Activity of the oral MEK inhibitor trametinib in RAS-mutant relapsed or refractory myeloid malignancies

Abstract

Background

RAS/RAF/MAPK activation is common in myeloid malignancies. Trametinib, a MEK1/MEK2 inhibitor with activity against multiple myeloid cell lines at low nanomolar concentrations, was evaluated for safety and clinical activity in patients with relapsed/refractory leukemias.

Methods

This phase 1/2 study accrued patients with any relapsed/refractory leukemia (phase 1) and relapsed/refractory acute myeloid leukemia (AML) or high-risk myelodysplastic syndromes (MDS) with NRAS or KRAS mutation; RAS wild-type (or unknown mutation status) AML, MDS, or chronic myelomonocytic leukemia (CMML); or CMML with NRAS or KRAS mutation (cohorts 1, 2, and 3 of phase 2, respectively). Trametinib was administered at the recommended phase 2 dose of 2 mg once daily.

Results

The most commonly reported treatment-related adverse events were diarrhea, rash, nausea, and increased alanine aminotransferase. Overall response rates were 20%, 3%, and 27% in cohorts 1, 2, and 3, respectively, indicating preferential activity among RAS-mutated myeloid malignancies. Repeated cycles of trametinib were well tolerated, with manageable or reversible toxicities, similar to results of other trametinib studies.

Conclusions

Selective, single-agent activity of trametinib against RAS-mutated myeloid malignancies validates its therapeutic potential. Combination strategies based on better understanding of the hierarchical role of mutations and signaling in myeloid malignancies are likely to improve response rate and duration. This study was registered at www.clinicaltrials.gov as NCT00920140.

Introduction

Activation of the extracellular signal-regulated kinase (ERK) 1/2 mitogen-activated protein kinase (MAPK) pathway is common in acute myeloid leukemia (AML), myelodysplastic syndromes (MDS), and chronic myelomonocytic leukemia (CMML).^1–4 Mutation in the RAS gene (25–40%) is the most frequent genetic event leading to MAPK activation in myeloid malignancies^5–7 In certain subgroups of AML (eg, AML with inversion 16 cytogenetic abnormality),⁸ the overall frequency of mutations of genes involved in the MAPK pathway is approximately 90%.⁹ In CMML, the frequency of RAS mutation is approximately 20% and is often associated with proliferative variant of disease.^10,11 Apart from mutations in the RAS family of proteins, other oncogenic mutations (eg, FLT3,^12,13 BCR‐ABL^14,15) that are not linearly involved also activate the MAPK pathway. Finally, loss of negative regulators of RAS, eg, SPRY4 (chromosome 5q) and NF1 (chromosome 17), can lead to a myeloid leukemia phenotype in mouse models.¹⁶

Direct inhibition of RAS has been challenging, but the mitogen-activated protein kinase kinase (MEK) 1 and 2 are potential targets in the RAS/RAF/MEK/MAPK pathway immediately upstream of the ERK1/ERK2 MAP kinases.¹⁷ Efficacy of MEK inhibitors has been demonstrated in preclinical models of AML.¹⁷ Trametinib (Mekinist; GlaxoSmithKline [GSK], Research Triangle Park, NC, USA) is a selective, oral, allosteric inhibitor of MEK1/MEK2 that inhibits ERK phosphorylation at low nanomolar concentrations. Trametinib is approved by the US Food and Drug Administration as a single agent or in combination (with dabrafenib [Tafinlar; GSK, Research Triangle Park, NC, USA]) for the treatment of unresectable or metastatic melanoma with BRAF V600E/V600K mutation.

Trametinib inhibits proliferation of myeloid cell lines at low nanomolar concentrations, supporting its clinical evaluation in myeloid malignancies.¹⁷ We report the results of the phase 1/2 study evaluating the safety, pharmacokinetic (PK), and clinical activity of trametinib monotherapy in patients with relapsed or refractory leukemias.

