Summary Basis of Decision for Livtencity

 

As described above, the clinical review of the New Drug Submission for Livtencity was conducted as per Method 3 described in the Draft Guidance Document: The Use of Foreign Reviews by Health Canada.

Clinical Pharmacology

Livtencity (maribavir) is an orally bioavailable benzimidazole riboside antiviral drug with a novel mechanism of action against human cytomegalovirus (HCMV). Maribavir inhibits the CMV UL97 viral protein kinase by preventing the binding of adenosine triphosphate (ATP) to the kinase ATP-binding site. By abolishing the phosphotransferase activity of UL97, maribavir inhibits essential processes like deoxyribonucleic acid (DNA) replication and nuclear egress of virions.

The clinical pharmacology component of this submission included data from human pharmacodynamic and pharmacokinetic studies, as well as in vitro studies.

Following oral administration, maribavir was rapidly absorbed with a median time to reach maximum concentration (T_max) between 1 and 3 hours. Maribavir exhibited linear pharmacokinetics with an approximately dose-proportional increase in exposure. Maribavir was highly bound to plasma protein (approximately 98%). Based on the population pharmacokinetic analysis, there was no evidence that age, gender, race, ethnicity, weight, transplant type, hepatic impairment, renal impairment, or moderate and weak cytochrome P450 (CYP)3A4 inhibitors affected the steady-state pharmacokinetics of maribavir to a clinically relevant extent.

The steady-state mean terminal half-life of maribavir was 4.32 hours in transplant patients, with a mean oral clearance of 2.85 L/h. Steady-state was achieved in 2 days and the accumulation ratio was 1.37 for the maximum concentration (C_max) and 1.47 for the area under the concentration versus time curve (AUC). Hepatic metabolism via CYP3A4 is the primary metabolic pathway of maribavir, with a minor contribution from CYP1A2.

During the first 24 hours after oral administration, maribavir and VP 44469 (maribavir's major metabolite) accounted for all the radioactivity observed in plasma (88% for maribavir and 12% for VP 44469). Maribavir and 5 metabolites were recovered in the urine. Maribavir accounted for only 1.8% and 5.7% of the dose in the urine and feces, respectively, suggesting that renal excretion was minimal. VP 44469 accounted for 34.0% and 7.2% of the dose in the urine and feces, respectively.

A single 1,200 mg dose of maribavir did not prolong the QTc interval to any clinically relevant extent. Mild, moderate, and severe renal impairment resulted in exposure increases of less than 9%. Moderate hepatic impairment was associated with a 26% and 35% increase in maribavir AUC and C_max, respectively. No dose adjustment is required for patients with mild or moderate hepatic impairment. The pharmacokinetic parameters of maribavir were not studied in patients with severe hepatic impairment.

As maribavir is primarily metabolized by CYP3A, concomitant administration of drugs that are CYP3A inducers is not recommended due to the potential decrease in maribavir efficacy. Maribavir is an inhibitor of p-glycoprotein (P-gp) and breast cancer resistance protein (BCRP) and it is expected that co-administration will increase the exposure of drugs that are sensitive substrates of CYP3A4, P-gp, and BCRP. Maribavir may increase the exposure of immunosuppressant drugs that are CYP3A and P-gp substrates, including tacrolimus, cyclosporine, everolimus, and sirolimus. Based on a physiologically based pharmacokinetic model, maribavir is also expected to increase exposure to statins that rely on BCRP for their elimination. The risks associated with maribavir potential for drug-drug interactions have been mitigated through labeling in the Livtencity Product Monograph.

The results of the pivotal comparative bioavailability studies demonstrated that following the administration of a single 400 mg dose of a developmental formulation of maribavir, moderate-fat meal decreased the C_max and the area under the concentration-time curve from zero to infinity (AUC_T) by approximately 28% and 14%, respectively. 

In addition, the rate (C_max) and extent (AUC_T) of absorption of maribavir was comparable between a 100 mg dose of a developmental formulation of maribavir administered crushed or whole under fasting conditions.

The rate (C_max) and extent (AUC_T) of absorption of maribavir was comparable between a 100 mg dose of a developmental formulation of maribavir administered following an antacid versus without an antacid under fasting conditions.

The clinical pharmacology data support the use of Livtencity for the recommended indication. For further details, please refer to the Livtencity Product Monograph, approved by Health Canada and available through the Drug Product Database.

Clinical Efficacy

The clinical efficacy of Livtencity for the authorized indication was primarily evaluated based on data from the pivotal Phase III Study 303.

