Summary Basis of Decision for Torisel ™

3.3.1 Pharmacodynamics

A pharmacodynamic (PD) study was conducted in healthy subjects to evaluate the exposure/response relationship of temsirolimus after a single intravenous administration. Fluorescence response associated to phosphorylated S6-ribosomal protein (p-S6RP) in CD3+ and CD19+ peripheral blood mononuclear cells was chosen to indirectly measure the effect of temsirolimus on mTOR. Data showed that the inhibition of S6RP phosphorylation was dose-dependent at the dose levels between 1-25 mg of temsirolimus. Following the clinical recommended 25 mg dose, the maximum inhibition of p-S6RP in CD3+ cells was 50% for at least 3 days. Temsirolimus was well-tolerated when given as a single intravenous dose up to 25 mg to healthy subjects. Overall, the most common (>10%) treatment-emergent adverse events were stomatitis (dose-related), acne, pain, and leucopenia.

3.3.2 Pharmacokinetics

Absorption

Following a 30-minute infusion of temsirolimus 25 mg, the maximum plasma concentration (C_max) of temsirolimus was reached by the end of the infusion period. The C_max and the drug exposure (AUC) of temsirolimus and its major metabolite (sirolimus) in whole blood increased in a dose-related but non-proportional manner. Sirolimus C_max accounted for approx. 1/10th of the temsirolimus C_max but contributed to approximately 72% of the total exposure due to the longer half-life of sirolimus (73 h vs. 18 h for temsirolimus).

Distribution

Temsirolimus and sirolimus exhibited preferential portioning into red blood cells. Distribution was extensive, increased with dose, and exceeded total body water. In vitro, temsirolimus binding to plasma proteins was approximately 87% at the concentration of 100 ng/mL.

Metabolism

Temsirolimus is primarily metabolized in the liver by the cytochrome P450 isozyme, CYP3A4. Temsirolimus is also a substrate of β-glycoprotein and potential inhibitor for CYP3A4, CYP2D6 and β-glycoprotein.

After intravenous administration of temsirolimus, sirolimus was observed to be the major metabolite. Sirolimus is considered to be equally potent to the parent drug.

Elimination

Elimination of temsirolimus was primarily via the feces and only a small percentage of the drug was excreted via the kidney (<5%). The clearance of temsirolimus from whole blood increased as the dose increased. Following repeated doses, exposure was slightly reduced from cycle to cycle for both temsirolimus and sirolimus with little or no accumulation observed between weekly treatment cycles.

Drug Interactions

Concomitant use of a CYP3A4/5 inducer, rifampicin, reduced sirolimus exposure by 56% and total exposure for temsirolimus and sirolimus (AUC_sum) by 41% with little effect on temsirolimus exposure.

A CYP3A4 inhibitor, ketoconazole, increased sirolimus exposure by 3.1-fold and total exposure by 2.3-fold with no apparent effects on the pharmacokinetics (PK) of temsirolimus.

Temsirolimus showed no effects on the PK of CYP2D6 substrate, desipramine, however, there was an 8-hour gap between the two peak concentrations of the drugs which may have reduced the magnitude of the interaction.

An increased risk of serious adverse events was demonstrated when temsirolimus was co-administered with other anti-neoplastic drugs, e.g., sunitinib, gemcitabine or 5-fluorouracil. Fatal cases of bowel perforation were observed when temsirolimus was combined with 5-fluorouracil. Synergistic interactions between temsirolimus and gemcitabine resulted in significant higher incidence rates of hematologic dose-limiting toxicities.

Special Populations and Conditions

No dosage adjustments are required based on patient age, gender, or weight. No PK data are available for pediatric patients. The PK effects caused by renal impairment were not studied. PK studies were not conducted in patients with moderate to severe liver impairment.

A small study with 7 Japanese patients showed that exposure levels were nearly doubled that of the non-Japanese patients at the same dose. A larger study with Asian patients is needed to demonstrate whether dose adjustment is required for patients of oriental origin.

Patients with higher hematocrit values treated with temsirolimus resulted in higher trough drug concentrations in the blood, leaving the peak concentrations unaffected.

The Phase I studies did not provide sufficient information to evaluate the cardiovascular safety of temsirolimus. Therefore, the effect of temsirolimus on QTc interval is unknown at this time. A thorough QT study will be submitted at a later date.

