Abstract
Objective: In patients with soft tissue sarcoma (STS) the histological response (tumour grade regression: TGR) to neoadjuvant chemoradiotherapy (CRT) may influence the outcome. The main aim of the study was to evaluate the predictive value of 11C-methionine (MET) and 18F-FDG PET/CT in patients with STS treated with neoadjuvant CRT, correlating TGR with SUVmax (standardized uptake value) percentage variation before and after CRT. Patients and Methods: Nine patients with STS already scheduled for a neoadjuvant CRT and surgery were enrolled. They underwent MET and FDG PET/CT in a one-day procedure before and after the CRT. Pre-therapy SUVmax and the percentage variation of SUVmax for MET and FDG were correlated with TGR according to the Huvos grade. Grades I-II were considered as partial responders (PR) and grades III-IV as complete responders (CR). Results: FDG pre-treatment mean SUVmax in PR patients was 7.1, while in CR patients it was 13.2. Pre-treatment mean MET SUVmax in PR patients was 7.5, while in CR patients it was 4.9. The mean percentage variation in FDG SUVmax, was -21.2% in PR patients and -74.5% in CR patients while that for MET SUVmax was 48% in PR patients and -53.9% in CR patients. Conclusion: According to this preliminary study, the percentage variation in FDG before and after CRT seems to discriminate between PR and CR better than MET.
- Soft tissue sarcoma
- PET
- FDG
- methionine
- TRG
- Huvos classification
Soft tissue sarcomas are a heterogeneous group of tumours that arise from tissues of mesenchymal origin. They comprise approximately 0.7-1% of adult malignancy (1) and 6.5% of all cancers in children younger than 15 years of age (2). About 45% of all soft tissue sarcomas are found in the extremities, especially in the lower limb, and 20% intra-abdominally (3).
The metastatic spread of sarcomas is mainly hematogeneous to the lungs (50%), bone, liver and brain (40%), and sometimes to the retroperitoneum and other soft tissue. Lymphatic diffusion to regional nodes may also occur (1, 3, 4). The presence or the absence of metastasis, the tumour histology and the malignancy grade mainly influence the treatment of choice.
Diagnostic imaging plays a key role in the evaluation of patients with sarcoma; the site and size of the primary tumour can be determined using magnetic resonance imaging (MRI) and conventional computed tomography (CT) but benign soft tissue masses and soft tissue sarcomas may appear to be very similar on clinical and radiological examination (1). Biopsy of the mass is the most specific method for diagnosis and grading and this is usually directed by anatomical imaging (1, 5). Positron emission tomography (PET) with 18F-FDG has been proposed as a tool which may be useful in the management of soft tissue sarcomas (2-17). In fact, several studies have also shown significant differences in FDG uptake values between low- and high-grade soft tissue sarcomas (3, 7, 8, 12-14).
Furthermore, FDG PET could be useful in the detection of recurrent (6-8, 11, 12) and metastatic disease (2, 6, 15, 16), or in guiding biopsy into the most metabolic area and finally in the prediction and/or monitoring response to therapy (3, 6, 9, 17). Nevertheless, it should be remembered that the use of 18F-FDG PET in soft tissue sarcoma (as in any other tumour) could be limited by the relatively high rate of false-positive results caused by inflammatory tissues.
For this reason some authors have addressed investigations to the potential use of some other radiopharmaceuticals which involve proteic or fatty acid metabolism such as fluorine-18alfa methyltyrosine (FMT), 11C-methionine (MET), 11C-choline, and 11C-tyrosine (18-20).
Hence, the main aim of this study was to investigate if PET/CT (with FDG or MET) correlates to tumour grade regression (TGR) in a group of patients with soft tissue sarcoma treated with neoadjuvant chemoradiotherapy and to determine the best tracer for this purpose.
Patients and Methods
Between September 2003 and February 2005, 9 patients (6 women and 3 men, aged between 16 and 64 years) were prospectively studied with different histological types of high-grade soft tissue sarcomas, mainly found in limbs (patient population characteristics are summarized in Table I). All patients underwent a pre-surgical neoadjuvant treatment with radio and chemotherapy. Before starting therapy, all patients underwent FDG and/or MET PET/CT: 7 patients underwent FDG and MET PET on the same day; 1 patient underwent only MET PET, and 1 patient only 18F-FDG PET/CT (Table II).
