Ceft-to-Ceft Study: Real-Life Experience with Ceftaroline and Ceftobiprole in Treatment of the Principal Infectious Syndromes in a Spanish Multicenter Hospital Cohort

Background: To compare the real-life effectiveness and safety of ceftaroline fosamil (ceftaroline-F) and ceftobiprole medocaril (ceftobiprole-M) for infections in hospitalized patients. Methods: This comparative, observational, retrospective, and multicenter Spanish study included patients receiving outpatient parenteral antimicrobial therapy (OPAT) and hospitalized patients treated for at least 48 h with ceftaroline-F or ceftobiprole-M between their first incorporation in the clinical protocol of each hospital and 31 July 2022. Results: Ceftaroline-F was administered to 227 patients and ceftobiprole-M to 212. In comparison to the latter, ceftaroline-F-treated participants were younger (63.02 vs. 66.40 years, OR 1.1; 95%CI: 1.001–1.05) and had higher rates of septic shock (OR 0.27; 95%CI: 0.09–0.81) and higher frequencies of targeted (57.7 vs. 29.7%; OR: 0.35; 95%CI: 0.18–0.69) and combined (89.0 vs. 45.8%, OR: 0.13; 95%CI: 0.06–0.28) therapies that were second line or more (82.4% vs. 64.6%%; OR 0.35; 95%CI: 0.18–0.69), and higher rates of infections due to Gram-positive cocci (92.7 vs. 64.7%, p = 0.001), bacteremia (51.9 vs. 21.7%, p = 0.001), infective endocarditis (24.2 vs. 2.4%, p = 0.0001), and mechanical ventilation-associated pneumonia (8.8 vs. 2.4%, p = 0.0001). Ceftobiprole-M was more frequently administered against polymicrobial infections (38.1 vs. 14.0%, p = 0.001), those produced by Gram-negative bacilli (19.7 vs. 6.0%, p = 0.0001), nosocomial pneumonia (33 vs. 10.6%, p = 0.0001), and skin and soft-tissue infections (25.4 vs. 10.1%, p = 0.0001). Patients treated with ceftaroline-F had a longer hospital stay (36 (IQR: 19–60) vs. 19.50 (IQR: 12–30.75, p = 0.0001) days), with no difference in infection-related mortality at 14 (13.2 vs. 8.0%, p = 0.078) or 28 (4.8 vs. 3.3%, p = 0.415) days or in dropout rate for adverse effects (2.2 vs. 0.9%; p = 1). Conclusions: The fifth-generation cephalosporins, ceftaroline-F and ceftobiprole-M, are safe and effective in real life, with no difference between them in health outcomes.


Introduction
The increase in antimicrobial resistance is one of the main public health threats worldwide, including in Europe and Spain [1].The rise of multiresistant microorganisms is associated with higher health costs, more frequent antibiotic treatment failures, quality of life impairment, and increased mortality [2].The latest World Health Organization report on antimicrobial resistance in 2020 described a 16.6% prevalence of carbapenem-resistant Pseudomonas aeruginosa and a 23.3% prevalence of methicillin-resistant Staphylococcus aureus (MRSA) in Spain, with the latter being above the European mean of 16.7% [3].These data underscore the need for a carbapenem-sparing strategy in our setting, using bactericides other than carbapenems in patients with sepsis or septic shock, such as fifth-generation cephalosporins: ceftaroline-F and ceftobiprole-M.
Clinical trials are evidently the gold standard for the approval and commercialization of novel pharmaceutical compounds; however, extrapolation of their outcomes is limited by the strict inclusion criteria and special conditions imposed [19].The aim of real-life studies (Real World Data (RWD)) is to bridge the knowledge gap between clinical trials and clinical practice.This type of study has served to reduce medical costs, improve outcomes, and accelerate the incorporation of novel therapies and technologies into routine clinical practice [20].
Given this background and the limited scientific evidence available on fifth-generation cephalosporins, the objective of this study was to compare real-life effectiveness and safety between ceftaroline-F and ceftobiprole-M in the treatment of the principal infectious syndromes in hospitalized patients in Spain.
All GPC isolates in the ceftaroline-F group (n = 113) were susceptible to vancomycin vs. 96.4% of GPC isolates in the ceftobiprole-M group (n = 53) (p = 0.106), with one Staphylococcus hominis and one MRSA being vancomycin resistant.
All tested microorganisms in the ceftaroline-F group (forty-three microorganisms: thirteen MRSA, 12

