ORIGINAL ARTICLE 
Annals of Nuclear Medicine Vol. 8, No. 3, 187-191, 1994 
FDG-PET for predicting the prognosis of malignant lymphoma 
Junichi OKADA,* Hiroshi OONISHI,* Kyosan YOSHIKAWA,** Jun ITAMI** Kimiichi UNO,** Keiko IMASEKI** and Noboru ARIMIZU** 
*Department of Radiology, Narita Red Cross Hospital 
**Department of Radiology, Chiba University Hospital 
To evaluate the usefulness of FDG-PET as a predictor of prognosis, 34 patients with untreated malignant lymphoma in the head and neck region were studied. After FDG-PET and treatment, they were observed from 15 to 50 months. Tumors which were aggressive and resistant to treatment tended to show high uptake of FDG. The survival rate of patients with high uptake of FDG, DAR >= 8, was lower than the rate of the other patients. It is considered to be useful to add FDG uptake of the tumor to other prognostic factors for predicting the prognosis. 
Key words: 18F-fluoro-deoxy-glucose (FDG), positron emission tomography, malignant lymphoma, prediction of prognosis 
INTRODUCTION 
IN OUR EARLY EXPERIENCE, higher uptake of 18F-fluoro-deoxy-glucose (FDG) and higher glucose metabolism was observed in lymphomas which showed poor response to treatment.1 The uptake of FDG correlated with the proliferative activity of the lymphoma, which was pathologically estimated.2 It was reported that the proliferative activity was helpful for grading and predicting prognosis.3,4 Consequently, positron emission tomography with FDG (FDG-PET) is considered to be promising as a predictor of cancer prognosis. 
FDG-PET is applied to various kinds of malignant tumors including musculo-skeletal tumors,5,6 liver tumors 7 and melanoma,8 lymphomas,1,2,9 etc. FDG-PET was used as an imaging procedure, and FDG uptake was compared to histological grading and tumor viability. In gliomas, FDG uptake may predict the survival rate of patients.10 However, relationship between the initial uptake of FDG and the long term survival rate has not been fully studied in malignant lymphoma. 
We have accumulated FDG-PET and clinical experience. Thirty-four patients, in whom FDG uptake of lymphomas had been measured before anti-cancer treatment, were observed for at least 15 months after the end of the treatment. In this study, the PET and clinical data of patients with malignant lymphoma were studied to evaluate the usefulness of FDG-PET as a predictor of the prognosis. 
PATIENTS AND METHODS 
Thirty-four patients, from 15 to 85 years old, with untreated malignant lymphoma in the head and neck region were studied. Three patients had Hodgkin's disease, and the other 31 patients had non-Hodgkin's lymphomas (NHLS). The 31 patients were diagnosed according to the Working Formulation for clinical usage. They were divided into low, intermediate, and high-grade malignancies by fresh biopsies 11 (Table 1). The pathological diagnosis, clinical staging by Ann Arbor classification, and mean tumor diameter at the beginning of the treatment of all patients were also listed in Table 1. After FDG-PET and the other examinations, they were treated by radiotherapy and/or chemotherapy, including Bleomycin, Vincristine, Cyclophosphamide, Adriamycin, Dexamethasone and Etoposide. When tumors disappeared on CT images by the treatment, we regarded this as "remission". When a small lesion remained after the treatment, a biopsy was performed at the lesion, and whether a remission occurred or not was pathologically diagnosed. They were observed for at least 15 months to fifty months after the treatment. The content of the treatment and the prognosis were listed in Table 1. 
FDG-PET Techniques 
In Chiba University Hospital. FDG was synthesized by CYPRIS and CUPID, a cyclotron system and drug-synthesizer system made by Sumitomo Heavy Industries. The patients were positioned to obtain PET images on the plane in which the tumors were shown at maximum size on X-ray CT. After transmission scan for attenuation correction, about 148 MBq (4 mCi) of FDG was injected intravenously in a fasting state longer than twelve hours. Beginning 60 minutes after the injection, a 5 minute scan was acquired by means of a Shimadzu-SET 130W (HEADTOME III) PET scanner, whose spatial resolution was 1.04 cm FWHM. 
FDG Activity was measured by setting a 10 x 10-mm square region of interest (ROI) on the PET image. A tumor ROI was set in the area where FDG activity was highest in the tumor. From the tumor activity, the differential absorption ratio (DAR) of FDG-to-tumor was calculated. The DAR was derived from the following equation 12. 
DAR = tumor activity / (injected dose / body weight). 
