Summary Basis of Decision for Rozlytrek

Clinical Pharmacology

Entrectinib (the medicinal ingredient in Rozlytrek) is a tyrosine receptor kinase inhibitor. It targets the tropomyosin receptor kinase (TRK) family proteins A, B and C (TRKA, TRKB, and TRKC), which are encoded by the neurotrophic tyrosine receptor kinase 1, 2 and 3 genes (NTRK1, NTRK2 and NTRK3), ROS proto-oncogene 1 receptor tyrosine kinase (ROS1; encoded by the ROS1 gene), and anaplastic lymphoma kinase (ALK; encoded by the ALK gene).

Chromosomal rearrangements can result in the fusion of NTRK1, NTRK2, NTRK3, ROS1, or ALK with other genes. These gene fusions give rise to oncogenic TRK, ROS1 and ALK fusion proteins with constitutive kinase activity. This in turn activates cell signalling pathways involved in cell proliferation and survival, promoting the development of tumours. Entrectinib has therefore been developed as an anti-cancer agent for the treatment of patients with tumours that harbour NTRK1, NTRK2, NTRK3, ROS1, or ALK gene fusions.

Entrectinib has been shown to inhibit TRK fusion proteins, as well as the proliferation of cell lines containing NTRK gene fusions, in a concentration-dependent manner. In three NTRK1-driven intracranial tumour models, including two high-grade glioma mouse models, exposure to entrectinib significantly inhibited tumour activity.

Following administration of a single 600 mg oral dose in adult cancer patients, entrectinib is rapidly absorbed reaching time-to-maximum plasma concentration (T_max) after approximately four to six hours. Based on population pharmacokinetic analysis, steady state is achieved within five days. The elimination half-life of entrectinib and its major active metabolite M5 is estimated to be 20 and 40 hours, respectively. Entrectinib is a central nervous system-penetrant molecule that crosses the blood-brain barrier.

Entrectinib is metabolized predominantly by cytochrome P450 (CYP) 3A4 enzymes (approximately 76%). Minor contributions from several other CYPs and UDP-glycosyltransferase 1-4 (UGT1A4) enzymes were estimated at less than 25% in total. The active metabolite M5 (formed by CYP3A4) and the direct N-glucuronide conjugate, M11 (formed by UGT1A4), are the two major circulating metabolites identified. Co-administration of strong and moderate CYP3A4 inhibitors with entrectinib should be avoided.

Following administration of a single dose of [¹⁴C]-labelled entrectinib administered orally to healthy subjects, the majority of radioactivity was excreted in feces (82.9%) with minimal excretion in urine (3.06%). In feces, 35.7% and 22.1% of the dose was excreted as unchanged entrectinib and M5, respectively, indicating that hepatic clearance is the major route of elimination.

In an ECG sub-study conducted in patients who received Rozlytrek 600 mg once daily, entrectinib exposure-dependent trends in change from baseline in QTcF were not apparent in the analysis of ECG data, which included 107 patients with concentrations at dates/times matching the QT assessment.

In pediatric patients, non-compartmental analysis and population pharmacokinetic modelling approaches suggested that the pharmacokinetics of entrectinib and M5 was comparable to adults. Data obtained from population pharmacokinetic analyses suggested that a dose of 300 mg/m² of Rozlytrek once daily in pediatric patients results in a similar systemic exposure attained in adults treated with 600 mg of Rozlytrek, once daily. A pediatric indication has not been granted by Health Canada for Rozlytrek, as safety has not been established in this patient population.

In patients with moderate to severe hepatic impairment, the safety and efficacy have not been studied. Based on the excretion profile of entrectinib, no dose adjustment is recommended for patients with mild hepatic impairment.

In patients with mild and moderate renal impairment, population pharmacokinetic data show that the pharmacokinetics of entrectinib is not significantly affected in renal impairment. No formal pharmacokinetic study has been conducted and no population pharmacokinetic data were collected in patients with severe renal impairment.