Adverse Effects
No adverse effects were observed in 96% of the patients treated with ceftaroline-F or 97.6% of those treated with ceftobiprole-M.In the former, mild effects were recorded in three patients (1.3%) and moderate effects in six (2.5%), while withdrawal of the antibiotic was mandated in five (2.1%).In the five patients treated with ceftobiprole-M, adverse effects were mild in three (1.4%),moderate in two (0.9%), and caused withdrawal of the antibiotic in two (0.9%) (Table 6).

Discussion
The patients in our cohort were mostly males with a high comorbidity index, and all were aged > 60 years.The ceftaroline-treated patients were younger, their clinical status was worse, and they more frequently received targeted and combined second-line treatments or more.Both ceftaroline and ceftobiprole were most often combined with daptomycin.Previous studies have demonstrated the clinical benefit of combining these fifth-generation cephalosporins with daptomycin in bacteriemia due to MRSA [21,22].Around one in four patients were hospitalized in medical departments, with 20% of ceftaroline-F-treated patients and 5% of ceftobiprole-M-treated patients being in the ICU.More than half of the infections were nosocomial or healthcare-related, and around one-third of them met the criteria for sepsis.Septic shock, reported by some authors to increase the mortality risk by up to 40% [23], was five-fold more prevalent among the ceftaroline-F-treated patients in our cohort.
Ceftaroline-F was most frequently prescribed to treat bacteremia, IE, ABSSTIs, or pneumonia.The IMVAP healing rate was 80%, similar to previous reports [24].In comparison to the rates obtained with ceftaroline-F treatment in the multicenter CAPTURE study of more than 500 patients with pneumonia, the cure rate was slightly higher in the present patients with CAP (89.2 vs. 79%), lower in those with chronic obstructive pulmonary disease (COPD) (15 vs. 40%) [24], and similar in those with NP (80%) and IMVAP (60%) [25].In the patients with IE treated with ceftaroline-F, the related mortality was 23.6%, lower than in the CAPTURE study (34.3%) [26], despite having a similar percentage of patients with IE due to SA (40 vs. 35.8%,respectively), and in surgery-indicated patients who did not undergo surgery (20% vs. 19.5%).It has previously been observed [27] that the isolation of MSSA and the indication but non-performance of surgery are associated with a higher mortality in patients with IE.
Ceftobiprole-M was prescribed against pneumonia in more than half of the patients, ABSSTIs in around one-fifth, and bacteremia in around one-sixth.Among the cases of pneumonia treated with ceftobiprole-M, around half were CAP and half NP, with only five cases of IMVAP.This fifth-generation cephalosporin has demonstrated non-inferior clinical effectiveness and safety with respect to its comparators in various clinical trials for the treatment of ABSSTIs [28], complicated S. aureus bacteriemia (ERADICATE study) [29], CAP, and NP [16,17].However, worse outcomes have been obtained in patients with IMVAP treated with ceftobiprole-M vs. ceftazidime plus linezolid [17], and ceftobiprole-M is, therefore, not indicated for its treatment [30].Recently, a real-life study has reported on the use of ceftobiprole-M for severe infections in a French ICU, finding that the main indication for this antibiotic was pneumonia (51%) with associated bacteremia (72%); clinical cure was achieved in around 80% of patients, with an in-hospital mortality rate of 32% [31].
