Summary Basis of Decision for Revolade ™

3.3.1 Pharmacodynamics

Following repeat dose administration of Revolade™, platelet counts were observed to increase in a dose-dependent manner. After ten days of dosing, peak values were reached a maximum 2 weeks after the initiation of dosing, and returned to baseline within approximately 2 weeks after the last dose of eltrombopag. Platelet function tests indicated no increase or decrease in platelet activation in Revolade™-treated ITP patients compared to untreated ITP patients or healthy volunteers.

3.3.2 Pharmacokinetics

Absorption

Revolade™ plasma levels increased in a dose proportional manner, with peak absorption concentration observed 2-6 hours following oral administration. The elimination half-life was between 21-32 hours.

Distribution

Following oral administration of a radiolabelled dose, Revolade™ accounted for most (64%) of the radioactivity circulating in plasma. In vitro data showed that Revolade™ was highly bound to human plasma proteins (>99%), predominately to albumin.

Metabolism

Revolade™ is metabolised through various pathways, but primarily through cytochrome P450 (CYP) 1A2 and CYP 2C8 and in conjugation with glucuronic acid. In vitro data and one clinical interaction study indicated that Revolade™ did not induce or inhibit CYP450 enzymes.

In vitro studies demonstrated that Revolade™ is not a substrate for the organic anion transporter polypeptide, OATP1B1, but is an inhibitor of this transporter. In vitro studies also demonstrated that Revolade™ is a breast cancer resistance protein (BCRP) substrate and inhibitor. Rosuvastatin being a substrate for those transporters, co-administration with Revolade™ resulted in a 55% increase in exposure, as measured by the area under the curve (AUC).

Elimination

The primary route of elimination was through the faeces (59%) and to a minor extent via the urine (31%). Urinary elimination was essentially complete within the first 48 hours post administration. Faecal elimination continued for several more days, with most of the dose recovered by 144 hours after dosing.

Drug-Drug Interactions

Co-administration of Revolade™ with rosuvastatin increased the maximum serum concentration (C_max) of rosuvastatin by 2.03-fold and the AUC by 55%. Therefore a reduced dose of rosuvastatin should be considered when administered concomitantly with Revolade™ and careful monitoring of possible side effects should be undertaken. Interactions may be expected with other statins (e.g., pravastatin, simvastatin, and lovastatin), however clinically significant interactions are not expected between Revolade™ and either atorvastatin or fluvastatin. When co-administered with Revolade™, a reduced dose of statins should be considered and carefully monitored. In addition, concomitant administration of Revolade™ and other OATP1B1 and BCRP substrates should be undertaken with caution.

Co-administration of Revolade™ with lopinavir/ritonavir (LPV/RTV) may cause a decrease in the concentration of Revolade™. Therefore, caution should be used when co-administering Revolade™ with LPV/RTV therapy; and platelet counts should be carefully monitored in order to ensure appropriate medical management of the dose of Revolade™.

Revolade™ combines with polyvalent cations such as aluminium, calcium, iron, magnesium, selenium and zinc. Antacids, dairy products and other products containing polyvalent cations such as mineral supplements should be administered at least four hours apart from Revolade™ dosing to avoid significant reduction in Revolade™ absorption.

Food-Drug Interactions

Administration of a single 50 mg dose of Revolade™ with a standard high-calorie, high-fat breakfast that included dairy products reduced plasma Revolade™ concentrations. Food low in calcium (<50 mg calcium) did not significantly impact plasma Revolade™ exposure, regardless of calorie or fat content.

Special Populations

Renal Impairment

A pharmacokinetic analysis investigating use of Revolade™ in patients with mild, moderate, and severe renal impairment compared to healthy volunteers following administration of a single 50 mg dose of Revolade™ showed that plasma Revolade™ AUC was 32%, 36%, and 60% lower and C_max was 30%, 19%, and 54% lower in patients with mild, moderate, and, severe renal impairment, respectively, compared to healthy volunteers. Caution is therefore warranted when administering Revolade™ in patients with mild to moderate renal impairment. Administration of Revolade™ is not recommended for use in patients with severe renal impairment (creatinine clearance <30 mL/min) unless the expected benefit outweighs the risk.

