Australian public assessment report for Terlipressin

V. Pharmacovigilance findings

The sponsor submitted a Risk Management Plan which was reviewed by the TGA’s Office of Product Review (OPR).

Safety specification

The sponsor provided a summary of Ongoing Safety Concerns which are shown at Table 7.

Table 7. Summary of the Ongoing Safety Concerns as specified by the sponsor

Important identified risks	Cardiovascular:	myocardial ischaemia
	Respiratory:	wheezing/bronchospasm, dyspnoea, pulmonary oedema
	Gastrointestinal:	vomiting, diarrhoea, abdominal pain, intestinal ischaemia
	Skin Disorders:	peripheral cyanosis, livedo reticularis
	Electrolyte disturbances:	hypomagnesaemia
Important potential risks	Cardiovascular:	Torsade de Pointes, QT prolongation and ventricular fibrillation
Important missing information	None identified

The OPR reviewer considered that the summary of the Ongoing Safety Concerns was acceptable.

Pharmacovigilance plan

The sponsor proposed to undertake routine pharmacovigilance activities for all of the ongoing safety concerns.³⁴

In addition, the sponsor has proposed to undertake enhanced pharmacovigilance through the provision of safety information and active encouragement to health professionals (physicians, nurses, hospital pharmacists) for the submission of spontaneous reports when adverse events are identified.

The OPR reviewer had no objection to the sponsor undertaking routine pharmacovigilance activities for all of the ongoing safety concerns. Furthermore, the proposal by the sponsor to actively encourage the submission of spontaneous reports when an AE is identified during the initial postmarketing period was supported by the OPR reviewer. Given terlipressin will only be used in hospitals where prescribing will routinely be by specific specialists, this is a very practical proposal and will assist in the early identification of any changes in the adverse event profile.

Risk minimisation activities

As no additional risk minimisation activities were proposed, there is no risk minimisation plan.³⁵

The sponsor has stated that post-authorisation, they will be providing relevant product literature to physicians, nurses, pharmacists who are involved with the management of HRS-1 patients within hospitals. These include:

The approved Lucassin PI;

Physician's guide to prescribing Lucassin; and

Contact details of local safety officer for adverse events reporting.

Although the OPR reviewer had no objection to the sponsor only undertaking routine risk minimisation, the sponsor stated that they will provide a ‘Physician’s guide to prescribing Lucassin’ to health professionals. As no additional information was provided it was not entirely clear whether this guide is intended as an educational tool/additional risk minimisation activity. It was recommended to the Delegate that the sponsor be required to provide a copy of the ‘Physician’s guide to prescribing Lucassin’; if a copy is not yet available, the sponsor should provide further details on the information that will be provided in the guide.

Furthermore, the sponsor should provide further information on the “local safety officer for adverse event reporting”. It is not clear if this refers to a contact person within each hospital where terlipressin will be administered or if the sponsor is referring to an employee within their organisation. In addition the sponsor should indicate if they will provide the details for reporting adverse events directly to the TGA.

Summary of recommendations

The OPR recommended that the implementation of a RMP satisfactory to the TGA is imposed as a condition of registration.

It was recommended to the Delegate that the sponsor be required to provide a copy of the ‘Physician’s guide to prescribing Lucassin’; if a copy is not yet available, the sponsor should provide further details on the information that will be provided in the guide. In addition the sponsor should provide further information on the “local safety officer for adverse event reporting”.

VI. Overall conclusion and risk/benefit assessment

The submission was summarised in the following Delegate’s overview and recommendations:

Quality

There were no quality objections to registration. Lucassin was discussed at the 140th PSC meeting in August 2011. The PSC agreed that the drug product should be labelled as containing 0.85 mg terlipressin, although the potential confusion with clinicians used to the dose being referred to as 1 mg was recognised. The PSC also made recommendations for amendments to the PI that were agreed to by the sponsor.

Nonclinical

The nonclinical evaluator considered the overall toxicological component of the submission inadequate for a satisfactory risk assessment and hence did not offer adequate support for the registration of terlipressin acetate for the proposed clinical use. The evaluator noted that terlipressin had been in clinical use in Europe for over 20 years and suggested there may be sufficient clinical experience to offset the deficiencies in the nonclinical component of the submission.

The maximum non-lethal doses administered were 20 mg/ kg IV in mice, < 2 mg/kg IV in rats and 0.5 mg/kg IV in dogs; > 7 times the proposed 2 mg clinical dose on a body surface area basis. Deaths were attributed to severe vasoconstriction, a pharmacological effect. Repeat dose toxicity studies to 28 days were conducted in rats and dogs. Effects were seen in the kidneys, lungs, and testes attributable to the pressor activity of terlipressin. The nonclinical evaluator noted that due to aspects of the repeat dose study design the full toxicological profile of terlipressin is unlikely to have been revealed.

Terlipressin was not genotoxic. No carcinogenicity studies were submitted. This was accepted due to the short duration of use proposed and the negative genotoxicity findings.