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

Clinical Efficacy

Three open-label, single-arm clinical studies in adult patients (18 years of age and older) with advanced cancers contributed data from 54 patients to a pre-specified integrated efficacy analysis evaluating Rozlytrek for the treatment of extracranial NTRK fusion-positive solid tumours. The three clinical studies were:

• ALKA, a Phase I dose-finding study, which contributed data from one patient
• STARTRK-1, a Phase I dose-finding/efficacy study, which contributed data from two patients
• STARTRK-2, a Phase II basket study evaluating efficacy, safety, and pharmacokinetics, which contributed data from 51 patients

Among the 54 adult patients, the median age was 57 years, and overall 37% of patients were 65 years of age or older. The most frequently represented solid tumour types were sarcomas, non-small cell lung cancer, mammary analogue salivary gland tumours, and breast cancers (the majority being secretory breast cancers). Tumours harbouring gene fusions of each of the NTRK genes were represented: over half of the patients (57%) had NTRK3 fusions, NTRK1 gene fusions were detected in 41% of patients, while an NTRK2 gene fusion was detected in a single patient. Most patients presented with metastatic disease at baseline. The most common sites of metastatic disease were lungs and lymph nodes. Baseline central nervous system metastases were documented in 12 patients (22%) as assessed by investigator. Twenty patients (37%) received Rozlytrek as the first systemic therapy after metastatic disease diagnosis, while 43% of patients had two or more prior systemic therapies. This population was considered to be representative of a general adult population with NTRK fusion-positive extracranial solid tumours.

In all three studies, Rozlytrek was administered orally in a continuous daily dosing regimen, in treatment cycles of 28 days. The dose used by the majority of adult patients included in the integrated analysis was 600 mg once daily. Computed tomography or magnetic resonance imaging was used to assess tumour response and disease progression based on Response Evaluation Criteria in Solid Tumours (RECIST) v1.1. The frequency of scheduled tumour assessments for adults was every 8 weeks starting at the end of Cycle 1 until documented disease progression by Blinded Independent Central Review (BICR). For patients with primary central nervous system tumours, intracranial objective tumour response or disease progression was assessed based on Response Assessment in Neuro-Oncology (RANO) criteria.

In all three studies, efficacy endpoints included objective response rate (ORR) and duration of response (DOR) as evaluated by the BICR according to RECIST v1.1. For the integrated efficacy analysis, these were the primary endpoints. In the integrated efficacy analysis, formal significance tests were not performed; therefore, p values were not reported. Instead, 95% two-sided confidence intervals (CI) for point estimate were utilized to estimate the magnitude of the effect. A response rate that excluded 30% or higher was to be considered clinically meaningful.

At the time of primary integrated analyses, the median duration of follow-up in responders from the time of first response was 13.1 months (range: 2.8, 21.0).

The results obtained from this analysis showed that objective responses, as assessed by the BICR, were observed in 31 of the 54 patients and therefore, the ORR was 57% (95% CI: 43%, 71%). Four patients (7%) achieved a complete response. The majority of objective responses were achieved at the first tumour assessment after commencing Rozlytrek treatment (end of Cycle 1).

Responses were durable, with a median DOR among responders, as assessed by the BICR, of 10.4 months (95% CI: 7.1, not evaluable). At the time of the clinical cut-off date, 17 of the 31 responders had responses lasting longer than 6 months. The Kaplan-Meier estimated 6-month event-free proportion was 69% (95% CI: 51%, 86%). The intracranial ORR in the seven patients with measureable central nervous system metastases was 57% (95% CI: 18%, 90%). These results are considered clinically meaningful.

Among the six adult patients with NTKR fusion-positive primary brain tumours, one patient had an objective response (partial response) with a duration of 2.8 months. This result are not considered to support the use of Rozlytrek in the treatment of primary brain tumours.

Pediatric Population

The Rozlytrek New Drug Submission also contained an extrapolation package provided to support a similar benefit-risk profile of Rozlytrek for the use in children with tumours harbouring NTRK 1, NTRK 2, and NTRK 3 fusions. The extrapolation package included a population pharmacokinetic model developed in the pediatric population. In addition, a Drug Safety Report on bone fractures was submitted to Health Canada during the course of the review.