Hepatic/Biliary Impairment

Results from a pharmacokinetic study conducted in patients with mild (Child-Pugh score of 5 to 6), moderate (Child-Pugh score of 7 to 9), and severe hepatic impairment (Child-Pugh score of 10 to 15) compared to healthy volunteers, following administration of a single 50 mg dose of Revolade™ showed that plasma Revolade™ AUC values were 41% higher in patients with mild hepatic impairment and 80% to 93% higher in patients with moderate to severe hepatic impairment. Therefore, Revolade™ should not be used in ITP patients with moderate hepatic impairment (Child-Pugh score 7 to 9) unless the expected benefit outweighs the risk. For patients with severe hepatic impairment, use of Revolade™ is contraindicated.

Following administration of Revolade™, abnormal liver function was also observed with increases in serum alanine aminotransferase (ALT), and aspartate aminotransferase (AST). These findings were mostly mild (Grade 1-2), reversible and not accompanied by clinically significant symptoms that would indicate impaired liver function. Across three placebo-controlled studies, one patient in the placebo group and one patient in the Revolade™ group experienced a Grade 4 liver test abnormality. Therefore, careful measuring of serum ALT, AST and bilirubin levels prior to initiating treatment with Revolade™, in addition to ongoing monitoring of these levels every 2 weeks during the dose adjustment phase, and monthly following the establishment of a stable dose is recommended. Further guidelines for monitoring liver enzymes have been provided in the Product Monograph.

Race

Based upon a population pharmacokinetic analysis, data showed that East Asian ITP patients had approximately 87% higher Revolade™ plasma values compared to non-East Asian ITP patients (mainly Caucasian in origin). Similar results were also observed between East Asian and non-Asian healthy volunteers. The pharmacokinetic difference observed between East Asian and Caucasian might suggest that multiple factors such as body weight and genetic differences in metabolizing enzymes and transporters may contribute to this observed difference. Evaluation of platelet response across ITP studies also suggests that East Asian ITP patients were more likely to achieve platelet counts >400 x 10⁹/L. In addition, there was a trend to a higher incidence of hepatobiliary laboratory abnormalities in this patient population. A starting dose of 25 mg once daily is recommended in patients of East Asian ancestry.

Geriatric

There is limited data on the use of Revolade™ in elderly patients (>65 years of age). Based on the clinical studies conducted, age did not appear to influence the pharmacokinetics of Revolade™. However, given this patient population may be more prone to decreased hepatic, renal, or cardiac function, and of concomitant disease or other drug therapy, caution should be used when prescribing Revolade™ to patients (>65 years of age).

Paediatric

The safety and efficacy of Revolade™ in paediatric patients have not been established.

3.3.3 Clinical Efficacy

A total of five studies in 493 adult subjects with previously-treated chronic ITP were presented in this submission: three double-blind, randomized, multicentre, placebo-controlled studies (TRA100773A, TRA100773B and RAISE) and two open-label multicentre studies (REPEAT and EXTEND).

TRA100773A Study

TRA100773A was a 6-week, dose-ranging Phase II study conducted in ITP patients who had either not responded to one or more prior therapies (refractory), or who had relapsed within three months of prior therapy (relapse). The goal of this study was to identify an optimal dose for Revolade™. The primary efficacy endpoint was the proportion of patients who achieved a platelet count ≥50 x 10⁹/L at Day 43 from a baseline platelet count <30 x 10⁹/L. The study design consisted of a 1:1:1:1 randomization to either a placebo, or one of three Revolade™ dosing arms (30 mg, 50 mg or 75 mg). Patients were administered study medication once daily for up to 6 weeks. Patients were also stratified based on use or non-use of ITP medications at baseline, splenectomy status and baseline platelet count (≤15 x 10⁹/L or >15 x 10⁹/L). Patients who attained a platelet count >200 x 10⁹/L were discontinued from the study to reduce the risk of developing thrombocytosis. Additionally, no dose adjustments were permitted for the duration of the study.

