Final Report Summary - AIDA (Preserving old antibiotics for the future : assessment of clinical efficacy by a pharmacokinetic/pharmacodynamic approach to optimize effectiveness and reduce resistance for off-patent antibiotics)
Executive Summary:
In an era of increasing emergence of drug resistance (EDR) and lack of new antibiotics, old off- patent antibiotics are increasingly being prescribed to patients. However, many of these were developed in an age before the advent of a structured process for drug assessment and approval, and the establishment of clinical efficacy and effectiveness in randomized controlled trials in particular. The AIDA project aimed to answer the question of clinical effectiveness and optimal dosing of off-patent antibiotics for infections caused by multiple drug resistant (MDR) bacteria in three randomized controlled clinical trials. In a multidisciplinary approach the exposure response relationships for each antibiotic was elucidated by including pharmacokinetic (PK), pharmacodynamic (PD) and microbiological studies, including emergence of drug resistance.
The studies in the AIDA project commenced in 2012. The design of each of the three trials was fine-tuned and subsequently submitted to Medical Ethical Committees. After approval, inclusion in each of the three trials was completed in 2017.
In the first trial the efficacy of colistin alone was compared to colistin plus meropenem for severe infections caused by carbapenem-resistant gram-negative bacteria in an open-label randomized controlled superiority clinical trial to answer the questions whether combination therapy is useful if the micro-organism is not susceptible to the second drug. In total, 406 patients were recruited.
In the second trial fosfomycin 3 g (single dose) was compared with nitrofurantoin for the treatment of uncomplicated lower urinary tract infection in women aged ≥ 18 years at high risk of antibiotic-resistant pathogens in a multi-center randomized superiority trial. Anecdotal data had indicated a low response rate of a single dose of oral fosfomycin. Supported by pharmacokinetic studies and computer simulations these results may trigger a change in usage and more studies to define the best dosing regimen of oral fosfomycin. At completion, 513 patients were recruited.
In the third trial, oral treatment with minocycline plus rifampicin was compared with oral treatment with linezolid for complicated skin and soft tissue infections (cSSTI) due to MRSA in a controlled randomized multi-center non-inferiority clinical trial. At completion, 126 patients were randomized.
Each of these three clinical studies was linked to a variety of laboratory and animal studies supporting and understanding clinical trial results. The analysis of the three trials has been finalized and the main results are currently being considered for publication in major journals. Pharmacokinetic studies indicated significant pharmacokinetic variation between subjects that will need to be taken into consideration. Pharmacodynamic analyses clearly indicate that some of the trial drugs are likely under-dosed in current clinical practice. For some agents such as fosfomycin, the supporting pharmacodynamic studies indicated that resistance emerges relatively fast.
The collaboration of researchers from different disciplines and the successful completion of the AIDA project highlights the capacity of academic groups to “re-develop” old antibiotics and integrate complex clinical studies with high quality non-clinical research. Based on the project results, dosage regimens will need to be updated and the use and indications of old antibiotics revisited. The results have already led to ongoing discussion with regulatory bodies, such as the EMA, among others.
Old antibiotics are off-patent and no formal responsibility is defined for updating the knowledge of these valuable drugs. AIDA produced vital integrated non-clinical and clinical data to strengthen our evidence-based confidence for using these old drugs in daily clinical practice.
Updates and more details of the results of the studies can be found at the website of AIDA, www.aida-project.eu.
Project Context and Objectives:
1. Introduction and overall description of the AIDA project
In an era of increasing emergence of drug resistance and lack of new antibiotics, old off-patent antibiotics are increasingly being prescribed to patients. However, many of these were developed in an age before the advent of a structured process for drug assessment and approval, and the establishment of clinical efficacy and effectiveness in randomized controlled trials in particular. Moreover, dosing regimens of these antimicrobial agents were developed more on a trial and error basis and information to optimize dosing regimens using exposure-response relationships is not readily available, if at all. It is unclear whether the current dosing regimens used are optimal or even efficacious. Even if comparative trials were performed in the past to determine whether one antibiotic was non-inferior or superior to another, oftentimes the dosing regimens have changed over time. This change is a significant problem in particular since old off-patent antibiotics are increasingly being prescribed to patients now that escalating resistance poses a specific challenge in antimicrobial treatment. On many occasions, microorganisms are now fully resistant to commonly used drugs. Examples include ESBL (Extended-Spectrum-Beta-Lactamase) producers, and recently, KPC (Klebsiella pneumoniae carbapenemases) or NDM-1 (New Delhi Metallo-beta-lactamase) producers. Old drugs, such as colistin and fosfomycin, are then used without any certainty that the correct dosing regimens are being applied. In addition, comparisons between treatment modalities to determine the best choice of treatment in randomized controlled trials (RCT) have not been established for many of these older antibiotics. A re-evaluation of these drugs is therefore urgently needed and includes establishing PK/PD relationships and the optimal dosing regimens to use.
In the AIDA project we aimed to answer the question of clinical effectiveness and optimal dosing of off-patent antibiotics for infections caused by multiple drug resistant (MDR) bacteria in three randomized controlled clinical trials in three work packages (WP1,2,3). In a multidisciplinary approach the exposure response relationships for each antibiotic was to be elucidated by including pharmacokinetic (PK), pharmacodynamic (PD) and microbiological studies, including emergence of drug resistance (WP4). The project addressed the optimization of treatment of infections caused by MDR pathogens that impose a major burden of disease in Europe and the rest of the world by selecting 5 off-patent-antibiotics - colistin, fosfomycin, nitrofurantoin, minocyclin and rifampicin that were increasingly being used without clear evidence with respect to their effectiveness, duration of therapy and issues of EDR. Common to all the clinical trials proposed was the selection of a clinically-relevant patient population; randomized controlled trial design to minimise bias; assessment of clinically-relevant outcomes for the individual patient, PK/PD assessment and correlation between PK/PD indices and clinical outcomes, including efficacy, toxicity and emergence of resistance. In summary, the main goal of the project AIDA was to gain preclinical and clinical data of 5 off-patent antimicrobials to allow optimizing dosing regimens. To that end, 3 clinical trials (WP1-3) were to be performed and preclinical studies (WP4) of each of the antibiotics were to supplement the evidence for optimal dosing and use. An important objective was the dissemination of results and was handled in a separate WP (WP5). Below, the specific objectives of each WP are briefly described.
2. Specific objectives for each work package
WP1 – Multicenter open-label RCT to compare colistin vs. alone colistin plus meropenem
The primary objective of this RCT was to demonstrate superiority of colistin in combination with meropenem vs. colistin alone in the treatment of bloodstream infections and ventilator-associated pneumonia (VAP) caused by carbapenem-resistant Acinetobacter sp., Pseudomonas aeruginosa or Enterobacteriaceae. Secundary objectives include: (differences in) change in functional capacity from baseline; superinfections, defined as clinically or microbiologically-documented infections by CDC criteria within 28 days; 28-day all-cause mortality; time to clinical success; microbiological failure, defined as the isolation of a bacterium at day 7 or more that is phenotypically identicalto the first isolate; colonization by carbapenem and/ or colistin-resistant Gram-negative bacteria; Clostridium-difficile-associated diarrhea; renal failure using the RIFLE criteria; an analysis of adverse advents, including, seizures, critical illness and other neuropathy
WP2 – Multicenter RCT to compare fosfomycin vs. nitrofurantoin for treatment of uncomplicated Urinary Tract Infections (UTI)
The primary objective of this RCT was to demonstrate the superiority of 5 days nitrofurantoin over once 3g fosfomycin for the treatment of lower, uncomplicated UTI in women at high risk of antibiotic-resistant uropathogens. The secondary objectives were to evaluate clinical and bacteriological cure at day 7; duration of symptoms; loss of working days; hospital admission during 28 days; incidence of pyelonephritis or urosepsis during this period and development of adverse events and emergence of antibiotic resistance.
WP3 - Multicentre RCT to compare minocycline plus rifampicin to linezolid against MRSA in cSSSI
The primary objective of this RCT was to demonstrate non inferiority between patients with acute complicated skin and soft tissue (cSSTI) infection due to MRSA treated with minocycline plus rifampicin compared to those treated with linezolid in terms of efficacy and safety.
WP4 – Preclinical studies and integration of PK/PD analyses of trial results : redevelopment of old antibioticsfrom a PK/PD point of view
The primary objective of this WP was to provide, within limits, a preclinical development package for each of the five off-patent antibiotics. This included preclinical pharmacodynamic studies in vitro and if necessary in animals to obtain the necessary pharmacodynamic targets for each drug, pharmacokinetic studies in patients or, if not feasible, in volunteers to estimate the exposure and the variation there-of, and building population pharmacokinetic models to allow estimates of exposures in the individual patient. The final goal was to integrate the PK/PD relationships of the clinical trials with those of the preclinical data, EDR and thereby provide a rational basis for improved treatment indications of use and also to suggest clinical breakpoints and epidemiological cut-off values.
WP5 - Dissemination of results, communication to the professionals and the public
The primary objective of this work package was to ensure that the results of the studies were communicated to the public and relevant bodies in a timely manner. In addition, general communications were to increase the public awareness of the problem of emergence of resistance and the lack of knowledge of older drugs that were increasingly being used.
3. Overreaching objectives of the AIDA project
Apart from the specific objectives in each WP, the AIDA project entailed several overreaching objectives. These included addressing key scientific and clinical questions regarding the best treatment options for important disease entities and enabling refinement of these options by additional PK/PD research. Importantly the trials, in particular those in WP1 and WP2, should set a model for clinical trial testing of new antibiotics. For instance, justified criticism has been raised on industry-conducted trials assessing new antibiotics in SSTI and abdominal infections. Indeed, these trials are often times designed and destined to show equivalence, but have very low applicability to the treatment of patients who are
subsequently treated with these antibiotics. Patients affected by MDR Gram-negative bacteria in hospitals are severely-ill and have multiple co-morbid conditions. In WP1 this patient population is specifically addressed. The trial in WP1 was therefore to be designed and conducted by clinicians. If successful, this trial design could serve as a model for regulatory bodies to be considered in updated guidance documents for conducting formal Phase 3 testing of new antibiotics.
Project Results:
1. Introduction
In the AIDA project we aimed to answer the question of clinical effectiveness and optimal dosing of five off-patent antibiotics for infections caused by multiple drug resistant (MDR) bacteria in three randomized controlled clinical trials in three work packages (WP1,2,3). In a multidisciplinary approach the exposure response relationships for each antibiotic was elucidated by including pharmacokinetic (PK), pharmacodynamic (PD) and microbiological studies, including emergence of drug resistance (WP4).
Below, the results of the AIDA project are presented. The overall results of the clinical trials are presented first; In the section thereafter these are put into context together with the main outcomes of the PK/PD studies. It should be noted that many of the outcome data are still confidential and will be published in scientific journals in the year(s) to come.
2. Design and outcomes of the clinical trials
During the first period of the project, the initial study proposals of each the three studies were fine-tuned. The proposals were subsequently submitted to the relevant Medical Ethical Committees and after approval the trials were initiated. The description of results follows this timeline for each trial.
2.1 WP1 – Multicenter open-label RCT to compare colistin vs. alone colistin plus meropenem
2.1.1 The final study protocol
After multiple discussions with AIDA partners and external advisors, we proceeded to a randomized controlled trial comparing optimally-dosed colistin, based on PK considerations vs. colistin combined with high-dose extended-infusion meropenem. The final study protocol was submitted and published in the BMJ open (Dickstein et al; 2016). The summary of the final study reads as follows:
INTRODUCTION: The emergence of antibiotic-resistant bacteria has driven renewed interest in older antibacterials, including colistin. Previous studies have shown that colistin is less effective and more toxic than modern antibiotics. In vitro synergy studies and clinical observational studies suggest a benefit of combining colistin with a carbapenem. A randomised controlled study is necessary for clarification.
METHODS AND ANALYSIS: This is a multicentre, investigator-initiated, open-label, randomised controlled superiority 1:1 study comparing colistin monotherapy with colistin-meropenem combination therapy for infections caused by carbapenem-resistant Gram-negative bacteria. The study is being conducted in 6 centres in 3 countries (Italy, Greece and Israel). We include patients with hospital-associated and ventilator-associated pneumonia, bloodstream infections and urosepsis. The primary outcome is treatment success at day 14, defined as survival, haemodynamic stability, stable or improved respiratory status for patients with pneumonia, microbiological cure for patients with bacteraemia and stability or improvement of the Sequential Organ Failure Assessment (SOFA) score. Secondary outcomes include 14-day and 28-day mortality as well as other clinical end points and safety outcomes. A sample size of 360 patients was calculated on the basis of an absolute improvement in clinical success of 15% with combination therapy. Outcomes will be assessed by intention to treat. Serum colistin samples are obtained from all patients to obtain population pharmacokinetic models. Microbiological sampling includes weekly surveillance samples with analysis of resistance mechanisms and synergy. An observational trial is evaluating patients who met eligibility requirements but were not randomised in order to assess generalisability of findings.
ETHICS AND DISSEMINATION: The study was approved by ethics committees at each centre and informed consent will be obtained for all patients. The trial is being performed under the auspices of an independent data and safety monitoring committee and is included in a broad dissemination strategy regarding revival of old antibiotics. TRIAL REGISTRATION NUMBER: NCT01732250 and 2012-004819-31
A full description of inclusion and exclusion criteria and other details can further be found in this paper.
