Cotellic - Scientific Information
|Ingredients:||Cobimetinib Fumarate, Croscarmellose Sodium, Lactose Monohydrate, Magnesium Stearate, Microcrystalline Cellulose, Polyethylene Glycol 3350 (macrogol 3350), Polyvinyl Alcohol, Talc, and Titanium Dioxide.|
|Proper name:||cobimetinib fumarate|
|Chemical name:|| methanone, [3,4–difluoro–2–[(2–fluoro–4–iodophenyl)amino]phenyl][3–hydroxy–3–
|Molecular formula:||C46H46F6I2N6O8 (2 C21H21F3IN3O2 . C4H4O4)|
|Structural formula: cobimetinib fumarate|
|Physicochemical properties:|| Cobimetinib is a fumarate salt appearing as white to off–
whitesolid and exhibits a pH dependent solubility. The
solubility of cobimetinib in aqueous media increases with
decreasing pH. In water (37°C), cobimetinib shows a
solubility of 0.72 mg⁄mL. In 0.1 M HCl (37°C), its
solubility is 48.21 mg⁄mL.
Study Demographics and Trial Design
Study GO28141 is a multicenter, randomized, double–blind, placebo–controlled, Phase III studyto evaluate the safety and efficacy of COTELLIC (cobimetinib) in combination withvemurafenib as compared to vemurafenib plus placebo, in patients with previously untreatedBRAF V600 mutation–positive unresectable locally advanced (stage IIIc) or metastaticmelanoma (stage IV).
Following confirmation of a BRAF V600 mutation using the cobas® 4800 BRAF V600mutation test, 495 patients with unresectable locally advanced or metastatic melanoma wererandomized to receive either:
- Placebo once daily on Days 1–21 of each 28–day treatment cycle and 960 mgvemurafenib twice daily on Days 1–28
- COTELLIC 60 mg once daily on Days 1–21 of each 28–day treatment cycle and 960 mgvemurafenib twice daily on Days 1–28
balanced between treatment arms. Key baseline characteristics included: 58% of patients weremale, 93% reported White race, median age was 55 years (range 23 – 88 years), 60% hadmetastatic melanoma stage M1c, Eastern Cooperative Oncology Group performance status was 0in 72% and 1 in 28%, serum lactate dehydrogenase was elevated in 45%, prior adjuvant therapyhad been administered to 10%, and <1% had previously treated brain metastases.
Overall, demographic characteristics and patient baseline disease characteristics were wellbalanced between treatment arms. Key baseline characteristics included: 58% of patients weremale, 93% reported White race, median age was 55 years (range 23 – 88 years), 60% hadmetastatic melanoma stage M1c, Eastern Cooperative Oncology Group performance status was 0in 72% and 1 in 28%, serum lactate dehydrogenase was elevated in 45%, prior adjuvant therapyhad been administered to 10%, and <1% had previously treated brain metastases.
Treatment continued until disease progression, death, unacceptable toxicity, or withdrawal ofconsent, whichever occurred earliest. Patients in the placebo plus vemurafenib arm were noteligible to cross over to the COTELLIC plus vemurafenib arm at disease progression.
Progression–free survival (PFS) as assessed by the investigator (Inv) using RECIST v1.1 was theprimary endpoint. Secondary efficacy endpoints included overall survival (OS), objectiveresponse rate (ORR), duration of response (DoR), and PFS as assessed by an independent reviewfacility (IRF). Global health status / health–related quality of life by patient–report, also asecondary efficacy endpoint, were measured for each treatment arm using the EuropeanOrganisation for Research on the Treatment of Cancer (EORTC) Quality of Life Questionnaire(QLQ–C30).
Overall, the median duration of follow–up for all patients was 7.3 months. 207 (41.8%) patientshad experienced a PFS event at the time of the final analysis of the primary endpoint.The combination of COTELLIC and vemurafenib showed a statistically significant increase inPFS (Inv) with a hazard ratio of 0.51 (95% CI 0.39–0.68, p <0.0001). The median PFS (Inv) was9.9 months in patients receiving COTELLIC and vemurafenib versus 6.2 months in patientsreceiving placebo and vemurafenib. The PFS benefit in the COTELLIC plus vemurafenib armwas observed across key characteristic subgroups, specifically disease stage, age, sex, ECOGperformance status, baseline lactate dehydrogenase level, and BRAF V600 mutation E and Kgenotypes. In the COTELLIC plus vemurafenib arm, 67.6% of patients had an objectiveresponse (complete response or partial response) versus 44.8% in the placebo plus vemurafenibarm. Median OS had not been reached in either treatment arm at the time of the pre–specifiedinterim OS analysis, and the analysis did not cross the pre–specified boundary for statisticalsignificance.
