Pazopanib as first line treatment for solitary fibrous tumours: the Royal Marsden Hospital experience

Since January 2013, 13 patients with progressive SFT have been treated at the Royal Marsden Hospital with continuous dosing pazopanib. This is a prospective collection of data of all consecutive patients treated in the same way in a single institution. Among the 11 patients assessable for response, the clinical benefit rate (partial response + disease stabilisation) was the same (82%) using both RECIST and Choi criteria. The outcome of the patients in our series suggests that pazopanib is an active treatment in advanced SFT. This is consistent with the results achieved with other anti-angiogenic therapies already reported in the literature. However, our series show some significative differences.

For instance, Stacchiotti et al. [16] focused on the treatment of SFT with sunitinib. Among a large cohort of 31 patients, they found a larger number with progressive disease compared to this series, and a lower overall response rate. However, interestingly, in that series the median PFS was 6.0 months whereas it was 4.7 months in our patients. Similarly to us, they found sunitinib to be active in SFT, with possible long lasting disease control.

The French Sarcoma Group recently reported their experience with sorafenib, another small molecule antiangiogenic drug [22]. Among five patients, they found no objective response, with a median overall survival of less than 20 months, but with some evidence of disease stabilisation.

Similar results were reported from a single cancer centre, 10 patients received sunitinib or pazopanib as a second or further line of treatment, with a median PFS of 5.2 months and overall an acceptable toxicity profile [23].

The MD Anderson Cancer Centre reported on the efficacy of an anti-angiogenetic drug combination for SFT [14]. Temozolamide plus bevacizumab produced a Choi partial response in 11 patients (79%) with hemangiopericytoma and malignant SFT, two patients (14%) had stable disease and one patient (7%) had progressive disease. The estimated median PFS was 9.7 months with a 6-month progression-free rate of 78.6%.

The role of chemotherapy has also been investigated. Our group [12] reported of 17 SFT patients treated with anthracycline-based chemotherapy, in whom there was only 1 partial response, and most of the patients had progressive disease. The median PFS was 4.2 months and OS was 14.6 months. The study was interpreted as showing that chemotherapy was of little value in advanced SFT, however the median PFS, OS and likelihood of response according to RECIST were little different from the results with pazopanib, emphasising the need for caution in interpreting the results from small series, and demonstrating the need for larger, prospective phase II trials in this disease.

Another group of authors has reported on the outcome of SFT treated with dacarbazine [13] showing the antitumor activity of this drug in SFT. Among 8 patients treated, they found 3 partial responses and only one progressive disease, with a median PFS of 7 months.

The same French group who reported on antiangiogenic drugs, also reported their experience with chemotherapy [23]. They treated 23 patients in first line with different chemotherapy regimens, mainly containing anthracycline (18 out of 23). The median PFS was 5.1 months, with 2 partial responses (9%) and 13 stabilizations (57%) as best response.

A comparison among the survival outcome from all the studies conducted so far for metastatic SFT is reported in Table 4.

In summary, and compared to the largest recent series reported, our data confirm that pazopanib has some activity against SFT. A number of agents appear to be capable of inducing genuine disease stabilisation. Our experience and that of the group at MD Anderson suggests that consideration should be given to evaluation of response using Choi criteria rather than RECIST. Several studies suggested that Choi could be a more sensitive tool to evaluate disease response in sarcoma, especially in gastrointestinal stromal tumours [21,24,25], and there is some evidence for its use in non GIST sarcomas [26,27]. A significant obstacle to the widespread introduction of these criteria is the limited reproducibility. However some of the challenges of Choi criteria such as the need to avoid areas of fistula and obtaining reproducible measurements on mobile small bowel lesions are more common in GIST and use in non GIST sarcomas is therefore often less problematic. The use of Choi criteria, with the double assessment of density and size, requires increased time of an experienced sarcoma radiologist to report all examinations for the same patient and consistent CT protocols at each follow-up time point. Implementation of Choi criteria in a non-oncologic hospital, with lack of expertise may not be possible. Anyway, Choi criteria were overall able to better appreciate the response, since they are based on changes in contrast enhancement on CT, which reflects tumour blood flow, as well as tumour size, which may not alter greatly with these agents. We assessed the activity of pazopanib both with RECIST and Choi. Since Choi criteria are more stringent in the definition of progression, we were not concerned about failing to identify treatment failure. and indeed when RECIST showed progressive disease this was confirmed by Choi. Similarly, in 4 cases (36%) of stable disease by RECIST, partial response could be detected by Choi, with a significant reduction of attenuation on CT (Figure 3). For future studies Choi criteria should be incorporated as part of the disease assessment as well as RECIST as a better indication of disease modulation by agents such as pazopanib.

In terms of toxicity, pazopanib appeared to be an acceptable treatment. Most of our patients experienced grade 1 to 2 toxicities and only a few of them had severe side effects The toxicity profile in our cohort of SFT is in line with the result from the phase III registration trial of pazopanib in soft tissue sarcoma [19]. The most common adverse events reported with pazopanib in treating sarcomas were fatigue, diarrhoea, nausea, weight loss, and hypertension. From our series, fatigue, diarrhoea, and skin reaction were the most common side effects. Interestingly, only a small number of grade 3 or 4 side effects were observed.

In terms of further research, a phase II trial, to be performed by the Spanish Sarcoma Group together with French and Italian Centres, is due to investigate the role of pazopanib in SFT (NCT 02066285). An American study in advanced sarcoma with dasatinib included patients with SFT (NCT 00464620) and final data are awaited. We are recruiting to a translational research study with pazopanib which will include patients with SFT utilizing tumour biopsies. In this way, we want to search for molecular targets or specific pathways able to help in understanding which patients could most benefit from these treatments.

Overall, our data confirm that anti-angiogenic agents have some activity against SFT. Pazopanib is licensed for the treatment of soft tissue sarcomas and hence is available for treating this particular subtype. However, in spite of the evidence that disease stabilisation and sometimes objective response can be achieved in a percentage of patients, median PFS and OS are remarkably similar across the various reports indicating the need for a more effective approach to the systemic management of this disease.