Results from the TRA100773A study demonstrated that Revolade™ effectively raised platelet counts compared to placebo following up to 6-weeks of treatment irrespective of ITP medication use and baseline platelet count. Also, a dose-dependent increase in response was observed, and the primary endpoint was achieved by 11% of patients on placebo in comparison to 28%, 70% , and 81% of patients on Revolade™ 30 mg, 50 mg, and 75 mg, respectively. Furthermore, patients treated with either Revolade™ 50 mg or 75 mg showed a statistical greater odds of responding at Day 43 than patients treated with a placebo. Given these observations, the 50 mg dose was determined to be the most efficacious starting dose.

TRA100773B Study

TRA100773B was a 6-week, Phase III study conducted in ITP patients who, like in the TRA100773A study, had experienced response to prior ITP therapy. Patients were randomized on a 1:2 ratio to a placebo versus Revolade™ 50 mg starting dose (lowest effective dose identified from TRA100773A). The study design of TRA100773B was similar to TRA100773A, with the exception that patients with platelet counts <50 x 10⁹/L on or after Day 22 were allowed to increase their dose of Revolade™ to 75 mg (or matching placebo). The same primary efficacy endpoint as in TRA100773A was used.

Treatment response from the TRA100773B study confirmed the results observed from the TRA100773A study, with 16% of patients on placebo and 59% of patients on Revolade™ achieving platelet counts ≥50 x 10⁹/L (p<0.001). In addition, the incidence and severity of symptoms associated with chronic ITP, such as bleeding, bruising, and petechiae (as assessed by the (World Health Organization [WHO] bleeding scale) was lower in the Revolade™ treatment group compared to the placebo group.

Although both TRA100773A and TRA100773B study results were positive for both patient populations with or without splenectomy, the design of both studies was not meant to assess Revolade™ versus splenectomy comparatively in the treatment of ITP. Therefore, the current data submitted was considered insufficient to determine the efficacy of Revolade™ in the population of non-splenectomized patients.

RAISE Study

RAISE was a 6-month Phase III study similar in design to TRA100773B with randomization being on a 1:2 ratio to a placebo versus Revolade™ 50 mg (starting dose). However patients were now instructed to follow specific dosing modifications based upon their individual platelet count response. The dose modification guidelines were as follows:

• increase the dose of study medication to a maximum of 75 mg or matching placebo once daily if the platelet count elevation was insufficient (e.g., <50 x 10⁹/L);
• maintain the dose of study medication if platelet counts were between 50-200 x 10⁹/L;
• reduce the dose of study medication if platelet counts had risen to values between 200-400 x 10⁹/L; or
• interrupt treatment with study medication if platelet counts exceeded 400 x 10⁹/L and restart at a lower dose when platelet counts returned to ≤150 x 10⁹/L.

The primary efficacy endpoint of the study was the odds of achieving a platelet count ≥50 x 10⁹/L and ≤ 400 x 10⁹/L during the 6-month treatment period, for patients receiving Revolade™ treatment compared to a placebo.

Results from the RAISE study confirmed the efficacy findings observed in the two short-term studies TRA100773A and TRA100773B that Revolade™ treatment did raise and maintain platelet counts to safe levels (between ≥50 x 10⁹/L and ≤ 400 x 10⁹/L). Fifty-four percent of patients treated with Revolade™ and 13% of placebo-treated patients achieved platelet counts between ≥50 x 10⁹/L and ≤ 400 x 10⁹/L following 6 weeks of treatment. Furthermore, 53% maintained these levels within the Revolade™ treatment group compared to 16% in the placebo group over a 6-month period. The incidence and severity of symptoms associated with chronic ITP (i.e., bleeding) was also clinically significantly lower in the Revolade™ treatment group compared to the placebo-treated group. In addition, there was also a greater reduction in the use of baseline concomitant ITP medications and use of rescue medication in the Revolade™ treatment group compared to the placebo group.