Study 303 was a Phase III, multicentre, randomized, open-label, active-controlled, superiority study designed to assess the efficacy and safety of Livtencity treatment compared to investigator-assigned treatment (IAT). The study included 352 hematopoietic stem cell transplant (HSCT) or solid organ transplant (SOT) recipients with CMV infections refractory to treatment with ganciclovir, valganciclovir, foscarnet, or cidofovir, including CMV infections with or without confirmed resistance to one or more anti-CMV agents. The median age of patients treated with Livtencity was 57 years (range: 19 to 79 years) versus 54 years (range: 19 to 77 years) for the IAT group. Overall, 60.5% of patients in the study population were male and 39.5% were female. The majority (75.6%) of patients were White, 13.4% were Black, 4.5% were Asian, and 6.5% were of other races or their data was missing. Some patients were confirmed to be genotypically resistant to standard treatment at randomization, while all subjects had current CMV infection that was refractory to the most recently administered standard anti-CMV treatment (ganciclovir, valganciclovir, foscarnet, or cidofovir). Key exclusion criteria included pregnancy or breastfeeding, concomitant human immunodeficiency virus (HIV) infection, and severe liver impairment (defined as liver enzymes more than five times the upper limit of normal).

Patients were stratified by transplant type (HSCT or SOT) and by screening viral load and were then randomized in a 2:1 allocation ratio to receive 400 mg of Livtencity twice daily (total number [n] = 117) or IAT (ganciclovir, valganciclovir, foscarnet, or cidofovir; n = 235) for an 8-week treatment period and a 12-week follow-up phase. The primary efficacy endpoint was confirmed CMV viremia clearance (plasma CMV DNA concentration below the lower limit of quantification [<LLOQ]; i.e., <137 IU/mL) at Week 8. The key secondary endpoint was CMV viremia clearance (undetectable CMV DNA levels) and control of CMV infection symptoms (in those who were symptomatic with CMV disease) at the end of Study Week 8 with maintenance of this treatment effect through Study Week 16. All randomized patients were included in the efficacy analysis, even if they did not complete the full 8 weeks of therapy, or if they changed therapy during the 8-week treatment phase (IAT only).

The pivotal study met its primary endpoint demonstrating that Livtencity was superior to IAT with respect to CMV viremia clearance at Week 8 (56% versus [vs.] 24%, respectively, p <0.001). Patients were able to stay on Livtencity treatment for approximately 50% longer than IAT with fewer discontinuations due to intolerances, adverse events (32% IAT versus [vs.] 13% Livtencity), or adverse reactions (1.7% IAT vs. 1.3% Livtencity).

The main reason for treatment failure in the Livtencity arm was virologic failure. For the IAT arm, the main reason for treatment failure was discontinuation of the drug/study, mainly due to adverse events. The overall proportion of virologic failures was similar between the two treatment arms, but a higher proportion of patients in the IAT arm failed to achieve undetectable CMV DNA levels (30% IAT vs. 20% Livtencity), whereas more patients in the Livtencity arm achieved undetectable CMV levels but then experienced CMV DNA breakthrough (6% IAT vs. 14% Livtencity). Only 31.6% of patients in the IAT arm completed 8 weeks of treatment compared to 77.9% in the Livtencity arm, reflecting a greater tolerability of Livtencity.

In a sensitivity analysis of patients in both arms who completed 8 weeks of treatment and excluded all patients in both treatment groups who discontinued study-assigned treatment early, the percentage of responders who achieved the primary efficacy endpoint in the IAT group increased substantially (from 24% to 60%). As a result, the Livtencity efficacy primary endpoint was no longer statistically significant. Livtencity is more effective than IAT for the primary endpoint, however increased tolerability may be driving a proportion of this efficacy finding. Given the known toxicities of the current standard of care treatments, increased tolerability is considered to be of substantial benefit in clinical care.

One study design limitation of the pivotal Study 303 was that it was open-label. This was chosen for two main reasons. Firstly, some IAT medications required intravenous administration and sham/blinding medications would be difficult given the differing routes of administration. Secondly, in the IAT arm, flexibility was needed for treating clinicians to adjust the medications based upon the patient’s clinical status, so as to comply with current treatment guidelines, while within the confines of the study design. Thus, clinicians would need to know what the patient was taking so that necessary adjustments could be made in a safe and ethical manner. This may have resulted in a potential selection bias by allowing the study patients and investigators to selectively switch arms (i.e., enter the Livtencity rescue arm) or discontinue from the study knowing what treatments the patients were taking. If a patient experienced an adverse event in the IAT group, the treatment may have been discontinued or switched to another agent, which would be considered a treatment failure in the primary efficacy endpoint analysis. The most common reason for treatment failure in the IAT arm was treatment discontinuation or switch due to an adverse event; thus, the efficacy results may be biased towards Livtencity. The sponsor explored this effect by conducting sensitivity analyses which excluded patients who discontinued treatment after several increasing time points (72 hours, 7, 14, 21, and 28 days). Statistically significant results for the primary endpoint that favoured Livtencity were still seen.