Chemotherapy consisted of 3 cycles, one cycle every three weeks after hematological examination, of epirubicine (60 mg/mq) plus ifosfamide (3,000 mg/mq) on the first and second day, and ifosfamide (3,000 mg/mq) alone on the third day.
Radiotherapy treatment was performed after chemotherapy with 3D conformal radiotherapy (3DCR), up to a total dose of 4,400 cGy in 22 days (200 cGy/day).
All patients underwent neoadjuvant chemotherapy, while only 6 patients were also subjected to radiotherapy.
After about 20 days from the end of the treatment, all patients underwent a second PET scan: 7 patients had both FDG and MET PET scan on the same day, 1 patient only MET PET and 1 only FDG PET (Table II). In every PET scan, standardized uptake value (SUV) of all lesions, expressed in g/ml, was evaluated and only the highest SUVmax value was recorded.
All scans were carried out on a dedicated PET/CT tomograph (Discovery, GE, Milwaukee, WI, USA), following standard procedures. PET acquisition was performed in the 6 hours'-fasted patient, after intravenous injection of 5.3 MBq/kg of 18F-FDG or 11C-MET; uptake time was 60 min for FDG and 10 min for MET. PET images were evaluated on the basis of visual inspection by three experienced readers; in all cases agreement among readers was obtained for the final report.
After post-therapy PET scan, all patients were addressed to surgery and every tumour mass was analyzed by the pathologist.
The Huvos grade was applied (which considers the percentage of necrosis after the treatment) (Table III) to evaluate the histological tumour response (TGR) (21). Grades I-II were considered as partial responders (PR) and grades III-IV as complete responders (CR).
After surgery, some patients completed the treatment with adjuvant chemo and/or radiotherapy. Chemotherapy included 2 cycles with the same schedule as the neoadjuvant therapy; radiotherapy was performed using 3DCR and/or brachytherapy (BRT): 4 patients underwent chemotherapy, 4 patients chemotherapy and external radiotherapy (RTE), and 1 patient interstitial BRT.
The objectives of this study were firstly to evaluate the correlation between the pre-treatment uptake of FDG and MET (measured by SUVmax) with the TGR; secondly, to evaluate the correlation between the variation expressed in percentage of the FDG and MET uptake before and after CRT with the TGR.
Results
Histological response criteria according to Huvos grade were: 4 patients grade I, 1 patient grade II, 2 patients grade III and 2 patients grade IV. Histological results are summarized in Table IV.
In order to meet the first objective of the study, pre-treatment FDG uptake (expressed as SUVmax value: FDG-SUV1) was correlated with histological response according Huvos grade. The mean SUV1 in 4 patients considered PR was 7.1 g/ml while that in 4 patients considered CR was 13.2 g/ml. Results are summarized in Figure 1. Pre-treatment MET uptake (MET-SUV1) was also correlated with histological response according to Huvos grade. The mean SUV1 in 4 patients considered PR was 7.5 g/ml while that in 4 patients considered CR was 4.9 g/ml. Results are summarized in Figure 2.
In order to determine the correlation between percentage variation of the FDG SUVmax and Huvos grade, only 8 patients could be considered. Results are summarized in Figure 3. In particular in PR patients, the percentage variation in SUVmax ranged between +44.4% and -27.9%, except in one case (-84.21%), with a mean percentage of -21.2% (standard deviation, SD=52.7%). In CR patients, the range was between -61.9% and -83.3, with a mean percentage of -74.5% (SD=10.3%).
Correlating pre and post neoadjuvant percentage variations in SUVmax and Huvos grade, the results are summarized in Figure 4. In particular in PR patients, the range was between +194.8% and -75.14% with a mean percentage of 48% (SD=139.7%); in CR patients, the range was between -15.78% and -72.2%, with a mean percentage of -53.9% (SD=25.7%).