Long-Term Studies

The REPEAT and EXTEND studies were ongoing global, open-label, single-group studies evaluating the effect of re-treatment (REPEAT) and long-term treatment (EXTEND) with Revolade™. Both open-label studies had different entrance criteria with regard to platelet count compared to the previous three double-blind studies. In REPEAT, patients were required to have a baseline platelet count between 20 x 10⁹/L and 50 x 10⁹/L. However, no platelet count entry criteria was specified for eligibility into the EXTEND study.

REPEAT Study

The REPEAT study was designed to evaluate the efficacy, safety, and tolerability of Revolade™ 50 mg (starting dose) administered once daily over three cycles of repeated, intermittent dosing. A cycle was defined as administering Revolade™ treatment for up to 6 weeks (on-therapy) followed by stopping Revolade™ treatment up to 4 weeks (off-therapy). The duration of both the on-therapy and the off-therapy periods of each cycle were determined by each patient's individual platelet count. The primary endpoint of this study was the proportion of patients who achieved a platelet count ≥50 x 10⁹/L and at least two times above the baseline count in cycle 2 or 3, given this response in cycle 1.

As in the TRA100773A and TRA100773B studies, patients in REPEAT interrupted treatment for the cycle if they achieved a platelet count >200 x 10⁹/L, or if they had completed 6 weeks of dosing. The next cycle of treatment with Revolade™ was initiated when their platelet counts fell <20 x 10⁹/L, or when they reached week 4 of the off-therapy period, and had platelet counts <50 x 10⁹/L. Patients who did not respond in the first cycle (platelet count ≥50 x 10⁹/L and at least two times above the baseline count) were not eligible to continue into the second or third cycle. Similar to the double-blind Phase III study design (TRA100773B and RAISE), the Revolade™ dose (or matched placebo) could be increased to 75 mg on or after Day 22 of each cycle if a patient's platelet count did not increase above 50 x 10⁹/L.

Of the 52 patients who responded in cycle 1, out of 66 patients enrolled, 80% achieved a platelet count of ≥50 x 10⁹/L and at least two times baseline on Day 43. Of the 52 patients who responded in cycle 1, 63% and 79% satisfied the response criteria by Day 8 and Day 15 in cycle 1, respectively. Forty-one patients (80%) responded again in cycle 2 and 38 (78%) patients responded again in cycle 3. A reduction in any bleeding (WHO Grade 1-4) and clinically significant bleeding (WHO Grade 2-4) during the treatment phases was demonstrated in each cycle. At the baseline visit of cycle 1, 50% and 19% of patients reported any bleeding and clinically significant bleeding, respectively. At the Day 43 visit of cycle 1, the proportion of patients bleeding was reduced; 12% and 0% of patients reported any bleeding and clinically significant bleeding, respectively. Similar results were found during the subsequent treatment cycles.

EXTEND Study

EXTEND is an ongoing long-term extension study designed to evaluate the safety and efficacy of Revolade™ in patients with chronic ITP who were previously enrolled in a Revolade™ study. The dosing instructions for the EXTEND study were similar to those in RAISE and were based upon each patient's individual platelet count response to treatment with Revolade™. In this study, patients were permitted to modify their dose of study medication, as well as, decrease or discontinue concomitant ITP medications, in addition to having access to rescue treatment as needed for standard care.

The study was designed in stages which depended on each patient's use of concomitant ITP medications. In stage 1, patients were administered a starting dose of 50 mg per day in order to reach a platelet count of ≥50 x 10⁹/L or ≥100 x 10⁹/L depending on whether they were taking a concomitant ITP medication at study entry. Concomitant ITP medications, if taken at study entry, were then tapered to a minimal dose or discontinued entirely (stage 2), while maintaining a platelet count of ≥50 x 10⁹/L. The Revolade™ treatment dose was then titrated to a minimal effective dose to maintain platelet counts of ≥50 x 10⁹/L (stage 3) and continued for as long as the patient continued to benefit (stage 4).