Treatment related reactions were seen at injection sites in rats and dogs. Trace to severe perivascular inflammation was seen at about 3 times the proposed clinical dose on a mg/ kg basis.

The data suggested injection site reactions may be seen in the clinical setting.

The proposed specifications for impurities in the drug substance and degradants in the drug product are either below the ICH qualification thresholds or have been qualified.

Clinical

Clinical evaluation

The clinical evaluator recommended that terlipressin not be registered because of an unfavourable risk benefit balance. Terlipressin did not improve survival to 90 days in patients with HRS in the pivotal clinical trial (see also Clinical Summary and Conclusion above for further discussion).

Pharmacology

Literature references and limited clinical trial data, including a population pharmacokinetics study of patients with HRS were submitted. Terlipressin is a vasopressin analogue. It is a prodrug, being converted to lysine vasopressin in the circulation after the N-triglycyl residue is cleaved by endothelial peptidases. This results in a ‘slow release’ of the vasoactive lysine-vasopressin¹.

Pharmacokinetic data for terlipressin in healthy subjects were obtained from 2 published papers. In these studies terminal t_½ was approximately 1 hour, clearance approximately 9 mL/kg/min and mean Vd between 0.6 and 0.9 L/kg. Terlipressin is metabolised to lysine-vasopressin via sequential cleavage of the 3 glycyl groups. Once formed lysine-vasopressin is rapidly eliminated via various peptidase-mediated routes.

A population PK analysis was performed using data from subjects with HRS Type 1 enrolled in Study OT-401. This model predicted clearance of 0.375 L/h/kg (6.25 mL/kg/min) and median terminal t½ of 1.01 hours, similar to the values for these parameters in healthy subjects.

Terlipressin is not metabolised in blood or plasma. After IV administration the glycyl residues are cleaved in a stepwise fashion by endogenous proteases releasing lysine-vasopressin (LVP), the active metabolite. LVP is rapidly eliminated via peptidase-mediated routes. The majority of terlipressin metabolism occurs in liver and kidney. A small amount of terlipressin is excreted unchanged in urine. Terlipressin did not induce cytochrome P450 isoenzymes in in vitro studies in human hepatocytes.

The effects of terlipressin on heart rate, blood pressure, systemic and hepatic hemodynamic effects, skin blood flow, cerebral blood flow, antidiuretic effect and effect on coagulation were assessed either in clinical trials or reported in published papers of clinical trials.

There were 2 efficacy studies: TAHRS and OT-0401. In TAHRS mean arterial pressure (MAP) did not change significantly from baseline to end of treatment. In OT-0401 systolic blood pressure increased to 4.2 mmHg (3.9%) and diastolic blood pressure by 2.9 mmHg (4.6%) at 2 hours post-dose and transient decreases in heart rate (3 beats/min [3.4%]) were observed in patients given terlipressin. Changes in mean arterial pressure from baseline to end of study are shown in the CER. In patients with reversal of HRS the increases in MAP were greater in those patients given terlipressin than in those given placebo. For patients who did not have reversal of HRS, those given terlipressin had a lesser reduction in MAP than those given placebo. The effect of terlipressin on MAP was in the region of 3-4 mmHg difference compared with placebo both patients with reversal of HRS and those without reversal. Literature studies reported mean reductions in heart rate ranging from 2-8% patients with HRS given terlipressin.

In patients with cirrhosis a single dose of terlipressin was associated with reductions in hepatic venous pressure gradient, hepatic blood flow, portal venous blood flow, splenic blood flow and perfusion pressure while renal blood flow and perfusion pressure increased. The extent of change for these parameters is summarised in the CER. Terlipressin was also reported to cause decreases in skin blood flow in healthy volunteers and increases in cerebral perfusion and intracranial pressure in patients with acute liver failure.

In published studies terlipressin generally decreased plasma rennin, aldosterone and noradrenaline and increased atrial natriuretic peptide in cirrhotic and HRS patients with hyperdynamic circulation. In study OT-0401 reductions of 16% in rennin and 19% in aldosterone from baseline to end of study were noted in patients given terlipressin and albumin but these changes were not statistically significant. No significant differences were seen in study TAHRS. The extent of change in these vasoactive hormones is shown in the CER.

One published study in healthy volunteers showed an antidiuretic effect of terlipressin 7.5 µg/kg commencing within 60 minutes of administration with a progressive increase in urine osmolality during the 5 hours of observation. In 2 small studies from 1979/80 terlipressin did not affect levels of plasminogen activator, factor VIII or factor VIII-related antigen.

Four published studies examined drug interactions with terlipressin. These studies are summarised in the CER. Octreotide, prazosin and nitroglycerin each in combination with terlipressin resulted in small additional improvements in the hepatic venous pressure gradient that were greater than those seen with terlipressin alone. Most of the changes were not statistically significant however these were small studies. Human Atrial Natriuretic Polypeptide (ANP) in combination with terlipressin did not result in additional improvements or a trend to improved hemodynamic function in patients with cirrhoses and ascites.

