Summary Basis of Decision for Xenleta

 

Clinical Pharmacology

Lefamulin, the medicinal ingredient in Xenleta, is a pleuromutilin antibacterial agent. It inhibits bacterial protein synthesis by interacting with the A‑ and P‑sites of the peptidyl transferase centre in the central part of domain V of the 23S ribosomal ribonucleic acid (RNA) of the 50S ribosomal subunit, preventing correct positioning of the transfer RNA.

The major pharmacokinetic aspects of absorption, distribution, metabolism, and elimination of Xenleta (following both oral and intravenous administration) have been well characterized in patients with community‑acquired pneumonia (CAP) and in healthy volunteers.

Lefamulin has a mean oral bioavailability of approximately 25%. Following oral administration, peak plasma concentration (C_max) is achieved in healthy individuals within 0.88 to 2 hours.

In healthy adults, findings based on the area under the concentration‑time curve (AUC) demonstrated exposure of lefamulin was comparable following oral and intravenous administration. Peak plasma concentration was significantly lower following oral administration, compared to intravenous administration; however, this is not expected to have an impact on efficacy since the antibacterial effect of lefamulin is most closely correlated to the AUC to MIC (minimum inhibitory concentration) ratio (see Clinical Efficacy section). In CAP patients, the AUC and C_max were higher compared to healthy individuals.

Lefamulin is rapidly distributed to the epithelial lining fluid (ELF) of the lungs, with therapeutic exposures achieved after a single dose, demonstrating that no loading dose is required to achieve or maintain therapeutic concentrations at the site of infection. The estimated ratio of ELF AUC to unbound plasma AUC is approximately 15. Steady state was achieved after 2 days in healthy adults, independent of the route of exposure (oral vs. intravenous).

Lefamulin metabolism is primarily driven by cytochrome P540 (CYP) 3A4. The main lefamulin metabolite in plasma, BC‑8041, showed no antibacterial activity. Following a single oral or intravenous administration of lefamulin, elimination was multiphasic and primarily via non‑renal clearance.

In patients with hepatic impairment, differences were noted in most pharmacokinetic parameters compared to healthy individuals, including C_max, clearance rates, and elimination half-life (t_1/2). In patients with moderate to severe hepatic impairment, the t_1/2 for lefamulin and the main metabolites increased, as did excretion via the urine. Slight decreases in plasma protein binding were also observed. The Xenleta Product Monograph adequately highlights the need for dosage adjustments for patients with hepatic impairment.

In patients with impaired renal function, pharmacokinetic parameters were comparable for individuals on and off dialysis and to those in healthy individuals.

A clear, statistically significant food interaction with lefamulin was demonstrated in various studies, with lower exposure, delayed t_1/2, and lowered C_max noted in the fed vs. fasted state. Following concomitant administration of a single oral dose of lefamulin with a high fat, high calorie breakfast, there was a decrease in AUC from time 0 to the last quantifiable concentration (AUC_T) and C_max by approximately 19.0% and 23.2%, respectively, when compared to administration under fasting conditions. This was appropriately reflected in the administration instructions in the Xenleta Product Monograph.

Drug‑Drug Interactions

Lefamulin is a substrate of CYP3A and a substrate and inhibitor of P‑glycoprotein (P‑gp). Overall, oral administration of lefamulin was observed to have a more significant potential for drug‑drug interactions compared to intravenous administration due to the involvement of P‑gp. More specifically, co‑administration of intravenous lefamulin with oral rifampin, a strong inducer of CYP3A and a substrate and inhibitor of P‑gp, resulted in a clinically significant reduction in plasma concentration of lefamulin, as measured by AUC from time zero to infinite time (AUC_0‑inf) and C_max. An even greater reduction was seen with orally administered lefamulin.

Additionally, co‑administration with oral ketoconazole, a strong CYP3A inhibitor and P‑gp inhibitor, resulted in a clinically significant increase in lefamulin plasma concentrations following oral administration of lefamulin.

No clinically significant differences in the pharmacokinetics of midazolam, a CYP3A substrate, were observed when administered concomitantly with intravenously administered lefamulin; however, there were profound increases in mean AUC_0‑inf and C_max when midazolam was orally administered concomitantly with and at 2 or 4 hours after oral administration of lefamulin. No clinically significant differences in the pharmacokinetics of digoxin, a P‑gp substrate, were observed when administered concomitantly with orally administered lefamulin.

QT Prolongation

The cardiac safety of Xenleta was investigated through all phases of clinical testing. The risk of QTcF (corrected QT interval by Fridericia) interval prolongation was evaluated in two Phase III studies known as Study 3101 and Study 3102 (see Clinical Efficacy section), where a concentration‑dependent QTcF prolongation effect of lefamulin was observed in CAP patients. The mean QTcF change from baseline was 13.6 ms (90% confidence interval [CI]: 11.7, 15.5 ms) following intravenous administration of 150 mg lefamulin twice daily and 9.3 ms (90% CI: 7.6, 10.9 ms) following oral administration of 600 mg lefamulin twice daily. By comparison, the mean change from baseline QTcF in the moxifloxacin arm was 16.4 ms (90% CI: 14.5, 18.3 ms) and 11.6 ms (90% CI: 10.0, 13.2 ms), respectively, on these days.

Further, the overall body of data was sufficient to draw the conclusion that administration of lefamulin increases the risk of QT interval prolongation at clinically relevant doses and, therefore, increases the possibility of adverse cardiac events. As a result, the risk of QT prolongation has been highlighted in a Serious Warnings and Precautions box in the Product Monograph for Xenleta.

Overall, the clinical pharmacological data support the use of Xenleta for the recommended indication.

For further details, please refer to the Xenleta Product Monograph, approved by Health Canada and available through the Drug Product Database.

Clinical Efficacy

The clinical efficacy of Xenleta (lefamulin) for the the treatment of adults with CAP caused by designated susceptible microorganisms was evaluated in two Phase III pivotal multicentre, randomized, double‑blind, double‑dummy, active‑controlled, parallel‑group, non‑inferiority studies known as Study 3101 and Study 3102.

Study 3101 enrolled adult (≥18 years of age) patients with CAP with a Pneumonia Outcomes Research Team (PORT) Risk Class III to V who required intravenous antibiotic therapy as initial treatment for the current episode of CAP and who were expected (but not required) to be hospitalized. Patients were randomized to receive 150 mg intravenous Xenleta twice a day for 5 to 10 days (number of patients [n] = 276) or 400 mg intravenous moxifloxacin once a day for 7 to 10 days (n = 275). After 3 days, patients in the Xenleta and moxifloxacin treatment arms could be switched to an oral formulation of 600 mg Xenleta twice a day or 400 mg moxifloxacin once a day, respectively.

If methicillin‑resistant Staphylococcus aureus (MRSA) was suspected at screening, patients randomized to moxifloxacin were to receive adjunctive 600 mg intravenous linezolid twice a day, with the option to switch to 600 mg oral linezolid after 3 days, while patients randomized to Xenleta received a placebo instead of linezolid. The PORT severity index, ranging from least severe (PORT Risk Class I) to most severe (PORT Risk Class V), is a validated prediction rule for 30‑day mortality and medical complications in patients with CAP. In Study 3101, approximately 72% of patients were PORT Risk Class III, 27% were PORT Risk Class IV, and 1% were PORT Risk Class V. Common comorbidities included hypertension (41%), asthma/chronic obstructive pulmonary disease (COPD) (18%), and diabetes mellitus (13%).

Study 3102 enrolled adult patients with CAP with a PORT Risk Class II to IV who were appropriate candidates for oral antibiotic therapy and did not require hospitalization. Patients were randomized to receive 600 mg oral Xenleta twice a day for 5 days (n = 370) or 400 mg oral moxifloxacin once a day for 7 days (n = 368). Approximately 50% of patients were PORT Risk Class II, 38% were PORT Risk Class III, and 11% were PORT Risk Class IV. Common comorbidities included hypertension (36%), asthma/COPD (18%), and diabetes mellitus (13%).

Baseline microbiology (pooled) was consistent with that of CAP, meaning Streptococcus pneumoniae was the most common baseline pathogen (61.9% overall), and polymicrobial infections were observed in 34.6% and 31.6% of patients in the Xenleta and moxifloxacin groups, respectively, the most common being Gram‑positive and fastidious Gram‑negative pathogens (15.9% and 14.5%, respectively).

In both studies, the primary efficacy endpoint was the percentage of patients with an Early Clinical Response (ECR) of responder at 72 to 120 hours after the first dose of study drug (the ECR responder rate) in the intent‑to‑treat (ITT) analysis set, which comprised all randomized patients. Response was defined as survival with improvement of at least two signs and symptoms that the patient presented with at baseline, no worsening of any sign or symptom, and no receipt of non‑study antibacterial treatment for CAP.

Both studies met their primary objective and demonstrated that Xenleta is non‑inferior to moxifloxacin in the treatment of adult patients with CAP. In Study 3101, the ECR responder rate was 87.3% in the Xenleta group and 90.2% in the moxifloxacin group (treatment difference ‑2.9%; 95% CI: ‑8.5, 2.8); 12.5% non‑inferiority [NI] margin. In Study 3102, the ECR responder rate was 90.8% in the Xenleta group and 90.8% in the moxifloxacin group (treatment difference 0.1%; 95% CI: ‑4.4, 4.5); 10% NI margin. The non‑inferiority (10% NI margin) of Xenleta to moxifloxacin was confirmed in the pooled analysis, with an ECR responder rate of 89.3% in the Xenleta group versus 90.5% in the moxifloxacin group (treatment difference ‑1.1%; 95% CI: ‑4.4, 2.2). The proportion of patients meeting ECR responder criteria by visit was similar between pooled treatment groups, with approximately 60% of patients in both groups meeting ECR responder criteria by Day 3, and approximately 91% meeting the ECR responder criteria by Day 5.

Additional statistical analyses conducted using data from both pivotal studies demonstrated that a favourable clinical response to Xenleta at 72 to 120 hours after the first dose was a good predictor for clinical success 5 to 10 days after the last dose (at the Test of Cure [TOC] visit). Negative predictive values, ranging from 45.0% to 52.1%, indicate that some patients may still recover with additional time on therapy. Of note, while the ECR responder rate in the ITT analysis set was also used as the primary endpoint by the United States Food and Drug Administration, it was used as a secondary endpoint by the European Medicines Agency.

Early Clinical Response responder rates were reassuring in the ITT population for the subgroups with PORT Risk Class IV and V and the subgroups with important co‑morbidities known to be associated with poorer outcomes (e.g., lung/heart disease and moderate renal impairment). However, in the subgroups of <65 years of age and American Thoracic Society (ATS) minor severity criteria, the 95% CIs did not cross zero. Differences in ECR responder rates in the ATS subgroup appeared to be primarily driven by an imbalance in the number of discontinuations of study drug in the Xenleta treatment group in Study 3101 for reasons other than lack of efficacy. The subgroup of <65 years of age appeared to have been confounded by results in the ATS equals Yes subgroup. Additionally, in the subgroup with baseline bacteremia, the ECR responder rates in the Phase III pool were 61.5% (8/13 patients) for Xenleta versus 83.3% (10/12 patients) for moxifloxacin. However, this numerical difference in response rate may be the result of chance and small numbers; it is also possible that confounding by pathogen types, differences in baseline disease severity, and duration of treatment may have contributed. In some patients, plausible alternative reasons for failure were identified. For all of the patients, follow-up blood cultures were either not obtained or negative. Of note, mortality among bacteremic patients was low across both treatment groups.

