In our patients with locally advanced, recurrent, or metastatic SFTs treated with cytotoxic chemotherapy, the tumor response to chemotherapy was poor. The objective response rate was 0%, although 28% of patients had disease stabilization for longer than 6 months. The median PFS duration, measured from the start of chemotherapy, was 4.6 months.
Our findings are similar to those of Constantinidou et al., who found that only 1 (8%) of 13 patients with SFTs treated with first-line doxorubicin- or ifosfamide-based chemotherapy demonstrated a partial response and that the median PFS duration was 4 months . In our study, only 2 patients survived and were disease-free at last follow-up. Both patients had received neoadjuvant chemotherapy; in 1 patient the disease progressed, and in the other, the tumor exhibited only 50% necrosis. It is thus more likely that surgery, not chemotherapy, had a primary role in the successful outcomes of these patients.
Stacchiotti et al. recently published the results of response to chemotherapy in 31 patients with solitary fibrous tumor . In this study all patients received an anthracycline and 23/31 received an anthracycline plus ifosfamide . Both our study and Stacchiotti’s study  used RECIST to measure response. In Stacchioti’s study  there were more patients with either a partial response 6 (20%) or progression 16 (53%) while in our study the majority of patients (16 (72%) had stable disease. Median progression-free survival in both studies are similar, 4 months in Stachhioti’s study  vs. 4.6 months in ours. Overall the findings in the two studies indicate that there is some clinical benefit of chemotherapy in solitary fibrous tumors.
The findings of our retrospective analysis were limited by our small sample size and thus by a lack of statistical power. The heterogeneity of our patients’ disease characteristics was another potential limitation. We also were only able to use RECIST, not Choi criteria, to assess tumor response to conventional chemotherapy [17, 20, 21]. Owing to the limitations of the medical records, especially the inconsistencies in how radiologic studies had been performed and archived, a Choi response assessment was not feasible for most of the patients. Among the 4 patients whose Choi responses were measurable (all of whom had been treated with first-line doxorubicin-based chemotherapy), only 1 demonstrated a Choi partial response (a size reduction of 13% and a computed tomography density reduction of 35%), and the other 3 showed Choi stable disease. Had we been able to assess the entire group’s tumor responses according to the Choi criteria, it is possible that those response rates would have been higher than the RECIST response rates. The reason for this is that in the Choi criteria the definition of a PR is a decrease in tumor size of 10% or more or a decrease in tumor density of 15% or more. This is in contrast to RECIST where the definition of a PR is a decrease of 30% or more in the sum of lesions. One of the limitations of using RECIST is that we might be overlooking a potentially beneficial treatment secondary to where the cut off for response is placed, 30% in RECIST vs. 10% in the Choi criteria. However, the relative short PFS seen in our group suggests that ultimately, conventional chemotherapy is not particularly clinically effective regardless of the best tumor response achieved. Our findings are also in contrast to previously reported cases that did show tumor responses to doxorubicin-based therapy, and this discrepancy likely reflects our study’s limited sample size [6, 22].
Nevertheless, to our knowledge this retrospective analysis is one of the largest published series of patients with SFTs treated with conventional chemotherapy agents. The lack of objective tumor responses to chemotherapy observed in our patients suggests that treating patients who have advanced SFTs with first-line conventional doxorubicin- or gemcitabine-based chemotherapy might not be very effective to achieve tumor shrinkage. Thus, it may be useful to consider other therapies.
Recently, several novel agents have shown promise in the treatment of advanced SFTs. We previously reported our experience using the combination of temozolomide plus bevacizumab, a monoclonal antibody directed against the activity of vascular endothelial growth factor (VEGF) . Fourteen patients with locally advanced, recurrent, or metastatic SFTs were treated, with 79% demonstrating a Choi partial response (i.e., tumor size reduction of ≥10%, computed tomography density reduction of ≥15%, or both) [20, 21] and 14% demonstrating a RECIST partial response. The median PFS duration using the Choi response criteria was 9.7 months, and the median PFS duration using RECIST was 10.8 months.
Several tyrosine kinase inhibitors have also shown promise in treating SFTs. Sunitinib, a multitargeted tyrosine kinase inhibitor whose targets include VEGF receptors 1–3 and platelet-derived growth factor receptors A and B, has shown therapeutic promise for SFTs. Stacchiotti et al. found that 6 (60%) of 10 patients with advanced, chemotherapy-refractory SFTs treated with sunitinib achieved a Choi partial response, and 5 maintained their response for more than 6 months. The tumor response rate to sunitinib when assessed using RECIST was 0% . Sorafenib, another anti-VEGF-receptor, multitargeted tyrosine kinase inhibitor, has also produced tumor responses in patients with SFTs [13, 14]. Insulin-like growth factor I is overexpressed in some SFTs, and treatment regimens using figitumumab, an anti-insulin-like growth factor I receptor monoclonal antibody, have also produced tumor responses in a few patients with advanced SFTs [12, 16]. These and other treatments might be better alternatives for the treatment of SFTs than conventional chemotherapy.