Efficacy

Dose finding was limited. The dose regimen selected for the pivotal study was determined after a review of doses used in published studies and on the advice of the Terlipressin Clinical Advisory Board. Data from the population PK analysis of Study OT-0401 showed no correlation between response to exposure to terlipressin, assessed as AUC and HRS reversal response.

Two studies provided information on safety and efficacy and additional information was available from published studies. Only 1 double blind, placebo controlled study of terlipressin in patients with HRS Type 1 has been conducted.

Note the dose of 1 mg terlipressin diacetate stated to have been given in the pivotal study was actually 0.85 mg of terlipressin free base. This dose is referred to as 1 mg for consistency.

Study OT-0401 was a randomised, double blind, placebo controlled study of IV terlipressin in patients with HRS Type 1. It was conducted between 2004 and 2006 at 35 sites in the USA Russia and Germany. The primary objective was to demonstrate that IV terlipressin is safe and effective in the treatment of patients with HRS Type 1. Secondary objectives were to demonstrate that terlipressin improves renal function and survival compared to placebo.

There were 2 initial primary endpoints:

Treatment success required an initial reduction of serum creatinine (SCr) to ≤ 1.5 mg/dL followed by a confirmatory SCr measurement of ≤ 1.5 mg/dL 48 hours after the initial HRS reversal and an additional SCr < 2.5 mg/dL at Day 14, without intervening liver transplant or dialysis. Patients who did not have an SCr collected at these 3 time points were considered non-responders.

HRS reversal defined as the number of patients who demonstrated reversal of HRS (SCr ≤ 1.5 mg/dL on at least 2 measurements obtained 48 ± 2 h apart), without intervening dialysis or liver transplantation divided by the total number of patients in the ITT population.

Following review of the study data, which did not show a statistically significant difference in treatment success between terlipressin and placebo and on discussion with the FDA, the primary endpoint was amended and additional SCr values collected from patients medical records were incorporated into a re-analysis of the data. The revised primary efficacy parameter was treatment success at Day 14 defined as the percentage of patients who were alive at Day 14 and who demonstrated a reversal of HRS (SCr ≤1.5 mg/dL on ≥ 2 measurements obtained 48±8 h apart), without dialysis or recurrence of HRS. Other efficacy endpoints included measures of renal function and overall and transplant free survival up to Day 180.

Patients received either terlipressin starting at 1 mg every 6 h (q6h), increasing to 2 mg q6h after 3 days if the patient did not achieve a ≥ 30% decrease in SCr or placebo. Most patients also received albumin, this was initially titrated to a specific albumin level but was amended during the study such that all subsequent patients received a standard albumin dose (100 g on Day 1 and 25 g on each subsequent day until the end of study drug administration).

Patients were required to meet the International Ascites Club (IAC) diagnostic criteria for HRS Type 1 with some additions to allow for a homogeneous HRS Type 1 population. Patients with ongoing shock, uncontrolled bacterial infection, fluid loss, intrinsic or parenchymal renal disease or who were either receiving nephrotoxic drugs or who had liver disease as a result of drugs that were also nephrotoxic (for example, paracetamol overdose) were excluded from study.

For the initial definition of treatment success, where missing SCr values at Day 14 were imputed as “not a treatment success” both the ITT and MITT analyses failed to show a statistically significant difference between terlipressin and placebo. These results are shown in the CER. For the revised definition of treatment success at Day 14, in the ITT population a total of 16/56 (28.6%) of patients given terlipressin and 7/56 (12.5%) given placebo met the revised criteria for treatment success. This difference was statistically significant (p=0.037). Reversal of HRS was achieved in 19/56 (33.9%) patients given terlipressin and 7/56 (12.5%) patients given placebo (p = 0.008) for the ITT population. The difference for the MITT population was also statistically significant.

The CER shows the number of patients in each group alive at each time point through to 180 days. Follow up to Day 180 showed no significant differences in survival rates at any of the time points assessed (Days 14, 30, 60, 90 and 120). Twenty four patients given terlipressin and 21 given placebo remained alive at Day 180. Additionally transplant-free survival was similar in the 2 groups to Day 180. Overall, terlipressin treated patients received their transplants later (mean 31 days) compared with the placebo treated patients (mean 21 days). Thirteen patients who had responded to treatment (10 terlipressin and 3 placebo) had died as of the Day 180 follow up with none of these deaths attributed to relapse of HRS, the majority died as a result of their underlying liver failure.

Eighteen terlipressin and 17 placebo treated patients received a liver transplant up to Day 180 (6 given terlipressin and 5 given placebo in the ITT analysis). Seven terlipressin treated and 5 placebo-treated patients who had not received liver transplants were alive at Day 180.

Study TAHRS was a randomised, open, controlled study of terlipressin in patients with hepatic cirrhosis and HRS Type 1 or Type 2. The primary objective of this study was to investigate the effects of treatment with terlipressin and albumin on the survival of patients with hepatic cirrhosis and HRS Type 1 or 2. The study also planned to evaluate whether the improvement in renal function, if it occurred, resulted in an increase in the probability of survival to transplantation and in a reduction of post-transplant complications.

