Koselugo Mechanism of Action

Pharmacology: Pharmacodynamics: Mechanism of action: Selumetinib is an orally available, potent and selective inhibitor of mitogen-activated protein kinase kinases 1 and 2 (MEK1/2) that is not competitive with respect to ATP. MEK1/2 proteins are critical components of the RAS-regulated RAF-MEK-ERK pathway, which is often activated in different types of cancers. Selumetinib blocks MEK activity and inhibits growth of RAF-MEK-ERK pathway activated cell lines. Therefore, MEK inhibition can block the proliferation and survival of tumour cells in which the RAF-MEK-ERK pathway is activated.
Pharmacodynamic effects: In genetically modified mouse models of NF1 that generate neurofibromas that recapitulate the genotype and phenotype of human type 1 neurofibromas, oral dosing of selumetinib inhibits ERK phosphorylation, reduces neurofibroma volume, proliferation, number and growth.
Cardiac electrophysiology: The effect of selumetinib on the QTc interval following a single 75 mg oral dose, in a placebo- and positive-controlled (moxifloxacin) study, in 48 healthy adults showed no clinically relevant effect on the QTc interval (<10 msec change). A pharmacokinetic-pharmacodynamic analysis predicted a <10 msec change at 150 mg dose (3-times higher than the recommended maximum dose of 50 mg in paediatric patients with NF1).
Clinical efficacy: SPRINT: The efficacy of KOSELUGO was evaluated in an open-label, multi-centre, single-arm study [SPRINT Phase II Stratum 1 (NCT01362803)] of 50 paediatric patients with NF1 inoperable PN that caused significant morbidity. Morbidities that were present in ≥20% of patients included disfigurement, motor dysfunction, pain, airway dysfunction, visual impairment, and bladder/bowel dysfunction. Inoperable PN was defined as a PN that could not be surgically completely removed without risk for substantial morbidity due to encasement of, or close proximity to, vital structures, invasiveness, or high vascularity of the PN. Patients received 25 mg/m² (BSA) twice daily, for 28 days (1 treatment cycle), on a continuous dosing schedule. Treatment was discontinued if a patient was no longer deriving clinical benefit, experienced unacceptable toxicity or PN progression, or at the discretion of the investigator.
The target PN, the PN that caused relevant clinical symptoms or complications (PN-related morbidities), was evaluated for response rate using centrally read volumetric magnetic resonance imaging (MRI) analysis per Response Evaluation in Neurofibromatosis and Schwannomatosis (REiNS) criteria. Tumour response was evaluated at baseline and while on treatment after every 4 cycles for 2 years, and then every 6 cycles.
Patients had target PN MRI volumetric evaluations and clinical outcome assessments, which included functional assessments and patient reported outcomes.
At enrolment, the median age of the patients was 10.2 years (range: 3.5 - 17.4 years), 60% were male, 84% were Caucasian.
Disease characteristics at baseline are provided in Table 1. (See Table 1.)

Click on icon to see table/diagram/image

The primary efficacy endpoint was Objective Response Rate (ORR), defined as the percentage of patients with complete response (defined as disappearance of the target PN) or confirmed partial response (defined as ≥20% reduction in PN volume, confirmed at a subsequent tumour assessment within 3-6 months), based on NCI centralised review. Duration of Response (DoR) was also evaluated.
Efficacy results are provided based on a data cut-off of March 2021, unless stated otherwise.
The primary endpoint, ORR was 68% (95% CI 53.3 - 80.5). An independent centralized review of tumour response per REiNS criteria (data cut-off June 2018) resulted in an ORR of 44% (95% CI 30, 59). The median time to onset of response was 7.2 months (range 3.3 months to 3.2 years).
The median DoR from onset of response was not reached; at the time of data cut-off the median follow-up time was 41.3 months from first dose. Of the 34 patients who had confirmed partial responses, 31 (91.2%) remained in response after 12 months, 26 (76.5%) remained in response after 24 months and 21 (61.8%) remained in response after 36 months. The probability to remain in response after 12, 24 and 36 months, estimated using the Kaplan-Meier method, was 100% (95% CI not estimated), 90.0% (95% CI 72.1 - 96.7) and 86.3% (95% CI 67.3 - 94.6), respectively. The median time from treatment initiation to disease progression while on treatment was not reached. (See Table 2.)

Click on icon to see table/diagram/image

At the time of data cut-off or last scan on treatment for patients who had discontinued treatment, 25 (50%) patients remained in confirmed partial response, 1 (2%) had unconfirmed partial responses, 12 (24%) had stable disease and 10 (20%) had progressive disease.
The median best percentage change in PN volume from baseline was -27.85% (range: -60.3% to 2.2%). Figure shows the best percentage change in target PN volume for each patient. (See figure.)

