Anti-inflammatory Action of Oral Clarithromycin in Community-acquired Pneumonia

Community-acquired pneumonia (CAP) is one of the most common bacterial infections and a leading cause of death globally since many patients deteriorate into sepsis and organ dysfunction. Traditional management relies on the prompt administration of antimicrobials that target the most common causative pathogens namely Streptococcus pneumoniae, Mycoplasma pneumoniae, Chlamydophila pneumoniae and Legionella pneumophila. Although there was some hesitancy in the former guidelines of the Infectious Diseases Society of America (IDSA) and of the American Thoracic Society (ATS) to suggest a clear-cut role of macrolides for the management of CAP, the new guidelines published by the ATS in 2019 clearly suggest for the management of all cases of CAP either treated as out-patients or as in-patients with a combination of β-lactam with macrolides.

The shift in the position of ATS expressed in the 2019 guidelines is coming from the growing body of evidence that the addition of a macrolide in the treatment regimen of CAP is accompanied by considerable survival benefit. This finding is mainly coming from the retrospective analysis of observational clinical studies in CAP. Results were also supported by the meta-analysis of these studies . Superiority coming from the macrolide use is mainly shown in cases of severe CAP i.e. situations with pneumonia severity index (PSI) greater than 2 that are most commonly caused by S.pneumoniae. The proposed benefit of macrolides is coming from their anti-inflammatory mode of action. This statement generates two main questions: a) what an anti-inflammatory mode of action consists of in the clinical setting; and b) is this a common property for all macrolides? Although it is easy to suggest that an antimicrobial modulates immune responses in vitro, it is extremely difficult to prove this in vivo. The main hurdle is coming from the fact that it is difficult to decipher to what extent clinical benefit is coming from the antimicrobial effect per se and to what extent this is due to modulation of the immune responses. In other terms, an anti-inflammatory effect is better shown in infections caused by pathogens that do not belong to the antimicrobial spectrum of macrolides. Our group has conducted two large scale randomized clinical trials (RCTs) where clarithromycin was co-administered intravenously along with β-lactams in either patients with ventilator-associated pneumonia (VAP) by multidrug-resistant Gram-negative pathogens or patients with severe Gram-negative infections like acute pyelonephritis, intraabdominal infections and primary Gram-negative bacteremia. The total number of patients enrolled in both studies was 800 and isolated pathogens did not belong to the antimicrobial spectrum of macrolides. Addition of clarithromycin provided overall survival benefit after 90-days in patients with VAP (57% survival versus 40% of placebo-treated comparators). Mortality by septic shock after 28-days was also considerably decreased in patients with severe Gram-negative infections (53.1% versus 73.1% of the comparators).

Although these findings point towards an anti-inflammatory mode of action of clarithromycin, they do not necessarily imply that a survival benefit similar to Gram-negative infections will apply in CAP. There is only one RCT to test the anti-inflammatory effect of clarithromycin in patients with CAP. This RCT was designed for non-inferiority and randomized patients were allocated into single β-lactam treatment or the combination with oral clarithromycin. The primary endpoint was clinical instability after seven days. Contrary to what investigators were expecting this was shown in 41.2% of non-macrolide treated patients and 33.6% of macrolide-treated patients (p: 0.070). It may be argued that if the study was powered for superiority, the study primary endpoint would have shown benefit from the addition of clarithromycin.

In recent publication coming from the research network of the Hellenic Sepsis Study Group (HSSG) 130 patients with CAP were treated with a combination of β-lactam and clarithromycin. They were compared with another 130 patients treated with a combination of β-lactam and azithromycin, with 130 patients treated with respiratory fluoroquinolone monotherapy and with 130 treated with β-lactam monotherapy. The study has a case-matching design and selection of cases of the three comparator groups were based on the group of patients treated with clarithromycin. Matching selection criteria were severity as assessed by the severity score of SOFA (sequential organ failure assessment), APACHE II (acute physiology and chronic health evaluation), PSI and CCI (Charlson's comorbidity index) and type of β-lactam. The 28-day mortality of the four groups was 20.8%, 33.8%, 32.3% and 36.2% respectively, showing a profound survival benefit with the intake of clarithromycin.

Based on the above analysis, it seems likely that an RCT that proves the attenuation of the high inflammatory burden of the host with CAP after addition of clarithromycin in the treatment regimen is missing. The need for this RCT is outscored in the recent guidelines of ATS . Such a type of RCT should take into consideration the SOFA score of the patients, the presence of the systemic inflammatory response syndrome (SIRS), the existence of elevated procalcitonin (PCT) in serum and the outcome of patients infected by macrolide-resistant S.pneumoniae. SOFA score is nowadays proposed as the sine qua non for severity. PCT more than 0.25 ng/ml is widely accepted as an index of systemic inflammation in the event of CAP to such an extent that decrease more than 80% or to levels lower than 0.25 ng/ml can be used as an index of therapy withdrawal. In such an RCT rapid resolution of the high inflammatory burden of the host should be highlighted in the achievement of the early treatment response of CAP after 72 hours that is recently appointed by the Food and Drug Administration and the European Medicines Agency as the primary endpoint goal of CAP.

This is an RCT that is aiming to prove that addition of oral clarithromycin to a β-lactam rapidly attenuates the high inflammatory burden of the host in CAP.

Inclusion criteria

• Adult patients (≥18 years)
• Male of female gender
• In case of non-menopausal women, unwillingness to become pregnant during the study period. Women of child-bearing potential will be screened by a urine pregnancy test before inclusion in the study.
• Written informed consent provided by the patients or by a first-degree relative in case of patients unable to consent
• Presence of at least two signs of SIRS (see below for definition)
• SOFA score ≥2 (see Appendix I)
• PCT ≥0.25 ng/ml
• Presence of at least two of the following signs: i) cough; ii) purulent sputum expectoration; iii) dyspnea; and/or iv) pleuritic chest pain
• Presence of CAP (see below for definition)

SIRS is defined by the presence of at least two of the following criteria:

• Core temperature >38 Celsius degrees or <36 Celsius degrees
• Heart rate >90 beats/minute
• Breath rate >20 breaths/minute or pco2<32 mmHg
• Total white blood cell count >12,000/mm3 or <4,000/mm3 or >15% bands

CAP is defined as the presence of auscultatory findings compatible with CAP and new consolidation in chest X-ray in a patient without any history of contact with the hospital environment or with health-care facilities the last 90 days.

Exclusion criteria

• Age below 18 years
• Denial of written informed consent
• Presence of infection by SARS-CoV-2 (COVID-19)
• Intake of any macrolide for the current episode of CAP under study
• Oral or intravenous intake of corticosteroids defined as any more than 0.4mg/kg daily intake of equivalent prednisone for the last 15 days
• Neutropenia defined as an absolute neutrophil count below 1,000/mm3
• Known infection by the human immunodeficiency virus
• Any chronic anti-cytokine treatment (e.g. antibodies against TNF for rheumatoid arthritis)
• Hospitalization for more than 2 days the last 90 days
• QTc interval at rest ECG ≥500 msec or history of known congenital long QT syndrome
• Concomitant administration with HMG-CoA reductase inhibitors (statins) that are extensively metabolized by CYP3A4, (lovastatin or simvastatin), and presence of any contraindications for the study drug
• Pregnancy or lactation. Women of child-bearing potential will be screened by a urine pregnancy test before inclusion in the study.