Statistical Analysis 
Student's t-test was used to compare mean values between the groups. Survival was measured from the beginning of treatment until death or the time of the last follow-up. The survival rate was estimated by the Kaplan-Meier method, and the statistical difference between survival curves was determined by the log-rank test.13 Multivariate analysis by means of the Cox proportional hazards model14 was used to test whether the following covariates had a significant impact on survival: DAR, age, histological grading by Working Formulation, tumor size, and staging by Ann Arbor classification. Covariates were included in the model by means of the maximum partial likelihood ratio test. Covariates were tested for their independent effects. 
RESULTS 
According to the prognosis after the treatment, all patients were divided into three groups. Group 1 (22 patients): complete remission was obtained and no sign of tumor relapse was observed. Group 2 (six patients): remission was obtained by the treatment, but relapse occurred. Group 3 (6 patients): remission was not obtained. Death occurred in five patients in Group 2 and in six patients in Group 3. The DAR of all patients was plotted in Figure 1. The mean value +- standard deviation for Groups 1, 2, and 3 was 6.4 +- 3.0, 7.0 +- 2.9, and 14.4 +- 5.5, respectively. The DAR of Group 3 was statistically higher than that of the other groups (p < 0.02) . The difference between Group 1 and Group 2 was not significant. 
The survival rate of patients was calculated by the Kaplan-Meier method. The patients with a DAR higher than 8 showed a lower survival rate than the patients with a lower DAR (Fig. 2). The difference between the two groups was significant (p < 0.005). We set the cutoff at 6, 7, 8, 9 and 10 and tested the differences between the two groups. The statistical difference was largest when we use the 8 cutoff. 
A Cox proportional hazards model was used to test the prognostic factors of DAR, age, pathological grading of malignancy, tumor size measured by X-ray CT and staging by the Ann Arbor classification (Table 2). DAR was confirmed as a prognostic factor as well as age and staging. 
DISCUSSION 
In our early experience, we reported that low accumulation of FDG was observed in low-grade NHL and high accumulation was observed in high-grade NHL. FDG-PET seemed to be useful for grading lymphomas.1 In Japan, the number of intermediate-grade NHL was much larger than the number of other lymphomas. Besides pathological grading, useful methods for grading and predicting prognosis were expected to be developed. Ki-67 immunostainning is one of the methods. A monoclonal antibody, Ki-67, reacts with a nuclear antigen present in cells in the G1, S, G2 and M phases of the cell cycle but not in the G0. The number of Ki-67 positive cells showed tumor proliferative activity and it was reported that patients with a high number had lower survival rates. This has been helpful in the prognosis of malignant lymphoma.3,4 In a previous report by us,2 the tumor accumulation of FDG correlated with Ki-67 positivity (r = 0.67). Ki-67 immunostainning was performed with fresh biopsy specimens. On the other hand, PET can be examined in vivo. 
In the study, we also obtained the value given by kinetic rates of glucose, k1k3 / (k2 + k3), by means of the graphic method demonstrated by Patlak and Gjedde et al.15,16 The value was closely connected with glucose metabolism.17 In our study, the kinetic rate correlated well with the DAR measured at 60 minute after injection in lymphomas. The DAR, which is obtained without arterial blood sampling, can be used as an indicator of glucose metabolism in the tumor.2 An optimal time point for measuring the tumor activity of FDG has not been defined.18 In all of our patients, the activities increased continuously for 60 minutes. Considering limited machine time and the good correlation between k1k3 / (k1 + k2) and the DAR measured at 60 minute after the injection, we have been using the latter value. 
To evaluate the usefulness of FDG-PET in predicting the prognosis, the tumor uptake of FDG before treatment was compared with the prognosis in 34 patients with lymphoma in the head and neck region. Six lymphomas, which were not controlled by the initial series of treatment, showed significantly higher uptake of FDG. Cancer death occurred in all cases during the period of observation. Remission was observed in the other 28 patients. 
In six of the 28 patients, tumor relapse was observed during the period of observation. Death occurred in five of the 6 patients. Before beginning the treatment, FDG-PET seemed to be helpful for detecting tumors which are very aggressive and resistant to treatment. However, in patients with remission, no significant difference in FDG uptake was observed between patients with and without tumor relapse. It was unable to differentiate relapsers from non-relapsers. 
When we set a cut-off value for DAR = 8, the survival rates of patients in whom tumor showed higher uptake of FDG (DAR >= 8) was significantly lower than those of the other patients. To the impact on survival of the uptake of FDG, we used the Cox proportional hazards model. Age, pathological grading by Working Formulation, tumor size and staging by the Ann Arbor system were included. Age, staging and DAR showed significant impact on survival. In clinical practice, many factors, for example age, stage, tumor size, pathological subtype, LDH, B symptom and the number of extranodal sites, are evaluated in predicting prognosis for a patient with malignant lymphoma. Many immunohistological tumor marker techniques have also been examined. One of them is Ki-67 staining. A useful method and a useful combination of some prognosticators are expected to be developed. In this study, the number of materials, especially Hodgkin's disease and NHL, except intermediate grade, is limited. However, it is considered to be useful to add FDG-PET to other prognostic factors for the prediction of prognosis. 