Click on icon to see table/diagram/image

Clinical Outcome Assessments: Pain intensity of the target PN was self-reported by patients ≥8 years of age using an 11-point Numeric Rating Scale (NRS-11). Based on mixed model repeated measures (MMRM) analysis, a clinically meaningful reduction in pain (defined as, ≥2-point decrease from baseline) was seen at pre-Cycle 25 (adjusted mean change from baseline: -2.17, 95% CI: -3.17, -1.17) and at pre-Cycle 49 (adjusted mean change from baseline: -2.43, 95% CI: -3.47, -1.40).
Parent-reported (all patients) and patient-reported (≥8 years of age) health-related quality of life (HRQoL) was assessed using the Peds-QL questionnaire. Based on MMRM analysis, a clinically meaningful improvement in HRQoL (clinically meaningful threshold 11.90) was reported by parents at pre-Cycle 13, with a mean change from baseline in Peds-QL total score of 12.82 (95% CI 9.00 - 16.63) and was sustained at all remaining measurement cycles.
Improvement in HRQoL was also reported by patients at pre-Cycle 13 with a mean change from baseline of 6.17 (95% CI 0.93 - 11.40). The improvement in patient reported Peds-QL total score was clinically meaningful (clinically meaningful threshold 10.33) at pre-Cycle 37, with a mean change from baseline in Peds-QL total score of 11.23 (95% CI 6.81 - 15.65) and was sustained at pre-Cycle 49. These results should be interpreted in the context of the open-label, single-arm study design and therefore taken cautiously.
Pharmacokinetics: At the recommended dosage of 25 mg/m² twice daily in paediatric patients (3 to ≤18 years old), the geometric mean (coefficient of variation [CV%]) C_max following the first dose and at steady state was 731 (62%) ng/mL and 798 (52%) ng/mL, respectively. The mean area under the plasma drug concentration curve (AUC_0-12h) following the first dose was 2009 (35%) ng·h/mL and the AUC_0-6h at steady state was 1958 (41%) ng·h/mL. Selumetinib AUC and C_max increases proportionally over a dose range from 20 mg/m² to 30 mg/m² (0.8- to 1.2-times the recommended dose). Minimal accumulation of ~1.1-fold was observed at steady state upon twice daily dosing.
In paediatric patients, at a dose level of 25 mg/m², selumetinib has an apparent oral clearance of 8.8 L/h, mean apparent volume of distribution at steady state of 78 L and mean elimination half-life of ~6.2 hours.
Absorption: In healthy adult subjects, the mean absolute oral bioavailability of selumetinib was 62%. Following oral dosing, selumetinib is rapidly absorbed, producing peak steady state plasma concentrations (t_max) between 1-1.5 hours post-dose.
Effect of food: In separate clinical studies, in healthy adult subjects and in adult patients with advanced solid malignancies at a dose of 75 mg, co-administration of selumetinib with a high-fat meal resulted in a mean decrease in C_max of 50% and 62%, respectively, compared to fasting administration. Selumetinib mean AUC was reduced by 16% and 19%, respectively, and the time to reach maximum concentration (t_max) was delayed by approximately 1.5 to 3 hours (see Dosage & Administration).
In healthy adult subjects at a dose of 50 mg, co-administration of selumetinib with a low-fat meal resulted in 60% lower C_max when compared to fasting administration. Selumetinib AUC was reduced by 38%, and t_max was delayed by approximately 0.9 hours (see Dosage & Administration).
In adolescent patients with NF-1 and inoperable PN treated with multiple doses of 25 mg/m² bid, co-administration of selumetinib with a low-fat meal resulted in 24% lower C_max when compared to fasting administration. Selumetinib AUC was reduced by 8%, and t_max was delayed by approximately 0.57 hours (see Dosage & Administration).
A population PK analysis including children and adolescent patients with NF-1 and inoperable PN, adult patients with advanced solid malignancies and healthy adult subjects taken from 15 studies showed that concomitant administration of a low or high fat meal resulted in a mean decrease in the exposure (AUC) of selumetinib when compared to fasted administration (23.1% and 20.7%, respectively) which was not considered clinically relevant.
Effect of gastric acid reducing agents on KOSELUGO: KOSELUGO capsules do not exhibit pH dependent dissolution. KOSELUGO can be used concomitantly with gastric pH modifying agents (i.e., H₂-receptor antagonists and proton pump inhibitors) without restrictions, except for omeprazole which is a CYP2C19 inhibitor.