2.1.2 Outcome of the clinical trial.
The trial was successfully completed in six medical centers in Israel, Greece and Italy. In total 406 patients with severe infections caused by the target carbapenem-resistant Gram-negative bacteria were recruited. The trial was powered to show superiority of the combination, defined as an absolute 15% advantage in clinical failure (a composite of death, clinical instability or bacteremia persistence at 14 days) and its results refuted within the 95% confidence intervals an advantage to combination therapy. Clinical failure at day 14, all-cause mortality at day 14 and 28 and microbiological failure were all very similar between the groups, with risk ratios ranging between 0.93 and 1.1 for these outcomes, all non-significant. Combination therapy did not decrease the probability of acquiring colistin-resistant bacteria. Adverse events rates were all similar, but for renal failure which was lower with combination therapy. The manuscript has been accepted for publication in Lancet Infect Dis. Further analyses using the colistin level and microbiological samples obtained in the trial are ongoing. These analyses will allow further precision in the assessment of monotherapy vs. combination therapy effects.
Thus, to answer the objective, we refuted the clinical superiority of colistin-meropenem combination therapy in the treatment of carbapenem-resistant Gram-negative bacteria. The trial addressed the patient population treated in clinical practice as reflected by clinical failure rates of 75.9% (308/406) at day 14 and 28-day mortality rates of 180/406 (44.3%), very similar to observational cohorts of patients with carbapenem-resistant Gram-negative bacteria. It is the first randomized controlled trial and its results disagree with the data available from observational studies. We summarized the data available from observational studies, and mainly their limitations leading to very low quality of evidence (Zusman et al. J Antimicrob Chemother 2017). Our trial provides a strong signal that treatment decisions for multidrug-resistant bacteria must be based on a randomized controlled trial.
Some other observations can be made. The objective was to prove (or disprove) the superiority of combination therapy among the assemblage of the three major carbapenem-resistant Gram-negative bacteria - Acinetobacter sp., Pseudomonas aeruginosa and Enterobacteriaceae. The distribution among these bacteria was dictated by the epidemiological trends in the trial centers. The trial largely included patients with Acinetobacter baumannii infections (312/406, 76.8%) and thus provides an answer for this pathogen. The trial remained unpowered to address the question among patients with carbapenem-resistant enterobactericeae and P. aeruginosa. We are collaborating with Prof. Keith Kaye, PI of a parallel NIH-funded trial, designed very similarly and collecting the same outcomes. The compilation of the two trials will hopefully advance the understanding regarding carbapenem-resistant enterobacteriaceae.
2.2 WP2 – Multicenter RCT to compare fosfomycin vs. nitrofurantoin for treatment of uncomplicated Urinary Tract Infections (UTI)
2.2.1 The final study protocol
A summary of the final study protocol reads as follows:
INTRODUCTION: Cystitis and lower UTI in women are common infections. The alarmingly high resistance rates exhibited by contemporary uropathogens necessitate the re-evaluation of older but still microbiologically active antibiotics in patients at high risk of antibiotic-resistant uropathogens. The objectives of this study were to demonstrate the superiority of 5 days nitrofurantoin over once 3g fosfomycin for the treatment of lower, uncomplicated UTI in women at high risk of antibiotic-resistant uropathogens and to evaluate clinical and bacteriological cure at day 7; duration of symptoms; loss of working days; hospital admission during 28 days; incidence of pyelonephritis or urosepsis during this period; development of adverse events and emergence of antibiotic resistance.
METHODS AND ANALYSIS: This is a Phase IV, open-label, data-analyst-blinded, randomized, multi-centre clinical superiority trial with a planned sample size of 600 patients, 300 in each arm in non-pregnant female patients aged ≥ 18 years with uncomplicated lower UTI comparing outcome of Nitrofurantoin 100 mg tid for five days with Fosfomycin 3 g (single dose). The main parameters of efficacy are clinical response at 14 and 28 days after treatment cessation and bacteriological response at 14 and 28 days post treatment cessation. Upon meeting inclusion criteria and providing informed consent, non-pregnant adult female patients with uncomplicated lower UTI will be randomized to receive either study drug. Clinical and bacteriologic assessments will be made at baseline and days 14 and 28 after therapy cessation; monitoring of adverse events will be performed throughout the study period. Questions will be asked at study visits regarding symptomatology, potential adverse events, use of additional medication, contacts with healthcare professionals, quality of life, daily activities, and days of work lost. Conventional tests for superiority in the intention-to-treat analysis and per-protocol analyses will be performed. We will conduct a pre-defined subgroup analysis for patients with UTIs caused by extended-spectrum beta-lactamase (ESBL) producing and/or fluoroquinolone-resistant bacteria.
ETHICS AND DISSEMINATION: The AIDA project has appointed a data-safety monitoring board (DSMB) that will be convened for the purpose of monitoring patient safety and treatment efficacy throughout the study. Board members are specialists in infectious diseases. TRIAL REGISTRATION NUMBER : NCT01966653
2.2.2 Outcome of the clinical trial
The trial was successfully completed. Recruitment closed when 513 patients, of nearly 1000 screened, were included. The major reasons for exclusion were : not meeting eligibility criteria (n=320); declined participation (n=108). Of the ITT population, 255 were allocated to receive nitrofurantoin and 258 fosfomycin. In both arms there was some loss due to immediate loss to follow-up/receipt not confirmed, 8 and 11 subjects respectively. The clinical mITT population, patients that received at least one dose of study medications thus consisted of 247 patients in the nitrofurantoin arm and 247 in the fosfomycin arm. The per protocol population with primary outcome available and no major protocol deviation was 237 patients in both arms.
A medical review, in blinded fashion, of the 29 trial participants whose clinical outcome had been deemed “indeterminate” was performed and subsequently a blinded statistical analysis then took place. These analyses yielded interesting results—that remain under embargo until their presentation at the European Congress of Clinical Microbiology and Infectious Diseases in Madrid in April 2018. The full manuscript is currently under peer review.
2.3 WP3 - Multicentre RCT to compare minocycline plus rifampicin to linezolid against MRSA in cSSSI
2.3.1 The final study protocol
The study protocol faced several challenges before a final version could be agreed on and exemplified the difficulties of performing clinical trials in the field. Specifically, after a scoping exercise of 38 researchers in Eastern Europe it appeared that the trial protocol had to be modified in light of these findings. The major issue was the decline in MRSA rate during the period the trial was conceived and to be executed and thereby the successful enrollment was jeopardized in such a way that successful completion seemed unlikely. Modifications in the protocol were made and the study moved to southern Europe. A summary of the final study protocol reads as follows:
OBJECTIVES : Primary objective: To demonstrate non-inferiority between patients treated with oral minocycline plus rifampicin and those patients treated with gold standard therapy of linezolid in terms of clinical cure at Test of Cure (TOC). Secondary objectives were to assess the safety profile between the two treatment groups; to assess microbiological eradication of MRSA from the site of infection between the treatment groups; to assess the risk of emergence of resistance through sensitivity testing; to demonstrate a relationship between minocycline plus rifampicin drug exposure and outcome, including emergence of resistance, using state of the art pharmacokinetic-pharmacodynamic tools; to demonstrate a reduction in health care costs and resource utilization associated with the use of minocycline plus rifampicin compared with linezolid.
METHODS AND ANALYSIS: This is a prospective, open label, randomized, controlled multicentre clinical trial with a full PK/PD evaluation to be performed in 195 evaluable patients, 130 patients in minocycline/rifampicin treatment group, 65 patients in linezolid treatment group based on non-inferiority testing with a 5% significance level, 87% clinical cure rate for the comparator, linezolid at a 2:1 ratio and non-inferiority limit of 15 percentage points for 90% power. The target population includes adult patients, with complicated skin and skin structure infection caused by MRSA, may be enrolled on the basis of clinical criteria and fulfilment of the inclusion and exclusion criteria. The study is predicted to take 30 months enrolment including a 12 month PK/PD study. The Investigational Medical Products - Dose/route/ regimen are: Minocycline 100mg 12hourly po, rifampicin 600mg 24hourly po, linezolid 600mg 12 hourly po, OR as local national Summary of Product Characteristics (SmPC) guidelines. Efficacy will be assessed by a clinical evaluation by a blinded Clinical Investigator supported by photographic data; clinical response of signs and symptoms of infection and microbiological assessment. Safety will be assessed by a clinical review and laboratory monitoring. Pharmacokinetic/ Pharmacodynamic assessments are performed by 4 plasma blood draws (over 24 hours) are required on Day 1, and 4 plasma blood draws (over 24 hours) are require on Day 5. Each sample will be a minimum of 5mls and will be separated as soon as possible, and then frozen between -20°C and -80°C. A Health Economic Evaluation includes Costs associated with the use of minocycline plus rifampicin compared with treatment with linezolid in 4 categories; visits to providers, medications, interventions and testing. Four different populations will be analysed; the intention-to-treat (ITT), the per protocol (PP), and the microbiologically evaluable subsets of the ITT and PP populations (mITT and mePP respectively). The ITT population will include all patients who were allocated a treatment by the minimization website regardless of how much treatment was actually taken. Patients will be considered clinically evaluable (PP) if they meet the inclusion criteria, have a clinical outcome of either Cure or Failure at TOC (Day 14), receive drug for at least 5 days and have not departed in any significant way from the protocol. If medication is stopped early by the physician, either because of adverse events or because the infection has completely cleared and further treatment is not deemed necessary, this will not be regarded as a protocol violation. All patients by definition must have a positive culture of MRSA at baseline so the microbiologically evaluable patients will include all patients who have undergone 14 days follow-up with evaluation of eradication or relapse. Where the infection has completely cleared at the TOC visit (Day 14) and no sample can be obtained for culture the infection will be regarded as eradicated.
ETHICS AND DISSEMINATION: The study was approved by ethics committees at each centre and informed consent will be obtained for all patients. The trial is being performed under the auspices of an independent data and safety monitoring committee and is included in a broad dissemination strategy regarding revival of old antibiotics. TRIAL REGISTRATION NUMBER : EudraCT 2014-001276-56.
2.3.2 Outcome of the clinical trial
The trial was successfully completed. In total, 1105 patients were screened, and in 213 patients an MRSA was isolated. Of these, 87 failed in initial screening due to a variety of exclusion critera. Thus, 126 patients were randomized. Of these, 3 did not receive study medication. Of the 123 patients that did receive medication, 81 received minocycline and 42 received linezolid. In the minocycline-rifampicin arm 54 completed the study and 27 were withdrawn, in the linezolid arm these numbers were 31 and 11, respectively. The major reason for withdrawal was No TOC visit (14), followed by death (12, 6 in each arm).
There was one deviation from the protocol; the protocol stated that the ITT population would include all patients who were allocated a treatment by the minimization website irrespective of the amount of treatment received. Three patients (two in the minocycline plus rifampicin group and one in the linezolid group) were allocated to treatment but withdrew from the study before any treatment was started. These patients were not included in the ITT or other analysis populations since they contributed no data. Therefore, a total of 126 patients were randomised but 123 patients received study medication, 81 in the minocycline plus rifampicin group and 42 in the linezolid group, and these patients formed the ITT population. All were assessed at the baseline visit (Table 2). A total of 54 patients in the minocycline plus rifampicin group completed the study and 31 in the linezolid group.
Table 2.3.1: Number (%) of patients assessed at each visit, ITT
Minocycline+Rifampicin Linezolid Total
Visit Day N=81 N=42 N=123
BL/1 1 81 (100%) 42 (100%) 123 (100%)
2 5 (±1) 7 2 (88.9%) 37 (88.1%) 109 (88.6%)
TOC 14 (±2) 54 (66.7%) 31 (73.8%) 85 (69.1%)
The ITT population was identical to the mITT population and also formed the safety population (Table 2.3.2). A total of 94 patients were in the CE population and 84 of these patients were microbiologically evaluable and included in the ME population.
Table 2.3.2 : Number of patients included in the efficacy analysis
Study population Minocycline+Rifampicin Linezolid
ITT, N= 81 42
CE, N= 59 35
mITT, N= 81 42
ME, N= 53 31
Patients 601/007, 606/004, and 613/001 all withdrew from the study before treatment began. These patients are not included in the ITT or other populations. The ITT and mITT populations are identical as every patient in the ITT population had MRSA confirmed.
There were recruitment challenges in the study that meant that the target number of patients was not reached. The low recruitment in Italy was due to the normal working of the Italian healthcare system, which did not easily allow the recruitment of patients into the study. These included primarily: a lack of referral of patients from other departments; the time taken to receive culture data was outside the protocol inclusion requirement ; the target patients were an elderly and predominantly community/GP based population, thus in violation of the inclusion criteria for hospitalization. A similar issue of delay in receiving culture data was also observed in Greece. However, this was overcome by central coordination and the introduction of molecular testing to provide rapid (within 24hr) results. Despite overcoming this challenge the incidence of MRSA was notably lower than predicted by published literature; therefore, the pool of patients available to the study was not as large as originally estimated. Based on the present study’s screening rates, the rate of MRSA in Greece was between 9% and 13%, far lower than the original estimates based on data from the 2012 EARSS Annual Report report (35%-40%, ECDC, 2013) and the reported rate for 2016 of 38.8% (ECDC, 2017).