3.3.3 Clinical Efficacy

Clinical data supporting the efficacy and safety of Torisel (temsirolimus) for treatment of metastatic renal cell carcinoma (RCC) were obtained primarily from one Phase II dose-finding study and one pivotal Phase III study. The Phase III study involved 626 patients, randomized to one of three treatment arms: Torisel 25 mg IV weekly (n=209), Interferon alfa (IFN-a) up to 18 MU IV three times weekly (n=207), and a combination of Torisel 15 mg IV weekly and IFN-a up to 6 MU IV three times weekly (n=210).

The supportive Phase II study led to the choice of dose for Torisel by demonstrating efficacy at a dose of 25 mg IV per week with no additional efficacy at a dose of 75 mg weekly and 250 mg weekly. Furthermore, the incidence of adverse events increased with the higher doses.

In the Phase III study, the primary efficacy endpoint was overall survival (OS). Secondary efficacy endpoints included progression free survival (PFS), objective response rate (ORR), clinical benefit rate, time to treatment failure (TTF), and the quality adjusted survival measurement (a measurement of time without symptoms of progression of disease or treatment toxicity using the Q-TWiST approach). The OS results showed an increase of 3.6 months for the patients treated with Torisel compared to the patients treated with IFN-a. The median OS was 10.9 months and 7.3 months for the Torisel group and IFN-a group, respectively (p=0.0078). There were also significant improvements in the secondary outcomes and an increase in quality adjusted survival time, without symptoms of progression of disease or toxicity, in the Torisel group compared with the IFN-a group. The Q-TWiST measurement was increased by an estimated 1.3 months in patients that were treated with Torisel compared to patients treated with IFN-a. The importance of Q-TWiST is to demonstrate that the increase in OS does not come at the cost of poor quality of life. The actual duration of improvement seen in this study of first-line treatment of poor prognosis RCC is small, but taken in the context of a statistically and clinically significant increase in OS, the improvement is worthwhile.

3.3.4 Clinical Safety

The clinical safety data was taken primarily from the Phase III, three-arm, randomized, open-label study described in section 3.3.3 Clinical Efficacy. Generally, Torisel (temsirolimus) was well tolerated. Adverse events (AEs) in the Torisel arm were less severe than in the IFN-a arm, consequently fewer treatment emergent adverse events (TEAEs) led to discontinuation, (18.3% vs 30.5%) or dose reduction (20.2% vs 39.5%). The incidence of TEAEs leading to dose-delay was similar for Torisel and IFN-a, at 51.4% and 47.0% respectively. Grade 3 or 4 asthenia and anemia were the most common events leading to modification of Torisel treatment.

The following adverse events are of special interest. Unless otherwise stated, the following AE percentages for Torisel were taken from the Torisel 25 mg IV arm in the Phase III study.

Hypersensitivity Reactions

Prior to the start of a Torisel infusion, an IV antihistamine was administered as hypersensitivity reactions were anticipated. Despite this, a number of patients developed symptoms of hypersensitivity and required an interruption of the infusion. All but one patient was able to continue treatment after management of the event. A protocol for managing drug-related events during the infusion is provided in the Product Monograph, and this seems a reasonable approach. No cases of anaphylaxis and hemodynamic or respiratory compromise were encountered in the Phase III study, or in other studies to the best of our knowledge. The development of an anaphylactic event related to temsirolimus or sirolimus is identified in the Product Monograph as an absolute contraindication for continuing with the agent.

Respiratory-related Events

Dyspnea and cough were the most common events, noted in 27.9% and 25.5% of patients respectively. Grade 3 or 4 dyspnea was reported in 8.7% of patients, and while a small number required dose modifications, no patients discontinued the study due to dyspnea. Of more concern was the apparent association of interstitial lung disease (ILD) in patients receiving Torisel. Reports of ILD have been made from patients receiving Torisel for a broad range of malignancies. At least one death is referred to in the submission. The actual incidence of ILD is unknown at this time. Some patients have been able to continue treatment with the addition of steroids and/or antibiotics, but the long-term results of this are unknown. This particular issue has been identified as important in the risk management program.

