Supplementary Table 1.Subvisible particle contents of oligomeric samples used in the study measured by Light Obscuration
Fraction / Subvisible particles (part./ mL) / calculated mass(% of total)
≥2 µm / ≥5 µm / ≥10 µm / ≥25 µm
UV-stress-induced / 1173 / 153 / 13 / 0 / 0.06 %
process-related oligomers / 5169 / 889 / 108 / 3 / 0.35 %
pH-stress-induced oligomers / 316 / 73 / 8 / 0 / 0.02 %
Because the risk of re-forming of small amounts of subvisibleparticles in the stressed fractions could not be excluded completely, subvisible particle measurements by light obscuration were performed(Suppl. table 1). Although a certain number of subvisible particles have indeed been re-formed after the final filtration step, these numbers are very low even compared to the figures normally seen in unstressed protein therapeutics. In addition to the numbers of particles measured, we also calculated the approximate mass of the protein contained in these particles (according to a procedure reported by Barnard et al., J Pharm Sci 100, (2):492-503). As reported in the table below the mass of protein contained in subvisible particles in all cases is merely a small fraction of a percent (0.06 % in the case of light-stress oligomers which induced breakage of tolerance in our experiments).
Most importantly, the process-related-stress-induced oligomer sample (which in our experiments did not break immune tolerance) contained by far the largest amount of subvisible particles – 5 to 10 fold larger than the number of particles (depending on the size range) and approximately 6 fold larger calculated protein mass as compared to the corresponding values measured in the UV-stress oligomer sample, the latter being the only one that broke tolerance in our experiments. These results allow us to conclude that the small amounts of subvisible particles present in the samples used in our experiments cannot be the driver of the effects reported in our manuscript.