The formal analysis of the study is still under embargo until formal publication of the results.
3. WP4 – Preclinical studies and integration of PK/PD analyses of trial results : redevelopment of old antibiotics from a PK/PD point of view
The primary objective of this WP was to provide, within limits, a preclinical development package for each of the five off-patent antibiotics. This included preclinical pharmacodynamic studies in vitro and if necessary in animals to obtain the necessary pharmacodynamic targets for each drug, pharmacokinetic studies in patients or, if not feasible, in volunteers to estimate the exposure and the variation there-of, and building population pharmacokinetic models to allow estimates of exposures in the individual patient. The final goal was to integrate the PK/PD relationships of the clinical trials with those of the preclinical data, EDR and thereby provide a rational basis for improved treatment indications of use and also to suggest clinical breakpoints and epidemiological cut-off values. Below, a brief summary of results is presented for each drug.
3.1 Nitrofurantoin
Nitrofurantoin is one of the oldest off-patent drugs available and has been used widely. Two meta-analyses were performed to assess efficacy and safety. In the first meta-analysis (Huttner et al, 2015) a systematic review and meta-analysis was conducted to assess nitrofurantoin's efficacy and toxicity in the treatment of lower UTI. A systematic review of all human controlled clinical trials published from 1946 to 2014 and assessing short-term (A review on the pharmacokinetics of nitrofurantoin was also finalized but is still under consideration. From the review it appeared that relatively few pharmacokinetic data were available. In addition, it did not appear to be feasible to determine the pharmacokinetics of nitrofurantoin in the patients in the clinical trial. Therefore, a separate study was conducted to assess nitrofurantoin pharmacokinetics in plasma and urine for two different dosing regimens. To that purpose, a nitrofurantoin assay was set up. The final results of these pharmacokinetic studies are under consideration.
Surprisingly little appeared to be know of the pharmacodynamics of nitrofurantoin. These were assessed primarily in time-kill curves (Fransen et al 2016;2017). Early-phase pharmacodynamics was analysed with a sigmoidal Emax model and the maximal killing rate, slope and EC50/MIC ratio were determined for species for Enterobacter cloacae, Klebsiella pneumoniae and Escherichia coli. Overall, no killing was observed at low sub-MIC concentrations, whereas regrowth was found at 0.5-1x MIC after a short decline in cfu. Surprisingly, the Hill slopes for these three species were significantly different (10.45 +/- 9.37 2.68 +/- 0.64 and 1.01 +/- 0.06 respectively), indicating a strong concentration-dependent early-phase antibacterial activity against E. cloacae. Also, EC50/MIC ratios were significantly lower for E. coli (0.24 +/- 0.08 mg/L) and K. pneumoniae (0.27 +/- 0.03 mg/L) as compared with E. cloacae (0.77 +/- 0.18 mg/L). These results led to the surprising conclusion of an unusual differential pattern of activity with concentration-dependent-type killing behaviour against E. cloacae and time-dependent killing behaviour against E. coli, which may have significant consequences on species-dependent dosing regimens. The results demonstrated that the pharmacodynamic properties of some drugs cannot be generalized within a family, here the Enterobacteriaceae. Additional studies were performed with a different pH, since the pH may differ in urine, and in urine, since the pharmacodynamic properties could be different in urine. A strong pH-dependent pharmacodynamic output of parameters were observed. At 24 h, pH 5.5-6.5 for E. coli and Klebsiella pneumoniae required significantly lower nitrofurantoin concentrations compared with pH 7.5 or 8.5. Although for E. cloacae similar strong decreasing trends were visible with decreasing pH, none of the tested pharmacodynamic parameters was significant. Overall, bactericidal activity against Enterobacteriaceae significantly increased at lower pH levels. Bactericidal activity of nitrofurantoin may be overestimated or underestimated, which may have implications for therapy and the interpretation of clinical breakpoints.
Urinary inhibitory titers were also determined using urine from the human volunteer study. These data are still not fully analyzed, but do indicate that active concentrations of nitrofurantoin in urine are not very high. This supports the frequent dosing of nitrofurantoin over the day.
3.2 Fosfomycin
Similar to nitrofurantoin, surprisingly little appeared to be known with respect to its pharmacokinetics and pharmacodynamics despite its long use. Similar to nitrofurantoin, it was not feasible to study fosfomycin pharmacokinetics in the patients treated in the UTI trial, thus a separate pharmacokinetic study was performed with as a main goal the concentration time course in urine and in particular the variation there-of. It was hypothetized that concentrations could be low in some patients, potentially explaining failures of fosfomycin therapy for UTI (Wijma et al, 2017). Three grams of fosfomycin trometamol was administered to 40 healthy female volunteers with an estimated mean glomerular filtration rate of > 90 mL/min/1.73m2 to evaluate its effectiveness. Urine samples were collected from every urination during 48 hours, and then twice daily for up to 7 days. Time, volume and pH were recorded. Concentrations were quantified with UPLC-MS/MS. Effectiveness was evaluated based on urinary concentrations and the target MIC of E. coli, the most common uropathogen. A high interindividual variability was found. Peak concentration was 1982.0 +/- 1257.4 mg/L, urinary half-life 12.4 +/- 5.7 hours and excretion rate over 48 hours 29.9 +/- 7.1 mg/h. Recovery was 44.5 +/- 12.6% after 48 h and 47.0 +/- 10.4% after 7 days. Concentrations remained above the EUCAST breakpoint of 32 mg/L in 100% of the volunteers over the first 24 h, 67.5% for 48 h and 30% for 72 h. A high urinary output was associated with low urinary concentrations and consequently reduced time > MIC, AUC0-7days/MIC and Cmax/MIC values. It was concluded that considerable interindividual variability observed in the pharmacokinetics of fosfomycin signifies a risk for inadequate drug exposure in a significant proportion of the population and that the current dosing regimen should therefore be reevaluated.
Just like nitrofurantoin, little information on the pharmacodynamics of fosfomycin was available at the start of the project, and several studies were undertaken to fill this gap. Time kill studies were undertaken to determine concentration effect relationships and emergence of resistance in vitro of fosfomycin (Fransen et al 2017). A large variation in Hill slope factors for Escherichia coli of 0.87 up to 4.02 indicates that the killing behaviour appears to be more time dependent for some strains but concentration dependent for others. In the fosfomycin-exposed cultures under low and high selective pressure (>/=2x MIC) the median resistance proportions between the resistant and total population increased from Enterobacter cloacae; MIC range 0.25–64 mg/L) were examined. Pathogen kill and emergence of resistance was assessed over 72 h. Observed in vitro fosfomycin concentrations accurately simulated urinary fosfomycin exposures (Tmax 3.8+0.5 h; Cmax 2630.1+245.7 mg/L; AUC0-24 33932.5+1964.2mgh/L). Fifteen of 24 isolates regrew, with significant rises in fosfomycin MIC (total population MIC50 4 to 64 mg/L, MIC90 64 to.1024 mg/L, P=0.0039; resistant subpopulation MIC50 128 to.1024 mg/L, MIC90 .1024 mg/L, P=0.0020). E. coli and E. cloacae isolateswere killed with PK/PD EI50 of fAUC0-24/MIC of 1922, fCmax/MI of149 and fTime.4%MIC of 44 h. In contrast, K. pneumoniae isolates were not reliably killed. It was concluded that using urinary fosfomycin exposures, E. coli and E. cloacae isolates with MIC > 16 mg/L, and all K. pneumoniae isolates, were not reliably killed. Emergence of resistance was significant. The results challenges fosfomycin dosing and clinical breakpoints, and questions the utility of fosfomycin against K. pneumoniae. Studies in an in vitro hollow fiber model simulating plasma concentrations corroborated these findings. The results are still under consideration, but the main conclusions were that fosfomycin appeared to be effective for Wild Type E. coli but not for other species. Summarizing, all three studies indicate that resistance against fosfomycin emerges fast and that treatment for E. coli Wild Type is likely effective.
Finally, a study was undertaken in an animal model of UTI infection (Jakobsen et al., 2014). In OF-1 mice,
efficacy of 6 fosfomycin dosage regimens (total dose of 0.05 - 1.6 mg/mouse; dosing frequency of 6-72 hours) was investigated using the murine model of ascending UTI to estimate the most predictive PK/PD indices. Infection was induced by inoculating 50 μl (5x106 CFU) E. coli (NU14; FOS MIC 0.75 mg/L) via a catheter (retracted after inoculation) transurethrally and treatment was initiated sc. 1 day post infection for 3 days with FOS or saline. Urine, bladder, and kidneys were collected 4 days post infection for CFU determination. All doses significantly reduced CFUs as related to control, untreated mice (% reduction, lowest to highest doses; median control counts); urine: 80 - 100% (control mice: Log 6,8 CFU/ml; bladder: ca. 50% (control mice: Log 5,7 CFU/bladder); kidneys: 50-100% (control mice: log 2,9 CFU/kidney). The optimal PKPD indices for urine CFU/ml were (R2): AUC/MIC, 0,82; and Time>MIC of 1.dose: 0,82. Low correlations were found for Time>MIC % in all tissues, and for all indices for kidney
counts. It was concluded fosfomycin was very effective in reducing urine counts. The optimal PKPD indices included AUC/MIC ratio (> 200), orTime>MIC of 1. Dose (>2 h). These data also indicate that relatively high AUC/MIC ratio’s are required for effectiveness of fosfomycin.
The results of these in vitro studies are being considered to explain the results observed in the UTI trial comparing fosfomycin against nitrofurantoin treatment.
3.3 Minocycline
Minocycline samples were taken from 79 patients of whom 54 (68%) had full samples for both the day 1 and day 5 profiles. The concentrations were analyzed using a newly validated method and are currently under consideration.
The pharmacodynamics of minocyclin was studied in an in vitro hollow fibre model, as there was little data to support present dosing regimens and no data on the risks of emergence of resistance (Noel et al. 2013). Clinical breakpoints, both In Europe and USA, are based on historic data. A series of dose ranging studies against 4 strains of S. aureus (MIN MIG 0,19-0,5mg/L, all susceptible by EUGAST breakpoints) was performed and antibacterial effect was measured by changes in viable count over 72h and changes in population profile by recovery on MICx4 or MICx8 plates after 24, 48 and 72h exposure. The fAUC/MIC at 24h for static and -1 log reductions in viable count were 11 ,5±6, 1 and18,1 ±8, 1 respectively, For -2 log reduction in viable count, the fAUC/MIC was >200 for two strains, and 33,1 and 42,8 for the two others, fAUC/MIC targets were modestly increased after 72h being 12,9±2,9, static effect; 17,7±52, -1 log drop, 210g drop values were> 75, and 32.1 and 18.2 for the other two strains. There were no changes in population profiles indicated by growth on MIGx4 or MIGx8 plates at 24, 48 or 72h. It was concluded that a suitable fAUC/MIC target related to static -1 log drop for minocycline is 10-15. This target was not associated with any change in population profiles and would suggest a clinical
breakpoint of 0.25-0.5mg/L for a dose of 100mg 12 hrly.
3.4 Rifampicin
Rifampicin samples were taken from 79 patients of whom 54 (68%) had full samples for both the day 1 and day 5 profiles. The concentrations were analyzed using a newly validated method and being evaluated in a population pharmacokinetic model. This are currently under consideration.
3.5 Linezolid
For the Linezolid arm samples were received on 41 patients of who 30 (73%) had full samples for both the day 1 and day 5 profiles. A total of 806 samples were received and assayed. The concentrations were analyzed using a newly validated method and are currently under consideration.
3.6 Colistin
The pharmacokinetics and pharmacodynamics of colistin were an important objective in AIDA. The population pharmacokinetic (PK) analysis of colistin and its prodrug colistin methanesulfonate, CMS was determined in patients participating in the AIDA study. For patients with creatinine clearance (CRCL) > 50 ml/min a 9 MU loading dose, followed by a 4.5 MU twice daily maintenance dose, was administered as 30min infusion. Based on an earlier colistin and CMS PK model (Mohamed, Karaiskos et al. 2012) two optimal PK sampling points (45min and 10h after start of the infusion) had been identified to determine individual exposure of colistin and CMS (Kristoffersson, Friberg et al. 2015). Samples were sent to the central laboratory in batches from the participating centres. CMS A+B and colistin A+B were analyzed by LC-MS/MS. Dialysis patients, patients with obvious dose or sampling time errors, and suspected hydrolysed samples were excluded from the initial analysis. Data analysis was performed in R and NONMEM7. 237 patients were included in the initial analysis. The median concentration for the 10h samples was 4.3mg/L (SD 2.7mg/L). There was a clear relationship between creatinine clearance (CRCL, Cockcroft-Gault formula) and colistin 10h concentrations. The 25th percentiles exceeded 2mg/L except for patients with CRCL>=120mL/min. In order to describe the data the Mohamed model required adjustment by allowing an increased fraction of available CMS. The colistin and CMS PK model by Karaiskos et al (Karaiskos, et al. 2015) using CRCL as covariate adequately described the data after reestimating CMS CL. Thus, in line with Karaiskos et al higher available CMS and colistin A+B concentrations were observed compared to older studies, and resulted in colistin concentrations above 2 mg/L in the majority of patients with a CRCL lower than 120mL/min. Patients with higher CRCL may require increased maintenance dosing. The population PK model will be applied to drive a PKPD analysis linking individual colistin exposure and microbiological data to patient outcomes.
