Reassessment of Historical Clinical Trials Supports the Effectiveness of Phage Therapy

Abstract

Phage therapy has become a hot topic in medical research due to the increasing prevalence of antibiotic-resistant bacteria strains. In the treatment of bacterial infections, bacteriophages have several advantages over antibiotics, including strain specificity, lack of serious side effects, and low development costs. However, scientists dismissed the clinical success of early clinical trials in the 1940s, slowing the adoption of this promising antibacterial application in Western countries. The current study used statistical methods commonly used in modern meta-analysis to reevaluate early 20th-century studies and compare them with clinical trials conducted in the last 20 years. Using a random effect model, the development of disease after treatment with or without phages was measured in odds ratios (OR) with 95% confidence intervals (CI). Based on the findings of 17 clinical trials conducted between 1921 and 1940, phage therapy was effective (OR = 0.21, 95% CI = 0.10 to 0.44, P value < 0.0001). The current study includes a topic review on modern clinical trials; four could be analyzed, indicating a noneffective therapy (OR = 2.84, 95% CI = 1.53 to 5.27, P value = 0.0009). The results suggest phage therapy was surprisingly less effective than standard treatments in resolving bacterial infections. However, the results were affected by the small sample set size. This work also contextualizes the development of phage therapy in the early 20th century and highlights the expansion of phage applications in the last few years. In conclusion, the current review shows phage therapy is no longer an underestimated tool in the treatment of bacterial infections.

Keywords: Felix d’Hérelle; bacteriophage therapy; clinical trials; effectiveness; historical evaluation; history of science; meta-analysis; phage therapy; systematic review.

FIG 1

Rising interest in phage therapy. Distribution of articles retrieved with the keyword “phage therapy” from MedLine during 1946 to 2021. The total number of publications steadily increased, reaching 686 articles per year in 2021 (red line), while clinical and randomized trials remained a small subset of the total, with a maximum of six publications per year in 2018 and 2020.

FIG 2

Phage therapy approaches. In the prêt-à-porter (“ready-to-use”) approach, environmental samples are collected, typically from sources such as ponds or sewage purification plants. Phages are purified from the mixture of microbes present in the samples and stored in biological banks. Several isolates are mixed in a single preparation that can be administered to the patients, hoping that at least one of the isolates will have the pathogenic bacterium as a host. In the sur-mesure approach, the starting material is a clinical sample, which is analyzed to characterize the pathogenic bacterium. Phages against the bacterium are selected either anew from environmental specimens or from banked isolates. The phagial preparation is administered to the patient, knowing that the phages will recognize the pathogenic bacterium. This targeted method has the disadvantage of being time-consuming, making it unsuitable for emergency medical care.

FIG 3

Outcomes of phage therapy. In passive therapy (left panel) there is no phagial replication and the pathogenic host is expected to be eradicated in a single event due a high ratio of phage particles to host cells of the intervention. Passive therapy is established when administering phages above a critical level V_C. A dose of phages at a concentration v_φ, given at time t_φ, will result in the decay of a certain proportion of viruses at a rate λ whereas the infectious particles will infect the bacteria at a rate δ. On the other hand, bacteria multiply at a rate μ and phages will establish a sustained infectious cycle only if the bacteria reach a minimal density at time T_P. The therapy’s success is essentially dependent on administering enough virions to infect 100% of the bacteria, because if even one bacterium escapes infection, it will multiply again. There is a critical time T_C beyond which the phages will not be able to infect the bacteria at a sufficient rate to avoid such an escape. If v_φ is below the inundation threshold V_I, however, the therapy will fail regardless of the administration time because there will be not enough virions to infect all bacteria. In active therapy (right panel) there is an increase in the number of phages, which expands at a rate β after infecting a bacterium. Active therapy is established when v_φ is below V_C, providing an environment with a low ratio of phage particles to host cells, but needs to account for the decay of virions and the replication of the bacteria. Thus, if v_φ is below the critical threshold V_F, not all bacteria will be infected. In addition, the lower t_φ is from T_P, the more phages will be lost to decay, and the fewer phages present at the start of the sustainable infectious cycle, the less likely it is that phages will catch up with bacterial expansion. As in passive therapy, if t_φ is beyond T_F the bacterial infection will not be cleared. The success of active therapy is also dependent on the fact that it occurs prior to the establishment of the immune response at time T_H, which can target both bacteria and phages.

FIG 4

Meta-analysis of the studies reported by Krüger and Scribner in their review published in 1941. (A) Random forest plot of the studies reporting the odds ratio (OR) of the intervention (experimental) over the control groups. An evident outlier is the Morison’s prophylactic study carried out between 1930 and 1935 in India. Removing this study increases the strength of the study, providing a random model with OR = 0.27 (95% CI = 0.16 to 0.45, P value < 0.0001) and heterogeneity characterized by I² = 86.9%. (B) Funnel plot of the studies shown in panel A, showing the trend line of the model.

FIG 5

Meta-analysis of the contemporary clinical studies employing phage therapy. (A) Random forest plot of the studies reporting the odds ratio (OR) of the intervention (experimental) over the control groups. The model showed an odds ratio above unity but not statistically significant (OR = 1.65, 95% CI = 0.76 to 3.58, P value 0.219) with heterogeneity characterized by I² = 71.7%. The works by Międzybrodzki et al. (2012) and Wright et al. (2009) (114, 117) might be considered outliers. Their removal increased the strength of the study: OR = 2.84 (95% CI = 1.53 to 5.26, P value = 0.0009) but raised heterogeneity (I² = 81.3%, 95% CI = 0% to 84.7%). (B) Funnel plot of the studies shown in panel A, showing the trend line of the model.