Background and importance Anti-PD1 immunotherapies such as nivolumab and pembrolizumab represent a revolution for their efficacy in the care of oncology pathology.

Aim and objectives Shown to be safe and well tolerated in clinical trials, our objective was to define and compare the tolerance of these two treatments in a real-life setting.

Material and methods We performed a retrospective study including all the patients treated with nivolumab or pembrolizumab from January 2015 to February 2021.

For each patient, data on undesirable effects (UE) were collected from the reports of the oncology 1-day hospitalisation, gathered in the patient’s computerised record.

A Fisher test was conducted for the statistical analysis.

Results The study cohort included 148 patients on nivolumab and 131 patients on pembrolizumab.

During the study period, 192 UE occurred with pembrolizumab and 331 UE with nivolumab, respectively; 28% and 15% of patients did not exhibit an UE (significant difference; p=0.007).

The most frequent UE with pembrolizumab were arthralgia (18%), dyspnea (18%), alteration of the general state (15%), anaemia and neutropenia (15%) and immune or infectious pneumopathy (15%).

The most frequent UE with nivolumab were pain (29%), severe asthenia (27%), alteration of the general state (21%), immune or infectious pneumopathy (18%), anorexia (15%), dermal toxicity (13%) and immune-mediated diarrhoea (10%).

A statistical difference was observed for haematologic toxicity (p=0.0066) with more UE for pembrolizumab. Conversely, nivolumab appeared to cause more asthenia (p=0.001), coughing spells (p=0.01) and anorexia (p<10^–4) than pembrolizumab.

The grade 4 adverse effects (mostly pulmonary or alteration of the general state) led to cessation of treatment for 33 patients on pembrolizumab and 44 on nivolumab (non-significant difference; p=0.2).

Conclusion and relevance Anti-PD1 has proved to be a huge benefit in term of efficacy and tolerance compared to conventional chemotherapies. However, as shown in our real-life study, adverse effects which can be major still occurred. Their harmfulness seems to be underestimated and requires awareness to be promoted among prescribers to improve patient care.

Conflict of interest No conflict of interest