Patients had HRS Type 1 (73.9%) or 2 (26.1%) with SCr > 2.0 mg/dL. They were not required to be candidates for liver transplant. Patients were randomised to receive either terlipressin with 20% human albumin or 20% human albumin alone. Doses of terlipressin were from 3 to 12 mg daily given in divided doses every 4 hours with 20 to 40 g daily of albumin. This was a supportive study because it was open, included patients with HRS Type 2, did not use the terlipressin dose regimen proposed for registration, allowed crossover rescue therapy in patients not responding to albumin alone and was of limited size. It was terminated after 4 years with 46 patients enrolled. At that time the estimated sample size required to demonstrate a significant treatment difference in survival to transplantation was 431 patients per group. To achieve this sample size would not have been possible within a reasonable time period so the study was terminated.

Published studies

Five randomised, placebo controlled studies, 5 non-randomised controlled studies, 1 case control study, 6 uncontrolled studies and 2 meta-analyses were presented to support efficacy and safety of terlipressin for the treatment of HRS Type 1. These studies/ analyses are summarised in the CER. A major difficulty in examining the literature was the inconsistency of the definition of responder and HRS reversal. These studies generally enrolled small numbers of patients with HRS Type 1 and reported small, not statistically significant differences in their efficacy endpoints between terlipressin and the comparator.

The Fabrizi meta-analysis published in 2006 initially considered data from 154 patients in 11 studies with 127 of these patients having HRS Type 1. This analysis was updated in 2009. These patients received from 1 to 6 mg/d terlipressin for from 2 – 26 days ± plasma expanders. Reversal of HRS was defined as a decrease in SCr to 1.5 mg/ dL or lower at the end of treatment with results presented for all patients (HRS Type 1 + HRS Type 2). A sub-analysis of 5 studies that included only patients with HRS Type 1 reported a pooled rate of HRS reversal of 0.53 (95%CI 0.41; 0.65). The update of this meta-analysis published in 2009, was not referred to in this submission.

The second meta-analysis, (Gluud 2006) included 3 randomised, controlled studies with a total of 51 patients. Two of the studies included only patients with HRS Type 1 and the third did not specify whether patients had Type 1 or Type 2 HRS. The terlipressin dose given was 1 mg bd and therapy duration varied from 2 to 15 days with maximum follow up to 14 days after treatment. Co-interventions included albumin 20 g/ day, fresh frozen plasma (150 mL qid), cimetidine (800 mg/d), sodium restriction, water restriction and dopamine infusion. 5/25 (20%) patients randomised to terlipressin and 15/23 (65%) randomised to the control group died. It was reported that terlipressin reduced mortality by 34% ((%%CI: -0.56 to -0.12).

There may have been some overlap in patients with the same patients included in more than one published study that was included in this meta-analysis. The clinical evaluator noted that control for bias in the Gluud meta-analysis was unclear.

Safety

Safety data from individual studies were not pooled due to design and enrolment differences between Studies OT-0401 and TAHRS. In these studies patients were exposed to terlipressin for a mean of 6.3 days with a maximum of 14 days in 0401 (n=56) and 7.8 days with a maximum of 32 days in TAHRS (n=23). In Study 0401 34 (60.7%) of patients given terlipressin withdrew compared with 45 (81.8%) given placebo.

The most frequent reasons for withdrawal were lack of efficacy (21.4% terlipressin versus 38.2% placebo); liver transplant (10.7% versus 9.1%) and death on treatment (10.7% terlipressin versus 5.5% placebo). In TAHRS the most frequent causes were lack of efficacy (4.3% for terlipressin versus 11% for placebo) and death on treatment (21.7% for terlipressin versus 13.0% for placebo).

In Study OT-0401 the incidence of adverse events and serious adverse events were similar in terlipressin and placebo groups but serious events and events considered treatment related were more frequent in the terlipressin group (treatment related AEs 32% versus 22% placebo; serious treatment related AEs 9% versus 2% placebo). In TAHRS adverse events were more frequently reported in patients given terlipressin (91% versus 71% placebo) as were events considered serious and/or treatment related. A much higher incidence of adverse events were considered treatment related in patients given in TAHRS compared with 0401 (32.1% for terlipressin in 0401 versus 78.3% in TAHRS) and treatment related and serious (8.9% for terlipressin in 0401 versus 56.5% in TAHRS).

Differences in frequency of events with incidence of ≥ 10% grouped by System Organ Class are shown in the CER. In 0401 events from the SOCs Respiratory, Thoracic and Mediastinal Disorders (39.3% versus 23.6% placebo) and Infections and Infestations (32.1% versus 20.0% placebo) were more frequent in patients given terlipressin than placebo. Individual adverse events reported more frequently in patients given terlipressin compared with placebo in 0401 were: anxiety (7% versus 2% placebo); hypomagnesaemia (7% versus 0% placebo); multi-organ failure (7% versus 0% placebo); sepsis (7% versus 2% placebo), wheezing, bradycardia, flatulence, pain extremities and pneumonia.

