Impact of a TDM-guided ECPA Program for Optimizing Pharmacodynamic Target Attainment of Continuous Infusion Beta-lactam-based Regimen

Primary aims

• Verify if the attainment of optimal beta-lactam PK/PD targets improves the rate of microbiological eradication in OLT recipients with documented Gram-negative infections;
• Verify if the attainment of optimal beta-lactam PK/PD targets improves the rate of clinical cure in OLT recipients with documented Gram-negative infections.

Secondary aims

• To identify factors independently predicting failure in attaining early optimal beta-lactam PK/PD targets;
• To identify the relationship between the attainment of optimal beta-lactam PK/PD targets and antibiotic resistance development and post-OLT MDR colonization occurrence at 90-days;
• To identify the association between the attainment of optimal PK/PD targets for beta-lactams and variation of inflammatory biomarkers (C-reactive protein [CRP], procalcitonin [PCT], and pro-/anti-inflammatory cytokines, namely IL-6, IL-1, IL-12, IL-8, IL-10, and TNF-α);
• To evaluate the attainment of optimal beta-lactams PK/PD targets at the site of infection in case of pneumonia, intrabdominal and/or biliary infections.

Optimal beta-lactam PK/PD targets will be defined as follows:

1. For beta-lactams not coupled with beta-lactamase inhibitors (i.e., cefepime, ceftazidime, meropenem, and cefiderocol):

• Optimal: steady state concentrations (Css)/MIC ratio > 4
• Quasi-optimal: Css/MIC ratio = 1-4
• Suboptimal: Css/MIC ratio < 1

1. For beta-lactam coupled with beta-lactamase inhibitors (i.e., piperacillin-tazobactam, ceftolozane-tazobactam, ceftazidime-avibactam, and meropenem-vaborbactam):

• Optimal: a Css/MIC ratio > 4 for beta-lactam coupled with the attainment of an optimal threshold for beta-lactamase inhibitors (Css/threshold concentration [CT] ratio > 1 for tazobactam and avibactam, or AUC/MIC > 24 for vaborbactam).
• Quasi-optimal: only one of the two optimal thresholds are attained.
• Suboptimal: none of the two optimal thresholds are attained.

1. For other antibiotics administered as combination therapy in beta-lactam-based regimens (i.e., fosfomycin, tigecycline, aminoglycosides):

• Optimal: an AUC/MIC > 83 (against Enterobacterales) or > 40.8 (against Pseudomonas aeruginosa) for fosfomycin; an AUC/MIC > 6.96 for tigecycline; a peak concentration [Cmax]/MIC ratio > 10-12 for aminoglycosides associated with the attainment of optimal PK/PD target for the selected beta-lactam (refer to point a or b).
• Quasi-optimal: attainment of optimal PK/PD target for beta-lactams (refer to point a or b) coupled with suboptimal PK/PD target of selected agent for combination therapy in beta-lactam-based regimens (i.e., fosfomycin, tigecycline, aminoglycosides) or attainment of quasi-optimal or suboptimal PK/PD target for the selected beta-lactam (refer to point a or b) coupled with optimal PK/PD target of selected agent for combination therapy in beta-lactam-based regimens (i.e., fosfomycin, tigecycline, aminoglycosides).
• Suboptimal: an AUC/MIC < 83 (against Enterobacterales) or < 40.8 (against Pseudomonas aeruginosa) for fosfomycin; an AUC/MIC < 6.96 for tigecycline; a peak concentration [Cmax]/MIC ratio < 10-12 for aminoglycosides coupled with the attainment of quasi-optimal or suboptimal PK/PD target for the selected beta-lactam (refer to point a or b).

The study will be a prospective observational cohort tissutal study. All consecutive adult OLT recipients admitted to the Internal Medicine ward of the IRCCS AOUBO for the treatment of severe organ failure will be enrolled in the study at the time of liver transplant. Subsequently all OLT recipients providing informed consent will be screened for identifying who will develop a documented infection (including bloodstream infections, pneumonia, intrabdominal and/or biliary infections) caused by Gram-negative pathogens (both Enterobacterales and non-fermenting isolates) in the first 90 days after transplantation and who will be treated with beta-lactam-based regimens (namely piperacillin-tazobactam, ceftazidime, cefepime, meropenem, ceftazidime-avibactam, ceftolozane-tazobactam, meropenem-vaborbactam, imipenem-relebactam, and cefiderocol) according to a real-time TDM-guided ECPA program. Concomitant therapy with fosfomycin, tigecycline, and aminoglycosides will be allowed. Only patients with documented Gram-negative infections will be included in the study and in the analysis. This approach would allow to avoid the possible inability of some subjects to provide a valuable informed consent during ICU stay.

Patients will undergo first TDM assessment for beta-lactam and allowed concomitant antibiotic therapies at day +1 after starting therapy and subsequent reassessments at day +3, +5, and +7. Beta-lactam dosing will be adjusted according to recommendations provided by well-trained MD Clinical Pharmacologists in a real-time clinical pharmacological advice.

It is noteworthy that the optimization of antimicrobial dosing according to a real-time TDM-guided ECPA represents the standard clinical practice at the IRCCS Azienda Ospedaliero-Universitaria of Bologna, with the implementation in the last three years of an hospital-wide program, as also published [14,29]. Patients included in the study will be treated according to current clinical practice, which entails the choice of antibiotic therapy at the discretion of intensive care physicians, internists, or infectious disease consultants independently from the inclusion in this study.

