In this prospective single center study, a significant association between early interim FDG PET2 parameters and TTP was observed. Patients with higher residual disease metabolic activity at 6 weeks after initiation of neo-CTX study had worse outcomes compared to those with no or only mild residual FDG activity. Additionally, patients showing an MTV reduction from PET1 to PET2 had a lower pathological response rate. However, none of the PET parameters were predictive of OS.
FDG PET/CT is important for staging and therapy response assessment of patients with high-grade bone and soft tissue sarcomas (10, 18). However, limited data are currently available regarding the potential role of early interim PET/CT performed 6 weeks after neo-CTX initiation as a predictor of patient outcome, especially in the pediatric population. Conflicting published results are probably due to heterogeneous patient populations, different therapy regimen and different time points of PET/CT evaluations. Costelloe et al. reported in a mixed population of pediatric and adult bone sarcoma patients that SUVmax and TLG values measured before and after neo-CTX provided predictive information about treatment response (19). In a small study of bone sarcoma patients, changes in FDG-SUVmax at the end of neo-CTX predicted histopathologic responders and non-responders (23). However, as both studies measured glucose metabolic parameters after completion of neo-CTX, the impact on managing these patients appears limited. In the present study glucose metabolism responses were measured early during neo-CTX. Earlier identification of non-responders to neo-CTX could lead to meaningful treatment changes. In the current study a significant association between the early interim FDG PET parameters and patient outcome was observed. However, no significant association between the baseline FDG PET parameters and TTP was observed. This contrasts with other studies reporting PET1-SUVmax as a prognostic biomarker (20, 21).
None of PET1, PET2 and ΔPET parameters were associated with OS. This is probably due to the relatively small sample size and the limited number of events in the current study population.
Histologic response to neo-CTX is known to be a prognostic indicator in bone and soft tissue sarcoma, especially in osteosarcoma. Patients with > 90% tumor necrosis in response to treatment have improved outcomes (15). In contrast to other studies (22–25) we did not observe a significant association between any baseline or early interim FDG PET parameters and the response to treatment assessed by the percentage of tumor necrosis (n = 29) or by MRI evaluation (n = 5). However we observed a significant association of ΔMTV with the pathological response to neo-CTX. Additionally, patients with higher PET2-SUVmax were less likely to be responders to neo-CTX although without statistical significance. These results suggest that tumor metabolic activity changes under neo-CTX as assessed on early interim FDG PET2 can be integrated into the clinical risk prognostic assessment.
FDG PET/CT is a whole-body imaging modality and can detect distant metastatic lesions. In our cohort patients with metastatic disease had worst outcome. Of note, all metastatic lesions showed partial or complete response on PET2. It is considered standard of care to treat distant metastases with local control methods (surgery and/or radiation). Whether or not the prognosis is altered with this aggressive approach is controversial.
The main limitation of the study is its small sample size and the heterogeneity of the included sarcoma sub-types. However, each tumor sub-type was treated under the same therapy protocol. Results of the study were comparable in the sub-population of patients with bone sarcoma only (osteo- and Ewing sarcoma). Of note, patients with rhabdomyosarcoma were not included in this study as they were enrolled on competing COG therapeutic clinical trials, which included PET imaging as an experimental aim. Further sub-analyses considering the different tumor subtypes were not feasible considering the small sample size and the limited number of events. Larger and more homogeneous cohorts will be required to determine whether early interim FDG PET/CT imaging can be useful for early treatment response predictions and prognostic information. However, such studies are difficult to conduct as bone and soft tissue sarcomas are rare neoplasms, especially when only considering pediatrics patients only. Another limitation is the lack of control for tumor necrosis. The data reported were obtained using the pathology and MRI clinical reports considered as reference in the treatment management of the patient. Finally, the comparison with the RECIST criteria for assessing the response to neo-CTX was not performed because of heterogeneous conventional imaging follow-up (modality, time points).