The above frequencies are quite different from those of TAHRS, suggesting the open nature of that study may have influenced adverse event reporting. In that study anxiety, hypomagnesaemia, and multiorgan failure were not reported at all and sepsis was reported in only 1 patient (given terlipressin) The largest differences were in the incidences of abdominal pain (21.7% versus 4.3% placebo) and diarrhoea (30.4% versus 8.7%). The most frequent events in patients given terlipressin were: diarrhoea, abdominal pain, acute pulmonary oedema, hepatic encephalopathy, intestinal ischemia, hepatic failure, HRS, dyspnoea and fluid overload.

Most ARs considered treatment-related were reported in only 1 patient. Adverse events leading to death at any time during study (to Day 180) are shown in the CER.

The most frequent causes of death associated with an adverse event in patients given terlipressin were: Hepatobiliary Disorders (29% terlipressin versus 38% placebo in 0401 and in 56% terlipressin versus 56% placebo in TAHRS); Infections and Infestations (10.7% terlipressin versus 1.8% placebo in 0401 and in 17.4% terlipressin versus 8.7% placebo in TAHRS). Deaths due to an adverse event in Renal and Urinary Disorders were reported in 4% terlipressin versus 7% placebo in 0401 and in no patients in TAHRS. Limited data from published studies concerned adverse events. Details of adverse events were generally sparse and information on deaths not consistently reported. The clinical evaluator noted that of 1433 patients reported on in published papers only 7 deaths were reported.

Uses other than HRS Type 1 were included in safety data from clinical trials. Overall the most frequently reported adverse events associated with terlipressin in published papers were: abdominal pain/cramps; pallor, increased bowel movements, hypertension and diarrhoea. Subsequent information from published papers also included pulmonary oedema as a more frequently reported adverse event. The WHO database included abdominal pain, chest pain substernal, vasospasm and headache as the most frequently reported adverse events associated with terlipressin.

The clinical evaluator considered the following events of regulatory importance: ischaemic, gastrointestinal, respiratory, cardiac events; infection and skin and subcutaneous tissues. In the clinical studies 1 patient given terlipressin in each of studies 0401 and TAHRS had myocardial infarction versus none in patients given placebo. No association between use of terlipressin and QT prolongation was apparent. In Study 0401 wheezing and bronchospasm were more frequent in patients given terlipressin (11% terlipressin versus none placebo) but was not reported in study TAHRS. In the clinical studies terlipressin was associated with a higher incidence of death due to infection but none of these deaths were attributed to use of terlipressin. There was no increase in reporting of skin and subcutaneous tissue vasoconstriction associated events in patients given terlipressin.

The clinical evaluator has noted that in the clinical trials, patients given terlipressin who did not have HRS reversal had lower survival rates than patients given placebo, though there was no statistical analysis of this and overall terlipressin made no statistically significant difference to survival over any period up to Day 180 assessed in the pivotal clinical study.

Sponsor response

The main contention of the sponsor’s response to the clinical evaluation report was that the most appropriate efficacy endpoint should be a measure of the effectiveness in reversal of HRS, rather than survival. The clinical evaluation report has placed most emphasis on survival (at any of the time points measured). It was stated that large numbers of patients would be required to demonstrate a statistically significant effect on survival and that, as this is a rare condition, this was not practical. The HRS endpoints used in the pivotal study had been negotiated with another regulatory agency (the FDA), though terlipressin does not have a marketing authority for the proposed indication in the USA.

The sponsor also proposed that a small increase in survival time may allow for liver transplant or for recovery from an episode of decompensated cirrhosis caused by a reversible event.

The sponsor also noted the clinical evaluator’s concern that the survival rate was lower in the non-responders given terlipressin than in patients given placebo (responders and non-responders). When the analysis was of non-responders only in each group the survival rates were comparable.

The sponsor produced a Kaplan-Meier plot showing overall survival for HRS reversal versus no HRS reversal in the pivotal study. This plot includes patients given terlipressin in both the HRS reversal and non-reversal groups. This plot shows that HRS reversal correlates with survival. This analysis groups those patients who received terlipressin and did not respond with reversal and those who received placebo and also did not have reversal of their HRS. Patients given placebo and had reversal were grouped with those who received terlipressin and also had reversal of HRS. This analysis is useful in showing the extent of correlation between reversal of an episode of HRS and short term survival. The difference between proportion of patients surviving (with and without reversal of HRS) is greatest at Day 14, where from the plot it appears that approximately 40% of patients without reversal and 5% of those with reversal have died.

Risk management plan

No additional risk minimisation activities were proposed and there was no risk minimisation plan. The sponsor proposed routine pharmacovigilance with enhanced pharmacovigilance through the provision of safety information and active encouragement to health professionals for the submission of spontaneous reports when adverse events are identified. The sponsor noted that terlipressin will be given in a hospital setting and has a well established, predictable safety profile.

