(P 165) REAL- WORLD EXPERIENCE OF TYROSINE KINASE INHIBITORS IN PATIENTS WITH RECURRENT BONE TUMOURS (BT): A CANSARCC STUDY

Objective: Survival after relapse in osteosarcoma (OS), Ewing Sarcoma (ES) and chondrosarcoma (CS) remains dismal. Recent reports suggest a potential role of tyrosine kinase inhibitors (TKI) including Regorafenib (R) and Cabozantinib (C) in the management of recurrent BT. We conducted a retrospective multi-centre pan-Canadian study to assess real-world outcomes with these novel treatments.

Methods: After ethics board approval, data from patients with recurrent BT treated in 10 different institutions were retrospectively extracted from the CanSaRCC (Canadian Sarcoma Research and Clinical Collaboration) Database including patient demographics, disease characteristics, and toxicity from drug (based on clinic notes), response rate (based on imaging) and outcome. Toxicity was graded according to CTCAE. Progression-free and overall survival were estimated using Kaplan-Meier and log-rank test. T-test, ANOVA and Wilcoxon rank analyses were used to compare means and medians between groups. Chi-square and Fisher’s test were used to compare demographics features. Univariate (UVA) and multivariate (MVA) Cox regression models were computed for variables correlating to outcomes.

Results: From 2018-2022, 56 pts received R or C (Table 1). Best response by histology are listed in Table 2. Median time to best response was 2.8 mo (range 1.0-17.2mo) for R and 2.0 mo (range 0.6-5.0 mo) for C. Dose reduction was required in 20 (35.7%) patients (R: 10/23, 43.5% and C: 10/33, 30.3%). For R, starting dose was 40 – 60 mg in 3 pts (1 dose reduced), 80 mg in 5 pts (3 dose reduced), and 120-160 mg in 15 pts (6 dose reduced). For C, starting dose was 40 mg in 16 pts (5 dose reduced) and 60 mg in 17 pts (5 dose reduced). Grade ≥ 3 toxicity was reported in 4/23 (17.4%) for R and in 4/33 (12.1%) for C, including bleeding (n=2 only C), mucositis (n=2, 1 R, 1C), hand foot syndrome (n=2 only R), and 1 patient each with elevated bilirubin (only R) and elevated LFTs (only C) (Table 2). With a median follow-up of 10.5 mo (n=23, range 1.8-29) for R, and 5.3 mo (n=33, range 1.9-21) for C, median OS was 12.2 mo (95% CI 7.2–13.8) and median PFS was 3.95 months (95% CI 2.8–5.75) (Figure 1). Swimmer plots present the length of treatment duration and tumour response for patients with recurrent BT by TKI and histology (Figure 2). By univariate analysis, age, line of therapy, gender, location of primary, pathology or R vs. C did not correlate with PFS.

Conclusion: This is the first retrospective study conducted through the national collaboration of sarcoma centers across Canada under a defined consortium agreement. This collaboration among 10 sites across Canada permits the unique opportunity to increase the sample size of patients with rare sarcoma for real-world data. Consistent with previous published studies, our pan-country/national, real-world analysis shows that TKIs have meaningful activity in the setting of recurrent BT with acceptable toxicities. Inclusion in earlier lines of treatment and/or maintenance therapy could be questions for future research.