Key co‑secondary endpoints were assessed by the Investigator at the TOC visit (5 to 10 days after the last dose) in both the modified ITT (mITT) analysis set, which comprised all randomized subjects who received any amount of study drug, and the Clinically Evaluable at TOC (CE‑TOC) analysis set. Of note, only a very small number of patients in the ITT analysis set were excluded from the mITT analysis set. The CE‑TOC analysis set included all randomized patients who met pre-defined key inclusion/exclusion criteria, minimum dosing criteria, and had no confounding use of prohibited antibiotics. In both the individual studies and in the pooled analysis, success rates for the Investigator-assessed Clinical Response (IACR) endpoints were similar between treatment groups, and were supportive of the findings with the primary endpoint (ECR responder rate). In Study 3102, a greater treatment difference was observed in the CE‑TOC analysis set (‑3.9%) compared with the mITT analysis set (‑1.6%).

The ECR responder rate in the microbiological ITT (microITT) analysis set also served as a secondary endpoint. The microITT analysis set was comprised of patients with at least one baseline bacterial pathogen known to cause CAP. Results in both the individual studies and the pooled analysis were comparable to those of the larger analysis sets, with an ECR responder rate (Xenleta vs. moxifloxacin) at 72 to 120 hours after the first dose of 87.4% vs 93.1% in Study 3101, 90.7% vs. 93.0% in Study 3102, and 89.3% vs 93.0% in the pooled analysis.

The by‑pathogen IACR success rate at TOC in the pooled microITT analysis set also served as a secondary endpoint. The by-pathogen IACR success rates at TOC were generally high and similar between treatment groups. For example, success rates for Streptococcus pneumoniae were 85.2% vs. 86.5%, Xenleta vs. moxifloxacin, respectively. However, a numerical decrease in the IACR success rate at TOC was observed for the Xenleta subgroup with a baseline pathogen of penicillin-susceptible Streptococcus pneumoniae (PSSP) (i.e. 74.5% vs. 98.2%, Xenleta vs. moxifloxacin, respectively). In both studies, several Xenleta patients with PSSP who experienced an ECR of non-responder, appeared to have reasonable explanations for failure of therapy other than potential reduced efficacy. An imbalance across treatment groups in baseline characteristics, including a higher rate of concomitant bacteremia and more patients meeting severity index criteria in the Xenleta group, was observed. Thus, the numerical decrease in response rates at TOC in the subgroup with a baseline pathogen of PSSP may have been due to small numbers and confounding by baseline disease severity. Microbiological responses reflected the clinical outcomes at TOC as sputum cultures or other diagnostic tests were infrequently repeated following clinical improvement (hence presumed eradication of pathogens).

Success rates among resistant pathogens were high in the Xenleta group, although the number of resistant pathogens was low.

The unique mechanism of action of lefamulin has been well described, and to date, no development of resistance to lefamulin has been observed in clinical studies evaluating treatment for up to 14 days. Global surveillance will be conducted to monitor potential changes over time.

Xenleta is not active against Enterobacteriaceae and Pseudomonas aeruginosa. In Study 3101, three patients treated with Xenleta experienced superinfection with pathogens outside the known spectrum of activity of Xenleta. In Study 3102, 4 out of 5 patients treated with Xenleta who experienced a serious adverse reaction of pneumonia were subsequently found to have a non‑susceptible microorganism. These cases were captured as clinical failures in the efficacy analysis. No pathogens met the definition of decreasing susceptibility for a post‑baseline pathogen to lefamulin. The Xenleta Product Monograph communicates that Xenleta is not active against Enterobacteriaceae and Pseudomonas aeruginosa in the Microbiology section and that appropriate measures should be taken in the event that superinfection occurs during therapy within the Warnings and Precautions section.