Patients and Methods

Patients

Patients aged ≥ 18 years with relapsed or refractory (R/R) leukemia were eligible for phase 1 of the study. In phase 2, patients were enrolled to 1 of 3 cohorts; R/RAML or high-risk MDS (HR-MDS) with either NRAS or KRAS mutation (cohort 1), R/R AML or HR-MDS with RAS wild-type (wt) or unknown mutation status (cohort 2), or CMML with NRAS or KRAS mutation, treatment naive or previously treated (cohort 3). Additional inclusion criteria included Eastern Cooperative Oncology Group performance status of 0 to 2, calcium phosphorus product ≤ 4.0 mmol²/L² (50 mg²/dL²), adequate organ function, and no chemotherapy or investigational anticancer drug(s) within 2 weeks. Exclusion criteria included prior MEK inhibitor exposure, concurrent malignancy, predisposing factors or history of retinal vein occlusion (RVO) or central serous retinopathy (CSR), retinal pathology considered a risk factor for RVO or CSR, presence of cardiac risk factors, or any unresolved toxicity (grade > 1) from previous therapy.

Study Design

This open-label, dose-escalation, nonrandomized, multicenter phase 1/2 study (GSK Study MEK111759, ClinicalTrials.gov identifier: NCT00920140) was conducted at 19 centers in the United States, Germany, Belgium, and France. The protocol was approved by each institution’s ethics committee or review board in accordance with the Declaration of Helsinki, applicable International Committee on Harmonization Guidelines, GSK standard operating procedures, and national or local regulatory requirements. Written informed consent was obtained from all patients prior to participation in any study-related assessments, procedures, or treatment.

For phase 1, a traditional 3 + 3 design was used, with a ≤ 50% increase in dose level between cohorts. At least 3 patients in each dose cohort were required to receive treatment for 28 days (in absence of a dose-limiting toxicity [DLT]) prior to escalating to the next dose level. The starting dose was a 3-mg loading dose (LD) of trametinib followed by continuous 1-mg once-daily dosing. The LD was later omitted due to toxicity concerns (eg, ocular toxicity) arising out of the first-time-in-human (FTIH) solid tumor study, due to the higher maximum observed concentration (C_max). The phase 2 part at the recommended phase 2 dose (RP2D) used a Bayesian design to allow for early termination for lack of efficacy. Initial design included only cohorts 1 and 2; however, cohort 3 (patients with CMML with RAS mutation) was added because of the relatively high frequency of RAS mutations in CMML and the early evidence of trametinib activity in RAS-mutated leukemias.

Mutation Analysis

RAS mutations were detected in DNA extracted from bone marrow (BM) aspirates. The presence of any of 7 KRAS mutations in codons 12 and 13 was identified by real-time polymerase chain reaction (PCR). Assay sensitivity was 1% (2% for G12S). For the NRAS mutation, sample DNA was subjected to 2 PCR reactions (exon 2 and exon 3 of NRAS) followed by bidirectional dye terminator sequencing. The lower limit of detection of the assay was 10% mutated DNA into unmutated DNA.

Safety Assessment

Adverse events (AEs) were graded according to the Common Terminology Criteria for Adverse Events, version 4.0. For grade 2 or 3 trametinib-related toxicities, dosing was resumed with or without dose reduction if the toxicity resolved to grade 1 or baseline. Permanent treatment discontinuation was required for any grade 4 trametinib-related toxicity as well as specified treatment-related liver toxicities.

PK Assessment

Plasma PK analysis was done in phase 1, predose and 24 hours postdose following single (day 1) and multiple (day 15) day dosing. In phase 2, blood samples were taken predose on day 1 of each cycle and predose on day 15 of cycle 1.

Analytical Methods

PK analysis used liquid-liquid extraction, followed by high-performance liquid chromatography/mass spectrometry/mass spectrometry analysis.

Statistical Analysis

In phase 1, preliminary safety and PK data were reviewed after completion of each cohort. Responses were assessed according to the International Working Group criteria for AML, CMML, and MDS.¹⁸ In phase 2, each cohort was evaluated using the same hypotheses testing framework and interim monitoring rule. The null hypothesis was that the response rate (defined as percentage of patients who achieved complete remission [CR], CR with platelet count < 100 × 10⁹/L (CRp), partial response [PR], or a morphological leukemia-free state [MLFS]) with trametinib would not be of sufficient interest (≤ 3%). The alternative hypothesis was that the response rate would be of sufficient interest (≥ 15%). Bayesian statistics were used to calculate the predictive and posterior probabilities of success (rejecting the null hypothesis) using a noninformative prior distribution beta (0.005, 0.005). Enrollment to any cohort could stop early after enrolling 25 patients due to futility if the predictive probability of success was < 5%. The probability of early termination due to futility was 73% if the true response rate was ≤ 3%. There was a 3% risk of making an incorrect decision to terminate the study early if the true response rate was ≥ 15%.

