Cediranib for alveolar soft part sarcoma: a randomised study in relation to clinical practice

In The Lancet Oncology, Ian Judson and colleagues, investigators of the Cediranib for Alveolar Soft Part Sarcoma (CASPS) trial, report on a randomised, placebo-controlled, phase 2 trial testing the tyrosine-kinase inhibitor cediranib in metastatic alveolar soft part sarcoma (ASPS).The study was formally positive, although the clinical benefit of cediranib was small. ASPS is a rare subtype of sarcoma that mostly affects young adults, with a high frequency of distant metastasis leading to poor long-term survival despite a typically indolent disease course. Activity of cediranib in metastatic ASPS has previously been shown in a phase 2 study in gastrointestinal stromal tumours and sarcomas, including six patients with ASPS, four of whom had a durable partial response and one had prolonged stable disease. Additionally, in a National Cancer Institute study of 46 patients (43 evaluable) with ASPS, 15 (35%) achieved a Response Evaluation Criteria in Solid Tumour (RECIST)-defined overall response, 26 (60%) stable disease, and 36 (84%) controlled disease (ie, stable disease and partial responses) at 24 weeks. In the CASPS trial, 32 patients with ASPS were treated with cediranib and 16 were given placebo, and after 24 weeks (or sooner if disease progression occurred) all patients on placebo were crossed over to cediranib. With a median follow-up of 34·3 months (IQR 23·7–55·6) at the time of data cutoff for these analyses (April 11, 2018), this study met its primary endpoint, which was based on tumour response, defined as percentage change in the median sum of the longest diameters of target marker lesions at 24 weeks. Judson and colleagues aimed to detect a 20% difference in favour of cediranib, and found a significant difference in the median sum of the diameters of target marker lesions of 22% at 24 weeks (−8·3% [IQR −26·5 to 5·9] in the cediranib group vs 13·4% [1·1 to 21·3] in the placebo group; one-sided p=0·0010), even though the number of patients enrolled was relatively low (overall n=48, evaluable population n=44). Unexpectedly, of the evaluable participants at week 24 (n=28), 11% (n=3) achieved a RECIST defined partial response and 50% (n=14) had stable disease, results that are inferior to previous phase 2 studies of cediranib.This randomised study treated fewer patients with ASPS with cediranib than the NCI phase 2 study. Our question is to what extent the randomised study design, which selected response as the primary endpoint in this specific population and tested this class of drugs, increased the reliability of the CASPS trial results compared with other uncontrolled, phase 2 studies. Concerns about spontaneous disease stabilisation and slow progression of metastatic ASPS were the basis for conceiving the placebo-controlled design and for requiring evidence of progression in the previous 6 months among the entry criteria. This design required a longer study duration than other uncontrolled, phase 2 studies and made cediranib available to fewer patients, although admittedly after 24 weeks the patients in the placebo group were switched to cediranib. However, 44% (n=7) of patients in the placebo group had stable disease at 24 weeks, suggesting that the requirement of disease progression in the previous 6 months did not add substantially to the study design. Patients in the placebo group did not show spontaneous regression, by contrast with the placebo group in a randomised trial testing sorafenib in treatment-refractory and advanced desmoid tumours, a mesenchymal neoplasm notable for its different natural history.(...)

The Lancet Oncology , commentaire en libre accès, 2018

Voir le bulletin