Given that patients in the IAT arm could receive different medications or combinations of medications, this treatment arm was heterogenic. When subgroup analyses were conducted based on IAT type, Livtencity retained its efficacy superiority, with the exception of the more than one IAT subgroup, which included patients treated with more than one standard anti-CMV viral agent. Given that treatment with more than one agent is the current standard of care described in the transplant society guidelines when CMV viremia clearance is not achieved, Livtencity does not show an efficacy advantage over the current standard of care with respect to CMV viremia clearance. The number of patients in the more than one IAT arm was small relative to the other subgroup analyses, and as such, caution with interpretation is needed.

The key secondary endpoint for Study 303 showed statistical superiority of Livtencity over IAT as 19% vs. 10% of patients achieved both CMV viremia clearance and CMV infection symptom control in the Livtencity and IAT groups, respectively (p = 0.013). However, the key secondary efficacy endpoint required that patients who had responded at Week 8 be included. Therefore any selection bias (as discussed above) that potentially lowered the response rate for the primary efficacy endpoint, would also affect the key secondary efficacy endpoint. The overall lower response rates seen for the secondary endpoint is to be expected, given the latent nature of the CMV virus.

The clinical course following viral recurrence for patients treated with Livtencity remains uncertain. When subgroup analyses were conducted (i.e., transplant type, baseline CMV DNA viral load, genotypic resistance to other anti-CMV agents, CMV syndrome/disease at baseline, and age), Livtencity retained its superior efficacy, albeit with wider confidence intervals. This was likely due to the smaller sample sizes, with the exception of the renal transplant subgroup (which had a very small number of patients, thus no clear interpretation can be made).

Overall, Livtencity at 400 mg twice daily is more effective than IAT in the treatment of transplant patients with resistant/refractory CMV infection. Efficacy may be driven in large part by increased tolerability, however given the known toxicities of the current agents, a more tolerable and still effective agent would provide more options for patients.

Resistance

Baseline confirmed genotypic resistance to the various IAT agents was approximately equal between the IAT and Livtencity arms. This was to be expected, as the study population inclusion criteria mandated refractory status (with or without resistance) to the standard IAT drugs. The highest proportion of resistance in both study arms was to ganciclovir/valganciclovir, which is also expected as these are the accepted first-line treatment agents for CMV in transplant populations. More than half of the patients in each group demonstrated resistance amino acid substitutions (RASs) for IAT agents at baseline (67% IAT and 56% Livtencity). In contrast to IAT, very few of the patients in each group demonstrated RASs for Livtencity at baseline (3.0% IAT and 0.5% Livtencity). 

In patients with baseline resistance to standard treatment, a higher proportion treated with Livtencity achieved the primary efficacy endpoint as compared to those treated with IAT (62.8% Livtencity vs. 20.3% IAT). The increased efficacy achieving the primary efficacy endpoint was lower for Livtencity in patients without baseline resistance (43.8% Livtencity vs. 32.4% IAT), indicating that Livtencity shows a greater effect when resistance is present, but does show some increased effect when the patients are refractory at baseline. Gene location for mutation analysis showed that mutations to multiple gene locations showed worsening performance for the primary efficacy endpoint for IAT patients, but better performance for Livtencity patients (7.7% vs. 69.2%). This indicated that for patients who are known to have resistance mutations in multiple locations, Livtencity is more effective at achieving CMV DNA clearance at the end of the 8-week treatment period.

A higher number of patients in the Livtencity group developed RASs to the typical IAT agent gene locations on study relative to those treated with IAT (12.9% Livtencity vs. 4.9% IAT). The rates of development of resistance with Livtencity treatment were higher and represent a risk to future treatment of the patient should their CMV levels recur.

A very low number of patients (4 in total in both treatment arms) showed a baseline resistance to Livtencity. None of these patients were able to meet the primary efficacy endpoint. As the numbers are small, it is difficult to make any conclusions with respect to effectiveness.