Complete efficacy results are summarized in the table below.
|COTELLIC + ZELBORAF
|Placebo + ZELBORAF
|Progression Free Survival
|Number of Events (%)||79 (32.0%)||128 (51.6%)|
|Median (Kaplan–Meier estimate
|(9.0, NE)||(5.6, 7.4)|
|Hazard ratio (95% CI)||0.51 (0.39; 0.68)|
|Key Secondary Endpoints|
|Number of Deaths (%)||34 (13.8%)||51 (20.6%)|
|Hazard ratio (95% CI)||0.65 (0.42, 1.00)
|Objective response rate (ORR)||167 (67.6%)||111 (44.8%)|
|95% CI for ORR||(61.4%, 73.4%)||(38.5%, 51.2%)|
|Difference in ORR %
|22.85 (14.13, 31.58)
p–value < 0.0001
|Best Overall Response|
|Complete Response||25 (10.1%)||11 (4.4%)|
|Partial Response||142 (57.5%)||100 (40.3%|
|Stable disease||49 (19.8%||105 (42.3%)|
|Best Overall Response|
|Complete Response||25 (10.1%)||11 (4.4%)|
|Partial Response||142 (57.5%)||100 (40.3%)|
|Stable disease||49 (19.8%)||105 (42.3%)|
|Duration of response (DoR)|
|Median DoR (months)||NE||7.3|
|95% CI for median||9.3, NE||5.8, NE|
a Assessed by the investigator (Inv) using RECIST v1.1
b Did not cross the pre–specified boundary for significance (p–value <0.000004)
NE = not evaluable
The effect on PFS was also supported by analysis of PFS based on the assessment by blindedindependent review.
Retrospective next generation sequencing to evaluate BRAF mutation genotypes was performedon tumour tissue from 400 of 495 patients and demonstrated that 344 (86%) tumors carried aV600E mutation and 56 (14%) had a V600K mutation. The distribution of V600E versus V600Kwas balanced between treatment arms. Although limited by the low number of patients withV600K compared to patients with V600E, Inv–assessed exploratory subgroup PFS analysis by BRAF V600 mutation genotype suggests treatment benefit for V600K patients treated withCOTELLIC in combination with vemurafenib.
Figure 1 Kaplan–Meier Curves of Progression–free Survival (Inv)–Intent to TreatPopulation
Quality of Life
As measured by the EORTC QLQ–C30, all functioning domains and most symptoms (appetiteloss, constipation, insomnia, nausea and vomiting, dyspnea, pain, fatigue) were similar betweenthe two treatment arms and did not demonstrate a clinically meaningful change (≥ 10 pointincrease or decrease from baseline)
In vitro, cobimetinib was a selective inhibitor and MEK1 and MEK2 and did not inhibit otherserine–threonine or tyrosine kinases (IC50 >10 μM, 5310 ng⁄mL). In a panel of secondarypharmacological targets consisting of receptors, transporters and enzymes, the in vitro bindingactivities of the following were inhibited by > 50% at a cobimetinib concentration of 10 μM(5310 ng⁄mL): adrenergic β2 receptor, muscarinic M1 and M2 receptors, opiate μ and κ receptors,serotonin 5–HT2B receptor, σ (non–selective) receptor, somatostatin receptor, Ca2+ion channel,Na+ channel, and dopamine transporter.
In vitro, cobimetinib inhibited the human ether–á–go–go related gene (hERG) ion channel currentwith an IC50 of 0.5 μM [266 ng⁄mL]). In a dog cardiovascular safety pharmacology study, therewere no changes in hemodynamics or ECG parameters, including QTc interval.
At clinically relevant concentrations in the rat CNS safety pharmacology study, there were no cobimetinib–related physiological or behavioural changes.
In a tissue distribution study conducted in rats, levels of C–cobimetinib–related radioactivity in the non–circumventricular CNS tissues protected by the blood:brain barrier were low.