One in ten patients with pneumonia in our study, including both ceftobiprole-Mand ceftaroline-F-treated patients, were coinfected with COVID-19.COVID-19 coinfection and suprainfection have been associated with a worse clinical prognosis in patients with respiratory infection, increasing the mortality risk by up to three-fold [32].
The infection-related mortality rate at 14 and 18 days did not differ between the ceftarolineand ceftobiprole-treated patients, and no between-group differences in age, sex, previous disease, infection type, or isolated microorganism were observed among those whose death was attributable to the infection.Mortality risk factors were older age, Charlson index, ICU stay, and the presence of sepsis and septic shock in both antibiotic groups.This is consistent with published evidence that age, comorbidity, ICU admission (and duration), and septic shock are independent prognostic factors for mortality in patients with sepsis [21,33,34].The mortality rate of patients with IE treated with ceftaroline-F was 23.6%, although SA was the causal agent in two-fifths of these patients, and surgery was prescribed but not feasible in one-fifth.A previous study of patients with IE due to SA in our setting described a higher mortality rate of 50% that was not influenced by methicillin resistance [35].
The readmission rate was significantly higher among patients treated with ceftobiprole-M but below 5% for both antibiotics (4.7 vs. 1.3%, p = 0.036).Seven of the ten ceftobiprole-M-treated patients readmitted had ABSSTIs, which have previously been associated with a recurrence rate of around 10% [36].However, interpretation of the results should take account of disparities between the antibiotic cohorts (in previous diseases, pathways, and microorganisms) that may diminish the comparability of outcomes.
No adverse effects were observed in 96% of patients receiving ceftaroline-F or in 97.6% of those receiving ceftobiprole-M.These data are similar to the findings of clinical trials [8,17,37] and real-life studies [22,38,39], in which the most frequent effects in these studies were gastrointestinal, cutaneous, and hepatic profile disorders, with the less common observation of myoclonia with ceftobiprole-M [38] and cytopenia with ceftaroline-F [40].Cytopenia was observed in three of our ceftaroline-F-treated patients.Currently, a retrospective, single-center study has just been published, conducted in an Italian hospital, which compared in real life the effectiveness of the use of ceftaroline-M vs. ceftobiprole in a reduced number of subjects (75 vs. 63 patients), with data similar to ours in terms of clinical efficacy and tolerability [41].
This study is limited by its retrospective design and associated selection bias, which may influence the results obtained and their interpretation.However, this is the first reallife comparative study of ceftaroline-F and ceftobiprole-M and has a wide patient sample from different centers, representing as far as possible the routine clinical utilization of these fifth-generation cephalosporins in Spain.