Apart from the pharmacokinetic studies a pharmacodynamic study was performed in an in vitro model of infection to determine its activity profile against K. pneumoniae. Simulating different concentration-time profiles and a range of exposures against strains with MICs of 0.5 to 4 mg/L, it was shown that the most important pharmacodynamic index associated a bactericidal effect was the AUC/MIC (R2 0.64-0.68) and that a value of at least 25 was required for a significant effect. Subsequent Monte Carlo simulations using reported data on colistin pharmacokinetics indicate that target attainment was only very moderate, even taking into account the dosing regimens currently used. It was also concluded that therapeutic drug monitoring of colistin might be warranted.
Potential Impact:
1. Main dissemination activities
The AIDA project was conceived with a work-package specifically dedicated to dissemination of results and raising awareness tot the use of off-patent antibiotics. There were four main target groups : the public in general, the health and scientific community, bio and pharma-industry and regulatory bodies. Dissemination activities were undertaken for each group.
1.1 The general public
For the general public there were a large number of press releases, including newspaper articles, background pieces and interviews both on specific issues of the project itself as well as the use of off-patent antibiotics in general. A good example is a press release of a symposium organized on off-patent antibiotics in 2013 –co-organized by dr Theuretzbacher on behalf of AIDA reads :
Conference Delegates Ask: Can Antibiotics From the Past Tackle the Growing Menace of Drug Resistant ‘Superbugs’ in the Present?
October 22: From today, Vienna plays host to a major international conference, convened by the European Society of Clinical Microbiology and Infectious Diseases (ESCMID), where researchers, academics and representatives from pharmaceutical companies and national medical agencies will explore the potential and pitfalls of using first generation, off patent antibiotics to tackle the growing global threat of antimicrobial resistance (AMR).
The three day conference, ‘Reviving Old Antibiotics’ has attracted participation from almost 300 delegates from 45 countries. They will explore the increasing threat of AMR, and the potential which exists for reverting to old and off patent antibiotics to treat today’s rapidly mutating and increasingly drug resistant ‘superbugs’.
In the past two decades AMR has emerged as one of the greatest threats to public health ever seen, with global agencies such as WHO and government institutions warning that without urgent remedial action the world is heading to a post-antibiotic age, where formerly treatable common infections will once again become fatal.
This crisis has come about as a result of a number of factors including over prescription of antibiotics; patient demand for unnecessary antibiotics to treat minor infections or viruses; unregulated access to antibiotics through pharmacies and online outlets; widespread use of antibiotics in agricultural practice to promote growth in livestock; poor hygiene practice in medical facilities; and the failure of the pharmaceutical industry to develop a new generation of drugs.
The European Centre for Disease Control estimates that 25,000 people in the EU currently die annually from infections that are resistant to multiple drugs. The economic cost associated with additional medical care for affected patients and loss of productivity is estimated at 1.5 billion Euros every year. The US Centre for Disease Control and Prevention estimates the combined cost of excess direct health care and lost productivity to the American economy to be between $55-70 billion.
“AMR threatens health care as we know it,” said Professor Gunnar Kahlmeter, ESCMID Past President.
“Many of the spectacular new practices and inventions in transplantation, cancer therapy, intensive and neonatal care and orthopaedics are at risk of having to be abandoned in those areas where multi-resistance is abundant. Once again mankind has failed in the basic ‘household management’ of one of its most precious and finest of resources.”
In the light of these challenges physicians are increasingly resorting to using old off-patent antibiotics that are still active against some drug resistant pathogens. These revived antibiotics were approved up to 55 years ago, however, and were not developed according to current standards of
science and practice. Continued use of these drugs without further study therefore carries significant downside consequences for individual patients as well as for the emergence of additional resistance.
Information about adequate dosages, indications, toxicity, and usage in special patient populations may be missing or not be adequate any more.
“We are using some old revived antibiotics now which were developed in the early penicillin era,” said Dr Ursula Theuretzbacher, founding President of ESCMID, and current Chair of its Scientific Programme Committee.
”We are potentially risking the lives of patients due to inadequate usage of these antibiotics and also losing them due to emergence of resistance.
“Strategies are urgently needed to “re-develop” these drugs according to modern standards, integrating the new knowledge into regulatory frameworks, and communicating the knowledge from research bench to bedside. The ESCMID conference in Vienna will help to communicate and add to this new knowledge.”
Areas of focus at the Vienna conference will include ‘Optimisation of Therapy’, ‘Extending the Life of Old Antibiotics’, ‘Breakpoints and Susceptibility Testing’ and ‘Regulating and Communicating New Knowledge’.
It is intended that the conference outcomes will provide the global community involved in the struggle against AMR with some practical solutions to the use of old generation antibiotics, including recommendations about priority research, identification of what knowledge gaps need to be filled and what policy changes and funding actions are required.
Other dissemination activities for the general public provided more detail., for instance how specifi off-patent antibiotics could or should be used. These releases were not only confined to the English language, for instance some of these were in german.
In addition to communication to the general public, contacts to patient advocacy and civil society groups were utilized through work within other EU funded projects (DRIVE-AB, COMBACTE-MAGNET). Patient advocacy groups are usually nationally organized and require local communication. Examples of communication with following organisations are: Doctors Without Borders, Médecins Sans Frontières Access Campaign, Gesundheitsladen Bielefeld, Brot für die Welt, Aussätzigen Hilfswerk Österreich, Deutschen Plattform Globale Gesundheit, Together 4 Diversity, Pew Charitable Trusts, Universities Allied for Essential Medicines, Association of European Academies.
1.2 The Health and Scientific community
As a scientific project, a large part of the dissemination activities were focussed on the health and scientific community. These included papers in scientific Journals, presentations at national and international meetings, organizations of workshops and symposia and newsletters.
1.2.1 Papers in Scientific Journals and presentations at meetings
The consortium has been very successful in publishing a significant number of papers in high-impact Journals. So far, 26 papers have been published, some are in press and it is expected that a considerable number of papers will follow in the next few years as more in depth analyses of the studies have been performed. Of note, the main results of the colistin versus colistin and meropenem trial will be published in the Lancet Infectious Diseases. Many of the results of AIDA where presented at international meetings such as the European Congres of Clinical Microbiology and Infectious Diseases (ECCMID) and the Interscience Conference of Antimicrobial Agents and Chemotherapy (ICAAC).
1.2.2 Workshops and Symposia
To further disseminate the current views on optimal use of off-patent antibiotics, a workshop was organized at around two thirds the project in conjunction with one of the yearly consortium meetings in Warschau in 2015. This was a co-sponsored meeting by the ESCMID and the European Pharmacology of Antibiotics Study Group (EPASG). During the three day course ‘Improving the dosage of old antibiotics’, delegates were brought up to date with current view points by members of the AIDA consortium.
Oral and poster presentations covering the AIDA topic as well as educational workshops further took place at the yearly meetings at the ECCMID (2011-2017 and the ICAAC (2011-2015). At the ECCMID, The AIDA project was regularly presented in the Network Corner of ECCMID and a poster presented.
A major event was the organization of a large symposium. This largest scientific and educational event of the AIDA project is the Conference on Reviving Old Antibiotics that took place on 22-24 October 2014 in Vienna. The Conference drew about 300 participants from 45 countries from a wide range of disciplines. Many of the world’s leading experts and researchers have accepted our invitation to actively participate in the meeting and to discuss how progress at the bench can be translated into actions in the field and highlight gaps in our knowledge. The last day of the conference was dedicated to an extensive discussion with experts and delegates who have identified gaps in our knowledge and agreed a number of key actions which need to be taken urgently.
The Vienna Conference was organized with administrative support of Europe’s largest professional interest group the European Society of Clinical Microbiology and Infectious Diseases (ESCMID), the ESCMID Study Group of Antibiotic Policies, the ESCMID PK/PD of Anti-Infectives Study Group and EUCAST. This broad alliance of professional interest groups will amplify the messages and results of the AIDA project in with ECCMID.
1.2.3 Newsletters
Newsletters with updates on the AIDA project were distributed regularly.
1.3 Bio- and Pharma Industry
It is of major importance that old antibiotics are redeveloped, and that new antibiotics are redeveloped in the context of emergence of drug resistance. A major role, in particular for the latter, is or should be played by industry. Increasing awareness and analyses of the current situation are of major importance. Some dissemination activities were undertaken in that direction.
1.4 Regulatory bodies
There were frequent communications with EMA to communicate the progress in the Field. The concept of AIDA is used as a model for “re-developing” old antibiotics. Several articles were published in peer-reviewed journals to highlight our project as a model for studying old antibiotics in academic settings. Face-to-face meetings with EMA and FDA as well as the PK/PD expert meeting at EMA showed the progress in the field, the awareness of regulators, and AIDA as a role model for “re-developing” old antibiotics. Representatives of AIDA were also present at the meeting organized by EMA in November 2015 to discuss PK/PD aspects in the (re) developments of antibiotics. These presentations were later-on used as a basis for the new PK/PD regulatory document for development of antibiotics.
2. Impact and Exploitation of Results
The concept of AIDA was used as a model for “re-developing” old antibiotics. Several articles were published in peer-reviewed journals to highlight our project as a model for studying old antibiotics in academic settings.
Face-to-face meetings with EMA as well as the PK/PD expert meeting at EMA showed the progress in the field, the awareness of regulators, and AIDA as a role model for “re-developing” old antibiotics. Extensive discussions with representatives of the extended the communication activities to the FDA who are very interested in our experiences with conducting clinical trials in severely sick patients with infections caused by extensively drug resistant Gram-negative bacteria. A presentation at the biennial meeting of the Transatlantic Task Force on Antimicrobial Resistance (TATFAR) was well received and widely discussed. F2F meetings with policy makers of several European countries (e.g. Sweden, Netherlands, Germany, Austria) to discuss the availability of old antibiotics and improve their usage enhanced AIDA’s visibility and the impact of the concept in general.
Ongoing dialogues with professional societies, especially the European Society for Clinical Microbiology and Infectious Diseases (ESCMID) and ESCMID study groups, International Society for Infectious Diseases (ISID), International Society of Chemotherapy (ISC), International Society for Anti-Infective Pharmacology (ISAP) contributed to raising awareness and to disseminate scientific findings.
In fruitful discussions with representatives of other EU funded projects (e.g. COMBACTE), with regulatory agencies (EMA and FDA), and pharmaceutical companies we could demonstrate how AIDA was successful in recruiting critically ill patients with infections caused by the most resistant pathogens though many pharmaceutical companies claim that it is not possible. Experiences gained by AIDA may influence the development of new antibiotics in a similar patient population. Our communication activities helped to broaden the discussions and provide input for development strategies of new antibiotics.
The concept of redeveloping antibiotics has a major impact on how antibiotics will be used in the future in several way. The results of the clinical trials provide evidence on the optimal use and indications of the trial drugs. For instance, the colistin trial is the first randomized controlled trial showing evidence that addition of a carbapenem is not useful, although current practice in many European countries. The scientific evidence now provided will help convince clinicians to alter their behaviour and reduce overall use of carbapenems and is thereby an important tool in reducing emergence of resistance. Furthermore, the PK/PD analyses of the trials will show how to optimize the use of the drugs and thereby increasing both efficacy and reduce adverse events, including emergence of resistance.
The PK/PD framework and supporting studies have provided further evidence of how the off-patent antibiotics should be used. The revision of the colistin SPC was aided by AIDA results and discussions. Discussions with the European Committee on Antimicrobial Susceptibility Testing (EUCAST) on the revision of clinical breakpoints have been initiated based on the AIDA results. These will lead to changes in clinical breakpoints, and there for the overall optimal use of these antibiotics in the EU, but also worldwide, as an increasing number of non-European countries are using EUCAST guidelines. The analyses of colistin and fosfomycin in particular have made or will make a significant impact.
The project has also led to collaboration with other groups. As a European consortium of scientists, there was the possibility to work closely together with a similar project funded by NIH and led by dr Keith Kaye, a collaboration that would have not been possible in an industry funded trial. This has led to significant achievements, one of which was the adjustment of the protocols to allow a combined analysis after finishing the studies. In addition, this collaboration was one of the drivers for three large international meetings on the use of Polymyxins, the third of which will take place after ECCMID, Madrid 2018. During these meetings standardization on the use, formulation and indications were discusses intensely and consensus documents have been published, e.g. the ‘Consistent global approach on reporting of colistin doses to promote safe and effective use’ in Clinical Infectious Diseases (2014) and ‘Framework for optimisation of the clinical use of colistin and polymyxin B: the Prato polymyxin consensus’ in the Lancet Infectious Diseases (2015).
3. Conclusions
Overall, the impact of the AIDA project has been and will be significant on several levels, including public, scientific, regulatory and clinical.