The primary objective of the extrapolation package was to support the efficacy and safety of Rozlytrek in children with NTRK fusion-positive tumours, through demonstration that the disease manifestation and the expected response to therapy in the adult and pediatric populations are similar. The pediatric extrapolation for safety and efficacy was based on 26 pediatric patients from the STARTRK-NG study and two additional patients treated with Rozlytrek via compassionate access who both had at least six months of follow-up. The pediatric population data were compared with data from 339 adult patients pooled across Rozlytrek studies ALKA, STARTRK-1, and STARTRK-2. The package demonstrated that the disease manifestation and the expected response to therapy in the adult and pediatric populations are similar. Population pharmacokinetic modelling approaches suggested that the pharmacokinetics of entrectinib and its active metabolite, M5, was comparable in adults and children. However, this package could not adequately support the safety of Rozlytrek in children. The limitations of and knowledge gaps in the current entrectinib pediatric safety profile could not be reliably addressed via the safety data in adults treated with entrectinib or via the safety data in adults and pediatric patients treated with larotrectinib.

The Drug Safety Report reported that the rate of bone fractures observed in the four main clinical trials of entrectinib was much higher in pediatric patients at 23% (7/30) than in adults patients (5%, 17/337); updated results continue to report a rate of 23% (9/39) in pediatric patients. In adult patients, some fractures occurred in the setting of a fall or other trauma to the affected area, while in the pediatric patients, fractures occurred with minimal or no trauma. Given this safety signal, Health Canada does not consider Rozlytrek to have an acceptable safety profile in the pediatric population.

Based on the limited data submitted to and reviewed by Health Canada, the safety of Rozlytrek in pediatric patients has not been established; therefore, Health Canada has not authorized an indication for pediatric use.

Efficacy Conclusion

Overall, the data reviewed provided promising evidence of the clinical effectiveness of Rozlytrek for the approved indication, when administered according to the recommended dosing regimen. The results of additional studies and post-market safety update reports will be submitted to verify the clinical benefit of Rozlytrek.

Indication

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

• Rozlytrek (entrectinib) is indicated for the treatment of adult and pediatric patients with neurotrophic tyrosine receptor kinase (NTRK) fusion-positive locally advanced or metastatic solid tumours, who have progressed following prior therapies, or as initial therapy when there are no acceptable standard therapies.

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

• Rozlytrek (entrectinib) is indicated for the treatment of adult patients with unresectable locally advanced or metastatic extracranial solid tumours, including brain metastases, that have a neurotrophic tyrosine receptor kinase (NTRK) gene fusion without a known acquired resistance mutation, and with no satisfactory treatment options.

The indication proposed by the sponsor was revised by Health Canada to reflect the following:

• the lack of sufficient safety data to characterize the increased risk of fractures and support use in the pediatric population
• the lack of sufficient efficacy data to support benefit in primary brain tumours
• the current limitations in the available data supporting the use of Rozlytrek, by stating that it should be used only in patients with no other satisfactory treatment options
• that Rozlytrek is not indicated for patients in whom a known acquired resistance mutation in the NTRK gene fusion has been identified; and
• the need to test for the NTRK gene fusion prior to initiation of treatment.

The approval of the tumour-agnostic indication is based on the strength of the scientific rationale supporting the hypothesis that the inhibition of TRK would cause shrinkage of tumours with NTRK gene fusions; the clinically meaningful and durable responses observed in a variety of tumour types harbouring a diverse array of NTRK fusions; the strength of supportive non-clinical data; and the recent authorization of larotrectinib (Vitrakvi), another TRK inhibitor, for a tumour-agnostic indication.

Notably, given the current small sample size of the integrated analysis set and the limited duration of follow-up, not all tumour types and NTRK gene fusion partners have been adequately characterized. The durability of responses has also not been fully characterized. In addition, the efficacy of Rozlytrek in primary brain tumours and its safety in pediatric patients have not been established. However, Health Canada considers that within a tumour-agnostic indication in adults, an extrapolation of efficacy findings to some extracranial tumour types and NTRK gene fusion partners is acceptable in the context of providing access to a promising new drug for patients with advanced TRK fusion cancers and for whom there are no satisfactory treatment options. Rozlytrek was therefore authorized by Health Canada under the NOC/c Guidance on the basis of the promising nature of the clinical evidence, and the need for further follow-up to confirm the clinical benefit. The results of additional studies and post-market safety update reports will be submitted to Health Canada to continue to monitor the clinical benefit of Rozlytrek.

