Twelve versus 36 months of adjuvant imatinib (IM) as treatment of operable GIST with a high risk of recurrence: Final results of a randomized trial (SSGXVIII/AIO)

LBA1

Background: Adjuvant IM administered for 12 months (mos) after surgery improves recurrence-free survival (RFS) of patients (pts) diagnosed with operable GIST. We compared 12 vs 36 mos of adjuvant IM as treatment of pts with a high risk for GIST recurrence after surgery. Methods: Pts with histologically diagnosed KIT-positive GIST were entered to this prospective, open-label, multicenter, randomized Phase III study (identifier NCT00116935 ). The risk of recurrence was estimated according to the modified Consensus Criteria. The primary objective was RFS. The secondary objectives included survival (OS) and treatment safety. The key exclusion criteria were ECOG PS >2, metastatic or inoperable GIST, and >12 wks from surgery to the study entry. IM was administered orally 400 mg/d. The sample size (n =200 in each group to obtain ≥110 events) was estimated by simulating log-rank tests assuming an overall hazard ratio (HR) of 0.44, a 20% drop-out rate, 2-sided type-I error rate of .05 and power 0.80. Analysis was based on the intention-to-treat population (ITT). Tumor histology was centrally reviewed. Results: 400 pts were entered to the study from Feb 2004 to Sep 2008. Three pts were excluded due to lack of consent from the ITT, which includes 15 pts who did not have GIST at a central review. The median FU time was 54 mos. RFS was longer in the 36-mo group compared to the 12-mo group (HR 0.46, 95% CI 0.32-0.65; p <.0001; 5-y RFS 65.6% vs 47.9%, respectively). Pts assigned to 36-mo of IM had longer OS (HR 0.45, 0.22-0.89; p =.019; 5-y OS 92.0% vs 81.7%). IM was generally well tolerated. The proportion of pts who discontinued IM during the assigned treatment period for reasons other than GIST recurrence was 25.8% in the 36-mo group and 12.6% in the 12-mo group. Exploratory efficacy subgroup analyses including KIT and PDGFRA mutation analysis data from 366 tumors will be presented. Conclusions: IM administered for 36 mos improves RFS and OS compared to 12 mos of administration as adjuvant treatment of GIST pts who have a high estimated risk of recurrence after surgery.

Non-metastatic Ewing's family tumors: High-dose chemotherapy with stem cell rescue in poor responder patients. Preliminary results of the Italian/Scandinavian ISG/SSG III protocol

10014

Background: Poor response to induction chemotherapy negatively affects prognosis for Ewing's sarcoma (ES) patients. To improve outcome, high-dose chemotherapy (HDCT) with stem cell rescue was added to conventional chemotherapy in patients poor responders after induction treatment. Methods: Non-metastatic ES patients aged <= 40 years were eligible. All patients received VAC (vincristin 2 mg-doxorubicin 80 mg/m2-cyclofosfamide 1,200 mg/m2) weeks 0 and 6, IVAc (ifosfamide 9g/m2-vincristin 2 mg-actinomycin 2 mg) week 3 and IE (ifosfamide 9g/m2-etoposide 450 mg/m2) week 9 as induction therapy. Poor response was histologically defined as persistence of microfoci of viable tumor cells and radiologically as persistence of soft tissue mass. As maintenance treatment good responders (GR) received three cycles of VAC-IVAc-IE. Poor responders (PR) received VAC (weeks 13,19), IE (week 22), CE (mobilizing cycle, cyclophoshamide 4g/m2, etoposide 600 mg/m2, week 16) and HDCT with BuMel (busulfan 4 mg/kg × 4 days orally and melphalan 140 mg/m2 with stem cell support week 25). Results: Starting from 1999, 296 patients were enrolled with a median age of 15 years (3–40). 54 % of the tumors were located to the extremities and 46% in the axial skeleton. 274 patients underwent local treatment: surgery in 222 (81%) patients (with post operative Rxt in 70), RxT alone in 52 (19%). 10 patients progressed during treatment. No toxic deaths were recorded. Response evalutation was available for 262 patients: 135 (52%) were PR. 116 of them completed protocol treatment and 19 did not receive HDCT (5 poor harvest, 5 medical contraindication, 9 refusal).With a median follow-up of 37 months (1–89) 5-year overall and event-free survival were 74 % and 65.5% respectively. 5-year event-free survival was 71% (95% CI 62–81%) for GR, 68% (95% CI 58–78%) for PR treated with HDCT and 35% (95% CI 10–60%) for PR who did not receive HDCT. Conclusions: The preliminary survival data show for PR similar outcome as for GR. The treatment including HDCT is feasible with no toxic death recorded. Longer follow-up will confirm its efficacy.

No significant financial relationships to disclose.