The general Context

Antibacterial drugs are facing increasing limitations in terms of effectiveness not only due to emergence of high-level resistance but also due to a moderate but significant decrease in bacterial susceptibility (low level resistance). These expose the patients to a risk of subtherapeutic dosages that explains treatment failures and further emergence of drug resistance (EDR). To mitigate these risks, clinicians tend to increase drug dosages and/or to combine antibiotics [1-3], which then expose the patients to potential toxicity while not necessarily minimizing efficiently EDR.

One obvious way to improve this situation is to refine clinical treatment interventions, in terms of dosages and duration. Real-time Therapeutic Drug Monitoring (TDM) of the antimicrobials in individual patients appears appealing. This approach should allow fine-tuning drug regimens (dosages and schedules of administration) to meet patient-specific pharmacological requirements for activity (pharmacodynamics) while, at the same, decreasing the risk of emergence of drug resistance (EDR) and controlling toxicity (toxicodynamics).

In the case of aminoglycosides and vancomycin, TDM (although not real time) has been successfully used since several years mainly to mitigate toxicity [4]. But, quite surprisingly, no routine drug monitoring is available for β-lactams [4], even though these represent the most widely used class of antimicrobials in both the hospital and the community setting.

Improving treatment approaches with β-lactams is particularly needed in nosocomial infections where decreased susceptibility of the etiological organisms is observed worldwide. This is of main concern in Hospital-Acquired Pneumonia [HAP] (including Ventilator-Associated Pneumonia [VAP]), defined as an infection contracted by a patient in a hospital at least 48–72 hours after being admitted. Despite recent advances in antimicrobial therapy, better supportive care and a wide range of prevention measures, HAP/VAP remains a significant cause of patient morbidity and mortality as well as health care costs. Together with Staphylococcus aureus, Gram (-) bacilli such as Enterobacteriaceae and Pseudomonas aeruginosa are the predominant organisms responsible for this infection. Severe underlying illnesses (eg. immunodeficiency or diabetes mellitus) predispose patients to the development of pneumonia and significantly increase mortality whereas patients with extremely severe conditions but no underlying disease are at lower risk.

β-lactams (penicillins, cephalosporins, and carbapenems) used alone or in combination with other antibiotics remain at the cornerstone of the current treatments in HAP/VAP patients. However, the underlying illnesses have a marked impact on β-lactams distribution properties and on patient’s excretory function [5], which creates much variability in their pharmacokinetics and the corresponding blood levels.

Thus within the specific HAP/VAP patients groups, there is a need for patient stratification and for individualized treatment with β-lactams. Failure to achieve this at an early stage of the treatment is probably a main cause of further failure with important consequences in terms of duration of treatment (with all economic consequences) and emergence of resistance.

Paediatric patients, and especially new-borns, represent another group of patients where pharmacokinetics are highly variable and where the applicability of adult PK/PD targets is still not well established [7]. There is a need in dose adaptation based on patient’s weight, body surface area and clinical maturation and developmental pharmacology [8]. Systematic monitoring is also complicated since collecting sufficient volume/number of samples is difficult in paediatric patients [9].

The most frequent administration route used for treating HAP/VAP patients with antibiotics is intravenous, with, for β-lactams, a majority of patients receiving intermittent administrations (e.g., one dose every 8h) although prolonged and continuous infusion are gaining popularity [6]. Adjusting treatment dosages and schedules to the specific patient’s conditions using pharmacokinetic models requires knowing both the drug volume of distribution (VD) and its total (renal and non-renal) clearance. But those are markedly perturbed in severely-ill patients and subject to rapid changes, which makes predictions based on population models imprecise. Thus, according to both the American Thoracic Society and several EU Societies, the treatment of HAP/VAP should be initiated empirically with the antibiotics used most often at the maximal registered dosages followed by de-escalation if possible. Most clinicians follow, in their daily practice, a “best guess” approach based on their own experience for deciding the most appropriate dosage regimen and the duration of the treatment.

