Size-based clinical response evaluation is insufficient to assess clinical response of sarcomas treated with isolated limb perfusion with TNF-α and melphalan.

Background: The clinical assessment of the response of sarcomas to preoperative treatment is usually defined using size-based evaluation standards. For nonresectable sarcomas, hyperthermic isolated limb perfusion with TNF-α and melphalan (TM-ILP) yields high response rates. Based on our experience, we assume that anatomic radiological response criteria are insufficient to assess the degree of regression after TM-ILP.
Methods: The clinical response of 35 sarcomas to TM-ILP was assessed by unidimensional, bidimensional, and tridimensional size-based anatomical criteria, and responders were identified according to the established thresholds. The same tumors were investigated for pathological response according to the Salzer-Kuntschik regression scale (>90% devitalization) and reviewed for cystic degeneration, hemorrhage, and predominant necrotic or fibrosclerotic regression phenotype.
Results: None of the clinical response criteria were able to reliably identify the pathologic responders. The extent of size changes showed no association with the pathological degree of regression. The number of clinical responders was low compared with the number of pathological responders (RECIST N = 1, WHO N = 3, volumetry N = 3, pathology N = 19). The occurrence of hemorrhage and/or cystic degeneration was more frequently observed in predominant necrotic sarcomas and was associated with an increase in tumor size after TM-ILP. Furthermore, we identified the fibrosclerotic phenotype of regression to be more significantly strongly associated with posttherapeutic shrinkage than necrosis.
Conclusions: Size-based clinical response evaluation is insufficient to assess clinical response in TM-ILP-treated sarcomas. The size changes of tumors after therapy reflect the type of regression rather than the extent of destruction.