Concerns about using phages as antibacterial agents can be distinguished into four categories: (1) phage selection, (2) phage host-range limitations, (3) the “uniqueness” of phages as pharmaceuticals, and (4) unfamiliarity with phages. See references 4 and 5 for additional discussion.

Not all phages make for good therapeutics.

Good therapeutic phages should have a high potential to reach and then kill bacteria in combination with a low potential to otherwise negatively modify the environments to which they are applied. These characteristics can be reasonably assured so long as phages are obligately lytic, stable under typical storage conditions and temperatures, subject to appropriate efficacy and safety studies, and, ideally, fully sequenced to confirm the absence of undesirable genes such as toxins.10,18 Note that a phage that is “obligately lytic” we define as not temperate and released from infected cells via lysis, that is, unable to display lysogeny and not released chronically. The use of temperate phages as therapeutics is problematic due to a combination of display of superinfection immunity,13 which converts phage-sensitive bacteria into insensitive ones, and the encoding of bacterial virulence factors, including bacterial toxins.8–10,18,27,29

In addition to avoiding temperate or toxin-carrying phages, the aim of phage characterization is to exclude as therapeutics those phages that display poor killing potential against target bacteria. Such low “virulence” can be due to poor adsorption properties, low potential to evade bacterial defenses, or poor replication characteristics.3 Also less desirable for therapeutics are those phages that display poor pharmacokinetics, that is, poor absorption, distribution, and survival in situ.3 Ideally phages should also display a low potential to transfer bacterial genes between bacteria (transduction).10,18

Phage characterization additionally can include virion morphology (via electron microscopy), protein profiles, or genotype characterization other than via full-genome sequencing (e.g., PFGE profiles of restriction digested genomes), etc.,18 though the costs associated with exhaustive phage characterization prior to phage use can be prohibitive. The general aim, therefore, should be to identify those phages that display good primary pharmacodynamics (that is, antibacterial virulence), minimal secondary pharmacodynamics (low potential to do harm to patients), and good pharmacokinetics (an ability to reach target bacteria in situ).3 Phages that do not adequately meet these criteria should in most circumstances not be employed as therapeutics. Minimally this should entail avoiding temperate phages and, ideally, full genome sequencing should be used to rule out virulence-factor carriage.