Results

Patient Characteristics

A total of 97 patients (phase 1 [N = 14]; phase 2 [N = 83]; AML, 75%; HR-MDS, 12%; CMML, 11%; ALL, 1%) were enrolled (Table 1). Cytogenetic information was unavailable in 26% of patients; 19% were diploid, and 20% had adverse risk cytogenetics (Table 1). Overall, 63% of patients enrolled had RAS mutations; 13 patients (13%) had KRAS mutations, and 54 patients (56%) had NRAS mutations (Table 1). Twenty-five patients enrolled in 1 center (The University of Texas MD Anderson Cancer Center) had mutation testing for FLT3, NPM1, and KIT (Supplemental Table 1). Among them, 2 patients had FLT3 D835 mutation (both NRAS mutated), and 1 had NPM1 mutation (NRAS mutated). Ninety-six percent of patients had received at least 1 prior therapy, with the majority of patients (94%) having received chemotherapy (Table 1). For 64 patients (66%), trametinib was the third or greater line of therapy.

Table 1.

Patient Demographics and Baseline Characteristics

Demographics and Characteristics	< 2 mg (n = 6)a	Cohort 1 (n = 50)	Cohort 2 (n = 30)	Cohort 3 (n = 11)	Total (N = 97)
Age, years
Median	71.0	69.0	60.5	67.0	67.0
Range	(53–85)	(37–84)	(21–87)	(56–79)	(21–87)
Sex, n (%)
Male	2 (33)	27 (54)	22 (73)	5 (45)	56 (58)
Female	4 (67)	23 (46)	8 (27)	6 (55)	41 (42)
Primary tumor type, n (%)
AML	5 (83)	40 (80)	26 (87)	0	71 (73)
MDS	1 (17)	10 (20)	1 (3)	0	12 (12)
CMML	0	0	0	11 (100)	11 (11)
Other, not specified	0	0	2 (7)	0	2 (2)
ALL	0	0	1 (3)	0	1 (1)
Genetic abnormality, n (%)
Unknown	1 (17)	10 (20)	4 (13)	10 (91)	25 (26)
Other	0	11 (22)	11 (37)	0	22 (23)
Normal	1 (17)	11 (22)	5 (17)	1 (9)	18 (19)
Complex cytogenetics (> 5)	2 (33)	3 (6)	4 (13)	0	9 (9)
del 7	0	3 (6)	4 (13)	0	7 (7)
del 5	0	4 (8)	0	0	4 (4)
t(8;21)	0	1 (2)	1 (3)	0	2 (2)
t(16;16) or inv(16)	0	0	1 (3)	0	1 (1)
RAS mutationb
KRAS mutated	0				13 (13)
NRAS mutated	1 (17)				54 (56)
Prior anticancer therapy
Any therapy	6 (100)	50 (100)	28 (93)	9 (82)	93 (96)
Chemotherapy	6 (100)	48 (96)	28 (93)	9 (82)	91 (94)
Prior lines of therapy
0–2	5 (83)	29 (58)	17 (57)	7 (64)	58 (60)
≥ 3	1 (17)	21 (42)	13 (43)	4 (36)	39 (40)

Abbreviations: ALL, acute lymphocytic leukemia; AML, acute myeloid leukemia; CMML, chronic myelomonocytic leukemia; MDS, myelodysplastic syndromes.

^aCohort < 2 mg: includes patients who received 0.5 mg of trametinib once daily, 3-mg loading dose trametinib/1 mg of trametinib once daily, or 1 mg of trametinib once daily.

^bThere were 6 patients identified as both NRAS and KRAS mutation positive.

Treatment

The first cohort of patients enrolled in phase 1 (n = 3) were given a 3-mg LD of trametinib followed by continuous 1-mg once-daily dosing. Two additional patients were treated at the 1-mg dose level without an LD. Nine patients were treated at the 2-mg daily dose level. The only DLT at the 2-mg dose level was a fatal intracerebral hemorrhage in the setting of refractory thrombocytopenia that was considered attributable to trametinib by the investigator. Based on safety, PK, and pharmacodynamic data from a trametinib study in solid tumors¹⁹ in which DLTs (grade 3 rash and diarrhea, grade 2 chorioretinopathy) were reported at the 3-mg dose level, a program-based decision was made by the study sponsor to select the 2-mg dose as the RP2D despite no additional DLT-defining events occurring in this study. The overall median duration of treatment was 56 days (range, 7–337 days).