Discussion
Jones et al. proposed a study with 9 pateints who were subjected to different kinds of neo-adjuvant treatment (chemotherapy/radiotherapy/hyperthermia) deducing that FDG PET may be of benefit in the monitoring of sarcoma response to neoadjuvant therapy (22).
Schuetze et al. demonstrated that patients with a baseline tumour SUVmax≥6 and a <40% decrease in FDG uptake were at high risk of systemic disease recurrence, while patients whose tumours had a ≥40% decline in the SUVmax in response to chemotherapy were at a significantly lower risk of recurrent disease and death after complete resection and adjuvant radiotherapy. In this way, they showed which patients could benefit from neo-adjuvant treatment (6). Bredella et al. proposed a study in which they found that FDG PET is better than MRI in evaluating the viable tumour after post-therapeutic changes (9). Only limited data are currently available concerning FDG PET for preoperative assessment of response to chemotherapy and they are essentially derived from retrospective studies. In all reported cases, there is a correlation between uptake intensity after chemotherapy and the percentage of viable cells (23, 24). The long-term prognosis of patients considered to be “good responders” by FDG PET is unknown at present and therefore cannot be compared to that of patients with a good histological response.
In a recent study, Evilevitch et al. (25) demonstrated that a change in tumour glucose metabolic activity is a significantly more accurate parameter than a change in size for assessing histopathological response to neoadjuvant therapy in patients with high-grade soft tissue sarcoma.
The present study has shown that the patients of the PR group (Huvos grade I-II) presented similar pre-therapy FDG and MET SUV1 values as well as similar standard deviation (SD): mean SUV1 FDG value was 7.1 g/ml, SD 5.6; mean MET SUV1 value was 7.5 g/ml, SD 6.7. Instead, patients in the CR group (Huvos grade III-IV) presented higher FDG than MET SUV1 values, with a smaller MET standard deviation (mean FDG SUV1: 13.2 g/ml, SD 5,7; mean MET SUV1: 4.9 g/ml, SD 2.1). The high grade of flogosis within the tumour mass may be the cause of the relatively high FDG SUV1 value in CR patients.
Discussing the second objective of the study, if a cut-off percentage decrease of SUVmax FDG of 45% is considered, it is possible to discriminate between CR and PR patients: in fact in the PR group, 3 out of 4 patients had a SUVmax decrease of less than 27.09% and in the CR group all patients presented a decrease of SUVmax of more than 61.9% [positive predictive value (PPV) 100%; negative predictive value (NPV 75%)]. Even if preliminary, these data could testify that there is a good correlation between metabolic (FDG SUVmax value) and histological (TGR) behaviour. On the other hand, it is not possible to identify an SUV1 cut-off value of percentage variation of MET SUVmax between CR and PR groups (PPV 75%; NPV 50%; with the 45% cut-off). In fact, similar percentage decreases in SUVmax both in PR and CR groups were observed (-67.53%, -75.14% and -64.7%, -72.2% respectively). Nevertheless, a remarkable augmentation of SUVmax percentage was present only in the PR group, for both MET (two patients: +140% and +194.8%) and FDG tracers (one patient: +44.4%).
Comparing results obtained with FDG and MET PET/CT scans prior to neoadjuvant CRT, the use of SUVmax to predict TGR is not recommended. In fact, neither radiotracer shows different uptake behaviour between PR and CR.
Instead the use of the percentage variation in FDG PET SUVmax between pre and post therapy could give a good differentiation between PR and CR patients. On the contrary, the percentage variation in MET PET SUVmax did not show a clear differentiation between PR and CR. Therefore, FDG PET (and in particular the percentage variation of the SUVmax value) is better than MET PET in discriminating PR and CR patients.
As several previous studies in soft tissue sarcoma have shown, FDG seems to be a good tracer for monitoring the response to therapy. It is clear that further studies with a larger number of patients are necessary to validate these preliminary assertions.
- Received July 23, 2008.
- Revision received December 2, 2008.
- Accepted December 10, 2008.
- Copyright © 2009 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved