Efficacy and effectiveness of Ceftaroline Fosamil in patients with pneumonia: a systematic review and meta-analysis

9pt?>Experimental and observational studies aimed to evaluate the efficacy/effectiveness of ceftaroline fosamil in adult hospitalized patients and outpatients with a diagnosis of pneumonia, including CAP, HAP, VAP, and HCAP, were selected. The search was conducted in three electronic databases: PubMed, Scopus, and Cochrane Central Register of Controlled Trials, without any time restrictions. Only publications written in English language were selected. Several key-words, combined using different strings according to the electronic database protocols, were used: “Ceftaroline”, “Ceftaroline Fosamil”, and “Respiratory Infections”. To increase the search sensitivity, the list of references of the selected articles, as well as published systematic or narrative reviews, were manually and carefully assessed to include manuscripts not cited in the search record lists. Abstracts of the main pulmonology, infectious diseases, or microbiology conferences were not searched based on the poor information they could provide on the selection criteria and on the main clinical and bacteriological findings. Furthermore, non-peer-reviewed articles of the grey literature were not considered based on their poor clinical and methodological reliability.

Only articles clearly describing the primary objective of this systematic review, i.e. efficacy/effectiveness of Ceftaroline fosamil in patients with pneumonia (CAP, HAP, VAP, and HCAP) were selected. At least one of the following efficacy/effectiveness-related outcomes were considered: 1) number of responders/number of subjects in group at day 4 after initiating therapy; 2) cure rate at the end of therapy (EOT); 3) cure rate at the test of cure (8–15 days after the end of therapy, TOC); 4) 14-day clinical success/cure (±1 day) from the diagnosis of pneumonia. The assessment of the outcomes and, then, the suitability of the article was carried out during the evaluation of the abstract or the full-text article if the information was not stated in the abstract. Studies were included if adult (≥18 years of age) patients, recruited in the ceftaroline fosamil or in the control arm, were at least 20.

The following exclusion criteria were adopted for the searched records: 1) papers written in languages other than English; 2) narrative or systematic reviews and meta-analyses; 3) abstracts presented in national and international conferences; 4) editorials, research letters, commentaries, correspondences; 5) case-reports or –series; 6) manuscripts focused only on efficacy/effectiveness of ceftaroline fosamil in infections other than respiratory.

Safety and tolerability profile of ceftaroline fosamil included only the collection of the adverse events.

Records were independently assessed by two researchers (LS and FT). They carefully evaluated titles and contents of the abstracts. In case of potential interesting articles, they retrieved and assessed the full-text. Inconsistencies during the first and second phase were solved by a third and senior reviewer (GS), who supervised the entire selection and review process.

Qualitative and quantitative data were extracted by the same reviewers (LS and FT) who selected the articles. Collection of the variables was decided during the preparation of the study protocol, as well as the implementation of an ad-hoc standardized form in an Excel format (Microsoft Office). Disparity during data collection was solved by a third reviewer (GS) by consensus. However, the final inter-rater agreement was approximately 100%. A random cross-check was carried out for ~ 20% of the selected citations.

The following variables were collected: response rate at day 4, response rate EOT, response rate TOC, 14-day clinical cure, publication year, epidemiological study design, country/ies where the study was carried out, study period, sample size (total, ceftaroline and control arm), age, gender, ethnic origin, severity of pneumonia, including the Pneumonia Severity Index (PSI) [2], lobar infiltration, pleural effusion, parenchymal or airway disease, previous episodes of pneumonia or bacteremia, previous exposure to antibiotics, asthma, etiology (i.e., S. pneumoniae, MSSA and MRSA), and adverse events. According to the Italian law on epidemiological studies based on anonymous and aggregated data, no ethical clearance was requested to the ethical committees of Milan and Sassari, Italy.

The systematic review and meta-analysis was carried out following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [8]. No significant inconsistencies and disparities were detected in the selection and data extraction phases. The agreement between LS and FT was higher than 97% and incongruences were solved by the intervention of GS.

A descriptive analysis of qualitative and quantitative variables was performed using proportions and central tendency/variability indicators (i.e., mean and standard deviation, SD), respectively. A meta-analysis was conducted for the efficacy/effectiveness-related outcomes. Forest plots were adopted to show the characteristics (i.e., between-study variability and sample size) of the single outcomes in comparison with the pooled estimates. Point and interval (95% confidence intervals, CI) estimates were used for studies’ and pooled outcomes. The inconsistency (I²) indicator was computed to show the variability across studies and was statistically tested with the chi-squared test for heterogeneity. Fixed or random-effects models were implemented according to the assumption that the true effect is or is not the same in all studies, respectively. Stratified analyses were conducted following the type of pneumonia (i.e., CAP, HAP, VAP, HCAP) or the etiology (i.e., S. pneumoniae, MSSA and MRSA). A two-tailed p-value less than 0.05 was considered statistically significant. The statistical software STATA version 14 (STATACorp LP, College Station, TX 77845, USA) was used to perform all statistical computations.

The search of the three electronic databases found 2364 records (Fig. 1). After the removal of duplicates, 999 citations were screened and only 14 were considered suitable for a qualitative and quantitative analysis.

