Elhalawani, Hesham M.

10030 Background: CABO is a multi-targeted kinase inhibitor of VEGFR2, c-MET, AXL and RET and has shown clinical evidence of activity in two prior studies of adult and pediatric patients with advanced OS. We report results of a feasibility cohort of CABO + MAP chemotherapy in patients with newly diagnosed high-risk OS. Methods: Patients <40 years, body surface area ≥ 0.8 m 2 and new diagnosis of high-grade OS with metastatic disease and resectable primary tumor were eligible. Patients were enrolled to two CABO dose levels (Dose Level 1 (DL1) 25 mg/m 2 /day, maximum 40 mg/day; Dose Level 2 (DL2) 35 mg/m 2 /day; maximum 60 mg/day) using a modified IQ 3+3 design. CABO was orally administered concomitantly with six 35-day cycles of MAP (cisplatin 60 mg/m 2 on days 1-2 (cycles 1-4), doxorubicin 37.5 mg/m 2 on days 1-2 (with dexrazoxane), methotrexate 12 g/m 2 on days 22, 29) and for 24 weeks as maintenance monotherapy. CABO was held after 7 weeks and resumed during 4 th MAP cycle to allow for primary tumor resection and metastatic surgeries. Dose-limiting toxicities (DLT) were assessed during weeks 1-6 of treatment; additional monitoring for post-surgical complications was conducted through completion of 4 th MAP cycle. The primary endpoint for this cohort was feasibility of combination therapy. Results: Ten patients (median age 11.5 years, 70% male) were enrolled and treated with CABO + MAP (DL1, n=4; DL2, n=6). Median cumulative doses of all MAP agents were maintained throughout treatment with no dose reductions. Concomitant dosing of CABO with MAP was tolerated with no protocol-specified DLTs occurring during the evaluation period. DL2 (35 mg/m 2 ) was established as the recommended dose for further study of combination therapy. Commonly reported adverse events included hematologic toxicities, febrile neutropenia, and alanine/aspartate aminotransferase (AST/ALT) elevation. Interruption of CABO was required in 6 patients for Grade 3+ AST/ALT elevation following methotrexate (n=3), wound complications (n=3), and Grade 3 pneumothorax (n=1). Wounds included two Grade 3 port site events and one Grade 1 primary tumor resection site; two patients later resumed CABO after healing without wound recurrence. Dose reduction of CABO was required in one patient for Grade 4 amylase elevation. Conclusions: The combination of CABO + MAP is feasible with no protocol-specified DLTs during the evaluation period. Generalizability is limited based on cohort size. Further evaluation of CABO + MAP is ongoing in a phase II/III randomized efficacy study in newly diagnosed standard and high-risk patients using CABO dose of 35 mg/m 2 /day (max 60 mg), with delayed initiation of CABO in patients with port placement and early safety monitoring for wound-related events, as well as full study monitoring of dose delivery, wound complications and systemic toxicity. Clinical trial information: NCT05691478 .

The translation of AI-generated brain metastases (BM) segmentation into clinical practice relies heavily on diverse, high-quality annotated medical imaging datasets. The BraTS-METS 2023 challenge has gained momentum for testing and benchmarking algorithms using rigorously annotated internationally compiled real-world datasets. This study presents the results of the segmentation challenge and characterizes the challenging cases that impacted the performance of the winning algorithms. Untreated brain metastases on standard anatomic MRI sequences (T1, T2, FLAIR, T1PG) from eight contributed international datasets were annotated in stepwise method: published UNET algorithms, student, neuroradiologist, final approver neuroradiologist. Segmentations were ranked based on lesion-wise Dice and Hausdorff distance (HD95) scores. False positives (FP) and false negatives (FN) were rigorously penalized, receiving a score of 0 for Dice and a fixed penalty of 374 for HD95. The mean scores for the teams were calculated. Eight datasets comprising 1303 studies were annotated, with 402 studies (3076 lesions) released on Synapse as publicly available datasets to challenge competitors. Additionally, 31 studies (139 lesions) were held out for validation, and 59 studies (218 lesions) were used for testing. Segmentation accuracy was measured as rank across subjects, with the winning team achieving a LesionWise mean score of 7.9. The Dice score for the winning team was 0.65 ± 0.25. Common errors among the leading teams included false negatives for small lesions and misregistration of masks in space. The Dice scores and lesion detection rates of all algorithms diminished with decreasing tumor size, particularly for tumors smaller than 100 mm3. In conclusion, algorithms for BM segmentation require further refinement to balance high sensitivity in lesion detection with the minimization of false positives and negatives. The BraTS-METS 2023 challenge successfully curated well- annotated, diverse datasets and identified common errors, facilitating the translation of BM segmentation across varied environments and providing the tools for future development of personalized volumetric reports to patients undergoing BM treatment.