This is most unfortunate because pharmacokinetic-pharmacodynamic [PK-PD] indices (drug exposure ensuring an optimal effect) of β-lactams are well known [10]. β-lactam antibiotics show a “time-dependent pattern” of antibacterial efficacy. Thus, the time during which the free drug concentration (unbound fraction) of the drug remains above the MIC (Minimum Inhibitory Concentration) is the dominant PK/PD index associated with bacterial killing (fT>MIC). Although the MIC of the causative organism may not be known before 24-48 h after isolation, its value can be estimated by using local epidemiological data. Alternatively, the clinician may use the value of the EUCAST clinical susceptibility breakpoints for the corresponding antibiotic/bacteria combination to guide dosing. During treatment, the clinician may also obtain new values of the MIC, which should help in further adjusting the dosage and the schedule of administration. These efforts, however, are largely useless if the clinician does not have the correct information concerning the actual free blood levels of β-lactams which are often unpredictable at the individual patient’s level. Knowledge of actual blood levels is therefore critical to ensure efficacy.

Optimal dosing is also a key parameter concerning emergence of drug resistance (EDR) and prevention of adverse side effects.

With respect to EDR, research made in the past decades collectively indicates that effective antimicrobial therapy cannot be solely relied upon drug potency or pathogen susceptibility but is a complex interplay of both factors. Firstly, by using standard treatment, infections due to less susceptible pathogens (higher MIC) are more difficult to treat than anticipated and clinical outcomes could be compromised as a result [11]. Secondly, the coexistence of subpopulations with decreased susceptibility increases the probability of these subpopulations surviving under inadequate antibiotic exposure, and eventually acquiring additional resistance mechanisms that increase resistance to higher levels [12]. Low to intermediate level resistance can be mitigated by using more aggressive dosing that achieves a higher PK/PD indices. Thus, monitoring and adjusting β-lactam blood levels to meet drifts of MICs of etiological micro-organisms may help reducing the probability of resistance emergence. However, until recently, monitoring β-lactam blood levels was considered unnecessary. For many years, bacteria could indeed be divided into “highly susceptible” or “fully resistant” ones (such as is the case for bacteria expressing β-lactamase(s) compared to their isogenic wild type strains). Now, the increasing recognition of the role played by low-level resistance mechanisms (causing moderate increases in MIC compared to wild type populations) has largely modified the situation, because it markedly reduces the efficacy margin of β-lactams that clinicians are using. Lastly, the studies that defined optimal PK/PD indices of β-lactams used short exposure times (usually 24h) which is not appropriate to evaluate the effects of drug exposure on resistance suppression. Longer duration of observation with repeated dosing of free β-lactam appears necessary to assess the influence of the dosage regimen on EDR. Obviously, this also applies to the clinical environment where treatment duration is usually for several days and where MICs of causative organisms often increase during treatment (MIC drift)[13]. This further strengthens the need to readjust β-lactams blood levels on a regular basis.

With respect to prevention of side effects, β-lactams have long been known to cause neurological disturbances (mainly convulsions) associated with their penetration in the central nervous system (CNS) [14]. While variable amongst β-lactams, and critically dependent on the permeability of the blood-brain barrier of individual patients, these effects have been clearly associated with elevated blood levels [14]. Additionally, achieving adequate drug concentrations at the site of infection is an ongoing concern, imposing higher doses to enhance target site penetration for deep tissues that may also trigger toxicity.

To sum-up, β-lactams TDM has a potentially crucial role in managing critically-ill patients such as HAP/VAPs leading to treatments that meet efficacy while controlling the EDR and adverse side-effects, at the level of the individual patient.

The implementation of β-lactams TDM is expected to represent an innovative clinical approach compared to both the empirical practices and the current “best-in-class” population-based PK-PD, none of them being able to meet specific needs of individual patients [15,16].

Actually, what is still missing today for really implementing β-lactam TDM is the possibility for the clinicians to obtain a rapid assessment of drug levels. All methods available so far rely on complex technologies (mainly HPLC and LC-MS-MS) that take several hours before results can be communicated. Since the patient’s situation is quickly changing over time, results that come late tend to be ignored. It is essential to provide the clinician with grounds for dosage adjustments as early as possible in the treatment process and at the level of the individual patient.

To this end, there is a clear unmet need to provide the clinicians with a direct, rapid (real-time) information about free β-lactam actual blood levels both at initiation and during therapy.


