CRAFITY score as a predictive marker for refractoriness to atezolizumab plus bevacizumab therapy in hepatocellular carcinoma: a multicenter retrospective study

Abstract

Background: Although atezolizumab plus bevacizumab (Atezo/Bev) therapy has been used as the preferred first-line treatment for advanced hepatocellular carcinoma (HCC), up to 26% of patients do not achieve disease control, suggesting alternative treatments might be more beneficial for such patients. We investigated key predictors for refractoriness to Atezo/Bev therapy, particularly in the first-line setting.

Methods: We retrospectively analyzed 302 patients with HCC who received Atezo/Bev therapy between October 2020 and September 2022 across nine hospitals in Japan. Refractoriness was defined as best overall response (BOR) of progressive disease or stable disease and a progression-free survival (PFS) of < 180 days (RECIST v1.1). Clinical benefit was defined as BOR of partial/complete response or stable disease with PFS of ≥ 180 days. Baseline characteristics and potential predictors, identified through literature review, were compared between these groups. Stratifications of overall survival (OS), and PFS were also assessed.

Results: Refractoriness was observed in 126 (41.7%) patients, while 154 (51.0%) achieved clinical benefit. Due to a significant association between the treatment line and refractory rate, the subsequent analysis focused on the first-line cohort (n = 214; 72 [33.6%] patients showed refractoriness). Among 13 potential predictors, the CRP and AFP in immunotherapy (CRAFITY) score had the best predictive performance, with refractory rates of 24.6%, 44.6%, and 57.9% in CRAFITY-0, 1, and 2 patients, respectively (p < 0.001). OS and PFS were also well-stratified by this scoring system.

Conclusions: Approximately one-third of patients were refractory to first-line Atezo/Bev therapy. The CRAFITY score demonstrated superior performance in predicting refractoriness.

Keywords: Alpha-fetoprotein; C-reactive protein; Immune checkpoint inhibitor; Liver cancer; Precision medicine.

Fig. 1

Survival outcomes stratified by best overall survival (BOR) (overall cohort). A Progression-free survival (PFS). B Overall survival (OS). In both figures, patients with a stable disease (SD) and PFS ≥ 180 days had similar survival outcomes to those with a partial response (PR). Meanwhile, patients with a SD and PFS < 180 days had similar outcomes to those with a progressive disease (PD). CR complete response, NE not evaluable

Fig. 2

Patient allocation (1st-line cohort). Patients were classified into the refractory or clinical benefit group based on their best overall response (BOR) and progression-free survival (PFS). Patients with a progressive disease (PD) were categorized as the PD group. Some patients were not included in either group. Atezo/Bev atezolizumab plus bevacizumab, SD stable disease, PR partial response, CR complete response, NE not evaluable

Fig. 3

Associations between the CRAFITY score and treatment outcomes with first-line atezolizumab plus bevacizumab therapy. A Best overall response (BOR) in patients stratified by CRAFITY score. Red and blue columns represent the refractory and clinical benefit groups, respectively. CR complete response, PR partial response, SD stable disease, PFS progression-free survival, NE not evaluable, PD progressive disease. B Overall survival (OS) stratified with the CRAFITY score. C PFS stratified with the CRAFITY score

Fig. 4

Associations between the CRAFITY score and treatment outcomes with second- or later line atezolizumab plus bevacizumab therapy. A Best overall response (BOR) in patients stratified by the CRAFITY score. Red and blue columns represent the refractory and clinical benefit groups, respectively. CR complete response, PR partial response, SD stable disease, PFS progression-free survival, NE not evaluable, PD progressive disease. B Overall survival (OS) stratified with the CRAFITY score. C PFS stratified with the CRAFITY score