Concomitant assessment of inflammatory biomarkers (i.e., C-reactive protein, procalcitonin, IL-6, IL-1, IL-12, IL-10, IL-8, and TNF-alfa) will be performed at the same timepoints of TDM samples. For patients treated with anti-Gram-negative combination regimens, TDM assessments and associated dosing adjustments will be performed also for fosfomycin, tigecycline, and aminoglycosides.

In case of pneumonia, intrabdominal, and/or biliary infections, concomitant samples will be collected from bronchoalveolar lavage, abdominal fluid, and/or Kehr's tube for determining beta-lactams and/or other antibiotics concentrations at the site of infection when feasible.

For each patient two different PK/PD indices will be calculated and analyzed: the attainment of the early optimal PK/PD target, assessed at day +1, and the attainment of the overall optimal PK/PD target, calculated by means of average beta-lactam and/or other antibiotics exposure at the different timepoints.

Study population All consecutive adult OLT recipients admitted to the Internal Medicine ward of the IRCCS Azienda Ospedaliero-Universitaria di Bologna for the treatment of severe organ failure who will be scheduled for liver transplantation will be enrolled in the study at the time of liver transplant, when they will be able to understand and sign the informed consent.

According to records of the post-transplant ICU and of the Internal Medicine ward for the management of severe organ failure of the IRCCS Azienda Ospedaliero-Universitaria of Bologna, approximatively 120-125 orthotopic liver transplants are performed per year, with a prevalence of early infection of 21% [3]. Consequently, we will expect to include 60-65 OLT recipients in the two years of the study.

When the sample size is 58, a logistic regression of a binary response variable, microbiological eradication (Y), on a binary independent variable, attainment of optimal beta-lactam PK/PD targets (X₁), will have 80,57% power to detect a change in the probability that Y=1 from a value of 0,43 at baseline to 0,90, assuming the R² from the regression of X₁ on other Xs is 0,2, 85% of samples are in group X₁=1, and that the two-sided test is made at the 5% level.

When the sample size is 62, a logistic regression of a binary response variable Y (attainment of clinical cure), on a binary independent variable, X₁, will have 80,22% power to detect a change in the probability that Y=1 from a value of 0,30 at baseline to 0,80, assuming the R² from the regression of X₁ on other Xs is 0,2, 85% of samples are in group X₁=1, and that the two-sided test is made at the 5% level.

Given the shortness of inpatients' follow-up, it was assumed that patients' drop-out will not occur. Both power analyses were carried out using nQuery 9.3.1 (ROT9 procedure).

The study will be composed by:

• a prospective recruitment period of 2 year after Ethics Committes' approvals and directorates' permissions to conduct the study (in any case not later than 30th June 2026);
• Follow-up period of 90-days from the occurrence of documented Gram-negative infections;
• Data collection and analysis, that will last 3 months For an overall duration of approximately 30 months Patients will undergo first beta-lactam TDM assessment at day +1 after starting therapy and subsequent reassessments at day +3, +5, and +7. Beta-lactam dosing will be adjusted according to recommendations provided by well-trained MD Clinical Pharmacologists in a real-time clinical pharmacological advice. Concomitant assessment of inflammatory biomarkers (i.e., C-reactive protein, procalcitonin, IL-6, IL-1, IL-12, IL-10, IL-8, and TNF-alfa) will be performed at the same timepoints of TDM samples. The assessment of cytokine serum levels will be performed at Unique Metropolitan Laboratory in Bologna. For patients treated with anti-Gram-negative combination regimens, TDM assessments and associated dosing adjustments will be performed also for fosfomycin, tigecycline, and aminoglycosides.

Notably, antibiotic TDM assessments will be performed by using the same blood sample collected for the routine daily monitoring of these patients, thus no additional blood samples will be collected.

In case of pneumonia, intrabdominal, and/or biliary infections, concomitant samples will be collected from bronchoalveolar lavage, abdominal fluid, and/or Kehr's tube for determining beta-lactams and/or other antibiotics concentrations at the site of infection when feasible. Notably, bronchoalveolar lavage represents the standard-of-care for the diagnosis of bacterial pneumonia and will be performed by ICU physicians according to clinical practice and independently from the inclusion of the patient in the study.

Notably, the proposed timepoints in which antibiotic TDM assessment will be performed are consistent with those currently adopted for standard clinical practice in special populations, such as OLT recipients (i.e., first assessment at 24 hours after starting antibiotic therapy at further reassessment every 48 hours). The same process is currently adopted also for the assessment of serum levels of common inflammatory biomarkers (i.e., C-reactive protein and procalcitonin).

In regard to the assessment of serum levels of pro/anti-inflammatory cytokines (i.e., IL-6, IL-1, IL-12, IL-10, IL-8, and TNF-alfa), these laboratory tests are not currently routinely performed, but only requested in specific cases at discretion of treating physician. However, the assessment of serum levels of pro/anti-inflammatory cytokines will be performed by using the same blood sample collected for the routine daily monitoring of these patients, thus no additional blood samples will be collected. Furthermore, the costs required for determinations of serum levels of pro/anti-inflammatory cytokines will be covered by dedicated funds included in the budget of the project.