Study Design
A real-life, retrospective, multicenter, observational, descriptive study was conducted on the use of ceftaroline-F and ceftobiprole-M in hospitalized patients or outpatient parenteral antimicrobial therapy (OPAT) with nosocomial/nosohusial or community-acquired infections.Patients were included between the date that these drugs were incorporated in the clinical protocol of the participating hospital and 31 July 2022.The study was approved by the provincial ethics committee of Granada (Ref.0095-N-22) and deemed exempt from the requirement for informed patient consent.

Treatment Description
This descriptive study did not involve any pharmacological intervention, and all treatments were prescribed by the attending physicians according to their habitual criteria.Inclusion criteria were age ≥ 18 years; at least 48 h of antibiotic treatment with ceftaroline-F or ceftobiprole-M (≥6 vials in patients with normal renal function and creatinine clearanceadjusted dosage in patients with kidney failure) between the date of drug approval at the hospital and 31 July 2022; and the completion of a 28-day follow-up, except for patients with IE or osteomyelitis (6-month follow-up).Exclusion criteria were pregnancy and allergy to beta-lactams or some formulation excipient.

Variables
Data were gathered on sex, age, hospital stay, prescribing hospital department, patient comorbidities, infection acquisition pathway, and the presence of sepsis or septic shock.The age-corrected Charlson index was calculated for each patient [42].
Information was also collected on the number of active infectious foci, the presence of bacteremia (primary/catheter-related) and IE (by modified Duke's diagnostic criteria [43] (defined/possible/rejected-suspicion), the type of IE (native valve/early prosthetic/late prosthetic/pacemaker or device/transcatheter aortic valve implantation, the performance of surgery (not indicated/indicated but not undergone/device extraction)), and the presence of respiratory infection (CAP/NP/invasive mechanical ventilation-associated pneumonia (IMVAP)), ABSSTI, urinary tract infection, central nervous system infection, spondylodiscitis, osteoarticular infection, intraabdominal infection, fever with no focus, and other infectious foci.For respiratory infections, data were recorded on coinfection by coronavirus SARS-2 2019 and suspected or confirmed fungal coinfection.
Treatment details were collected on the prescribed antibiotic (ceftaroline-F or ceftobiprole-M); the mode of administration, monotherapy or a combination with other antibiotics against the same infection, empirical or targeted, first-line or rescue treatment (in the latter case, specifying previous antibiotic therapy for the same infection and its duration); reason for antibiotic rescue (previous antibiotic treatment failure/receipt of indicative microbiological results/emergence of toxicity or adverse effects with previous treatments) and appropriate empirical antibiotic therapy (i.e., effective against isolated microorganisms or good clinical outcome if none are isolated); and the total days of antibiotic administration, total administered dose, and the presence, type, and severity of adverse effects.
The following health outcomes were recorded: clinical cure, total crude mortality at 14 and 28 days (6 months for patients with IE or osteomyelitis), infection-related mortality, readmission for the same motive, and infection recurrence.

Definitions
Nosocomial/Nosohusial infection: infection in patients hospitalized for >72 h (nosocomial) or healthcare-related infection (day hospital, nursing home, day center for the elderly).Septic shock: refractory hypotension accompanied by hypoperfusion and organ dysfunction despite adequate fluid resuscitation [45].Immunosuppression: congenital or acquired immunodeficiency (including oncohematological patients) or secondary to immunosuppressive treatment [46].Recurrence/relapse: the second microbiologically confirmed episode of the infection within one month of discharge [47].Severity of adverse effects: mild = no requirement for antidote or treatment; moderate = requirement for treatment modification (e.g., dose adjustment, combination with other drugs) but not an interruption, with possible need for a longer hospital stay or specific treatment prescription; severe = life threatening, mandating an interruption of drug administration and the prescription of a specific treatment; or lethal = directly or indirectly contributing to death.
Data were gathered from the different electronic records systems of participating hospitals by obtaining the records of patients receiving the drugs from the hospital Pharmacy Departments.Data were anonymized before being entered into the SPSS 20.0 database in accordance with the Helsinki Declaration and Spanish data protection legislation (Law 3/2018 5 December).

Sample Size
This descriptive comparative study included all patients meeting study eligibility criteria treated with ceftaroline-F or ceftobiprole-M at participating centers during the study period.

Statistical Analysis
In a descriptive analysis, absolute and relative frequencies (%) with a 95% confidence interval (CI) were calculated for qualitative variables, means with standard deviation (SD) for quantitative variables with a normal distribution (Kolmogorov-Smirnov test), and medians with an interquartile range (IQR) for those with a non-normal distribution.In bivariate analyses, the chi-square test was used to compare qualitative variables, Student's t-test was used for quantitative variables with normal distributions, and the Mann-Whitney U test was used for quantitative variables with non-normal distributions.
A logistic regression multivariate analysis was performed to examine between-group differences in epidemiologic variables, medical history of interest, drug dose, and drug administration (alone/in combination) and its prescription as empirical, targeted, or rescue therapy (with reason for rescue) by entering all variables that were statistically significant in bivariate analysis.Another regression analysis was carried out to evaluate mortality-related factors in the two antibiotic groups.

Conclusions
The fifth-generation cephalosporins, ceftaroline-F and ceftobiprole-M, are safe and effective in real life, with no difference between them in health outcomes.

Supplementary Materials:
The following supporting information can be downloaded at https: //www.mdpi.com/article/10.3390/antibiotics12121692/s1,Table S1: Combined antibiotic therapy with ceftaroline or ceftobiprole; Table S2: Previous antibiotic therapy before prescription of ceftaroline or ceftobiprole.Informed Consent Statement: This study was exempt from the need to obtain patient consent due to its retrospective design and large size.

Table 7 .
Bivariate and multivariate analysis of infection-related mortality.