No Livtencity resistance developed in patients treated with IAT, while 21% of Livtencity-treated patients developed resistance to maribavir during the treatment period and 8% during the subsequent observation period. This indicates that Livtencity has a relatively low barrier to resistance in this refractory patient population during both the 8-week treatment period and 12-week observational period used in this study. In the population of Livtencity-treated patients who developed resistance during the treatment phase of pivotal Study 303, only one patient achieved the primary efficacy endpoint. For those who did not meet the primary efficacy endpoint (41/42 patients), 57% did achieve virologic response at some point during treatment. This indicates that if Livtencity resistance develops during the course of treatment, considerable efficacy is lost, including that at the end of the 8-week on-treatment period. 

When the genetic location of the treatment-emergent maribavir RASs were analyzed, 57% of them showed cross-resistance to ganciclovir/valganciclovir as well. This is concerning as a high proportion of these treatment-emergent maribavir RASs patients are also resistant to the other standard agents, leaving them with limited treatment options. For patients in whom resistance develops, discussions with their treating clinician on their remaining treatment options is needed.

Indication

The New Drug Submission for Livtencity was filed by the sponsor with the following indication:

Livtencity (maribavir) is indicated for the treatment of adults with post-transplant cytomegalovirus (CMV) infection/disease who are refractory and/or resistant to one or more prior antiviral therapies.

To support safe and effective use of the product, Health Canada approved the following indication:

Livtencity (maribavir) is indicated for the treatment of adults with post-transplant cytomegalovirus (CMV) infection/disease who are refractory (with or without genotypic resistance) to one or more prior antiviral therapies.

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

Clinical Safety

The clinical safety of Livtencity was evaluated based on data from the pivotal, Phase III, multicentre, randomized, open-label, active-controlled, superiority study (Study 303) described in the Clinical Efficacy section. Additional supportive data was provided from two supportive Phase II studies (Study 202 and Study 203). Both studies were randomized, dose-ranging, parallel group studies to assess Livtencity for the treatment of post-transplantation CMV infection.

The safety analysis from the pivotal study included 234 patients who received at least one 400 mg dose of Livtencity twice daily and 116 patients who received IAT which consisted of monotherapy or dual therapy with ganciclovir, valganciclovir, foscarnet, or cidofovir for an 8-week treatment phase. Treatment for all patients began following a diagnosis of resistant/refractory CMV. Adverse events (AEs) were collected during the treatment phase and the follow-up phase through Week 20. The mean exposures (standard deviation) for Livtencity and IAT were 48.6 days (13.82) and 31.2 days (16.91), respectively. Livtencity-treated patients received a maximum duration of treatment of 60 days.

The most commonly reported adverse reactions occurring in at least 1% of patients in the Livtencity group were: taste disturbance (44.8%), nausea (8.5%), vomiting (7.7%), immunosuppressant drug concentration level increased (6.4%), diarrhea (3.8%), abdominal pain (2.1%), neutropenia (1.7%), acute kidney injury (1.7%), anemia (1.3%), and decreased appetite (1.3%). The most commonly reported serious adverse reactions were acute kidney injury (1.3%), and nausea, fatigue, immunosuppressant drug concentration level increased, pyrexia, treatment failure, hepatic failure, gastroenteritis, hepatic enzyme increased, and vomiting occurring at <1%.

Serious adverse reactions occurred less frequently in the Livtencity group than in the IAT group (5.1% vs. 14.7%, respectively). No patients in the Livtencity group experienced serious, drug-related neutropenia or febrile neutropenia. In contrast, 7.1% of patients treated with ganciclovir/valganciclovir had serious related febrile neutropenia. In addition, 1% of patients in the Livtencity group and 11% in the foscarnet group experienced serious drug-related renal disorders (acute kidney injury and renal impairment). Adverse reactions led to discontinuation in 1.7% of the IAT group compared to 1.3% in the Livtencity group.

Taste disturbance (comprised of the reported preferred terms ageusia, dysgeusia, hypogeusia, and taste disorder) occurred in 44.8% of patients treated with Livtencity. These events rarely led to discontinuation of Livtencity (0.9%) and, for most patients, resolved while patients remained on therapy (37%) or within a median of 7 days (Kaplan-Meier estimate, 95% Confidence Interval: 4-8 days) after treatment discontinuation.