Gastrointestinal Events

Mucositis and stomatitis were common in patients treated with Torisel, occurring in 24.6% and 20.1% of the patients, respectively. Grade 3 or 4, or serious AEs were uncommon at 1-3%, and while dose modifications were required for these patients, none of the patients discontinued treatment due to these adverse events. Other significant GI adverse events leading to discontinuation or dose modification included nausea, vomiting, anorexia, and diarrhea. Outside of the Phase III study, there were two reports of bowel perforation in patients who received 5-fluorouracil in combination with temsirolimus. This adverse event has been addressed in the Product Monograph.

Skin-related Events

Rash of all types was very common, occurring in over 47% of the patients taking Torisel. The majority of these were classified as "generalized rash" and most were considered to be treatment related. Acne was the most common specific rash, occurring in 10.1% of the patients. The incidence of Grade 3 or 4 rash-related events was very low (5% or less).

Hematological Events

Anemia was the most frequent hematologic TEAE in patients receiving Torisel, with over 94% of patients reporting at least one low haemoglobin value on treatment. Almost 20% of these values were Grade 3 or 4. Transfusion of at least one unit of blood was given to 30.3% of the patients receiving Torisel.

Thrombocytopenia was also common, and it occurred in >40% of the patients receiving Torisel. Of these only 3 patients (1.4%) reported Grade 3 or 4 thrombocytopenia. In the Phase I dose-escalation study of temsirolimus given weekly, thrombocytopenia was found to be dose-limiting.

Bleeding events were frequent (24.5%), but generally Grade 1 or 2. The most common bleeding event was epistaxis which was reported in 25 (12.0%) of the patients.

Thrombosis was uncommon in all arms of the study, approximately 4-6% overall. In particular, deep vein thrombosis and pulmonary embolus occurred in 3 patients (1.4%) and 2 patients (1.0%), respectively, in the Torisel arm.

Leukopenia and neutropenia were less common and rarely led to dose modification.

Infection-related Events

Infection-related TEAEs were reported in 32% of patients taking Torisel compared with 4.5% of patients taking IFN-a. The most common types of infection recorded were respiratory, including pneumonia in 7.8% and bronchitis in 3.7% of the patients. These two categories included infections classified as Grade 3 or 4, in approximately half of the cases.

Clinical Laboratory Related Events

The incidence of hyperglycemia and hyperlipidemia was very high. With 15.9% of the cases of hyperglycemia Grade 3 or 4 (>13.9 mmol/L), it is likely that a proportion of patients will need specific treatment for diabetes while receiving Torisel. The long-term outcome of patients who develop hyperglycemia on treatment is not known.

Similarly, hypercholesterolemia and hypertriglyceridemia were very common, occurring in 87% and 83% of patients, respectively, receiving temsirolimus. Treating physicians may have to institute specific therapy for these conditions while treating with Torisel. There is the potential for increased exposure to statins while concurrently receiving Torisel. Consequently, the risk of developing rhabdomyolysis may be increased.

Other specific clinical chemistry abnormalities noted in this study included hypophosphatemia (49% overall, 18.3% Grade 3 or 4). Hypophosphatemia can contribute to muscle weakness, an overall decline in clinical status, and may contribute to rhabdomyolysis.

Renal Events

Renal insufficiency was a common TEAE with Torisel. Elevated creatinine levels were reported in 57.2% of the patients treated with Torisel and 3.4% of these patients had elevations greater than 3x the upper limit of normal.

Acute kidney failure and kidney failure occurred in 4 patients receiving Torisel. However, in the Torisel arm, 39 AEs were reported that infer the development of renal insufficiency. In addition, hyperkalemia (which was likely associated with renal insufficiency) was reported in 10 of the patients treated with Torisel.

Renal failure is referred to as an AE requiring special monitoring in the Product Monograph.

Cardiovascular Events

Little in the way of cardiotoxicity was noted in this study, however, patients with significant cardiac history were excluded from randomization. Chest pain was reported in 16% of the patients treated with Torisel as compared to 9% in the IFN-a arm. Two myocardial infarctions were reported in both the Torisel arm and in the IFN-a arm. There was no clear increase in the incidence of congestive heart failure, but it is possible that at least some of the 42% of patients with edema were in fact demonstrating cardiac failure.

The issue of QT interval prolongation is not finished. There were no clear signals of QT prolongation in this study and it was not possible to make a statement based on those findings. A thorough QT study has been started by the sponsor, and the results will be made available. Presently, the Product Monograph contains a general warning about the concurrent use of QT prolonging drugs, and a precaution for patients with congenital long QT syndromes.