Should Lucassin be approved, the sponsor intends to provide healthcare professionals with a Physicians Guide to prescribing in addition to the PI and contact details of the local safety officer. The Physicians Guide had not been made available at the time of completion of the RMP evaluation. The RMP evaluator accepted the proposed level of risk management activity.

Risk benefit analysis

Delegate considerations

Six months after an episode of HRS-1 very few patients are likely to be alive without a liver transplant. Any survival benefit appears to be mostly during the first 14 days after commencement of treatment and occurs only in a minority of patients. In the pivotal study, 7 terlipressin treated patients and 5 placebo treated patients who had not received liver transplants were alive at Day 180. Thus the major clinical benefit of terlipressin would be to extend the duration of survival prior to transplant, however it does not appear to do this for very long or for the majority of patients.

For the total population given terlipressin there was a mean difference of 9.7 days (median difference 7.5 d) in time to liver transplant in the overall patient population who received transplants (18 given terlipressin and 17 given placebo). These times are likely to vary depending on the availability of livers for transplant for an individual patient and are therefore not necessarily reproducible. No statistical analysis was performed on these data.

While it is clear that terlipressin has an effect in reversing HRS-1, no effect on survival was demonstrated at any time point assessed during the study for the overall patient population.

HRS-1occurs in end stage liver disease. As noted in the sponsor’s response to the clinical evaluation report, HRS-1 is often one of a series of multiple concurrent life threatening complications in end stage liver disease. Demonstrating a survival benefit from treating the HRS-1 component amidst other concomitant life-threatening pathologies presents a challenging task. This was only partially met in the data submitted.

In the response to the clinical evaluation report the sponsor has shown that reversal of HRS correlates with a short term improvement in rate of survival (up to 90 days) with the maximal difference in proportion of survivors at day 14 of treatment.

In the pivotal study terlipressin was shown to reverse HRS-1 in 19/56 (33.9%) patients given terlipressin versus 7/56 (12.5%) patients given placebo (p = 0.008) for the ITT population. The absolute difference in HRS reversal was approximately 20% and it is these patients who may have some short term survival benefit from treatment, with the difference in survival rates being most apparent at 14 days from commencement of treatment. For 80% of patients with HRS Type 1 there was no benefit from treatment with terlipressin.

It seems likely that for approximately 20% of patients terlipressin results in a few additional days to weeks of survival without a liver transplant. The clinical benefit of such a small increase in survival time depends on whether this additional time is likely to result in a clinically significant increase in the availability of a liver for transplant. Therefore the clinical benefit of terlipressin will vary with the availability of livers for transplant and it is thus not possible to estimate how many patients will receive transplants (and have increased probability of longer term survival) because of the use of terlipressin. Where few livers are available the benefit would be negligible.

The Delegate proposed to reject Lucassin (terlipressin) for treatment of hepatorenal syndrome (HRS) Type 1 because HRS-Type 1 occurs in the setting of end stage hepatic failure. Treatment of HRS-Type 1 does not affect the underlying hepatic failure. Although terlipressin is better than placebo in reversing HRS-1 it has not been shown to increase survival to a clinically significant extent in either the total population given terlipressin or in any subgroup of patients with HRS Type 1.

While a short term increase in transplant free survival probably occurs for those 20% of HRS-Type 1 patients who respond to treatment, the medium to longer term survival of these patients will depend on the availability of livers for transplantation. The probability of a suitable liver becoming available during the additional days probably (but not statistically proven) to be gained from use of terlipressin would be variable but very likely to be extremely low in Australia at present.

While terlipressin may well improve renal function in patients with HRS before transplantation to avoid dialysis and significant renal impairment as stated by Professor McCaughan, this was not the proposed indication and evidence towards that use was not the subject of this submission.

The advice of the ACPM was particularly requested on:

Whether survival is the appropriate endpoint for assessment of efficacy or whether, as proposed by the sponsor, some measure of reversal of HRS Type 1 is more appropriate.

Given overall medium to longer term survival of a patient with HRS Type 1 is dependent on a liver transplant could Lucassin be registered with a limited indication permitting use in HRS Type 1 only when it is likely a liver will become available for transplant to that patient in the near future?

If the latter is acceptable, what limitations could be placed in the indications to reflect this limited patient access? Given the lack of ability to predict the availability of suitable livers for transplant to any individual patient how could such a limited access system be managed?

Response from sponsor

The sponsor noted that the Delegate proposed to reject the application for Lucassin (terlipressin) in the treatment of hepatorenal syndrome (HRS) Type 1 (HRS-l), because "it has not been shown to increase survival to a clinically significant extent". Although the Delegate acknowledged that "terlipressin is better than placebo in reversing HRS-1", the ultimate tests applied for the determination of clinical significance were based on two factors other than HRS reversal: (i) extent of transplant free survival benefits, which the Delegate described as "additional days" and "short term"; (ii) likelihood of liver availability during the extended survival period.