In the respiratory safety pharmacology study conducted in the rat, the highest dose of 300 mg⁄kgelicited a small gradual decrease of 18% in the respiratory rate compared to pre–dose values (5–fold human clinical exposure based on unbound Cmax). In the single dose toxicology study in therat, a dose of 300 mg⁄kg led to mortality within 3 days postdose.
The nonclinical toxicology profile of cobimetinib was investigated in the rat and dog in studiesof up to 13 weeks in duration. Cobimetinib administration resulted in findings attributedprimarily to its pharmacologic mechanism of action (inhibition of cell proliferation in tissueswith high proliferative rates including gastrointestinal, integument, and hematopoietic andlymphopoietic systems). These effects occurred in animals at total systemic cobimetinibexposures generally below those achieved at the oral therapeutic dose of 60 mg⁄day in cancerpatients (Cmax=273 ng⁄mL; AUC0–24=4340 ng hr⁄mL).
Repeated–dose toxicity studies in rats and dogs identified degenerative changes in the bonemarrow, gastrointestinal tract, skin, thymus, adrenal gland, liver, spleen, lymph node, ovary andvagina at plasma exposures below clinical efficacious levels. These changes were generallyreversible following the non–dosing recovery period, with the exception of lymphocytedepletion/necrosis and plasma cell hyperplasia in mandibular lymph nodes in rats. Dose–limitingtoxicities included skin ulcerations, surface exudates, and acanthosis in the rat and chronic activeinflammation and degeneration of the esophagus associated with varying degrees ofgastroenteropathy in dogs.
Mortality was observed in rats and dogs at exposures below that observed in humans. In the 4–week rat study, animals treated at 1 mg⁄kg⁄day and 3 mg⁄kg⁄day survived to scheduled necropsy.Mortality was observed in animals treated at 10 mg⁄kg (≥0.7–fold human clinical exposure basedon AUC). In the 13–week study in dogs, mortality was observed within the first 2 weeks ofdosing at the dose level of 3.0 mg⁄kg⁄day (≥2.0–fold human clinical exposure based on AUC).One female dog that received 1.0 mg⁄kg⁄day was euthanized on Day 58 (~0.5 human clinicalexposure based on AUC).
No carcinogenicity studies have been conducted with COTELLIC.
Cobimetinib was not mutagenic nor did it induce structural or numerical chromosomalaberrations in vitro. Cobimetinib did not induce the formation of micronuclei in vivo in the bonemarrow of rats that had been treated with cobimetinib.
Reproductive and Developmental Toxicology
No dedicated fertility studies in animals have been performed with COTELLIC.
n the repeat–dose toxicology studies, degenerative changes were observed in reproductivetissues including increased apoptosis⁄necrosis of corpora lutea and vaginal epithelial cells infemale rats (~2.5–fold human clinical plasma exposure based on AUC). After a single dose athigher exposures, degenerative changes were also observed in the seminal vesicles of male ratsand epididymal epithelial cells of male rats and dogs (≥8–fold human clinical plasma exposurebased on AUC).
In a repeat dose toxicity study in juvenile rats, cobimetinib systemic exposures were 2 to11 foldhigher on Day 1 than on Day 28 when exposures were similar to those in the pivotal toxicitystudies in adult rats. Daily oral doses of 3 mg⁄kg (approximately 0.13–0.5 times the adult humanAUC at the recommended dose of 60 mg) administered beginning postnatal Day 10 wereassociated with mortality on postnatal Day 17, the cause of which was not defined. The samedose did not lead to mortality in adult animals. In juvenile rats, cobimetinib administrationresulted in similar changes as seen in the pivotal toxicity studies in adults, including reversibledegenerative changes in the thymus and liver, decreased spleen and thyroid⁄parathyroid weights,increased phosphorus, bilirubin and red cell mass, and decreased triglycerides.
When administered to pregnant rats, cobimetinib caused embryolethality and fetal malformationsof the great vessels and skull at systemic exposures approximately 0.9 to 1.4–fold human clinicalplasma exposure based on AUC.
Cobimetinib was shown to be potentially phototoxic after UVA⁄UVB irradiation in vitro usingcultured murine fibroblasts. There was no evidence of cutaneous or ocular phototoxicity inLong–Evans pigmented rats exposed to a single dose of cobimetinib (~7–fold human clinicalplasma exposure based on AUC) and subsequent exposure to UVA and UVB radiation from axenon lamp (to simulate sunlight).