List of Websites:
Project public website address: http://www.aida-project.eu/
Contact details:
Prof. dr. Johan W. Mouton
Erasmus University Medical Center
Department of Medical Microbiology and Infectious Diseases (MMIZ)
P.O. Box 2040
3000 CA Rotterdam
The Netherlands
Email: j.mouton@erasmusmc.nl
In an era of increasing emergence of drug resistance (EDR) and lack of new antibiotics, old off- patent antibiotics are increasingly being prescribed to patients. However, many of these were developed in an age before the advent of a structured process for drug assessment and approval, and the establishment of clinical efficacy and effectiveness in randomized controlled trials in particular. The AIDA project aimed to answer the question of clinical effectiveness and optimal dosing of off-patent antibiotics for infections caused by multiple drug resistant (MDR) bacteria in three randomized controlled clinical trials. In a multidisciplinary approach the exposure response relationships for each antibiotic was elucidated by including pharmacokinetic (PK), pharmacodynamic (PD) and microbiological studies, including emergence of drug resistance.
The studies in the AIDA project commenced in 2012. The design of each of the three trials was fine-tuned and subsequently submitted to Medical Ethical Committees. After approval, inclusion in each of the three trials was completed in 2017.
In the first trial the efficacy of colistin alone was compared to colistin plus meropenem for severe infections caused by carbapenem-resistant gram-negative bacteria in an open-label randomized controlled superiority clinical trial to answer the questions whether combination therapy is useful if the micro-organism is not susceptible to the second drug. In total, 406 patients were recruited.
In the second trial fosfomycin 3 g (single dose) was compared with nitrofurantoin for the treatment of uncomplicated lower urinary tract infection in women aged ≥ 18 years at high risk of antibiotic-resistant pathogens in a multi-center randomized superiority trial. Anecdotal data had indicated a low response rate of a single dose of oral fosfomycin. Supported by pharmacokinetic studies and computer simulations these results may trigger a change in usage and more studies to define the best dosing regimen of oral fosfomycin. At completion, 513 patients were recruited.
In the third trial, oral treatment with minocycline plus rifampicin was compared with oral treatment with linezolid for complicated skin and soft tissue infections (cSSTI) due to MRSA in a controlled randomized multi-center non-inferiority clinical trial. At completion, 126 patients were randomized.
Each of these three clinical studies was linked to a variety of laboratory and animal studies supporting and understanding clinical trial results. The analysis of the three trials has been finalized and the main results are currently being considered for publication in major journals. Pharmacokinetic studies indicated significant pharmacokinetic variation between subjects that will need to be taken into consideration. Pharmacodynamic analyses clearly indicate that some of the trial drugs are likely under-dosed in current clinical practice. For some agents such as fosfomycin, the supporting pharmacodynamic studies indicated that resistance emerges relatively fast.
The collaboration of researchers from different disciplines and the successful completion of the AIDA project highlights the capacity of academic groups to “re-develop” old antibiotics and integrate complex clinical studies with high quality non-clinical research. Based on the project results, dosage regimens will need to be updated and the use and indications of old antibiotics revisited. The results have already led to ongoing discussion with regulatory bodies, such as the EMA, among others.
Old antibiotics are off-patent and no formal responsibility is defined for updating the knowledge of these valuable drugs. AIDA produced vital integrated non-clinical and clinical data to strengthen our evidence-based confidence for using these old drugs in daily clinical practice.
Updates and more details of the results of the studies can be found at the website of AIDA, www.aida-project.eu.
Project Context and Objectives:
1. Introduction and overall description of the AIDA project
In an era of increasing emergence of drug resistance and lack of new antibiotics, old off-patent antibiotics are increasingly being prescribed to patients. However, many of these were developed in an age before the advent of a structured process for drug assessment and approval, and the establishment of clinical efficacy and effectiveness in randomized controlled trials in particular. Moreover, dosing regimens of these antimicrobial agents were developed more on a trial and error basis and information to optimize dosing regimens using exposure-response relationships is not readily available, if at all. It is unclear whether the current dosing regimens used are optimal or even efficacious. Even if comparative trials were performed in the past to determine whether one antibiotic was non-inferior or superior to another, oftentimes the dosing regimens have changed over time. This change is a significant problem in particular since old off-patent antibiotics are increasingly being prescribed to patients now that escalating resistance poses a specific challenge in antimicrobial treatment. On many occasions, microorganisms are now fully resistant to commonly used drugs. Examples include ESBL (Extended-Spectrum-Beta-Lactamase) producers, and recently, KPC (Klebsiella pneumoniae carbapenemases) or NDM-1 (New Delhi Metallo-beta-lactamase) producers. Old drugs, such as colistin and fosfomycin, are then used without any certainty that the correct dosing regimens are being applied. In addition, comparisons between treatment modalities to determine the best choice of treatment in randomized controlled trials (RCT) have not been established for many of these older antibiotics. A re-evaluation of these drugs is therefore urgently needed and includes establishing PK/PD relationships and the optimal dosing regimens to use.
In the AIDA project we aimed to answer the question of clinical effectiveness and optimal dosing of off-patent antibiotics for infections caused by multiple drug resistant (MDR) bacteria in three randomized controlled clinical trials in three work packages (WP1,2,3). In a multidisciplinary approach the exposure response relationships for each antibiotic was to be elucidated by including pharmacokinetic (PK), pharmacodynamic (PD) and microbiological studies, including emergence of drug resistance (WP4). The project addressed the optimization of treatment of infections caused by MDR pathogens that impose a major burden of disease in Europe and the rest of the world by selecting 5 off-patent-antibiotics - colistin, fosfomycin, nitrofurantoin, minocyclin and rifampicin that were increasingly being used without clear evidence with respect to their effectiveness, duration of therapy and issues of EDR. Common to all the clinical trials proposed was the selection of a clinically-relevant patient population; randomized controlled trial design to minimise bias; assessment of clinically-relevant outcomes for the individual patient, PK/PD assessment and correlation between PK/PD indices and clinical outcomes, including efficacy, toxicity and emergence of resistance. In summary, the main goal of the project AIDA was to gain preclinical and clinical data of 5 off-patent antimicrobials to allow optimizing dosing regimens. To that end, 3 clinical trials (WP1-3) were to be performed and preclinical studies (WP4) of each of the antibiotics were to supplement the evidence for optimal dosing and use. An important objective was the dissemination of results and was handled in a separate WP (WP5). Below, the specific objectives of each WP are briefly described.
2. Specific objectives for each work package
WP1 – Multicenter open-label RCT to compare colistin vs. alone colistin plus meropenem
The primary objective of this RCT was to demonstrate superiority of colistin in combination with meropenem vs. colistin alone in the treatment of bloodstream infections and ventilator-associated pneumonia (VAP) caused by carbapenem-resistant Acinetobacter sp., Pseudomonas aeruginosa or Enterobacteriaceae. Secundary objectives include: (differences in) change in functional capacity from baseline; superinfections, defined as clinically or microbiologically-documented infections by CDC criteria within 28 days; 28-day all-cause mortality; time to clinical success; microbiological failure, defined as the isolation of a bacterium at day 7 or more that is phenotypically identicalto the first isolate; colonization by carbapenem and/ or colistin-resistant Gram-negative bacteria; Clostridium-difficile-associated diarrhea; renal failure using the RIFLE criteria; an analysis of adverse advents, including, seizures, critical illness and other neuropathy
WP2 – Multicenter RCT to compare fosfomycin vs. nitrofurantoin for treatment of uncomplicated Urinary Tract Infections (UTI)
The primary objective of this RCT was to demonstrate the superiority of 5 days nitrofurantoin over once 3g fosfomycin for the treatment of lower, uncomplicated UTI in women at high risk of antibiotic-resistant uropathogens. The secondary objectives were to evaluate clinical and bacteriological cure at day 7; duration of symptoms; loss of working days; hospital admission during 28 days; incidence of pyelonephritis or urosepsis during this period and development of adverse events and emergence of antibiotic resistance.
WP3 - Multicentre RCT to compare minocycline plus rifampicin to linezolid against MRSA in cSSSI
The primary objective of this RCT was to demonstrate non inferiority between patients with acute complicated skin and soft tissue (cSSTI) infection due to MRSA treated with minocycline plus rifampicin compared to those treated with linezolid in terms of efficacy and safety.
WP4 – Preclinical studies and integration of PK/PD analyses of trial results : redevelopment of old antibioticsfrom a PK/PD point of view
The primary objective of this WP was to provide, within limits, a preclinical development package for each of the five off-patent antibiotics. This included preclinical pharmacodynamic studies in vitro and if necessary in animals to obtain the necessary pharmacodynamic targets for each drug, pharmacokinetic studies in patients or, if not feasible, in volunteers to estimate the exposure and the variation there-of, and building population pharmacokinetic models to allow estimates of exposures in the individual patient. The final goal was to integrate the PK/PD relationships of the clinical trials with those of the preclinical data, EDR and thereby provide a rational basis for improved treatment indications of use and also to suggest clinical breakpoints and epidemiological cut-off values.
WP5 - Dissemination of results, communication to the professionals and the public
The primary objective of this work package was to ensure that the results of the studies were communicated to the public and relevant bodies in a timely manner. In addition, general communications were to increase the public awareness of the problem of emergence of resistance and the lack of knowledge of older drugs that were increasingly being used.
3. Overreaching objectives of the AIDA project
Apart from the specific objectives in each WP, the AIDA project entailed several overreaching objectives. These included addressing key scientific and clinical questions regarding the best treatment options for important disease entities and enabling refinement of these options by additional PK/PD research. Importantly the trials, in particular those in WP1 and WP2, should set a model for clinical trial testing of new antibiotics. For instance, justified criticism has been raised on industry-conducted trials assessing new antibiotics in SSTI and abdominal infections. Indeed, these trials are often times designed and destined to show equivalence, but have very low applicability to the treatment of patients who are
subsequently treated with these antibiotics. Patients affected by MDR Gram-negative bacteria in hospitals are severely-ill and have multiple co-morbid conditions. In WP1 this patient population is specifically addressed. The trial in WP1 was therefore to be designed and conducted by clinicians. If successful, this trial design could serve as a model for regulatory bodies to be considered in updated guidance documents for conducting formal Phase 3 testing of new antibiotics.
Project Results:
1. Introduction
In the AIDA project we aimed to answer the question of clinical effectiveness and optimal dosing of five off-patent antibiotics for infections caused by multiple drug resistant (MDR) bacteria in three randomized controlled clinical trials in three work packages (WP1,2,3). In a multidisciplinary approach the exposure response relationships for each antibiotic was elucidated by including pharmacokinetic (PK), pharmacodynamic (PD) and microbiological studies, including emergence of drug resistance (WP4).
Below, the results of the AIDA project are presented. The overall results of the clinical trials are presented first; In the section thereafter these are put into context together with the main outcomes of the PK/PD studies. It should be noted that many of the outcome data are still confidential and will be published in scientific journals in the year(s) to come.
2. Design and outcomes of the clinical trials
During the first period of the project, the initial study proposals of each the three studies were fine-tuned. The proposals were subsequently submitted to the relevant Medical Ethical Committees and after approval the trials were initiated. The description of results follows this timeline for each trial.
2.1 WP1 – Multicenter open-label RCT to compare colistin vs. alone colistin plus meropenem
2.1.1 The final study protocol
After multiple discussions with AIDA partners and external advisors, we proceeded to a randomized controlled trial comparing optimally-dosed colistin, based on PK considerations vs. colistin combined with high-dose extended-infusion meropenem. The final study protocol was submitted and published in the BMJ open (Dickstein et al; 2016). The summary of the final study reads as follows:
INTRODUCTION: The emergence of antibiotic-resistant bacteria has driven renewed interest in older antibacterials, including colistin. Previous studies have shown that colistin is less effective and more toxic than modern antibiotics. In vitro synergy studies and clinical observational studies suggest a benefit of combining colistin with a carbapenem. A randomised controlled study is necessary for clarification.
METHODS AND ANALYSIS: This is a multicentre, investigator-initiated, open-label, randomised controlled superiority 1:1 study comparing colistin monotherapy with colistin-meropenem combination therapy for infections caused by carbapenem-resistant Gram-negative bacteria. The study is being conducted in 6 centres in 3 countries (Italy, Greece and Israel). We include patients with hospital-associated and ventilator-associated pneumonia, bloodstream infections and urosepsis. The primary outcome is treatment success at day 14, defined as survival, haemodynamic stability, stable or improved respiratory status for patients with pneumonia, microbiological cure for patients with bacteraemia and stability or improvement of the Sequential Organ Failure Assessment (SOFA) score. Secondary outcomes include 14-day and 28-day mortality as well as other clinical end points and safety outcomes. A sample size of 360 patients was calculated on the basis of an absolute improvement in clinical success of 15% with combination therapy. Outcomes will be assessed by intention to treat. Serum colistin samples are obtained from all patients to obtain population pharmacokinetic models. Microbiological sampling includes weekly surveillance samples with analysis of resistance mechanisms and synergy. An observational trial is evaluating patients who met eligibility requirements but were not randomised in order to assess generalisability of findings.
ETHICS AND DISSEMINATION: The study was approved by ethics committees at each centre and informed consent will be obtained for all patients. The trial is being performed under the auspices of an independent data and safety monitoring committee and is included in a broad dissemination strategy regarding revival of old antibiotics. TRIAL REGISTRATION NUMBER: NCT01732250 and 2012-004819-31
A full description of inclusion and exclusion criteria and other details can further be found in this paper.
