Safety and efficacy of S-1-based treatment in colorectal cancer patients who switched to S-1 after development of cardiotoxicity on fluoropyrimidine-based adjuvant or neoadjuvant treatment

Background Fluoropyrimidines (FP) are the cornerstone of adjuvant therapy in colorectal cancer (CRC). Treatment options are limited for the 4%-6% of patients who experience cardiotoxicity on FP treatment. The CardioSwitch study in patients with solid tumours showed that switching to S-1-based treatment is feasible, as an alternative to permanent FP discontinuation (1). Purpose This analysis aimed to evaluate safety and efficacy data for patients with localized CRC who were switched to S-1 after experiencing cardiotoxicity during FP treatment in the adjuvant or neoadjuvant setting. Methods This retrospective study included patients with stage II-III CRC who had a history of cardiotoxicity during FP treatment in the perioperative setting. Primary endpoints were recurrence of cardiotoxicity after switch to S-1-based therapy. Secondary endpoints were safety and survival, and these data were compared with data for all 413 adjuvant patients treated between 2010 and 2022 in the Uppsala region, Sweden. Results Eighty-nine patients were included from eight centres in three countries. Median age was 66 (range 28-82) years, 51% were female, and 92% had ECOG PS 0-1. The primary tumour location was right colon in 26%, left colon in 24%, and rectum in 54%. Baseline cardiovascular comorbidities were present in 45%. Initial treatment that caused cardiotoxicity was capecitabine-based in 92%, infusional 5-FU in 6%, and bolus 5-FU in 2%. Of these, oxaliplatin was added in 39%. Initial cardiotoxicity was grade 3-4 in 42% and occurred in the first 2 cycles in 93%. Cardiotoxicity occurred during neoadjuvant CRT or chemotherapy in 22% and during adjuvant therapy in 78% (Figure 1). Switch to S-1-based treatment occurred during neoadjuvant therapy in 15% and during adjuvant therapy in 85%. S-1 monotherapy was administered to 46%, 42% received S-1 plus oxaliplatin (SOX), and 12% received S-1 CRT. Four patients (4%) experienced recurrent cardiotoxicity on S-1-based treatment, the worst level was grade 1 in two and grade 2 in two. Adjuvant treatment completion rate was 96%: two patients discontinued due to recurrent tachycardia, one due to diarrhoea, and one due to ileitis/tuberculosis. Grade 2-4 non-haematologic and/or 3-4 haematologic toxicity occurred in 34%. Baseline characteristics of the patients with recurrence on S-1 included cardiovascular comorbidity in two, and all had ECOG PS 0-1, rectal cancer, and received adjuvant treatment. The five-year DFS rate was 69% and 5-year OS rate was 83%, with median follow-up 69 months, in line with population-based data of 73% and 76%, respectively, from the Uppsala cohort (Figure 2). Conclusions S-1-based neoadjuvant CRT and/or adjuvant treatment is effective, safe and feasible after cardiotoxicity on other FP-based treatment and enables patients to continue curative treatment as indicated.