The use of phage therapeutic products has gained popularity in Belgium these last years, with treatments provided to more than 100 patients, against more than 10 different bacterial species. Therapeutic phage products are produced  in a bacterial host strain, hence pyrogen contamination (e.g. cell wall debris) is a major safety concern as these pro-inflammatory agents can cause unwanted clinical reactions with potential detrimental outcome. The most common and potent pyrogens are endotoxins, components of the cell wall of gram-negative bacteria which are released upon bacterial lysis.

In our routine phage quality control (QC) workflow, endotoxins are quantified using the Limulus amoebocyte lysate (LAL) assay and the animal-friendly alternative, the recombinant factor C (rFC) test. However, other (non-endotoxin) pyrogenic substances including compounds of the gram-positive bacterial cell wall, flagellin, etc. are not detected by those tests. As phages are being produced in a growing variety of bacterial host strains, including gram-positive bacteria, an expansion of the QC workflow is pivotal to limit the risk of non-endotoxin contamination. These pyrogens can be detected by the monocyte activation test (MAT), which employs human blood cells to quantify pyrogenicity in vitro. In this work, we discuss and compare the applicability of the LAL, rFC and MAT tests for pyrogen testing of clinical grade phage products based on the analysis of various productions manufactured by the Queen Astrid Military Hospital in Brussels. With a few exceptions, the results on different phages demonstrate that in general pyrogenic contamination is low for most samples.