2.1.2 Outcome of the clinical trial.
The trial was successfully completed in six medical centers in Israel, Greece and Italy. In total 406 patients with severe infections caused by the target carbapenem-resistant Gram-negative bacteria were recruited. The trial was powered to show superiority of the combination, defined as an absolute 15% advantage in clinical failure (a composite of death, clinical instability or bacteremia persistence at 14 days) and its results refuted within the 95% confidence intervals an advantage to combination therapy. Clinical failure at day 14, all-cause mortality at day 14 and 28 and microbiological failure were all very similar between the groups, with risk ratios ranging between 0.93 and 1.1 for these outcomes, all non-significant. Combination therapy did not decrease the probability of acquiring colistin-resistant bacteria. Adverse events rates were all similar, but for renal failure which was lower with combination therapy. The manuscript has been accepted for publication in Lancet Infect Dis. Further analyses using the colistin level and microbiological samples obtained in the trial are ongoing. These analyses will allow further precision in the assessment of monotherapy vs. combination therapy effects.
Thus, to answer the objective, we refuted the clinical superiority of colistin-meropenem combination therapy in the treatment of carbapenem-resistant Gram-negative bacteria. The trial addressed the patient population treated in clinical practice as reflected by clinical failure rates of 75.9% (308/406) at day 14 and 28-day mortality rates of 180/406 (44.3%), very similar to observational cohorts of patients with carbapenem-resistant Gram-negative bacteria. It is the first randomized controlled trial and its results disagree with the data available from observational studies. We summarized the data available from observational studies, and mainly their limitations leading to very low quality of evidence (Zusman et al. J Antimicrob Chemother 2017). Our trial provides a strong signal that treatment decisions for multidrug-resistant bacteria must be based on a randomized controlled trial.
Some other observations can be made. The objective was to prove (or disprove) the superiority of combination therapy among the assemblage of the three major carbapenem-resistant Gram-negative bacteria - Acinetobacter sp., Pseudomonas aeruginosa and Enterobacteriaceae. The distribution among these bacteria was dictated by the epidemiological trends in the trial centers. The trial largely included patients with Acinetobacter baumannii infections (312/406, 76.8%) and thus provides an answer for this pathogen. The trial remained unpowered to address the question among patients with carbapenem-resistant enterobactericeae and P. aeruginosa. We are collaborating with Prof. Keith Kaye, PI of a parallel NIH-funded trial, designed very similarly and collecting the same outcomes. The compilation of the two trials will hopefully advance the understanding regarding carbapenem-resistant enterobacteriaceae.
2.2 WP2 – Multicenter RCT to compare fosfomycin vs. nitrofurantoin for treatment of uncomplicated Urinary Tract Infections (UTI)
2.2.1 The final study protocol
A summary of the final study protocol reads as follows:
INTRODUCTION: Cystitis and lower UTI in women are common infections. The alarmingly high resistance rates exhibited by contemporary uropathogens necessitate the re-evaluation of older but still microbiologically active antibiotics in patients at high risk of antibiotic-resistant uropathogens. The objectives of this study were to demonstrate the superiority of 5 days nitrofurantoin over once 3g fosfomycin for the treatment of lower, uncomplicated UTI in women at high risk of antibiotic-resistant uropathogens and to evaluate clinical and bacteriological cure at day 7; duration of symptoms; loss of working days; hospital admission during 28 days; incidence of pyelonephritis or urosepsis during this period; development of adverse events and emergence of antibiotic resistance.
METHODS AND ANALYSIS: This is a Phase IV, open-label, data-analyst-blinded, randomized, multi-centre clinical superiority trial with a planned sample size of 600 patients, 300 in each arm in non-pregnant female patients aged ≥ 18 years with uncomplicated lower UTI comparing outcome of Nitrofurantoin 100 mg tid for five days with Fosfomycin 3 g (single dose). The main parameters of efficacy are clinical response at 14 and 28 days after treatment cessation and bacteriological response at 14 and 28 days post treatment cessation. Upon meeting inclusion criteria and providing informed consent, non-pregnant adult female patients with uncomplicated lower UTI will be randomized to receive either study drug. Clinical and bacteriologic assessments will be made at baseline and days 14 and 28 after therapy cessation; monitoring of adverse events will be performed throughout the study period. Questions will be asked at study visits regarding symptomatology, potential adverse events, use of additional medication, contacts with healthcare professionals, quality of life, daily activities, and days of work lost. Conventional tests for superiority in the intention-to-treat analysis and per-protocol analyses will be performed. We will conduct a pre-defined subgroup analysis for patients with UTIs caused by extended-spectrum beta-lactamase (ESBL) producing and/or fluoroquinolone-resistant bacteria.
ETHICS AND DISSEMINATION: The AIDA project has appointed a data-safety monitoring board (DSMB) that will be convened for the purpose of monitoring patient safety and treatment efficacy throughout the study. Board members are specialists in infectious diseases. TRIAL REGISTRATION NUMBER : NCT01966653
2.2.2 Outcome of the clinical trial
The trial was successfully completed. Recruitment closed when 513 patients, of nearly 1000 screened, were included. The major reasons for exclusion were : not meeting eligibility criteria (n=320); declined participation (n=108). Of the ITT population, 255 were allocated to receive nitrofurantoin and 258 fosfomycin. In both arms there was some loss due to immediate loss to follow-up/receipt not confirmed, 8 and 11 subjects respectively. The clinical mITT population, patients that received at least one dose of study medications thus consisted of 247 patients in the nitrofurantoin arm and 247 in the fosfomycin arm. The per protocol population with primary outcome available and no major protocol deviation was 237 patients in both arms.
A medical review, in blinded fashion, of the 29 trial participants whose clinical outcome had been deemed “indeterminate” was performed and subsequently a blinded statistical analysis then took place. These analyses yielded interesting results—that remain under embargo until their presentation at the European Congress of Clinical Microbiology and Infectious Diseases in Madrid in April 2018. The full manuscript is currently under peer review.
2.3 WP3 - Multicentre RCT to compare minocycline plus rifampicin to linezolid against MRSA in cSSSI
2.3.1 The final study protocol
The study protocol faced several challenges before a final version could be agreed on and exemplified the difficulties of performing clinical trials in the field. Specifically, after a scoping exercise of 38 researchers in Eastern Europe it appeared that the trial protocol had to be modified in light of these findings. The major issue was the decline in MRSA rate during the period the trial was conceived and to be executed and thereby the successful enrollment was jeopardized in such a way that successful completion seemed unlikely. Modifications in the protocol were made and the study moved to southern Europe. A summary of the final study protocol reads as follows:
OBJECTIVES : Primary objective: To demonstrate non-inferiority between patients treated with oral minocycline plus rifampicin and those patients treated with gold standard therapy of linezolid in terms of clinical cure at Test of Cure (TOC). Secondary objectives were to assess the safety profile between the two treatment groups; to assess microbiological eradication of MRSA from the site of infection between the treatment groups; to assess the risk of emergence of resistance through sensitivity testing; to demonstrate a relationship between minocycline plus rifampicin drug exposure and outcome, including emergence of resistance, using state of the art pharmacokinetic-pharmacodynamic tools; to demonstrate a reduction in health care costs and resource utilization associated with the use of minocycline plus rifampicin compared with linezolid.
METHODS AND ANALYSIS: This is a prospective, open label, randomized, controlled multicentre clinical trial with a full PK/PD evaluation to be performed in 195 evaluable patients, 130 patients in minocycline/rifampicin treatment group, 65 patients in linezolid treatment group based on non-inferiority testing with a 5% significance level, 87% clinical cure rate for the comparator, linezolid at a 2:1 ratio and non-inferiority limit of 15 percentage points for 90% power. The target population includes adult patients, with complicated skin and skin structure infection caused by MRSA, may be enrolled on the basis of clinical criteria and fulfilment of the inclusion and exclusion criteria. The study is predicted to take 30 months enrolment including a 12 month PK/PD study. The Investigational Medical Products - Dose/route/ regimen are: Minocycline 100mg 12hourly po, rifampicin 600mg 24hourly po, linezolid 600mg 12 hourly po, OR as local national Summary of Product Characteristics (SmPC) guidelines. Efficacy will be assessed by a clinical evaluation by a blinded Clinical Investigator supported by photographic data; clinical response of signs and symptoms of infection and microbiological assessment. Safety will be assessed by a clinical review and laboratory monitoring. Pharmacokinetic/ Pharmacodynamic assessments are performed by 4 plasma blood draws (over 24 hours) are required on Day 1, and 4 plasma blood draws (over 24 hours) are require on Day 5. Each sample will be a minimum of 5mls and will be separated as soon as possible, and then frozen between -20°C and -80°C. A Health Economic Evaluation includes Costs associated with the use of minocycline plus rifampicin compared with treatment with linezolid in 4 categories; visits to providers, medications, interventions and testing. Four different populations will be analysed; the intention-to-treat (ITT), the per protocol (PP), and the microbiologically evaluable subsets of the ITT and PP populations (mITT and mePP respectively). The ITT population will include all patients who were allocated a treatment by the minimization website regardless of how much treatment was actually taken. Patients will be considered clinically evaluable (PP) if they meet the inclusion criteria, have a clinical outcome of either Cure or Failure at TOC (Day 14), receive drug for at least 5 days and have not departed in any significant way from the protocol. If medication is stopped early by the physician, either because of adverse events or because the infection has completely cleared and further treatment is not deemed necessary, this will not be regarded as a protocol violation. All patients by definition must have a positive culture of MRSA at baseline so the microbiologically evaluable patients will include all patients who have undergone 14 days follow-up with evaluation of eradication or relapse. Where the infection has completely cleared at the TOC visit (Day 14) and no sample can be obtained for culture the infection will be regarded as eradicated.
ETHICS AND DISSEMINATION: The study was approved by ethics committees at each centre and informed consent will be obtained for all patients. The trial is being performed under the auspices of an independent data and safety monitoring committee and is included in a broad dissemination strategy regarding revival of old antibiotics. TRIAL REGISTRATION NUMBER : EudraCT 2014-001276-56.
2.3.2 Outcome of the clinical trial
The trial was successfully completed. In total, 1105 patients were screened, and in 213 patients an MRSA was isolated. Of these, 87 failed in initial screening due to a variety of exclusion critera. Thus, 126 patients were randomized. Of these, 3 did not receive study medication. Of the 123 patients that did receive medication, 81 received minocycline and 42 received linezolid. In the minocycline-rifampicin arm 54 completed the study and 27 were withdrawn, in the linezolid arm these numbers were 31 and 11, respectively. The major reason for withdrawal was No TOC visit (14), followed by death (12, 6 in each arm).
There was one deviation from the protocol; the protocol stated that the ITT population would include all patients who were allocated a treatment by the minimization website irrespective of the amount of treatment received. Three patients (two in the minocycline plus rifampicin group and one in the linezolid group) were allocated to treatment but withdrew from the study before any treatment was started. These patients were not included in the ITT or other analysis populations since they contributed no data. Therefore, a total of 126 patients were randomised but 123 patients received study medication, 81 in the minocycline plus rifampicin group and 42 in the linezolid group, and these patients formed the ITT population. All were assessed at the baseline visit (Table 2). A total of 54 patients in the minocycline plus rifampicin group completed the study and 31 in the linezolid group.
Table 2.3.1: Number (%) of patients assessed at each visit, ITT
Minocycline+Rifampicin Linezolid Total
Visit Day N=81 N=42 N=123
BL/1 1 81 (100%) 42 (100%) 123 (100%)
2 5 (±1) 7 2 (88.9%) 37 (88.1%) 109 (88.6%)
TOC 14 (±2) 54 (66.7%) 31 (73.8%) 85 (69.1%)
The ITT population was identical to the mITT population and also formed the safety population (Table 2.3.2). A total of 94 patients were in the CE population and 84 of these patients were microbiologically evaluable and included in the ME population.
Table 2.3.2 : Number of patients included in the efficacy analysis
Study population Minocycline+Rifampicin Linezolid
ITT, N= 81 42
CE, N= 59 35
mITT, N= 81 42
ME, N= 53 31
Patients 601/007, 606/004, and 613/001 all withdrew from the study before treatment began. These patients are not included in the ITT or other populations. The ITT and mITT populations are identical as every patient in the ITT population had MRSA confirmed.
There were recruitment challenges in the study that meant that the target number of patients was not reached. The low recruitment in Italy was due to the normal working of the Italian healthcare system, which did not easily allow the recruitment of patients into the study. These included primarily: a lack of referral of patients from other departments; the time taken to receive culture data was outside the protocol inclusion requirement ; the target patients were an elderly and predominantly community/GP based population, thus in violation of the inclusion criteria for hospitalization. A similar issue of delay in receiving culture data was also observed in Greece. However, this was overcome by central coordination and the introduction of molecular testing to provide rapid (within 24hr) results. Despite overcoming this challenge the incidence of MRSA was notably lower than predicted by published literature; therefore, the pool of patients available to the study was not as large as originally estimated. Based on the present study’s screening rates, the rate of MRSA in Greece was between 9% and 13%, far lower than the original estimates based on data from the 2012 EARSS Annual Report report (35%-40%, ECDC, 2013) and the reported rate for 2016 of 38.8% (ECDC, 2017).