In summary, tumors which were aggressive and resistant to treatment tended to show high uptake of FDG. The survival rate for the patients with high uptake of FDG was lower than that for the other patients. FDG uptake had a significant impact on survival in the Cox proportional hazards model. FDG-PET is considered to be useful for predicting prognosis before treatment. 
ACKNOWLEDGMENTS 
This work was supported by Grant-in-Aid for No. 02151072 from the Ministry of Education Science and Culture of Japan. 
REFERENCES 
1. Okada J, Yoshikawa K, Imazeki K, Minoshima S, Uno K, Itami J, et al. The use of FDG-PET in the detection and management of malignant lymphoma: Correlation of up-take with prognosis. J Nucl Med 32: 686-691, 1991. 
2. Okada J, Yoshikawa K, Itami M, Imaseki K, Uno K, Itami J, et al. Positron emission tomography using fluorine-18-fluorodeoxyglucose in malignant lymphoma: A compari-son with proliferative activity. J Nucl Med 33:325-329, 1992. 
3. Schrape S, Jones DB, Wright DH. A comparison of three methods for the determination of the growth fraction in non-Hodgkin's lymphoma. Br J Cancer 55: 283-286, 1987. 
4. Hall PA, Richards MA, Gregory WM, d'Ardenne AJ. Lister TA. Stansfeld A. The prognostic value of Ki-67 immunostaining in non-Hodgkin's lymphoma. J Pathol 1554: 223-235, 1988. 
5. Kern KA, Brunetti A, Norton JA, et al. Metabolic imaging of human extremity musculoskeletal tumors by PET. J Nucl Med 29: 181-186, 1988. 
6. Griffeth LK, Dehdashti F. Mcguire AH, et al. PET evaluation of soft-tissue masses with fluorine-2-deoxy-D-glucose. Radiology 182: 185-194, 1992. 
7. Okazumi S, Isono K, Enomoto K, et al. Evaluation of liver tumors using fluorine-18-fluorodeoxyglucose PET: characterization of tumor and assessment of effect of treatment. J Nucl Med 33: 333-339, 1992. 
8. Gritters LS, Francis IR, Zasadny KR, Wahl RL. Initial assessment of positron emission tomography using 2-fluorine-18-fluoro-2-deoxy-D-glucose in the imaging of malignant melanoma. J Nucl Med 34: 1420-1427, 1993. 
9. Leskinen-Kallio S. Ruotsalainen U, Nagren K, Teraes M, Joensusu H. Uptake of carbon-11-methionine and fluoro-deoxyglucose in non-Hodgkin's lymphoma: a PET study. J Nucl Med 32: 1211-1218, 1991. 
10. Patronas NJ, Di Chiro G, Kufta C, et al. Prediction of survival in glioma patients by means of positron emission tomography. J Neurosurg 62: 816-822, 1985. 
11. National Cancer Institute. Study of classifications on non-Hodgkin's lymphomas: summary and description of a working formulation for clinical usage. Cancer 49: 2112-2135, 1982. 
12. Fukuda M, Matuzawa T, Ito M. Clinical evaluation of cancer diagnosis with 18F-2-fluoro-D-glucose. Its usefulness in liver and pancreas cancers. CYRIC Annual Report, Tohoku University, pp. 244-250, 1983. 
13. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 53: 457-481, 1958. 
14. Cox DR. Regression models and life tables. J Royal Stat Soc Series B 34: 187-229 1972. 
15. Patlak CS, Blasberg RG, Frenstermacher JD. Graphical evaluation of blood-to-brain transfer constants from multiple-time uptake data. J Cereb Blood Flow Metab 3: 1-7, 1983. 
16. Gjedde A, Wienhard K, Heiss WD, et al. Comparative regional analysis of 2-fluorodeoxyglucose and methylglucose uptake in brain of four stroke patients with special reference to the regional estimation of the lumped constant. J Cereb Blood Flow Metab 5: 163-178 1985. 
17. Wienhard K, Pawlik G, Herholz K, Wagner R, Heiss WD. Estimation of local cerebral glucose utilization by positron emission tomography of [18F]2-fluoro- 2-deoxy-D-glucose: a critical appraisal of optimization procedures. J Cereb Blood Flow Metab 5: 115-125, 1985. 
18. Fischman AJ. Alpert NM. FDG-PET in oncology: There's more to it than looking at pictures. J Nucl Med 34: 6-11, 1993.