The concept of MON4STRAT

Our hypothesis is that the β-lactam-based treatment of critically ill patients such as HAP/VAPs should be improved provided there is an approach to track from the very beginning of the treatment:

  • patient-specific deviations of free β-lactam blood levels in comparison to population-based PK-PD targets ;
  • the risk of EDR alongside corresponding MICs increase;
  • the risk of potential (neuro)toxicity;

and, accordingly, to re-adjust the dosage in quasi-real time.

We support that this is only achievable provided clinicians have access to a user-friendly, cost-effective, rapid and specific method of accurately determining the actual free (unbound) β-lactam concentration in small volumes of a particular patient’s blood sample.

Accordingly, the concept of the MON4STRAT Project is to develop a novel and more rational approach to the treatment of HAP/VAP patients that combines (i) the knowledge of the β-lactam blood levels acquired on a real-time basis for individual patients, with (ii) best-in-class PK-PD and EDR models gained from population studies and optimized for minimizing EDR and adverse effects.

We propose to demonstrate that such approach will allow:

  • to gain a better understanding on the inter-relationships between efficacy, individual exposure-response, resistance and adverse effects;
  • to develop reliable knowledge-based treatment algorithms helping practitioners to administer to the patient the right dose on an individual basis;
  • to improve efficacy for new as well as old β-lactams (alone or in combination) while potentially reducing treatment duration and curbing EDR.

Rapid determination of actual β-lactam blood levels is key to the success of this approach.


Preliminary information

In this context, and within the framework of a nationally funded Project (Belgium, Walloon Region), ULG, UCL and WOW, 3 partners of this MON4STRAT Project, have developed a patented assay [European patent application no. EP11185288.5, priority date: 14/10/2011] that allows the highly specific, quantitative, real-time (5 minutes) measurement of non-protein bound (free) β-lactam concentration(s) in blood sample(s) (i) within a microbiologic, therapeutic and toxicologic meaningful range (typically 1 to 500 mg/L), (ii) with an operational instrument that can be used either at the bed-side or in the laboratory, (iii) capable of assaying anti-Gram(+), anti-Gram(-) and broad-spectrum β-lactams and their combinations (with specific assay for each of them), (iv) intrinsically insensitive to non-β-lactam drugs and, therefore, with a low potential for interference by other medications (including aminoglycoside and/or fluoroquinolone antibiotics commonly co-administered with β-lactams) or endogenous constituents observed in abnormal concentrations in sick patients. The instrument has demonstrated its ability for specific and reproducible assay of the main β-lactams used for the treatment of HAP/VAP using samples spiked with the corresponding antibiotics.

What is needed beyond the preliminary data

Even if the method and the accompanying instrumentation represent an interesting technical achievement opening the route to β-lactam TDM at patient-specific level, this is not enough.

To be useful in the clinical practice, especially for critically ill patients such as HAP/VAPs, and for offering dose readjustment options tailored to the patient, additional efforts are needed in terms of the following developments and deliverables:

  • (i) Final customization of assay procedures to the actual conditions of bed-side use as encountered in daily practice within hospital units caring for HAP/VAP patients;
  • (ii) Monitoring and managing the deviations between the actual blood levels that will be observed in patients and the PK/PD targets related to efficacy, minimization of EDR and prevention of toxicity (fT>MIC, Cmax, Ctrough, schedule of administration) and their corresponding thresholds;
  • (iii) Using this information as evidence-based data together with algorithmic models for deciding how to optimize dosages and schedules during the treatment.

This is why the MON4STRAT consortium brings together experts in complementary disciplines and fields. Thus, biochemists and experimental pharmacologists (at the origin of the patented assay; ULG and UCL) and engineers (responsible for the construction and operation of the device; WOW) will closely collaborate with (a) PK-PD experts (UCL, UHOUS, EXPRIMO), antimicrobial resistance experts (UHOUS, SERMAS) and clinicians (ICAN, SERMAS, ULB, UTARTU, UDSL, UHOUS) to launch an integrated program.