The formal analysis of the study is still under embargo until formal publication of the results.
3. WP4 – Preclinical studies and integration of PK/PD analyses of trial results : redevelopment of old antibiotics from a PK/PD point of view
The primary objective of this WP was to provide, within limits, a preclinical development package for each of the five off-patent antibiotics. This included preclinical pharmacodynamic studies in vitro and if necessary in animals to obtain the necessary pharmacodynamic targets for each drug, pharmacokinetic studies in patients or, if not feasible, in volunteers to estimate the exposure and the variation there-of, and building population pharmacokinetic models to allow estimates of exposures in the individual patient. The final goal was to integrate the PK/PD relationships of the clinical trials with those of the preclinical data, EDR and thereby provide a rational basis for improved treatment indications of use and also to suggest clinical breakpoints and epidemiological cut-off values. Below, a brief summary of results is presented for each drug.
3.1 Nitrofurantoin
Nitrofurantoin is one of the oldest off-patent drugs available and has been used widely. Two meta-analyses were performed to assess efficacy and safety. In the first meta-analysis (Huttner et al, 2015) a systematic review and meta-analysis was conducted to assess nitrofurantoin's efficacy and toxicity in the treatment of lower UTI. A systematic review of all human controlled clinical trials published from 1946 to 2014 and assessing short-term (A review on the pharmacokinetics of nitrofurantoin was also finalized but is still under consideration. From the review it appeared that relatively few pharmacokinetic data were available. In addition, it did not appear to be feasible to determine the pharmacokinetics of nitrofurantoin in the patients in the clinical trial. Therefore, a separate study was conducted to assess nitrofurantoin pharmacokinetics in plasma and urine for two different dosing regimens. To that purpose, a nitrofurantoin assay was set up. The final results of these pharmacokinetic studies are under consideration.
Surprisingly little appeared to be know of the pharmacodynamics of nitrofurantoin. These were assessed primarily in time-kill curves (Fransen et al 2016;2017). Early-phase pharmacodynamics was analysed with a sigmoidal Emax model and the maximal killing rate, slope and EC50/MIC ratio were determined for species for Enterobacter cloacae, Klebsiella pneumoniae and Escherichia coli. Overall, no killing was observed at low sub-MIC concentrations, whereas regrowth was found at 0.5-1x MIC after a short decline in cfu. Surprisingly, the Hill slopes for these three species were significantly different (10.45 +/- 9.37 2.68 +/- 0.64 and 1.01 +/- 0.06 respectively), indicating a strong concentration-dependent early-phase antibacterial activity against E. cloacae. Also, EC50/MIC ratios were significantly lower for E. coli (0.24 +/- 0.08 mg/L) and K. pneumoniae (0.27 +/- 0.03 mg/L) as compared with E. cloacae (0.77 +/- 0.18 mg/L). These results led to the surprising conclusion of an unusual differential pattern of activity with concentration-dependent-type killing behaviour against E. cloacae and time-dependent killing behaviour against E. coli, which may have significant consequences on species-dependent dosing regimens. The results demonstrated that the pharmacodynamic properties of some drugs cannot be generalized within a family, here the Enterobacteriaceae. Additional studies were performed with a different pH, since the pH may differ in urine, and in urine, since the pharmacodynamic properties could be different in urine. A strong pH-dependent pharmacodynamic output of parameters were observed. At 24 h, pH 5.5-6.5 for E. coli and Klebsiella pneumoniae required significantly lower nitrofurantoin concentrations compared with pH 7.5 or 8.5. Although for E. cloacae similar strong decreasing trends were visible with decreasing pH, none of the tested pharmacodynamic parameters was significant. Overall, bactericidal activity against Enterobacteriaceae significantly increased at lower pH levels. Bactericidal activity of nitrofurantoin may be overestimated or underestimated, which may have implications for therapy and the interpretation of clinical breakpoints.
Urinary inhibitory titers were also determined using urine from the human volunteer study. These data are still not fully analyzed, but do indicate that active concentrations of nitrofurantoin in urine are not very high. This supports the frequent dosing of nitrofurantoin over the day.
3.2 Fosfomycin
Similar to nitrofurantoin, surprisingly little appeared to be known with respect to its pharmacokinetics and pharmacodynamics despite its long use. Similar to nitrofurantoin, it was not feasible to study fosfomycin pharmacokinetics in the patients treated in the UTI trial, thus a separate pharmacokinetic study was performed with as a main goal the concentration time course in urine and in particular the variation there-of. It was hypothetized that concentrations could be low in some patients, potentially explaining failures of fosfomycin therapy for UTI (Wijma et al, 2017). Three grams of fosfomycin trometamol was administered to 40 healthy female volunteers with an estimated mean glomerular filtration rate of > 90 mL/min/1.73m2 to evaluate its effectiveness. Urine samples were collected from every urination during 48 hours, and then twice daily for up to 7 days. Time, volume and pH were recorded. Concentrations were quantified with UPLC-MS/MS. Effectiveness was evaluated based on urinary concentrations and the target MIC of E. coli, the most common uropathogen. A high interindividual variability was found. Peak concentration was 1982.0 +/- 1257.4 mg/L, urinary half-life 12.4 +/- 5.7 hours and excretion rate over 48 hours 29.9 +/- 7.1 mg/h. Recovery was 44.5 +/- 12.6% after 48 h and 47.0 +/- 10.4% after 7 days. Concentrations remained above the EUCAST breakpoint of 32 mg/L in 100% of the volunteers over the first 24 h, 67.5% for 48 h and 30% for 72 h. A high urinary output was associated with low urinary concentrations and consequently reduced time > MIC, AUC0-7days/MIC and Cmax/MIC values. It was concluded that considerable interindividual variability observed in the pharmacokinetics of fosfomycin signifies a risk for inadequate drug exposure in a significant proportion of the population and that the current dosing regimen should therefore be reevaluated.
Just like nitrofurantoin, little information on the pharmacodynamics of fosfomycin was available at the start of the project, and several studies were undertaken to fill this gap. Time kill studies were undertaken to determine concentration effect relationships and emergence of resistance in vitro of fosfomycin (Fransen et al 2017). A large variation in Hill slope factors for Escherichia coli of 0.87 up to 4.02 indicates that the killing behaviour appears to be more time dependent for some strains but concentration dependent for others. In the fosfomycin-exposed cultures under low and high selective pressure (>/=2x MIC) the median resistance proportions between the resistant and total population increased from Enterobacter cloacae; MIC range 0.25–64 mg/L) were examined. Pathogen kill and emergence of resistance was assessed over 72 h. Observed in vitro fosfomycin concentrations accurately simulated urinary fosfomycin exposures (Tmax 3.8+0.5 h; Cmax 2630.1+245.7 mg/L; AUC0-24 33932.5+1964.2mgh/L). Fifteen of 24 isolates regrew, with significant rises in fosfomycin MIC (total population MIC50 4 to 64 mg/L, MIC90 64 to.1024 mg/L, P=0.0039; resistant subpopulation MIC50 128 to.1024 mg/L, MIC90 .1024 mg/L, P=0.0020). E. coli and E. cloacae isolateswere killed with PK/PD EI50 of fAUC0-24/MIC of 1922, fCmax/MI of149 and fTime.4%MIC of 44 h. In contrast, K. pneumoniae isolates were not reliably killed. It was concluded that using urinary fosfomycin exposures, E. coli and E. cloacae isolates with MIC > 16 mg/L, and all K. pneumoniae isolates, were not reliably killed. Emergence of resistance was significant. The results challenges fosfomycin dosing and clinical breakpoints, and questions the utility of fosfomycin against K. pneumoniae. Studies in an in vitro hollow fiber model simulating plasma concentrations corroborated these findings. The results are still under consideration, but the main conclusions were that fosfomycin appeared to be effective for Wild Type E. coli but not for other species. Summarizing, all three studies indicate that resistance against fosfomycin emerges fast and that treatment for E. coli Wild Type is likely effective.
Finally, a study was undertaken in an animal model of UTI infection (Jakobsen et al., 2014). In OF-1 mice,
efficacy of 6 fosfomycin dosage regimens (total dose of 0.05 - 1.6 mg/mouse; dosing frequency of 6-72 hours) was investigated using the murine model of ascending UTI to estimate the most predictive PK/PD indices. Infection was induced by inoculating 50 μl (5x106 CFU) E. coli (NU14; FOS MIC 0.75 mg/L) via a catheter (retracted after inoculation) transurethrally and treatment was initiated sc. 1 day post infection for 3 days with FOS or saline. Urine, bladder, and kidneys were collected 4 days post infection for CFU determination. All doses significantly reduced CFUs as related to control, untreated mice (% reduction, lowest to highest doses; median control counts); urine: 80 - 100% (control mice: Log 6,8 CFU/ml; bladder: ca. 50% (control mice: Log 5,7 CFU/bladder); kidneys: 50-100% (control mice: log 2,9 CFU/kidney). The optimal PKPD indices for urine CFU/ml were (R2): AUC/MIC, 0,82; and Time>MIC of 1.dose: 0,82. Low correlations were found for Time>MIC % in all tissues, and for all indices for kidney
counts. It was concluded fosfomycin was very effective in reducing urine counts. The optimal PKPD indices included AUC/MIC ratio (> 200), orTime>MIC of 1. Dose (>2 h). These data also indicate that relatively high AUC/MIC ratio’s are required for effectiveness of fosfomycin.
The results of these in vitro studies are being considered to explain the results observed in the UTI trial comparing fosfomycin against nitrofurantoin treatment.
3.3 Minocycline
Minocycline samples were taken from 79 patients of whom 54 (68%) had full samples for both the day 1 and day 5 profiles. The concentrations were analyzed using a newly validated method and are currently under consideration.
The pharmacodynamics of minocyclin was studied in an in vitro hollow fibre model, as there was little data to support present dosing regimens and no data on the risks of emergence of resistance (Noel et al. 2013). Clinical breakpoints, both In Europe and USA, are based on historic data. A series of dose ranging studies against 4 strains of S. aureus (MIN MIG 0,19-0,5mg/L, all susceptible by EUGAST breakpoints) was performed and antibacterial effect was measured by changes in viable count over 72h and changes in population profile by recovery on MICx4 or MICx8 plates after 24, 48 and 72h exposure. The fAUC/MIC at 24h for static and -1 log reductions in viable count were 11 ,5±6, 1 and18,1 ±8, 1 respectively, For -2 log reduction in viable count, the fAUC/MIC was >200 for two strains, and 33,1 and 42,8 for the two others, fAUC/MIC targets were modestly increased after 72h being 12,9±2,9, static effect; 17,7±52, -1 log drop, 210g drop values were> 75, and 32.1 and 18.2 for the other two strains. There were no changes in population profiles indicated by growth on MIGx4 or MIGx8 plates at 24, 48 or 72h. It was concluded that a suitable fAUC/MIC target related to static -1 log drop for minocycline is 10-15. This target was not associated with any change in population profiles and would suggest a clinical
breakpoint of 0.25-0.5mg/L for a dose of 100mg 12 hrly.
3.4 Rifampicin
Rifampicin samples were taken from 79 patients of whom 54 (68%) had full samples for both the day 1 and day 5 profiles. The concentrations were analyzed using a newly validated method and being evaluated in a population pharmacokinetic model. This are currently under consideration.
3.5 Linezolid
For the Linezolid arm samples were received on 41 patients of who 30 (73%) had full samples for both the day 1 and day 5 profiles. A total of 806 samples were received and assayed. The concentrations were analyzed using a newly validated method and are currently under consideration.
3.6 Colistin
The pharmacokinetics and pharmacodynamics of colistin were an important objective in AIDA. The population pharmacokinetic (PK) analysis of colistin and its prodrug colistin methanesulfonate, CMS was determined in patients participating in the AIDA study. For patients with creatinine clearance (CRCL) > 50 ml/min a 9 MU loading dose, followed by a 4.5 MU twice daily maintenance dose, was administered as 30min infusion. Based on an earlier colistin and CMS PK model (Mohamed, Karaiskos et al. 2012) two optimal PK sampling points (45min and 10h after start of the infusion) had been identified to determine individual exposure of colistin and CMS (Kristoffersson, Friberg et al. 2015). Samples were sent to the central laboratory in batches from the participating centres. CMS A+B and colistin A+B were analyzed by LC-MS/MS. Dialysis patients, patients with obvious dose or sampling time errors, and suspected hydrolysed samples were excluded from the initial analysis. Data analysis was performed in R and NONMEM7. 237 patients were included in the initial analysis. The median concentration for the 10h samples was 4.3mg/L (SD 2.7mg/L). There was a clear relationship between creatinine clearance (CRCL, Cockcroft-Gault formula) and colistin 10h concentrations. The 25th percentiles exceeded 2mg/L except for patients with CRCL>=120mL/min. In order to describe the data the Mohamed model required adjustment by allowing an increased fraction of available CMS. The colistin and CMS PK model by Karaiskos et al (Karaiskos, et al. 2015) using CRCL as covariate adequately described the data after reestimating CMS CL. Thus, in line with Karaiskos et al higher available CMS and colistin A+B concentrations were observed compared to older studies, and resulted in colistin concentrations above 2 mg/L in the majority of patients with a CRCL lower than 120mL/min. Patients with higher CRCL may require increased maintenance dosing. The population PK model will be applied to drive a PKPD analysis linking individual colistin exposure and microbiological data to patient outcomes.