Immunosuppressant drug level increase occurred in 6.4% of patients treated with Livtencity. Livtencity has the potential to increase the drug concentrations of immunosuppressant drugs that are CYP3A and/or P-gp substrates with narrow therapeutic ranges (including tacrolimus, cyclosporine, sirolimus, and everolimus). Frequent monitoring of immunosuppressant drug levels is recommended throughout treatment with Livtencity, especially following initiation and after discontinuation of Livtencity. Adjustments of the dose may be needed.

Supportive Studies

Maribavir was generally well tolerated in the two Phase II supportive studies: Study 202 and Study 203.

In Study 203, 119 patients were exposed to Livtencity 400, 800, or 1,200 mg twice daily (BID) for up to 12 weeks. Overall, 95 to 100% of the patients who received Livtencity experienced treatment-emergent adverse events (TEAEs) while Livtencity-related TEAEs were reported for 67.2% of patients. Treatment-related severe adverse events were observed for 12 patients (10.1%) who received Livtencity compared to one patient in the valganciclovir group (3.4%). Similarly, AEs leading to discontinuation of the study drug were reported for 27 patients (22.7%) in the overall Livtencity group and 5 patients (12.5%) in the valganciclovir group.

The most commonly reported TEAE was dysgeusia (40.3% of patients in the overall Livtencity group) with no dose-related trend for the proportion of patients or the severity. Gastrointestinal disorders, which included diarrhea, vomiting, and nausea, were reported at a higher frequency in the overall Livtencity group compared to the valganciclovir group (42.9% vs. 25%). Accordingly, of the 10 patients who required a Livtencity dose reduction, 5 were due to gastrointestinal-associated toxicity. Another TEAE of special interest was increased immunosuppressant drug levels, which was observed in ten patients (8.0%) who received Livtencity. High or toxic levels of tacrolimus were reported for nine patients. Cyclosporine intoxication was reported for one patient. The incidence of increased immunosuppressant drug levels was higher in the higher dose groups. No significant neutropenia or nephrotoxicity was observed at Livtencity doses of up to 1,200 mg BID for up to 12 weeks.

Study 202 enrolled a population analogous to that of the pivotal Phase III Study 303. A total of 120 patients were exposed to 400, 800, or 1,200 mg of Livtencity BID for up to 24 weeks. All patients who received Livtencity experienced TEAEs, while maribavir-related TEAEs were reported for 77.5% of patients. Treatment-related severe AEs were observed for 20 patients (16.7%) and AEs leading to discontinuation of the study drug were reported for 41 patients (34.2%). Seventeen patients (14.2%) prematurely discontinued the study drug due to symptomatic CMV infection or disease. Thirty-two deaths were reported in this study. One death due to multi-organ failure in the group that received 800 mg of BID was considered by the investigator as possibly related to Livtencity therapy.

The most commonly reported TEAE was dysgeusia (65.0%) with no dose-related trend for the proportion of patients or the severity. Immunosuppressant drug level increases related to the administration of Livtencity were reported for 12 patients (10%) and occurred more frequently in the higher dose group. High or toxic levels of tacrolimus were reported for 11 patients and one patient had elevated sirolimus levels. One patient had discontinued Livtencity due to acute kidney injury secondary to increased tacrolimus levels. Other commonly reported TEAEs related to Livtencity included gastrointestinal disorders (nausea [20.8%], diarrhea [7.5%], vomiting [6.7%]), rash (6.7%), and anemia (5.8%). Twenty patients (16.7%) had AEs leading to interruption of the study drug, with nausea, dysgeusia, rash, and anemia reported as the AEs leading to Livtencity interruption in more than one patient.

Overall, Livtencity at a doses up to 1,200 mg BID was well-tolerated by HCMV-infected SOT or SCT recipients. The most commonly reported adverse reactions related to Livtencity treatment were dysgeusia, gastrointestinal disorders (nausea, vomiting, and diarrhea), immunosuppressant drug level increase, and anemia. These adverse reactions are described in the Livtencity Product Monograph and the possibility of increased immunosuppressant drug levels has been highlighted. There was no evidence of drug-related nephrotoxicity or neutropenia, which are adverse reactions associated with other anti-CMV agents. Altogether, the safety data from the Phase II studies provided supportive evidence for a favorable safety profile for Livtencity.

In conclusion, the CMV resistant/refractory transplant population is one with significant medical complexity and morbidity. Overall, in the studies evaluated, Livtencity treatment was not associated with treatment-limiting toxicities, which is in contrast to the currently available anti-CMV therapies. Appropriate warnings and precautions are in place in the approved Livtencity Product Monograph to address the identified safety concerns.

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