Distribution: The mean apparent volume of distribution at steady state of selumetinib across 20 to 30 mg/m² ranged from 78 to 171 L in paediatric patients, indicating moderate distribution into tissue.
In vitro plasma protein binding is 98.4% in humans. Selumetinib mostly binds to serum albumin (96.1%) than α-1 acid glycoprotein (<35%).
Biotransformation/Metabolism: In vitro, selumetinib undergoes Phase 1 metabolic reactions including oxidation of the side chain, N-demethylation, and loss of the side chain to form amide and acid metabolites. CYP3A4 is the predominant isoform responsible for selumetinib oxidative metabolism with CYP2C19, CYP1A2, CYP2C9, CYP2E1 and CYP3A5 involved to a lesser extent. In vitro studies indicate that selumetinib also undergoes direct Phase 2 metabolic reactions to form glucuronide conjugates principally involving the enzymes UGT1A1 and UGT1A3. Glucuronidation is a significant route of elimination for selumetinib Phase 1 metabolites involving several UGT isoforms.
Following oral dosing of ¹⁴C-selumetinib to healthy male subjects, unchanged selumetinib (~40% of the radioactivity) with other metabolites including glucuronide of imidazoindazole metabolite (M2; 22%), selumetinib glucuronide (M4; 7%), N-desmethyl selumetinib (M8; 3%), and N-desmethyl carboxylic acid (M11; 4%) accounted for the majority of the circulating radioactivity in human plasma. N-desmethyl selumetinib represents less than 10% of selumetinib levels in human plasma but is approximately 3- to 5-times more potent than the parent compound, contributing to about 21% to 35% of the overall pharmacologic activity.
Elimination: In healthy adult volunteers, following a single oral 75 mg dose of radiolabelled selumetinib, 59% of the dose was recovered in faeces (19% unchanged) while 33% of the administered dose (<1% as parent) was found in urine by 9 days of sample collection.
Interactions: In vitro, selumetinib is not a reversible inhibitor of CYP1A2, CYP2A6, CYP2C8, CYP2C19, CYP3A4 or CYP2E1, not an inducer of CYP1A2 and CYP2B6, and did not cause time-dependant inhibition of CYP2C9, CYP2D6 or CYP3A4/5. In vitro, selumetinib is a reversible inhibitor of CYP2C9, CYP2B6, CYP2D6, UGT1A3, UGT1A4, UGT1A6 and UGT1A9; however, these effects are not expected to be clinically relevant.
Interactions with transport proteins: Based on in vitro studies, selumetinib is a substrate for breast cancer resistance protein (BCRP) and P-glycoprotein (P-gp) transporters but is unlikely to be subjected to clinically relevant drug interactions at the recommended paediatric dose. Based on in vitro studies, selumetinib is not a substrate for OATP1B1, OATP1B3 or OCT1 transporters. In vitro, selumetinib is an inhibitor of BRCP, OATP1B1, OATP1B3, OCT2, OAT1, OAT3, MATE1 and MATE2K but does not inhibit P-gp or OCT1. There in vitro inhibitory effects are not expected to be clinically relevant effect with the exception of OAT3 where a clinically relevant effect on the pharmacokinetics of concomitantly administered substrates of OAT3 cannot be excluded.
Special populations: Renal impairment: The exposure of 50 mg oral selumetinib was investigated in adult subjects with normal renal function (n=11) and subjects with ESRD (n=12). The ESRD group showed 16% and 28% lower C_max and AUC, respectively, with the fraction of unbound selumetinib being 35% higher in ESRD subjects. As a result, the unbound C_max and AUC ratios were 0.97 and 1.13 in the ESRD group when compared to the group with normal renal function. A small increase, approximately 20% AUC, in the N-desmethyl metabolite to parent ratio was detected in the ESRD group when compared to the normal group. As exposure in ESRD subjects was similar to those with normal renal function, investigations in mild, moderate and severe renally impaired subjects were not performed. Renal impairment is expected to have no meaningful influence on the exposure of selumetinib (see Dosage & Administration).
Hepatic impairment: Adult subjects with normal hepatic function (n=8) and mild hepatic impairment (Child-Pugh A, n=8) were dosed with 50 mg selumetinib, subjects with moderate hepatic impairment (Child-Pugh B, n=8) were administered a 50 or 25 mg dose, and subjects with severe hepatic impairment (Child-Pugh C, n=8) were administered a 20 mg dose. Selumetinib total dose normalised AUC and unbound AUC were 86% and 69% respectively, in mild hepatic impairment patients, compared to the AUC values for subjects with normal hepatic function. Selumetinib exposure (AUC) was higher in patients with moderate (Child-Pugh B) and severe (Child-Pugh C) hepatic impairment; the total AUC and unbound AUC values were 159% and 141% (Child-Pugh B) and 157% and 317% (Child-Pugh C), respectively, of subjects with normal hepatic function (see Dosage & Administration).
