinical trials in relapsed prostate cancer: defining the target. J. Natl. Cancer Inst. 88:1623–34; 1996. 6. Eisenberger, M.A.; Nelson, W.G.; How much can we rely on the level of prostate-specific antigen as an end point for evaluation of clinical trials? A word of caution! J. Natl. Cancer Inst. 88:779–81; 1996. 7. Pantronas, N.J.; Chiro, G.D.; Kufta, C.; et al. Prediction of survival in glioma patients by means of positron emission tomography. J. Neurosurg. 62:816–22; 1985. 8. Rege, S.; Maass, A.; Chaiken, L.; et al. Use of positron emission tomography with fluorodeoxyglucose in patient with extracranial head and neck cancers. Cancer 73: 3047–58; 1994. 9. Nolop, K.B.; Rhodes, C.G.; Brudin, L.H.; et al. Glucose utilization in vivo by human pulmonary neoplasms. Cancer 60:2682–9; 1987. 10. Kato, T.; Fukatsu, H.; Ito, K.; et al. F-18-FDG-PET in pancreatic cancer: an unsolved problem. Eur. J. Med. 22:32–9; 1995. 11. Barker, F.G.; Chang, S.M.; Valk, P.E.; Pounds, T.R.; Prados, M.D. 18-fluorodeoxyglucose uptake and survival of patients with suspected recurrent malignant glioma. Cancer 79:115–26; 1997. 12. Ahuja, V.; Coleman, R.E.; Herndon, J.; Patz, EF. The prognostic significance of fluorodeoxyglucose positron emission tomography imaging for patients with nonsmall cell lung carcinoma. Cancer 83:918–24; 1998. 13. Oshida, M.; Uno, K.; Suzuki, M.; et al. Predicting the prognoses of breast carcinoma patients with positron emission tomography using 2-deoxy-2- fluoro[18F]-d-glucose. Cancer 82:2227–34; 1998. 14. Benard, F.; Sterman, D.; Smith, R.J.; et al. Prognostic value of FDG PET imaging in malignant pleural mesothelioma. J. Nucl. Med. 40:1241–5; 1999. 15. Oyama, N.; Akino, H.; Suzuki, Y.; et al. The increase of ac-
104
16. 17.
18.
19.
20.
Molecular Imaging and Biology, Volume 4, Number 1
cumulation of 18-fluorodeoxyglucose in untreated prostate cancer. Jpn. J. Clin. Oncol. 29:623–62; 1999. Sobin, L.H. TNM classification of malignant tumors. 5th ed. New York, NY: Wiley-Liss, 1997. Gleason, D.F.; the Veterans Administration Cooperative Urological Research Group. Histologic grading and clinical staging of carcinoma of the prostate in urologic pathology. In: Tannenbaum, M., ed. The Prostate. Philadelphia, PA: Lea & Febiger; 1997: 171–97. Hamacher, K.; Coenen, H.H.; Stocklin, G. Efficient stereospecific synthesis of no-carried-added 2-[18F]-fluoro2-deoxy-D-glucose using aminopolyether supported nucleophilic substitution. J. Nucl. Med. 27:235–8; 1986. DeGrado, T.R.; Turkington, T.G.; Williams, J.J.; et al. Performance characteristics of whole-body PET scanner. J. Nucl. Med. 35:1398–406; 1994. Effert, P.J.; Bares, R.; Handt, S.; et al. Metabolic imaging of
21.
22.
23.
24.
untreated prostate cancer by positron emission tomography with18fluorine-labeled deoxyglucose. J. Urol. 155: 994–8; 1996. Laubenbacher, C.; Hofer, C.; Avril, N.; et al. F-18 FDG PET for differentiation of local recurrent prostate cancer and scar [abstract]. J. Nucl. Med. 36 (suppl):198P; 1995. Shreve, P.D.; Grossman, H.B.; Gross, M.D.; Wahl, R.L. Metastatic prostate cancer: initial findings of PET with 2- deoxy2-[F-18] fluoro -D- glucose. Radiology 199:751–6; 1996. Yeh, S.D.J.; Imbriaco, M.; Garza, D.; et al. Twenty percent of bony metastases of hormone resistant prostate cancer are detected by FDG-PET whole body scanning [abstract]. J. Nucl. Med. 36(suppl ):198P; 1995. Agus, D.B.; Golde, D.W.; Sgouros, G.; et al. Positron emission tomography of a human prostate xenograft: association of changes in deoxyglucose accumulation with other Measures of outcome following androgen withdrawal. Cancer Res. 58:3009–14; 1998.