The Delegate further sought particular advice from the ACPM on: "whether survival is the appropriate endpoint for assessment of efficacy or whether ... some measure of reversal of HRS Type 1 is more appropriate"; and, whether/how Lucassin could be registered with a limited indication permitting use only when it is "likely a liver will become available for transplant to that patient in the near future."

The sponsor disagreed with the Delegate's conclusions on the grounds that: (a) HRS reversal is the appropriate endpoint for HRS-l, which has been largely overlooked; (b) the Delegate's over-emphasis on the survival data is disproportionate to the fact that it is supportive evidence, and that definitive treatment differences for such measures cannot be practically characterised; (c) the premise of the assessment which includes liver availability, is flawed; (d) the conclusions reached are unreasonable and substantially under-represent the true efficacy of the product.

In terms of the indication, the sponsor maintained that approval of Lucassin for the original proposed indication is justified based on the evidence submitted and the following contentions:

Lucassin is indicated for the treatment of patients with hepatorenal syndrome (HRS) Type 1.

Accordingly, the indication statement in the proposed Product Information (PI) was not modified.

Should the TGAI ACPM consider it necessary to provide further guidance on the use of the product in order to grant approval the sponsor proposed the following alternative indication statement for consideration:

Lucassin is indicated for the treatment of patients with hepatorenal syndrome (HRS) Type 1.

Consultation with a local transplant unit is recommended to discuss suitability of treatment and liver transplant referral/assessment.

No further changes were proposed to the dosage and administration information.

The key elements of the sponsor’s contentions are summarised below.

HRS reversal versus survival

In the HRS-l setting, renal measures based around HRS reversal are the appropriate primary endpoints, not survival. The Delegate's deliberation reflects an over-emphasis on the supportive survival data, with insufficient weight given to the primary outcomes based around HRS reversal. This is unreasonable as it is inconsistent with the broad acceptance of HRS reversal as a clinically relevant endpoint in HRS-l and disregards the fact that survival treatment differences cannot be practically characterised.

The accepted therapeutic goal of improving renal function in HRS-l, that is, HRS reversal, is unequivocal.

The benefit and clinical relevance of the rapid correction of acute renal failure in the setting of liver disease is well supported by the literature, clinical guidelines and a recent statement from the Australian Liver Association ('ALA Statement’).

Furthermore, while survival outcomes cannot feasibly be investigated in HRS-l (see below), HRS reversal is a measurable endpoint that has been shown to correlate strongly with improved survival. For the liver transplant setting, the data suggests that HRS reversal can provide clinically significant additional survival time to allow organ procurement which can be life saving for the patient.

The over-emphasis by the Delegate on the supportive survival data is unreasonable considering that the submitted studies were not intended nor were they powered to detect survival treatment differences; and that studies of such power are practically impossible to execute for the proposed orphan HRS-l indication.

Given that survival outcomes cannot be robustly investigated in this indication, the sponsor contended that the primary HRS reversal data should rightly be the principal determinant of the drug's efficacy.

The premise of the efficacy assessment is unreasonable; the conclusions reached underestimated treatment benefits

The Delegate's negative recommendation essentially means that patients with HRS-l should not be treated with terlipressin, and instead receive IV albumin alone. The consequence of this would be an increased likelihood of mortality within 30 days, or if the patient receives a liver transplant, emerge from the transplant with progressive renal disease, increased morbidity and mortality. Clearly, this is clinically undesirable.

Liver availability features prominently in the Delegate's deliberation. However, the sponsor contended that the regulatory assessment of any drug should be based upon its intrinsic properties demonstrated in the target population, rather than on an unpredictable external factor (liver availability) over which no one has any control. Based on the Delegate’s reasoning, it would be extremely difficult to have any drug approved for HRS-l where liver availability happens to be very low during a particular time period. Thus, the sponsor contended that liver availability is an unreasonable test to apply for the determination of regulatory approval. For a critically ill patient presenting with HRS-1, good medical practice dictates that the best treatment be given to stabilise the patient, irrespective of liver availability.

It is also problematic that, in determining that the clinical benefit of terlipressin is not of a "clinically significant extent", the Delegate has not quantitatively defined with justification, as to what would constitute a clinically significant benefit.

The Delegate's interpretation of the efficacy data reflects an underestimation of the clinical benefits achieved with terlipressin.

True extent of the clinical benefits with terlipressin

Of the pivotal study OT-0401, the Delegate commented that, "it is clear that terlipressin has an effect in reversing HRS-1 ... The absolute difference in HRS reversal was approximately 20% ". In essence, the data shows that patients treated with terlipressin were approximately three times more likely to achieve HRS reversal (primary endpoint) than compared to placebo (33.9% versus 12.5%).

When responder analyses were undertaken, a strong correlation was shown to exist between the primary endpoint of HRS reversal (as well as Treatment Success) and survival. Of this, the Delegate described the survival gain variously as, ''few additional days to weeks of survival without a liver transplant" and “additional days".