Six [9‐14] (42.9%) studies were clinical trials, published in the time period 2010–2015, and 7 [15‐21] (50.0%) were observational retrospective studies (6, 85.7%, cohort studies and 1, 14.3%, case-control study), published between 2014 and 2016 (Table 1). Only one [22] (7.1%) study was a retrospective analysis of previous clinical trials. In the majority of the cases (13/14, 92.9%), studies were carried out from 2007 to 2014 in USA; only the clinical trial of Zhong et al. [14] was conducted in five Asian countries. The efficacy/effectiveness of the ceftaroline fosamil arm was compared with that of a control group in 8 (57.1%) studies [9‐15, 22]; in the remaining 6 studies [16‐21] no comparators were chosen. The dosage of ceftaroline fosamil was the same across all 14 studies [9‐22] (i.e., 600 mg every 12 h). The ceftriaxone was the most frequently (7/8, 87.5%) prescribed antibiotic in controlled studies (Additional file 1: Table S1).

The sample size of the studies with a control arm ranged from 45 to 1153 patients; in particular, the ceftaroline fosamil arm ranged from 23 to 580 patients, whereas the control arm sized from 22 to 573 (Table 2). The mean (SD) age of the ceftaroline and the control group was ~ 60 (15) years, ranging from 58.8 (16.1) to 66.1 (14.7) years and from 58.8 (16.4) to 65.8 (13.9) years, respectively. The proportion of males was higher than 60%, both in the ceftaroline and in the control arm, in all studies, with the only exception of that of Arshad et al. [15], where the percentage of males was ~ 50% in both treatment groups. Three (37.5%) studies [9, 12, 22] did not describe the ethnic origin of the cohort; 3 studies [10, 11, 15] showed a highest (> 80%) prevalence of white patients in both arms, whereas the studies of Zhong et al. [14] and Arshad et al. [15] had a high proportion of Asian and black patients, respectively.

The PSI was heterogeneous (Table 3); the majority of the patients were diagnosed as risk class III or IV. Only the study of Arshad et al. [15] did not provide a pneumonia severity classification. Half of the patients in the ceftaroline fosamil and control arm were diagnosed as PSI risk class III. However, the study of Jandourek et al. [9] recruited two third of the patients diagnosed as PSI risk class IV. A description of a multi-lobar infiltration, as well as of a pleural effusion, was performed by two (25.0%) studies [9, 12]. At least one chronic parenchymal or airway disease (including COPD, bronchiectasis, and interstitial fibrosis) affected a proportion of patients ranging from 20.0 to 33.2% per single arm. Asthma was described only by four (50.0%) studies [10, 11, 13, 14] and was found in less than 10% of the patients. Previous episodes of pneumonia were recorded in one fifth of the treatment group; however, this information was provided only by three (37.5%) studies [10, 11, 13]. Four (50.0%) studies [10‐12, 22] found that 1/3–1/2 of the cohort was previously treated with antibiotics. The prevalence of bacteremia was very low (< 10%) per single arm in 4 (50.0%) studies [10, 11, 13, 14]. Only the study of Jandourek et al. [9] found a bacteremia prevalence of 100%.

Six [16‐21] out fourteen (42.9%) studies evaluated the efficacy/effectiveness of ceftaroline fosamil without a control arm. The sample size was heterogeneous, from 21 to 528 patients (Table 4). The mean (SD) age was high, ranging from 60 (18) to 64.3 (1.7) years. The proportion of males ranged from 48.3 to 57.5%.

The prevalence of patients with chronic parenchymal or airway diseases was significantly higher if compared with that found in the cohort of patients recruited in controlled studies (40.7–47.5%) (Table 5). Furthermore, in three out four studies (75%) [16‐18] one fifth of those treated with ceftaroline fosamil suffered of congestive heart failure. However, the percentage of previous episodes of pneumonia was similar (range: 24.6–25.4%). Three (50.0%) studies [16, 18, 19] described previous antimicrobial exposure, which was higher than 82%.

The overall efficacy/effectiveness of ceftaroline fosamil in all types of pneumonia (i.e., CAP, VAP, HCAP, HAP) cases was 81.2% (95% CI: 79.9–82.6; I²: 1.2%) (Fig. 2). The pooled treatment success rate was equal to 81.3% (95% CI: 80.0–82.7; I²: 7.7%) in patients with CAP, recruited in 12 (85.7%) studies (Fig. 3); a similar pooled clinical success (83.0%, 95% CI: 65.0–95.0; I²: -) was found for patients with VAP, HCAP, and HAP, enrolled in 2 (14.3%) studies (Fig. 4).

In the 8 (57.1%) controlled studies [9‐15, 22] the pooled relative risk of clinical cure was 1.1 (95% CI: 1.1–1.2; I²: 0.0%) (Fig. 5).

The stratified analysis according to the microbiological diagnosis showed a high clinical and microbiological success: it was 82.6% (95% CI: 78.6–86.4; I²: 0.0%) in patients with a lung infection caused by S. pneumoniae, 93.0% (95% CI: 77.0–100.0; I²: 0.0%) in those with a MDR S. pneumoniae infection (Figs. 6 and 7). The cases of pneumonia caused by MSSA and MRSA showed a pooled success rate of 72.3% (95% CI: 64.5–79.4; I²: 0.0%) and 71.7% (95% CI: 59.7–82.3; I²: 67.9%), respectively (Figs. 8 and 9).

Only 4 (28.6%) studies [10, 11, 13, 14] described the safety profile of the ceftaroline fosamil and control group: the percentage of adverse events ranged from 39.9 to 53.7% in the ceftaroline fosamil arm, whereas it ranged from 42.7 to 47.2% in the control arm (Table 6). The most frequently reported adverse events were: diarrhea, headache, insomnia, nausea, phlebitis, hypertension, and hypokalemia; however, their point prevalence was less than 5% in the ceftaroline fosamil group. Mortality rate in ceftaroline-treated arm is summarized in the Table 7.