Toxicity

Diarrhea was experienced by 52% of patients (grade 1/2, 48%; grade 3, 4%; grade 4, 0%). Skin-related toxicities, including rash, were seen in 57% of patients (grade 1, 36%; grade 2, 15%; grade 3, 5%; grade 4, 0%). While 15% of patients experienced blurring of vision, visual acuity was reduced in 4%, and CSR was not seen.

The study mandated periodic assessment of cardiac ejection fraction (EF) at baseline and every 8 weeks thereafter. Asymptomatic reduction in EF from baseline was seen in 75% of patients: < 20% decrease in 62% and ≥ 20% decrease in 13%. In 9% of patients, EF dropped to < 40%. One event of EF decrease was considered treatment related (grade 3) and was reversible with treatment interruption. All other EF changes were not considered drug related and clinically significant, as these mostly occurred in the setting of infection or sepsis. Grade 2 or greater QT prolongation was reported in 5 patients (5%) without any related arrhythmia.

Overall, 77 patients (79%) experienced ≥ 1 treatment-related AE, 47% of which were grade 1 or 2, 22% grade 3, 10% grade 4, and 1% grade 5. Of these events, the most commonly reported treatment-related AEs were diarrhea (30%), rash (25%), nausea (13%), and elevated alanine aminotransferase (11%). Treatment-related AEs are summarized by dose and maximum toxicity in Table 2.

Table 2.

Treatment-Related Adverse Events

Adverse Event	All Grades		Grade 3		Grade 4
	No.	%	No.	%	No.	%
Occurring in 2 or more patients receiving < 2 mg trametiniba
Individuals with any event	5	83	0	0	0	0
Diarrhea	3	50	0	0	0	0
Nausea	3	50	0	0	0	0
Vomiting	2	33	0	0	0	0
Fatigue	2	33	0	0	0	0
Occurring in 2 or more patients receiving 2 mg trametinib (RP2D)b
Individuals with any event	72	79	21	23	10	11
Diarrhea	26	29	2	2	0	0
Rash	24	26	3	3	0	0
Nausea	10	11	1	1	0	0
ALT increased	11	12	6	7	0	0
AST increased	8	9	2	2	2	2
Fatigue	7	8	1	1	0	0
Left ventricular dysfunction	6	7	3	3	0	0
Stomatitis	3	3	1	1	0	0
Alkaline phosphatase increased	3	3	1	1	0	0
Anemia	3	3	1	1	1	1
Thrombocytopenia	3	3	2	2	1	1
Atrial fibrillation	2	2	1	1	0	0
Cerebrovascular accidentc	2	2	0	0	1	1

Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; RP2D, recommended phase 2 dose.

^aTreatment-related events reported in only 1 patient receiving < 2 mg of trametinib included the following: constipation, alopecia, dry skin, erythema, exfoliative rash, pruritus, glaucoma, and decreased appetite; all were reported as grade 1 or 2.

^bTreatment-related grade 3 or 4 events reported in only 1 patient receiving 2 mg of trametinib included the following: esophageal pain, gastrointestinal hemorrhage, face edema, edema, systolic blood pressure increased, ejection fraction increased, gamma-glutamyl transferase increased, hemoglobin decreased, acute myocardial infarction, convulsion, hypokinesia, tremor, hyperglycemia, hyponatremia, hypophosphatemia, febrile bone marrow aplasia, pancytopenia, and ischemia (all grade 3) and atrioventricular block, myocardial infarction, hyperuricemia, bone marrow failure, lymphopenia, neutropenia, streptococcal bacteremia, and graft-vs-host disease in skin (all grade 4).

^c1 cerebrovascular event was reported as grade 5 (fatal).

Dose interruptions due to AEs were reported in 48% of patients, with 5% requiring a dose reduction. Forty percent of patients with dose interruptions had trametinib treatment resumed at the full dose after resolution of the event. Three of the events requiring dose reduction (rash, left ventricular dysfunction, and elevated aspartate aminotransferase) were reported as treatment related. Discontinuation of treatment due to AEs (irrespective of attribution) occurred in 25 patients (26%).