Indication

Health Canada approved the following indication for Xenleta:

• Xenleta (lefamulin) is indicated for the treatment of adults with community‑acquired pneumonia (CAP) caused by: Streptococcus pneumoniae including multi‑drug resistant S. pneumoniae (MDRSP), Staphylococcus aureus (methicillin susceptible isolates), Haemophilus influenzae, Haemophilus parainfluenzae, Moraxella catarrhalis, Legionella pneumophila, Mycoplasma pneumoniae, and Chlamydophila pneumoniae.
   
• Streptococcus pneumoniae including multi‑drug resistant Streptococcus pneumonia refers to isolates resistant to two or more of the following antibiotics/antibiotic classes: penicillins, cephalosporins, macrolides, tetracyclines, lincosamides, fluoroquinolones and folate‑synthesis inhibitors.
   
• To reduce the development of drug‑resistant bacteria and maintain the effectiveness of Xenleta and other antibacterial drugs, Xenleta should be used only to treat infections that are proven or strongly suspected to be caused by susceptible bacteria. When culture and susceptibility information are available, they should be considered in selecting or modifying antibacterial therapy. In the absence of such data, local epidemiology and susceptibility patterns may contribute to the empiric selection of therapy.

Health Canada revised the proposed indication slightly to specify that Xenleta is authorized for the treatment of adults, to reflect the patient population in the pivotal studies.

For more information, refer to the Xenleta Product Monograph, approved by Health Canada and available through the Drug Product Database.

Clinical Safety

The safety of Xenleta was evaluated primarily in 641 patients with CAP who received at least one dose of Xenleta in Studies 3101 and 3102 (see Clinical Efficacy section) at the proposed dosing regimen.

In a pooled analysis of Studies 3101 and 3102, the rates (Xenleta vs. moxifloxacin) of adverse reactions (34.9% vs. 30.4%), serious adverse reactions (5.6% vs. 4.8%), treatment discontinuations due to adverse reactions (3.1% vs. 3.3%), and deaths within 28 days (1.2% vs. 1.1%) were similar between treatment groups.

The most common (≥2%) adverse reactions with oral administration were diarrhea (12%), nausea (5%), vomiting (3%), and hepatic enzyme elevation (2%). The most common adverse reactions with intravenous administration of Xenleta were administration site reactions (7%), hepatic enzyme elevation (3%), nausea (3%), hypokalemia (3%), insomnia (3%), and headache (2%).

Local infusion site reactions were the most frequently reported adverse reactions associated with intravenous administration of Xenleta (7.0% vs 2.6%, Xenleta vs. moxifloxacin, respectively). The majority of administration site reactions were mild or moderate and only one patient discontinued study drug. With a view to improving local tolerability, a general precaution was incorporated into the Administration section of the Xenleta Product Monograph to recommend careful adherence to the instructions for dilution and infusion, and that the recommended dose and infusion rate should not be exceeded. Xenleta (lefamulin for injection) is recommended to be administered by intravenous infusion over 60 minutes after admixture of the entire 15 mL vial of Xenleta in a 250 mL solution of 10 mM citrate buffered 0.9% sodium chloride.

Gastrointestinal adverse reactions were the most frequently reported adverse reactions in the Xenleta and moxifloxacin treatment groups (13.1% vs. 10.1%, respectively). The difference between treatment groups (Xenleta vs. moxifloxacin) was driven primarily by gastrointestinal adverse reactions associated with the oral administration of Xenleta in Study 3102, most notably diarrhea (12.2% vs. 1.1%). In Study 3101, the opposite was observed with regard to the frequency of diarrhea (0.7% vs. 7.7%). Among Xenleta‑treated patients in Study 3101, gastrointestinal events were more likely to start during oral treatment (7.7%) than during intravenous treatment (3.7%), with the largest observed difference being associated with nausea.

