Background and importance There are several types of renal cell tumours, the most frequent being clear cell renal carcinoma (ccRCC) which represents 80% of malignant renal tumours in adults. Pembrolizumab, in combination with axitinib, is indicated for the firstline treatment of advanced RCC in adults.

Aim and objectives To describe and analyse the effectiveness and safety of pembrolizumab and axitinib in a tertiary hospital clinical practice.

Material and methods An observational retrospective study was conducted in all patients diagnosed with ccRCC and treated with pembrolizumab and axitinib from March 2019 to October 2020. All patients gave their informed consent. Data sources were the electronic medical records. Variables analysed were: sex, age, PDL-1, prior lines of treatments, IMDC risk and presence of metastases at the start therapy, duration of treatment and interruption causes, grade and type of toxicities and best TAC response.

Results 13 patients were included, 76.9% men, mean age 60.4±8.8 years. Among the four patients with tumour samples that were evaluated for PD-L1 expression, 75% had a score ≥1. Three patients were treated with at least one previous line. Previous treatments were: sunitinib (n=3) cabozantinib (n=1) or nivolumab (n=1). IMDC risk classification: 7.7% favourable, 53.8% intermediate and 38.5% poor risk. Presence of metastases: lung (7/13), bone (5/13), liver (3/13), ganglionar (2/13), cerebral (1/13) and unknown (2/13). All patients were treated with pembrolizumab 200 mg every 3 weeks and axitinib 5 mg twice daily until progression, unacceptable toxicity or death. Mean duration of treatment was 28.7 weeks. 46% are continuing with active treatment. Discontinuation causes included: death (n=3), adverse effects (n=3) and progression (n=1). Toxicities included: asthenia grade (G)1–3 (n=11), anorexia G1–2 (n=6), diarrhoea G1–4 (n=5), liver profile alterations G1–3 (n=3), hyperthyroidism G1–3 (n=3), abdominal pain G1–2 (n=3), palmar–plantar erythrodysesthesia G2–3 (n=2), pruritus G1 (n=1), dizziness and paraesthesia G1 (n=1), vomiting G1 (n=1), thrombopenia G2 (n=1) and arthralgias G1 (n=1). Best TAC responses obtained were: 50% stable disease, 25% partial response and 12.5% progressive disease. In five patients the response rate was not evaluated.

Conclusion and relevance Effectiveness in our patients resulted in a higher objective response rate than that in the KEYNOTE-426 trial. The combination treatment was well tolerated. To rationalise the use of novel medicines and optimise efficiency, measuring health results is crucial.

Conflict of interest No conflict of interest