Potential cytotoxicity, risk of triggering allergy and inflammation is a key issue decisive in use of nanomaterials in biomedical patents. In our study, we decided to check a potential effect of graphene on activation of the complement system as a result of contact with human serum and whether the parameters such as nanoparticle size, dose and state of aggregation have influence on immunogenicity of graphene. To that end, we incubated three types of graphene with human serum collected from healthy donors: A02 with nanoparticle size of 8 nm, A03 ‐12 nm i A04‐ 60 nm, at the concentration of 0.1 μg/mL and dose of 50 and 100 μl. We also incubated graphene in the form of plate deposited on the surface of silicon oxide (Gr‐SiO3). The structure of graphene was examined with a scanning electron microscope Phenom Word G2 PRO. The level of activation of the complement system was examined with the immunoenzymatic method determining the concentration of the C5a protein. That agent forms as a result of conversion of C5 into C5a and C5b, component common for three ways of activation of the complement system. The results show that graphene A02 and A03 with small size of nanoparticles induces splitting of C5 more strongly compared to A04. In the case of Gr‐SiO3 we observed a statistically insignificant drop in the value in comparison with the control sample and to A04 at the dose of 50 μl (p>0.05). In relation to the remaining samples, a statistically significant difference was present in all cases (p <0.05). In the case of graphene suspension, differences in activation of the complement system were statistically significant between the dose of 50 and 100 μl in all three examined types (p <0.05). As regards the size of nanoparticles, significant differences were also visible in all three types of graphene (p <0.05). A02 was equalled at the dose of 50 and 100 μl, it resulted in increased level of activation of the complement system by ~6.33% and ~15.08% (p <0.05) compared to the control sample. In the case of A03, we observed a statistically significant difference (p <0.05) only in the dose of 100 μl (~8,24%). Calculation of the concentration of C5a from the value of absorbance of standard curve showed that the control concentration of C5a in serum not incubated with graphene was about 7.31 ng/ml. Increasing that value by 15.08% equals the concentration of C5a being ~ 8.42 ng/ml, which falls within normal limits for C5a protein in humans (8.34 ± 2.05 ng/ml). The results show that even though graphene with small particle size caused increased activation of the complement system, the level of activation was within normal limits in healthy individuals. Our results show that graphene, both as nanoparticle suspension and graphene plate, does not activate the complement system in a significant manner. Additionally, we showed that graphene in the form of plate is the most compatible form as regards human serum. Our results emphasise that the parameters of graphene such as dose, size and state of aggregation can significantly influence its immunostimulating properties.Support or Funding InformationWUM grant: 2M2‐W1‐16This abstract is from the Experimental Biology 2018 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.