Combination Therapy of Medications from the Perspective of PK / PD

9 January, 2020

Professor Zhang Jing

Chief Pharmacist

Institute of Antibiotics, Huashan Hospital, Fudan University


Key Laboratory of Clinical Pharmacology of Antibiotics, National Health Commission


Q1: What do you think of the significance of pharmacokinetic/pharmacodynamic (PK/PD) theory for clinical antibacterial regimen under the situation of rapid changes in bacterial resistance and serious lag in the development of new antibacterial drugs?

A1: First, PK/PD theory should be applied through the first, second and third phase of the clinical trial for new antibacterial drug registration, and the screening of the appropriate regimen, including dose, interval and route of administration. After the antibacterial drugs are approved and applied in clinical practice, pharmacist will adjust the regimen according to the different PK characteristics of the local bacteria in terms of their sensitivity and resistance to antimicrobial drugs and the physiological and pathological conditions of patients (such as ICU patients, CRRT patients, etc.); so, the PK/PD theory has significance for the formulation of the regimen of the antibacterial drugs to achieve the maximum bactericidal activity in vivo.


Q2: What do you think is the relationship between breakpoint and PK/PD and why breakpoint has been changed in recent years?

A2: First, the formulation of drug susceptible breakpoint includes three parts; 1. Epidemiological cut-off values, to distinguish the minimum inhibitory concentration of flora with and without acquired resistance / mutant resistance mechanism, which is usually the upper limit of the minimum inhibitory concentration of wild flora; 2. PK/PD cut-off values: According to the PK / PD index and target value which are most related to the clinical efficacy, the upper limit of MIC distribution range when the probability of reaching the standard exceeds 90% is simulated1,2; 3. The clinical cut-off values are also used to distinguish the pathogens with good prognosis from those with failure in treatment. Due to the change and development of drug resistance, the breakpoint needs to be modified. 1. Modification of the breakpoint can be more in line with the current development of bacterial drug resistance and the actual clinical situation to facilitate the selection of reasonable treatment drugs. 2. The modified breakpoint can better show the real efficacy of antibiotics in the treatment of infection caused by sensitive bacteria with the currently recommended treatment protocol. 3. With the deepening of the research and understanding of clinical response and PK parameters, the breakpoint will be more reasonable for different types of infection, infected population and drug delivery mode.3


Q3: How does the current clinical PK/PD target value of commonly used antibiotics formulated?

A3: In the target indication patient population, the method of sparse point blood sampling is used to carry out the study of group PK (PPK) of antibiotics in the infected patients, and the PPK model is established. The PK/PD index of the patient is calculated according to the MIC of the pathogens of the patient, and the quantitative effect relationship model is established with the clinical efficacy (cure or failure) of the group of patients to determine the clinical PK/PD target value. At present, there is no large clinical trials or meta-analysis on PK/PD and its target value of commonly used antibacterial drugs in China, while data from Europe, America or Japan are not necessarily suitable for Chinese population.4


Q4: Although single drug antibacterial is the best choice for clinical treatment, it is inevitable to use combination therapy. How do you think the PK/PD theory can suggest the formulation of concomitant medication treatment protocol?

A4: combination therapy of clinical antibacterial drugs is also based on PK / PD theory, and relevant indicators include the minimum inhibitory concentration, fractional inhibitory concentration index, the number of colony count decreased with the time change of the concomitant medication in vitro and in vivo dynamic bactericidal concentration, the role of biofilm, etc. In addition, the PK of drugs in vivo will be affected by drug interaction, action of drug metabolizing enzyme of liver, liver and kidney metabolism and elimination mode after the concomitant medication.2,4


Q5: How does the bacteriostatic index of fractional inhibitory concentration index (FICI), biofilm action and concentration in vivo play a role in clinical treatment?

A5: When FICI≤0.5, it indicates synergistic effect. When 0.5<FICI<1, it is partial synergistic effect and when FICI=1, it is the summing effect. Taking levofloxacin and polymyxin as examples, studies have shown that FICI of the two drugs =0.37, which suggests that the combination of the two drugs is highly sensitive to bacteria.5 The concentration in vivo is also different due to the drug concentration reaching different tissue. Taking levofloxacin as an example, after intravenous injection of 7mg/kg in rat, the serum AUC0-∞ (μg·h/mL) =15.6±2.6, while the lung AUMC0-∞ (μg·h2/mL) =41.5±19.2, and the prostate AUMC0-∞ (μg·h2/mL) =83.2±58.2. Therefore, it can be expected to have a good therapeutic effect on pulmonary infection and prostatitis, which suggests that we should not only look at the blood concentration, but also pay attention to the drug concentration at the infection target organ when we choose combination therapy regimen.6


  1. 李耘, 郑波, 吕媛, 等. 新抗菌药物临床试验折点制定方案专家共识[J]. 中国临床药理学杂志, 2015(11).
  2. 周晓坤, 马佳, 姚冬梅, 等. 合理利用药敏试验选择抗菌药[J]. 中国合理用药探索, 2013, 10(4):50-52.
  3. 何礼贤. 抗生素体外药敏试验敏感度折点的修订及其临床意义[J]. 微生物与感染, 2012, 7(2):132-132.
  4. 中国医药教育协会感染疾病专业委员会. 抗菌药物药代动力学/药效学理论临床应用专家共识[J]. 中华结核和呼吸杂志, 2018(6):409-446.
  5. Wei W , Yang H , Hu L , et al. Activity of levofloxacin in combination with colistin against Acinetobacter baumannii: In vitro and in a Galleria mellonella model[J]. Journal of Microbiology, Immunology and Infection, 2015:S168411821500907X.
  6. Zimmermann ES, et al. Antimicrob Agents Chemother. 2015 Nov 30;60(2):946-54.