Trametinib Plasma Concentrations

Trametinib was rapidly absorbed, with a median time to C_max of 1.75 to 3 hours across all doses (Supplemental Table 2) and a dose-proportional increase in area under the plasma concentration-time curve over the dosing interval but a less than proportional increase in C_max after 15 days. Trametinib accumulated upon repeat dosing (accumulation ratios range, 6.32–11.1), with a mean effective elimination half-life of approximately 96.4 to 174 hours (4–7 days; Supplemental Table 2). The interpatient percent coefficient of variation (CV%) for PK parameter estimates was generally moderate to high (34.6%–80.3% in the 2-mg dose cohort).

The mean predose plasma level of trametinib 2 mg was 12.22 ng/mL on day 15 of cycle 1 and 10.72 ng/mL on day 1 of cycle 2. In phase 2, trametinib trough concentrations were similar on day 1 of subsequent cycles in patients who continued beyond cycle 2, indicating steady-state levels. In general, these levels were above the target efficacious trough concentration of 10 ng/mL observed in the FTIH study (Figure 1).²⁰

Figure 1. Observed mean trametinib trough concentrations per cycle.

The solid line represents the efficacious trough concentration (10 ng/mL) determined in the first-time-in-human (FTIH) study. This figure compares the observed mean trough concentrations per cycle with the efficacious trough concentration determined in the FTIH study. Cycles with data for ≥ 3 patients (cycles 1–9) are represented.

Clinical Response

Among all patients treated at the 2-mg dose level, the overall median duration of treatment was 8.3 weeks (range, 1–48 weeks). The median duration of treatment was 10 weeks (range, 2–37 weeks) in cohort 1, 5 weeks (range, 1–48 weeks) in cohort 2, and 12 weeks (range, 3–44 weeks) in cohort 3. Fourteen responses were seen at the RP2D (Table 3). There were 10 responders with RAS mutation in cohort 1 (AML/MDS), 1 responder with RAS wt/unknown in cohort 2 (AML/MDS/CMML), and 3 responders with RAS mutation in cohort 3 (CMML).

Table 3.

Investigator-Reported Best Response in Patients Receiving Trametinib 2 mg

	Cohort 1a (AML/MDS With RAS Mutation; n = 50)	Cohort 2a (AML/MDS/CMML With RAS wt Mutation; n = 30)	Cohort 3a (CMML With RAS Mutation; n = 11)
Best response, n (%)
CR	4 (8)	0	1 (9)
CRp	1 (2)	0	1 (9)
Marrow CR	1 (2)	0	1 (9)
MLFS	3 (6)	0	0
PR	1 (2)	1 (3)	0
HI	4 (8)	5 (17)	0
SD	26 (52)	6 (20)	8 (73)
PD/TF	4 (8)	14 (47)	0
Unknown/not evaluable	2 (4)	0	0
Missing	4 (8)	4 (13)	0
CR + CRp + marrow CR + MLFS + PR, n (%)
Response	10 (20)	1 (3)	3 (27)
95% CI	8.9–31.1	0.0–9.8	1.0–53.6

Abbreviations: AML, acute myeloid leukemia; CMML, chronic myelomonocytic leukemia; CR, complete remission; CRp, complete remission with platelets < 100 × 10⁹/L; HI, hematologic improvement; MDS, myelodysplastic syndromes; MLFS, morphologic leukemia-free state; PD, progressive disease; PR, partial response; SD, stable disease; TF, treatment failure; wt, wild-type.

^aPatients in all cohorts received treatment at the recommended phase 2 dose (2 mg once daily).

In cohort 1 (n = 50), the overall response rate (ORR) was 20%, with 4 CRs (response durations = 4, 8, 14, and 48 weeks), 1 CRp (response duration = 8 weeks), 1 marrow CR (mCR; response duration = 12 weeks), 3 MLFS (response durations = 4, 5, and 10 weeks), and 1 PR (response duration = 5 weeks). There was only 1 PR (3%) in cohort 2 (n = 30). The ORR was 27% in cohort 3 (n = 11), with 1 CR, 1 CRp, and 1 mCR, with response durations of 8, 4, and 25 weeks, respectively. Thus, among RAS-mutated patients, the ORR was 21% (13/61 patients). The characteristics of patients who achieved CR are summarized in Supplemental Table 3. Among the 61 patients with RAS mutations, the median overall survival (OS) in the 13 responders was 6.6 months (95% CI, 4.4–11.5 months) and in the 48 nonresponders was 5.2 months (95% CI, 4.0–6.0 months; Figure 2). The median OS in patients with MDS or AML who had RAS mutations (cohort 1, 4.9 months [95% CI, 4.0–5.7 months]) was similar to that in patients with MDS or AML without RAS mutations (cohort 2, 3.0 months [95% CI, 1.8–7.4 months]; Figure 2). Additional biological activity of trametinib as evidenced by a ≥ 50% reduction in bone marrow blasts was seen in 34%, 7%, and 36% of patients in cohorts 1, 2, and 3, respectively. Similar reductions of ≥ 50% in peripheral blasts were reported in 54% (cohort 1), 30% (cohort 2), and 45% (cohort 3) of patients.