Ethnicity: Following a single-dose, selumetinib exposure appears to be higher in Japanese, non-Japanese-Asian and Indian healthy adult volunteers compared to Western adult volunteers. However, there is considerable overlap with Western subjects when corrected for body weight or BSA (see Dosage & Administration).
Adult patients (>18 years old): The PK parameters in adult healthy volunteers and adult patients with advanced solid malignancies, are similar to those in paediatric patients (3 to ≤18 years old) with NF1.
In adult patients with solid malignancies, at a single dose of 75 mg selumetinib, geometric mean (%GCV) C_max and AUC were 1307 (76%) ng/mL and 4736 (37%) ng·h/mL, respectively. Peak plasma concentrations of selumetinib were achieved 1.5-hour post-dose with a mean elimination half-life of 7.8 hours. C_max and AUC increased dose proportionally over a 25 mg to 100 mg dose range, and administration of 75 mg selumetinib twice daily resulted in minimal accumulation of ~1.2-fold.
Toxicology: Preclinical safety data: Mutagenicity: Selumetinib showed no mutagenic or clastogenic potential in vitro but produced an increase in micronucleated immature erythrocytes (chromosome aberrations) in mouse micronucleus studies, predominantly via an aneugenic mode of action. The free mean exposure (C_max) at the No Observed Effect Level (NOEL) was approximately 27-times greater than clinical-free exposure at the maximum recommended human dose (MRHD) of 25 mg/m².
Carcinogenicity: Selumetinib was not carcinogenic in a 6-month study in rasH2 transgenic mice at free exposures 24-times (females) and 16-times (males) the free clinical AUC at MRHD and in a 2-year carcinogenicity study in rats at free exposures 2.9-times (females) and 3.7-times (males) the clinical-free AUC at MRHD.
Repeat-dose toxicity: In repeat-dose toxicity studies in mice and rats, the main effects seen after selumetinib exposure were in the skin, scabs associated with microscopic erosions and ulceration in rats at a free exposure similar to the clinical exposure (free AUC) at the MRHD; inflammatory and ulcerative GI tract findings in mice associated with secondary changes in the liver and lymphoreticular system at free exposures approximately 28-times the clinical-free exposure at the MRHD; and growth plate (physeal) dysplasia in male rats at a free exposure 11-times the clinical-free exposure at the MRHD. GI findings showed evidence of reversibility following a recovery period. Reversibility for skin toxicities and physeal dysplasia were not evaluated.
Reproductive toxicology: Fertility: In a 6-month mouse study, selumetinib did not affect male mating performance at any dose up to 20 mg/kg twice daily corresponding to approximately 22-times the human clinical exposure based on free AUC at the MRHD. In female mice exposed to selumetinib at 12.5 mg/kg twice daily, mating performance and fertility were not affected, but the number of live foetuses was slightly reduced. Following a three-week treatment withdrawal period, no effects were apparent on any parameter. The No Observed Adverse Effect Level (NOAEL) for both maternal toxicity and effects on reproductive performance was 2.5 mg/kg twice daily (approximately, 3.5-fold human free exposure at the MRHD).
Embryofoetal toxicity: In embryofoetal development studies in mice, selumetinib caused a reduction in the number of live foetuses due to an increase in post-implantation loss, a reduction in mean foetal and litter weights, increased occurrence of open eye and cleft palate at dose levels that did not induce significant maternal toxicity. These effects were seen at an exposure >3.5-fold the clinical exposure at MRHD based on free AUC and indicate that selumetinib may have potential to cause defects in the foetus.
Pre- and postnatal development: Administration of selumetinib to pregnant mice from gestation Day 6 through to lactation Day 20 resulted in reduced pup body weights, and fewer pups met the pupil constriction criterion on Day 21 post-partum. The incidence of malformations (prematurely open eye(s) and cleft palate) was increased at all dose levels. Malformations occurred at maternal concentration (C_max) 0.4-fold below the mean free clinical concentration at MRHD.
Selumetinib and its active metabolite were excreted in the milk of lactating mice at concentrations approximately the same as those in plasma.