As usual, some of the most interesting presentations at ECCMID were in the late-breakers “clinical trials” session. Four of 5 presentations were on treatment or prevention of S. aureus infection, the other one on oral treatment in patients with refractory fungal disease. With all respect to fungi, the meat was in the aureus, with nothing less than a Shakespearian tragedy.
The foreplay was a U.S. based study evaluating a truly new concept: a specific anti-S. aureus lysin (named exebacase), derived from a bacteriophage, administered intravenously as adjunctive therapy in patients with S. aureus bacteremia (SAB), including endocarditis. This was a phase 2 double-blind placebo-controlled RCT with 121 patients (43 MRSA), of which 73 received the lysin, for on average 33 days. The primary endpoint “improvement/resolution of signs/symptoms at day 14” was met in 70.4% (50/71) and 60.0% (27/45) of the patients with and without the lysin (116 patients in mITT analysis, p=0.314). In patients with MRSA infection the primary endpoints was met in 74.1% (20/27) and 31.3 (5/16) of the patients with and without the lysin (p=0.01). No safety issues reported. I think this is the first large-scale RCT with a lysin derived from bacteriophages, with a promising result. Very welcome in the evidence-free zone of bacteriophages, see. Can’t wait for the phase 3 study.
Similarly impressive was the CAMERA2 study, executed in multiple countries (between Israel and Australia) and addressing the question whether addition of a beta-lactam antibiotic (7 days) to vancomycin (99%) or daptomycin (4%) improves outcome in patients with MRSA bacteremia. After studying 352 patients the trial was stopped by the DSMB because of futility: no evidence of improved outcome (90d mortality, persistent bacteremia at d5, microbiological relapse/failutre), but a 30% incidence of acute kidney failure among patients receiving combi treatment (compared to 9% in standard care). Almost all patients that develop kidney failure had received flucloxacillin (6×2 gr/day). As only few patients had received cefazolin not 100% sure that all damage was done by flucloxacillin. A very relevant research question, clearly answered, book closed.
And then the tragedy: two knights beckoned to the hand of the young and handsome princess. The knights are monoclonal antibodies and the princess is the prevented episode of S. aureus HAP/VAP. Knight 1 has 1 antibody (named suvratoxumab), that neutralizes alpha-toxin, but knight 2 (named ASN100) has 2 antibodies that neutralize 6 cytokines, including alpha-toxin. Injected in lab animals they both did very well, and some might even say that Knight 2 did a little better. Thus they needed a trial to demonstrate their capacity to prevent S. aureus HAP/VAP in ICU patients carrying S. aureus in the respiratory tract at the time of ICU-admission. They both wanted a control group with a 25% incidence of S. aureus HAP/VAP (to demonstrate reduction). Knight 2 decided to rely on conventional culture results (with classical turn-around times) to identify eligible subjects (S. aureus carriers), where Knight 1 used PCR-based testing providing an answer within hours. Knight 2 also decided to allow study patients to be treated with anti-staphylococcal antibiotics, which Knight 1 did not allow. After several years of enrolment in double-blind placebo-controlled RCT Knight 2 had a control group with an incidences of S. aureus HAP/VAP of 4%-9% (depending on definition). With so little to prevent Knight 2 failed. For Knight 1 the control group had an incidence of 26% and a very good looking reduction of S. aureus HAP/VAP of 31.9% (and even more in patients <65 years of age; 47.4% reduction).
What happened to Knight 2? The company merged with a company focussing on treatment of rare diseases of the immune system and rare cancers. I‘m afraid we may never hear from ASN100 again. Knight 1 – suvratoxumab – will now go into phase 3 (I hope) and might become the first anti-S. aureus monoclonal! A stunning example of how important study design and execution is. The “may-be-more-promising” combination of antibodies was bound to fail because of a control group.
I do report a conflict of interest as I was involved in the SAATELITE study, which was executed within the COMBACTE-NET consortium.