Apart from the pharmacokinetic studies a pharmacodynamic study was performed in an in vitro model of infection to determine its activity profile against K. pneumoniae. Simulating different concentration-time profiles and a range of exposures against strains with MICs of 0.5 to 4 mg/L, it was shown that the most important pharmacodynamic index associated a bactericidal effect was the AUC/MIC (R2 0.64-0.68) and that a value of at least 25 was required for a significant effect. Subsequent Monte Carlo simulations using reported data on colistin pharmacokinetics indicate that target attainment was only very moderate, even taking into account the dosing regimens currently used. It was also concluded that therapeutic drug monitoring of colistin might be warranted.
Potential Impact:
1. Main dissemination activities
The AIDA project was conceived with a work-package specifically dedicated to dissemination of results and raising awareness tot the use of off-patent antibiotics. There were four main target groups : the public in general, the health and scientific community, bio and pharma-industry and regulatory bodies. Dissemination activities were undertaken for each group.
1.1 The general public
For the general public there were a large number of press releases, including newspaper articles, background pieces and interviews both on specific issues of the project itself as well as the use of off-patent antibiotics in general. A good example is a press release of a symposium organized on off-patent antibiotics in 2013 –co-organized by dr Theuretzbacher on behalf of AIDA reads :
Conference Delegates Ask: Can Antibiotics From the Past Tackle the Growing Menace of Drug Resistant ‘Superbugs’ in the Present?
October 22: From today, Vienna plays host to a major international conference, convened by the European Society of Clinical Microbiology and Infectious Diseases (ESCMID), where researchers, academics and representatives from pharmaceutical companies and national medical agencies will explore the potential and pitfalls of using first generation, off patent antibiotics to tackle the growing global threat of antimicrobial resistance (AMR).
The three day conference, ‘Reviving Old Antibiotics’ has attracted participation from almost 300 delegates from 45 countries. They will explore the increasing threat of AMR, and the potential which exists for reverting to old and off patent antibiotics to treat today’s rapidly mutating and increasingly drug resistant ‘superbugs’.
In the past two decades AMR has emerged as one of the greatest threats to public health ever seen, with global agencies such as WHO and government institutions warning that without urgent remedial action the world is heading to a post-antibiotic age, where formerly treatable common infections will once again become fatal.
This crisis has come about as a result of a number of factors including over prescription of antibiotics; patient demand for unnecessary antibiotics to treat minor infections or viruses; unregulated access to antibiotics through pharmacies and online outlets; widespread use of antibiotics in agricultural practice to promote growth in livestock; poor hygiene practice in medical facilities; and the failure of the pharmaceutical industry to develop a new generation of drugs.
The European Centre for Disease Control estimates that 25,000 people in the EU currently die annually from infections that are resistant to multiple drugs. The economic cost associated with additional medical care for affected patients and loss of productivity is estimated at 1.5 billion Euros every year. The US Centre for Disease Control and Prevention estimates the combined cost of excess direct health care and lost productivity to the American economy to be between $55-70 billion.
“AMR threatens health care as we know it,” said Professor Gunnar Kahlmeter, ESCMID Past President.
“Many of the spectacular new practices and inventions in transplantation, cancer therapy, intensive and neonatal care and orthopaedics are at risk of having to be abandoned in those areas where multi-resistance is abundant. Once again mankind has failed in the basic ‘household management’ of one of its most precious and finest of resources.”
In the light of these challenges physicians are increasingly resorting to using old off-patent antibiotics that are still active against some drug resistant pathogens. These revived antibiotics were approved up to 55 years ago, however, and were not developed according to current standards of
science and practice. Continued use of these drugs without further study therefore carries significant downside consequences for individual patients as well as for the emergence of additional resistance.
Information about adequate dosages, indications, toxicity, and usage in special patient populations may be missing or not be adequate any more.
“We are using some old revived antibiotics now which were developed in the early penicillin era,” said Dr Ursula Theuretzbacher, founding President of ESCMID, and current Chair of its Scientific Programme Committee.
”We are potentially risking the lives of patients due to inadequate usage of these antibiotics and also losing them due to emergence of resistance.
“Strategies are urgently needed to “re-develop” these drugs according to modern standards, integrating the new knowledge into regulatory frameworks, and communicating the knowledge from research bench to bedside. The ESCMID conference in Vienna will help to communicate and add to this new knowledge.”
Areas of focus at the Vienna conference will include ‘Optimisation of Therapy’, ‘Extending the Life of Old Antibiotics’, ‘Breakpoints and Susceptibility Testing’ and ‘Regulating and Communicating New Knowledge’.
It is intended that the conference outcomes will provide the global community involved in the struggle against AMR with some practical solutions to the use of old generation antibiotics, including recommendations about priority research, identification of what knowledge gaps need to be filled and what policy changes and funding actions are required.
Other dissemination activities for the general public provided more detail., for instance how specifi off-patent antibiotics could or should be used. These releases were not only confined to the English language, for instance some of these were in german.
In addition to communication to the general public, contacts to patient advocacy and civil society groups were utilized through work within other EU funded projects (DRIVE-AB, COMBACTE-MAGNET). Patient advocacy groups are usually nationally organized and require local communication. Examples of communication with following organisations are: Doctors Without Borders, Médecins Sans Frontières Access Campaign, Gesundheitsladen Bielefeld, Brot für die Welt, Aussätzigen Hilfswerk Österreich, Deutschen Plattform Globale Gesundheit, Together 4 Diversity, Pew Charitable Trusts, Universities Allied for Essential Medicines, Association of European Academies.
1.2 The Health and Scientific community
As a scientific project, a large part of the dissemination activities were focussed on the health and scientific community. These included papers in scientific Journals, presentations at national and international meetings, organizations of workshops and symposia and newsletters.
1.2.1 Papers in Scientific Journals and presentations at meetings
The consortium has been very successful in publishing a significant number of papers in high-impact Journals. So far, 26 papers have been published, some are in press and it is expected that a considerable number of papers will follow in the next few years as more in depth analyses of the studies have been performed. Of note, the main results of the colistin versus colistin and meropenem trial will be published in the Lancet Infectious Diseases. Many of the results of AIDA where presented at international meetings such as the European Congres of Clinical Microbiology and Infectious Diseases (ECCMID) and the Interscience Conference of Antimicrobial Agents and Chemotherapy (ICAAC).
1.2.2 Workshops and Symposia
To further disseminate the current views on optimal use of off-patent antibiotics, a workshop was organized at around two thirds the project in conjunction with one of the yearly consortium meetings in Warschau in 2015. This was a co-sponsored meeting by the ESCMID and the European Pharmacology of Antibiotics Study Group (EPASG). During the three day course ‘Improving the dosage of old antibiotics’, delegates were brought up to date with current view points by members of the AIDA consortium.
Oral and poster presentations covering the AIDA topic as well as educational workshops further took place at the yearly meetings at the ECCMID (2011-2017 and the ICAAC (2011-2015). At the ECCMID, The AIDA project was regularly presented in the Network Corner of ECCMID and a poster presented.
A major event was the organization of a large symposium. This largest scientific and educational event of the AIDA project is the Conference on Reviving Old Antibiotics that took place on 22-24 October 2014 in Vienna. The Conference drew about 300 participants from 45 countries from a wide range of disciplines. Many of the world’s leading experts and researchers have accepted our invitation to actively participate in the meeting and to discuss how progress at the bench can be translated into actions in the field and highlight gaps in our knowledge. The last day of the conference was dedicated to an extensive discussion with experts and delegates who have identified gaps in our knowledge and agreed a number of key actions which need to be taken urgently.
The Vienna Conference was organized with administrative support of Europe’s largest professional interest group the European Society of Clinical Microbiology and Infectious Diseases (ESCMID), the ESCMID Study Group of Antibiotic Policies, the ESCMID PK/PD of Anti-Infectives Study Group and EUCAST. This broad alliance of professional interest groups will amplify the messages and results of the AIDA project in with ECCMID.
1.2.3 Newsletters
Newsletters with updates on the AIDA project were distributed regularly.
1.3 Bio- and Pharma Industry
It is of major importance that old antibiotics are redeveloped, and that new antibiotics are redeveloped in the context of emergence of drug resistance. A major role, in particular for the latter, is or should be played by industry. Increasing awareness and analyses of the current situation are of major importance. Some dissemination activities were undertaken in that direction.
1.4 Regulatory bodies
There were frequent communications with EMA to communicate the progress in the Field. The concept of AIDA is used as a model for “re-developing” old antibiotics. Several articles were published in peer-reviewed journals to highlight our project as a model for studying old antibiotics in academic settings. Face-to-face meetings with EMA and FDA as well as the PK/PD expert meeting at EMA showed the progress in the field, the awareness of regulators, and AIDA as a role model for “re-developing” old antibiotics. Representatives of AIDA were also present at the meeting organized by EMA in November 2015 to discuss PK/PD aspects in the (re) developments of antibiotics. These presentations were later-on used as a basis for the new PK/PD regulatory document for development of antibiotics.
2. Impact and Exploitation of Results
The concept of AIDA was used as a model for “re-developing” old antibiotics. Several articles were published in peer-reviewed journals to highlight our project as a model for studying old antibiotics in academic settings.
Face-to-face meetings with EMA as well as the PK/PD expert meeting at EMA showed the progress in the field, the awareness of regulators, and AIDA as a role model for “re-developing” old antibiotics. Extensive discussions with representatives of the extended the communication activities to the FDA who are very interested in our experiences with conducting clinical trials in severely sick patients with infections caused by extensively drug resistant Gram-negative bacteria. A presentation at the biennial meeting of the Transatlantic Task Force on Antimicrobial Resistance (TATFAR) was well received and widely discussed. F2F meetings with policy makers of several European countries (e.g. Sweden, Netherlands, Germany, Austria) to discuss the availability of old antibiotics and improve their usage enhanced AIDA’s visibility and the impact of the concept in general.
Ongoing dialogues with professional societies, especially the European Society for Clinical Microbiology and Infectious Diseases (ESCMID) and ESCMID study groups, International Society for Infectious Diseases (ISID), International Society of Chemotherapy (ISC), International Society for Anti-Infective Pharmacology (ISAP) contributed to raising awareness and to disseminate scientific findings.
In fruitful discussions with representatives of other EU funded projects (e.g. COMBACTE), with regulatory agencies (EMA and FDA), and pharmaceutical companies we could demonstrate how AIDA was successful in recruiting critically ill patients with infections caused by the most resistant pathogens though many pharmaceutical companies claim that it is not possible. Experiences gained by AIDA may influence the development of new antibiotics in a similar patient population. Our communication activities helped to broaden the discussions and provide input for development strategies of new antibiotics.
The concept of redeveloping antibiotics has a major impact on how antibiotics will be used in the future in several way. The results of the clinical trials provide evidence on the optimal use and indications of the trial drugs. For instance, the colistin trial is the first randomized controlled trial showing evidence that addition of a carbapenem is not useful, although current practice in many European countries. The scientific evidence now provided will help convince clinicians to alter their behaviour and reduce overall use of carbapenems and is thereby an important tool in reducing emergence of resistance. Furthermore, the PK/PD analyses of the trials will show how to optimize the use of the drugs and thereby increasing both efficacy and reduce adverse events, including emergence of resistance.
The PK/PD framework and supporting studies have provided further evidence of how the off-patent antibiotics should be used. The revision of the colistin SPC was aided by AIDA results and discussions. Discussions with the European Committee on Antimicrobial Susceptibility Testing (EUCAST) on the revision of clinical breakpoints have been initiated based on the AIDA results. These will lead to changes in clinical breakpoints, and there for the overall optimal use of these antibiotics in the EU, but also worldwide, as an increasing number of non-European countries are using EUCAST guidelines. The analyses of colistin and fosfomycin in particular have made or will make a significant impact.
The project has also led to collaboration with other groups. As a European consortium of scientists, there was the possibility to work closely together with a similar project funded by NIH and led by dr Keith Kaye, a collaboration that would have not been possible in an industry funded trial. This has led to significant achievements, one of which was the adjustment of the protocols to allow a combined analysis after finishing the studies. In addition, this collaboration was one of the drivers for three large international meetings on the use of Polymyxins, the third of which will take place after ECCMID, Madrid 2018. During these meetings standardization on the use, formulation and indications were discusses intensely and consensus documents have been published, e.g. the ‘Consistent global approach on reporting of colistin doses to promote safe and effective use’ in Clinical Infectious Diseases (2014) and ‘Framework for optimisation of the clinical use of colistin and polymyxin B: the Prato polymyxin consensus’ in the Lancet Infectious Diseases (2015).
3. Conclusions
Overall, the impact of the AIDA project has been and will be significant on several levels, including public, scientific, regulatory and clinical.
List of Websites:
Project public website address: http://www.aida-project.eu/
Contact details:
Prof. dr. Johan W. Mouton
Erasmus University Medical Center
Department of Medical Microbiology and Infectious Diseases (MMIZ)
P.O. Box 2040
3000 CA Rotterdam
The Netherlands
Email: j.mouton@erasmusmc.nl