Because the approach is based on a rapid delivery of practical information to the clinician for facilitated decision, it represents a major breakthrough in the field. It will also directly address current and future European Public Health needs. If successful, the approach developed here for β-lactams could be applied to other anti-infective drugs and allow for a paradigm shift in the treatment of patients.



(PM: PubMed Identifying Number giving immediate access to the corresponding abstract (and full text if available))
1. Nicolau DP. Pharmacodynamic optimization of beta-lactams in the patient care setting. Crit Care 2008; 12 Suppl 4, S2. PM:18495059.
2. Tamma PD, Cosgrove SE, Maragakis LL. Combination therapy for treatment of infections with gram-negative bacteria. Clin Microbiol Rev 2012; 25, 450-70. PM:22763634.
3. Kanj SS, Kanafani ZA. Current concepts in antimicrobial therapy against resistant gram-negative organisms: extended-spectrum beta-lactamase-producing Enterobacteriaceae, carbapenem-resistant Enterobacteriaceae, and multidrug-resistant Pseudomonas aeruginosa. Mayo Clin Proc 2011; 86, 250-9. PM:21364117.
4. Roberts JA, Norris R, Paterson DL et al. Therapeutic drug monitoring of antimicrobials. Br J Clin Pharmacol 2012; 73, 27-36. PM:21831196.
5. Udy AA, Roberts JA, De Waele JJ et al. What’s behind the failure of emerging antibiotics in the critically ill? Understanding the impact of altered pharmacokinetics and augmented renal clearance. Int J Antimicrob Agents 2012; 39, 455-7. PM:22483562.
6. Walker MC, Lam WM, Manasco KB. Continuous and extended infusions of beta-lactam antibiotics in the pediatric population. Ann Pharmacother 2012; 46, 1537-46. PM:23115223.
7. Lutsar I, Metsvaht T. Understanding pharmacokinetics/pharmacodynamics in managing neonatal sepsis. Curr Opin Infect Dis 2010; 23, 201-7. PM:20179595.
8. Tayman C, Rayyan M, Allegaert K. Neonatal pharmacology: extensive interindividual variability despite limited size. J Pediatr Pharmacol Ther 2011; 16, 170-84. PM:22479159.
9. Mahmood I, Duan J. Population pharmacokinetics with a very small sample size. Drug Metabol Drug Interact 2009; 24, 259-74. PM:20408503.
10. MacGowan A. Revisiting Beta-lactams – PK/PD improves dosing of old antibiotics. Curr Opin Pharmacol 2011; 11, 470-6. PM:21862409.
11. Esterly JS, Wagner J, McLaughlin MM et al. Evaluation of clinical outcomes in patients with bloodstream infections due to Gram-negative bacteria according to carbapenem MIC stratification. Antimicrob Agents Chemother 2012; 56, 4885-90. PM:22777044.
12. Tam VH, Ledesma KR, Schilling AN et al. In vivo dynamics of carbapenem-resistant Pseudomonas aeruginosa selection after suboptimal dosing. Diagn Microbiol Infect Dis 2009; 64, 427-33. PM:19631096.
13. Riou M, Carbonnelle S, Avrain L et al. In vivo development of antimicrobial resistance in Pseudomonas aeruginosa strains isolated from the lower respiratory tract of Intensive Care Unit patients with nosocomial pneumonia and receiving antipseudomonal therapy. Int J Antimicrob Agents 2010; 36, 513-22. PM:20926262.
14. Miller AD, Ball AM, Bookstaver PB et al. Epileptogenic potential of carbapenem agents: mechanism of action, seizure rates, and clinical considerations. Pharmacotherapy 2011; 31, 408-23. PM:21449629.
15. Sime FB, Roberts MS, Peake SL et al. Does Beta-lactam Pharmacokinetic Variability in Critically Ill Patients Justify Therapeutic Drug Monitoring? A Systematic Review. Ann Intensive Care 2012; 2, 35. PM:22839761.
16. Roberts JA, Joynt GM, Choi GY et al. How to optimise antimicrobial prescriptions in the Intensive Care Unit: principles of individualised dosing using pharmacokinetics and pharmacodynamics. Int J Antimicrob Agents 2012; 39, 187-92. PM:22226651.