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

Clinical Safety

The safety profile of Rozlytrek was based on data derived from 339 adult patients with heterogeneous advanced cancers across three single-arm clinical studies (ALKA, STARTRK-1, and STARTRK-2), described in the Clinical Efficacy section, and data derived from 16 pediatric patients in the STARTRK-NG study. The overall integrated safety analysis set included 355 patients.

The median duration of treatment with Rozlytrek in the overall integrated safety population was 5.5 months corresponding to a median of 7.0 cycles. The median treatment duration was 5.8 months (range: 0.0, 42.1) in adult patients and 1.9 months (range: 0.2, 12.7) in pediatric patients. The majority of pediatric patients in the safety population had solid tumours with no NTRK fusions, therefore the shorter duration of exposure in pediatric patients was observed in the context of a higher incidence of disease progression.

Almost all patients (99.4%) experienced at least one adverse event. The most frequently reported adverse events (in >25% of patients) were constipation (45.9%), dysgeusia (43.7%), fatigue (37.7%), dizziness and diarrhea (34.6% each), nausea (34.5%), dyspnea (29.9%), peripheral edema (28.5%), and anemia (27.9%).

Serious adverse events occurred in 38.6% of patients, with the most frequently reported being pneumonia (3.9%), dyspnea (3.7%), pleural effusion (3.4%), pulmonary embolism (2.3%), and pyrexia (2.0%). Treatment-related serious adverse events occurred in 8.5% of patients, with the most frequently reported being cognitive disorders (1.1%).

Grade 3 and 4 adverse events were experienced by 60.3% of patients. The most frequently reported events were anemia (10.7%), increased weight (6.5%), dyspnea (5.6%), fatigue (4.2%), pneumonia, hypoxia, increased aspartate aminotransferase (AST) and pulmonary embolism (3.4% each), increased alanine aminotransferase (ALT) and pleural effusion (3.1% each), hypophosphatemia (2.8%), neutropenia, decreased neutrophil count, syncope (2.5% each), urinary tract infection (2.3%), diarrhea, hypokalemia, hyponatremia, increased lipase, and hypotension (2.0% each). Treatment-related Grade 3 and 4 adverse events occurred in 31.0% of patients.

Most adverse events requiring intervention were managed with dose interruption (45.9% of patients) or dose reduction (28.2% of patients). Adverse events leading to discontinuation of Rozlytrek were reported in 8.5% of patients.

Grade 5 adverse events occurred in 20 patients (5.6%). The majority of Grade 5 adverse events were reported in the context of worsening of underlying diseases or complications of the underlying malignancy. None of the 20 deaths due to an adverse event were determined by the investigator as related to Rozlytrek. However, given the nature of single-arm Phase I studies, a potential causal relationship between Rozlytrek and these deaths cannot be excluded and as such, all Grade 5 events have been included in the Rozlytrek Product Monograph.

Adverse Events of Special Interest

Adverse events of special interest for the integrated safety analysis included neurologic toxicity, changes in weight, congestive heart failure, increased creatinine and other renal events, eye disorders, QTc interval prolongation, elevated liver laboratory tests and other liver abnormalities, pneumonitis events, and hematologic events. These adverse events were selected on the basis of previous clinical experience, mechanism of action and safety profiles of drugs with similar targets. Additional adverse events of special interest identified during the review included skeletal fractures and hyperuricemia.

Neurologic toxicity

Consistent with Rozlytrek central nervous system activity and the known involvement of TRK receptors in the neuronal development and maintenance of the central and peripheral nervous system, neurologic adverse events have been observed in patients treated with Rozlytrek. Neurologic adverse events were reported in 88.7% of patients. The most frequently reported adverse events were dysgeusia (43.7%), dizziness (34.6%), paresthesia (20.6%), headache (17.7%), and muscular weakness (12.1%). Cognitive disorders were reported in 25.9% of patients, peripheral sensory neuropathy in 17.7% of patients, and syncope in 3.9% of patients. Patients receiving Rozlytrek treatment should be monitored and counselled on potential cognitive changes as well as other central nervous system events (including mood changes).

Changes in weight

Weight gain is likely an on-target effect of Rozlytrek, as tropomyosin receptor kinase B appears to be associated with appetite control. Weight gain occurred in 24.8% of patients, with Grade 3 events occurring in 6.5% of patients. However, given the single-arm studies and small sample sizes, it is difficult to assess the contribution of Rozlytrek to weight changes.

