Background: There has been a lack of prospective clinical trial data on subsequent treatment after immune checkpoint inhibitor (ICI)-based treatment in HCC. To address this question, we designed and completed a phase II multicentred study to evaluate the use of cabo in HCC after prior ICI-based treatment. Methods: This is an investigator-initiated clinical trial involving 3 academic centres in Hong Kong and Korea. Key eligibility criteria include confirmed diagnosis of HCC; refractoriness to prior ICI-based treatment; duration of prior ICI-based treatment ≥2 months; Child's A liver function; ECOG performance status 0-2; maximal 2 lines of therapy. All patients (pts) were started cabo at 60mg once daily to progressive disease or intolerable toxicity. The primary endpoint is progression-free survival (PFS). Other secondary endpoints including response rate (RR) according to the RECIST, overall survival (OS) and treatment-related adverse event (TRAE). Total sample size is 48 (one-sided alpha 0.05; power 80%). Results: Total 48 pts were recruited from Oct 2020 to May 2022. Data were frozen on 1 Aug 2022 for analyses. The median follow-up was 11.2m. Seven (14.9%) pts remained on treatment; 1pt was found to ineligible and excluded from the analysis. The median age is 61 (Range 36-83). Aetiology of HCC: HBV 34 (72.3%); HCV 2 (4.3%); alcoholic 8 (17%); Most pts have BCLC stage C (93.6%). All pts have Child’s A liver function. Total 39 (83%) pts receive ICI-based first-line therapy, and 19 (40.4%) pts have prior atezolizumab-bevacizumab (AB) treatment. The median PFS is 4.1m (95% CI 3.3-5.3). Regarding RR, PR and SD occurs in 3 (6.4%) and 36 (76.6%) pts, respectively. The median OS is 9.9m (95% CI 7.3-14.4) and the 1-year survival rate is 45.3%. Amongst the 19 pts receiving prior AB treatment, the median OS is 14.3m (95% CI: 5.5-14.4) with 1-year survival rate of 50.7%. For non-AB regimen, the median OS is 8.9m (95% CI 6.1-NR) with 1-year survival rate of 42.0%. Commonest G3-4 TRAE is thrombocytopenia (6%); hypertension (4%) and ALT elevation (4%).The median dose of cabo is 41.6mg per day (range: 24.3-60.0 mg). Conclusions: Post-ICI use of cabo is associated with a median PFS of 4.1m and median OS of 9.9m. AB-cabo sequence appears to have more favourable outcome than non-AB-cabo sequence. No new safety signals were observed. This clinical trial reports the first prospective post-ICI survival data on TKI. Clinical trial information: NCT04588051.