Two hundred and forty-nine patients completed ≥6 months of treatment, 210 completed ≥12 months of treatment, 138 patients completed ≥2 years of treatment, and 24 patients completed ≥3 years of treatment. The median follow-up was 100 weeks. The majority of patients had baseline platelet counts of <30 x 10⁹/L (70%). The median daily dose of Revolade™ following 6 months of therapy was 50 mg.

At baseline, 56% of patients had any bleeding (WHO bleeding Grades 1-4) and 16% had clinically significant bleeding. The proportion of patients with any bleeding and clinically significant bleeding decreased from baseline by approximately 50% for the majority of assessments up to one year.

Sixty-five percent of patients who reduced a baseline medication permanently discontinued or had a sustained reduction of their baseline ITP medication and did not require any subsequent rescue treatment. Ninety-six percent of these patients maintained this discontinuation or reduction for at least 24 weeks. Fifty-four percent of these patients completely discontinued at least one baseline ITP medication, and 49% of patients permanently discontinued all baseline ITP medications, without subsequent rescue treatment.

Results from the EXTEND study therefore showed that the majority of patients who responded in a previous Revolade™ study and received study medication in EXTEND responded again in EXTEND, indicating that a prior response to Revolade™ is predictive of a subsequently rise in platelets when patients are re-treated with Revolade™. The use of concomitant ITP medication was also significantly reduced, if not permanently discontinued, without need of any subsequent rescue treatment.

3.3.4 Clinical Safety

The safety data presented was collected from three double-blind, placebo-controlled studies (TRA100773A, TRA100773B, and RAISE), two open-label studies (REPEAT and EXTEND), one non-treatment observational study, thirteen completed pharmacology studies, and one study conducted and halted before completion in non-ITP patients with chronic liver disease. Across all studies, Revolade™ was well-tolerated with similar incidences overall of adverse events (AEs), serious adverse events (SAEs), and AEs leading to discontinuation from the study medication.

In the 6-week studies (TRA100773A and TRA100773B), the most common AEs (≥5%) reported in the Revolade™ 50 mg treatment group were headache, nasopharyngitis, nausea, fatigue, and arthralgia. Nausea was reported more frequently in the Revolade™ treatment group while headache and arthralgia were reported more frequently in the placebo treatment group. No dose-dependent patterns in the incidence or severity of AEs were noted across treatment groups. No apparent organ-specific toxicity profile emerged during comprehensive analyses of the pooled data.

In the 6-month RAISE study, the most common AEs observed in ≥10% of patients in the Revolade™ treatment group were headache, diarrhoea, nausea, nasopharyngitis, fatigue and upper respiratory tract infection. The rates for these AEs were similar in the placebo treatment group.

The incidence of SAEs (18% and 11%) and AEs leading to withdrawal (7% and 9%) was similar in both the Revolade™ treatment group and placebo treatment group. In addition, patients treated with Revolade™ had a significant reduction in bleeding AEs and in side-effects typically associated with corticosteroid use.

There were eight deaths reported across all studies: four occurred on-therapy, and four occurred between 42 days and 6 months after discontinuation of treatment with Revolade™. All eight deaths were unrelated to the study medication.

A safety issue raised in both the non-clinical and clinical program was the risk of developing liver toxicity during treatment with Revolade™. In the non-clinical studies, dogs showed increased liver enzymes and/or elevated bilirubin levels along with microscopic changes in the liver. In clinical studies, hepatobiliary laboratory abnormalities, such as increases in ALT, AST, and bilirubin serum levels were observed more frequently in ITP patients treated with Revolade™ compared to the placebo treatment group. Also, more hepatobiliary AEs were observed in the Revolade™ group compared to the placebo group. These findings were mostly mild (Grade 1-2), reversible and not accompanied by clinically significant symptoms that would indicate impaired liver function. Nonetheless, a cautionary statement has been inserted in the Product Monograph when administering Revolade™ to patients with moderate hepatic impairment; and Revolade™ is contraindicated for those patients with severe hepatic impairment.