Congestive heart failure

Congestive heart failure was reported in 3.4% of patients, with symptomatic Grade 3 events in 2.3% of patients and serious congestive heart failure events in 2.0% of patients. Given the incidence rate and rapidity of onset of congestive heart failure, there is a concern that congestive heart failure may be linked to Rozlytrek. One of the sponsor's post-marketing commitments refers to further assessing the causality between Rozlytrek and cardiac adverse events.

Increased creatinine and other renal events

Increased creatinine and other renal events were reported in 38.6% of patients and was the most common event leading to dose interruption in 3.9% of patients, but none led to treatment discontinuation. Few events of acute kidney injury were reported in 3.4% of patients, with most events being of Grade 1 or 2 severity. In these patients, other concurrent conditions, including dehydration and sepsis were alternative causes for acute kidney injury.

Eye disorders

Eye disorders were reported in 27.3% of patients. The most frequently reported events were blurred vision, photophobia, and diplopia.

QTc interval prolongation

The adverse event of QTc interval prolongation was reported in 2.0% of patients. Across the clinical studies, 2.8% of patients with at least one post-baseline electrocardiogram assessment experienced a QTc interval increase over baseline of greater than 60 ms and 1.7% patients had a QTc interval greater than 500 ms.

Elevated liver laboratory tests and other liver abnormalities

Abnormal liver function tests and liver dysfunction adverse events were reported in 21.1% of patients, including elevations of AST and ALT. No adverse event of drug-induced liver injury was reported. Hepatotoxicity is a known risk of TRK inhibitors. While most AST and ALT increases are of a low-grade, liver function tests should be monitored during Rozlytrek treatment.

Pneumonitis

Pneumonitis events were reported by 2.3% of patients. Given the presence of lung cancer in some of these patients and the single-arm study design of the clinical studies, causality cannot be established between Rozlytrek and these events.

Hematologic events

Hematologic adverse events were reported in 38% of patients. The most frequently reported hematologic toxicities were anemia and neutropenia.

Skeletal fractures

Skeletal fractures were reported in association with Rozlytrek; risks were higher in children than in adults. A cumulative review of the clinical studies and drug safety databases were completed for all fracture cases across the entire Rozlytrek development program. It included an updated population of 367 patients (337 adults, 30 pediatric patients). In total, 34 patients with fracture adverse events were identified: 27 adult patients (5.4%) and 7 pediatric patients (23%). Updated numbers reported 9 fractures in 39 pediatric patients (23%). In adult patients, some fractures occurred in the setting of a fall or other trauma to the affected area, while in the pediatric patients, fractures occurred in patients with minimal or no trauma. Health Canada does not consider this risk of fractures in the pediatric population to be acceptable.

Hyperuricemia

Hyperuricemia was reported as an adverse event in 9% of patients. While the majority of these events were Grade 1 (81%), the remaining 19% were Grade 4. In 32 patients with an event of hyperuricemia, 6% required a dose reduction, 6% required a dose interruption, and 34% required interventions to reduce uric acid levels (urate-lowering drugs). One event of Grade 4 hyperuricemia occurred in a patient with tumour lysis syndrome; this patient ultimately died. Based on the need for monitoring and intervention, hyperuricemia has been included in the Warnings and Precautions section of the Rozlytrek Product Monograph.

Safety Conclusion

The safety profile of Rozlytrek in adults is considered acceptable in the setting of a life-threatening disease with no satisfactory treatment options. The primary serious risks that have been identified with the use of Rozlytrek are neurologic toxicity (including cognitive disorders, peripheral sensory neuropathy, and syncope), congestive heart failure, QT interval prolongation, eye disorders, skeletal fractures, hyperuricemia, elevated liver laboratory tests, anemia, and neutropenia. Overall, the safety profile of Rozlytrek was generally comparable between subgroups with the exception of skeletal fractures, which occurred more frequently in children as compared to adults (23% in pediatric patients versus 5% in adults). Given this risk, Health Canada does not consider the safety profile of Rozlytrek in the pediatric population to be acceptable. Therefore, Health Canada has not authorized an indication for Rozlytrek for pediatric use.

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