Background Regorafenib has been approved for the treatment of patients with previously treated metastatic colorectal cancer (mCRC), and may be considered as a treatment for selected patients.

Purpose To evaluate the efficacy and safety of regorafenib treatment in patients with mCRC.

Material and methods Retrospective observational study of mCRC patients treated with regorafenib (September 2015 to 2017). Collected variables: age, sex, ECOG, KRAS gene status, treatment line, number of cycles and dose reduction. Efficacy endpoints were progression-free survival (PFS) and overall survival (OS) obtained by the Kaplan–Meier method. Adverse effects (AE) were collected for safety profile assessment. Descriptive statistical analysis was performed using the SPSS^® Statistics program V22.0.

Results Thirteen patients (7 males and 6 females) were included. The median age was 57 years (41–77). Initial ECOG was: 0 in 38.46%, 1 in 38.46% and 2 in 23% of patients. The KRAS gene was mutated in 50.8% of patients, wild-type in 30.8% and 3.8% undetermined. The treatment was regorafenib 160 mg once daily for 21 days, every 28 days. In six patients (46.2%) it was prescribed as a third-line treatment and in seven patients (53.8%) as the fourthth line or later therapy. Dose reduction was performed in 30.8% of patients. The mean number of cycles was 2.75±1.22 cycles. The median PFS was 3 months (95% CI: 2.52 to 3.47) and the median OS was 8.3 months (95% CI: 1.07 to 15.51). All patients had AE of some grade and 33.3% of grade 3–4. The most common AE were: hand-foot skin reaction (HFSR) (n=7), hypertension (n=5) and nausea (n=4) and grade 3–4: HFSR (n=1), hypertension (n=1), neutropaenia (n=1) and mucositis (n=1). The causes of treatment discontinuation were: progression (n=9), deterioration of general health (n=3) and toxicity (n=1). At the end of the study, none of the patients continued treatment.

Conclusion The SLP obtained in our study is greater than that described in the pivotal trial CORRECT (3 versus 1.9 months). This was possibly due to the longer time it took to determine the radiological response. The SG was also higher (8.3 versus 6.4 months), taking into account the limitation of the sample size. The AE described were similar to those published in the literature.

References and/or Acknowledgements 1. Lancet. 2013;381:303–12.

No conflict of interest