Renal tubular toxicity was observed in a 2-year oral carcinogenicity study in mice at doses of 25, 75, and 150 mg/kg/day. The exposure at the lowest dose was 1.2 times the human clinical exposure based on AUC. Renal tubular toxicity however, was not observed in the placebo-controlled clinical studies. In addition, no cases of grade 3 or 4 creatinine elevations were observed within these clinical studies. However, based on these observations, a cautionary statement was inserted in the Product Monograph regarding administration of Revolade™ treatment to patients who are renally impaired. Given there is limited data with the use of Revolade™ in patients with severe renal impairment, treatment with Revolade™ is not recommended in these patients unless the expected benefit outweighs the risk.

A total of fourteen patients treated with Revolade™ experienced a thromboembolic event. These events were not associated with abnormally elevated platelet counts. In the placebo-controlled studies, there were four patients in the Revolade™ group who developed such events, compared to none in the placebo group. Therefore the following cautionary statement is included in the Product Monograph: "Caution should be used when administering Revolade™ to patients with known risk factors for thromboembolism including but not limited to inherited (e.g. Factor V Leiden) or acquired risk factors (e.g. ATIII deficiency, antiphospholipid syndrome), advanced age, patients with prolonged periods of immobilisation, malignancies, contraceptives and hormone replacement therapy, surgery/trauma, obesity and smoking."

Review of the safety data revealed no apparent evidence of clinically relevant bone marrow alterations or clinical findings in patients treated with Revolade™, based on the blood smear and bone marrow data available to date. Several patients had reticulin and/or collagen findings without any clinical signs or symptoms that would indicate bone marrow dysfunction.

Transient decreases in platelet counts following discontinuation or interruption of Revolade™ occurred in a minority of patients in both treatment arms, but generally they were not associated with clinically meaningful bleeding events. The incidence of post-therapy bleeding AEs within 4 weeks following discontinuation of therapy was similar or less in the Revolade™ group compared to the placebo group in all three double-blind studies.

In non-clinical studies, the formation of cataracts was observed in rodents at exposure levels similar to those achieved in humans. However, in the placebo-controlled studies conducted in humans there was no apparent increase in the incidence of cataracts with up to six months treatment. Despite this observation, routine monitoring of patients for possible cataract development has been recommended.

In the clinical studies, there was no confirmation of potential cardiotoxicity as seen in non-clinical studies. In addition, there were no meaningful differences in the safety profile of Revolade™ with regard to age, sex, and race with the exception of more hepatobiliary laboratory abnormalities in Asians compared to Caucasians.

During the clinical review, the sponsor was issued a Notice of Deficiency (NOD) based on the following safety concerns. The most robust safety data from controlled clinical trials spanned over a six month duration which did not provide sufficient time-course analysis in support of long-term treatment. In addition, no animal model could be used to adequately perform purposeful on-target toxicity studies. Furthermore, the following safety risks and concerns which included thrombotic/thromboembolic complications, bone marrow abnormalities or cytopenia(s), hepatotoxicity, hematologic malignancies (myelodysplastic syndrome), drug interactions, ophthalmologic risks, worsening thrombocytopenia upon discontinuation, unknown long-term risks (given only controlled studies that were ≤ 6 months in duration) required further delineation, characterization and management. It was also difficult to determine given the design of the pivotal trials, where Revolade™ treatment would optimally ‘best fit' in the ITP population (splenectomy vs non-splenectomy) and current strategies of care.

In response to these deficiencies, the following recommendations were made: To limit Revolade™ treatment for no more than one year of continuous therapy and closely monitor the risks and uncertainties, as outlined above, through implementation of a rigorous Risk Management Plan. In addition, given the benefit-risk balance was not able to be evaluated in patients who were non-splenectomized the risk profile could not be stated as positive for this patient population. Accordingly, the proposed indication in this patient population has been restricted to the use of Revolade™ treatment as a second-line alternative in non-splenectomized patients for whom splenectomy is contraindicated.