Clostridium difficile-associated disease has been reported with the use of many antibacterial agents, including Xenleta. Clostridium difficile-associated disease was expanded upon in the Warnings and Precautions section of the Xenleta Product Monograph, along with recommendations for the initiation of appropriate therapeutic measures when the diagnosis is suspected or confirmed.

Adverse reactions indicative of increased hepatic enzymes were reported in 2.3% of patients in the Xenleta group. Those affected were typically asymptomatic with reversible clinical laboratory findings (e.g., alanine aminotransferase [ALT] values) that peaked within the first week of dosing, with decline to within or near normal levels within 2 to 4 weeks. Discontinuation of Xenleta due to non‑serious hepatobiliary adverse reactions occurred in two patients. However, the duration of therapy was short. A small number of patients experienced relatively severe (i.e., >10 times the upper limit of normal) elevations in ALT and aspartate aminotransferase, and the majority of these cases were associated with a positive dechallenge back to normal. No Xenleta‑treated patient met the laboratory criteria for potential cases of Hy's Law. Of note, patients with evidence of significant hepatic disease were excluded from pivotal studies. Given that a potential for hepatic injury could not be ruled out, precautionary text was included in the Warnings and Precautions section of the Xenleta Product Monograph, with a recommendation to perform serum chemistry testing inclusive of liver enzymes and bilirubin as clinically warranted.

Lefamulin is primarily metabolized by CYP3A4. In a hepatic impairment study in which non-infected patients with moderate or severe hepatic impairment were administered intravenous drug, biologically active lefamulin concentrations increased with the degree of hepatic impairment. An increased rate of adverse reactions (hepatobiliary and atrial fibrillation) was observed among Xenleta patients with elevated liver enzymes at baseline compared with normal liver enzymes at baseline. Risk was mitigated via communication within a Hepatic Impairment subsection of the Warnings and Precautions section of the Xenleta Product Monograph, including a recommendation for monitoring patients with hepatic impairment for adverse reactions associated with oral and intravenous Xenleta throughout the treatment period.

 

 

Eight patients receiving Xenleta compared with 4 patients receiving moxifloxacin experienced serious respiratory, thoracic, and mediastinal disorders (fatal and non‑fatal); 5 (Xenleta) vs. 2 (moxifloxacin) of these reactions appeared to be either related to the underlying pneumonia or worsening thereof. This difference may be due to small numbers, chance, and confounding factors affecting baseline disease severity. This did not appear to be a direct safety issue, and all of these patients were captured in the efficacy results as IACR failures.

A Serious Warnings and Precautions box in the Xenleta Product Monograph highlights the risk of QT interval prolongation. In both pivotal studies, a concentration‑dependent QT prolongation effect was observed. The mean change from baseline QTcF following intravenous administration of Xenleta was 13.6 ms (90% CI: 11.7, 15.5 ms), compared to 16.4 ms (90% CI: 14.5, 18.3 ms) in the moxifloxacin treatment group. Following oral administration of Xenleta, the mean change from baseline QTcF was 9.3 ms (90% CI: 7.6, 10.9 ms), compared to 11.6 ms (90% CI: 10.0, 13.2 ms) in the moxifloxacin treatment group. Furthermore, the oral administration of Xenleta (lefamulin tablets) is contraindicated in patients taking sensitive CYP3A4 substrates that prolong the QT interval (e.g., pimozide). Concomitant administration of oral Xenleta with sensitive CYP3A4 substrates such as pimozide may result in increased plasma concentrations of these drugs, leading to QT interval prolongation and cases of torsades de pointes.

Based on findings from animal studies, Xenleta may cause fetal harm when administered to pregnant women, and lefamulin was concentrated in the milk of lactating rats. Xenleta should be used during pregnancy only if the potential benefit justifies the potential risk to the mother and fetus. A woman breastfeeding her baby should pump and discard human milk for the duration of treatment with Xenleta and for 2 days after the final dose.

For more information, refer to the Xenleta Product Monograph, approved by Health Canada and available through the Drug Product Database.