The sponsor argued that none of the stated durations accurately reflects the significantly higher rates of transplant free survival for HRS reversal compared to no HRS reversal observed to 180 days in both the terlipressin and placebo groups. The significance of this survival benefit in HRS reversal responders is reinforced by a recently plotted Kaplan-Meier plot, which pools the data across the two treatment groups. The key findings from this plot were that: (i) The differential in transplant-free survival for HRS reversal versus no HRS reversal, which reached 37% at Day 14, was sustained to Day 180 (40% differential); (ii) The survival gains in patients with HRS reversal were far in excess of "a few days" or "weeks"; (iii) half of the patients achieving HRS reversal were still alive at 180 days without a transplant, compared with a dismal 10% where no HRS reversal was achieved.

The above takes on even greater significance when one considers the data in the ANZ Liver Transplant Registry (http://www.anzltr.org/statistics.html), which indicates that in 2010, the median time to transplant was 84 days.

Benefit risk conclusion

Patients presenting with HRS-l are critically ill and face real danger of imminent mortality. In clinical practice, the presentation of a very sick patient with acute renal failure associated with HRS-l are indicators for rapid assessment of suitability for liver transplantation and, if found suitable, then rapid elevation up the list for receiving donor organ. According to the data from the ANZ Liver Transplant Registry, the median time to transplant was 84 days in 2010.

In the setting of HRS-1 terlipressin has been shown to be efficacious in reversing the acute renal failure that is the central, life threatening feature of the condition (HRS reversal). By reversing HRS, the transplant-free survival of patients can be significantly extended by up to 180 days. These are highly clinically significant benefits which refute the Delegate's conclusion of 'few additional days to weeks of survival". The clinical implications of these benefits are that:

(i) For HRS-1 patients awaiting donor organ who respond to terlipressin treatment, their chance of successful bridging to curative transplantation is greatly improved, considering a median time to transplant of 84 days.

(ii) For patients yet to have been assessed for transplant suitability, terlipressin allows time for this to be expedited, along with procurement of donor organ.

(iii) Where transplantation is not possible, reversal of HRS provides additional time to allow clinically significant recovery of the underlying decompensated liver disease, particularly in the setting of decompensated cirrhosis provoked by a reversible event, for example, alcoholic hepatitis.³⁶

In terms of safety, this was not raised as an issue in the Delegate's closing remarks. As previously discussed in the sponsor's response to the CER, the adverse events (AEs) of terlipressin are predictable and recognisable and, since terlipressin is used within hospitals under a high vigilance setting, drug related AEs can be anticipated, recognised, and promptly managed.

Given these findings, the Delegate's negative recommendation on Lucassin is unreasonable which, if upheld, would be counterproductive and deleterious to patient care and their survival outcomes.

The efficacy/safety findings clearly weigh in favour of a positive benefit risk profile, further reinforced by the critical unmet need in an orphan HRS-l population that terlipressin would fulfil.

Advisory committee considerations

The Advisory Committee on Prescription Medicines (ACPM) (which has succeeded ADEC), having considered the evaluations and the Delegate’s overview, as well as the sponsor’s response to these documents, advised the following:

Efficacy

Overall, the small data set has demonstrated that this product has sufficient efficacy when the end point of HRS reversal is considered; however, the measurement of survival is the most appropriate clinical end point for hepatorenal syndrome Type 1. Few patients presenting with HRS 1 survive beyond 180 days without a liver transplant. It is noted that there are a range of external factors impacting on the timing for transplant organ availability and therefore the difficulty of using survival as an end point.

The evidence supports limiting the indication to include only the patient population who are actively being considered for a liver transplant.

The sponsor should be encouraged to conduct further studies on the likelihood and duration of survival and to include data points for patients who do not respond to this product.

Safety

Despite the significant side effect profile of this product it has a record of safe use that can be attributed to its restriction to use by experienced health professionals. This restriction must be continued.

Indication

The ACPM considered this product to have a positive benefit risk profile for the indication of:

Lucassin is indicated for the treatment of patients with hepatorenal syndrome (HRS) Type 1 who are actively being considered for a liver transplant.

The ACPM also made a recommendation concerning the PI but this is beyond the scope of this AusPAR.

The ACPM advised that the implementation by the sponsor of the recommendations to the satisfaction of the TGA, in addition to the evidence of efficacy and safety provided for Lucassin would support the safe and effective use of this product.

Outcome

Based on a review of quality, safety and efficacy, TGA approved the registration of Lucassin terlipressin 0.85 mg powder for Injection vial, indicated for:

The treatment of patients with hepatorenal syndrome (HRS) Type 1 who are actively being considered for a liver transplant.

Specific conditions applying to these therapeutic goods

The implementation in Australia of the terlipressin Risk management Plan (RMP) version 2, dated 27 July 2011, and any subsequent revisions, as agreed with the TGA and its Office of Product Review.