Figure 2. Kaplan-Meier overall survival curves.

(A) An ad hoc analysis was conducted to summarize the overall survival (OS) in responders (median, 6.6 months [95% CI, 4.4–11.5 months]) and nonresponders (median, 5.2 months [95% CI, 4.0–6.0 months]) with a RAS mutation. (B) In the planned analysis, OS was summarized by cohort (cohort 1: median, 4.9 months [95% CI, 4.0–5.7 months]; cohort 2: median, 3.0 months [95% CI, 1.8–7.4 months]; cohort 3: median, 14.5 months [8.6 months-not reached]). The OS curves are presented. AML indicates acute myeloid leukemia; CMML, chronic myelomonocytic leukemia; MDS, myelodysplastic syndromes; wt, wild-type.

Six responding patients (all from MD Anderson Cancer Center) had DNA samples available at enrollment and follow-up for next-generation sequencing for a limited set of cancer-relevant genes. None acquired FLT3, KIT, or NPM1 mutations. One patient acquired 1 mutation each in PTPN11 and FLT1, and another acquired a mutation in WT1.

Statistical Consideration

Response rates of 20% (95% CI, 8.9%–31%) and 27% (95% CI, 1.0%–54%) in cohorts 1 and 3, respectively, were deemed to be of sufficient clinical importance, with a posterior probability of success of 1. Accrual to cohort 2 was stopped after enrolling 30 patients due to lack of efficacy, with a 3% response rate and a posterior probability of success of 0.39.

Discussion

This is the first study linking clinical responses of a MEK1/MEK2 inhibitor to RAS mutations in myeloid malignancies. Single-agent trametinib activity in the relapsed/refractory setting underscores the relevance of MAPK signaling in RAS‐mutated myeloid malignancies. Although the strongest activity was seen in the CMML cohort, early closure of this cohort due to slow accrual impaired a more robust evaluation of activity. PK assessment indicated that trametinib trough concentration, above the effective biological concentration as predicted from preclinical studies, was achieved consistently at the RP2D. These results are consistent with trametinib activity in metastatic melanoma, where most activity was seen in the subgroup of patients with BRAF mutations.^20–22

The AEs were manageable with appropriate dose modifications and symptomatic therapy. Diarrhea was frequent, but dose interruptions or discontinuations due to diarrhea were seldom needed. While LVEF reductions were of concern, only 6% of patients experienced grade 3 or 4 events. Ocular toxicity was limited, with no events of RVO or CSR reported. Trametinib was also successfully resumed in half of the patients with significant treatment-related liver enzyme elevations.

The responses among patients with NRAS or KRAS mutation are comparable to responses with other kinase inhibitors in AML/MDS with relevant mutations.²³ However, a higher response rate among patients with RAS mutations did not translate into a survival benefit for those with RAS mutations or those who responded. The limited duration of responses indicates that kinase mutations such as RAS emerge later during leukemogenesis; this is supported by recent studies of the AML genome.^24–26 Given that many of the patients in this study were refractory to multiple therapies, it was very likely that multiple subclones with variable dependence on MAPK activation were in existence at treatment initiation. The lack of RAS allelic burden data in this study limits a definitive assessment of such subclones. Whether additional mutations in the RAS gene or complementary pathways emerge at the time of loss of response to trametinib is currently being investigated.

As epigenetic mutations are considered “early” mutations in myeloid leukemias, epigenetic modulators (eg, hypomethylating agents) with MEK inhibitors could potentially improve rate and duration of response. The promising activity of hypomethylating agents plus FLT3 inhibitors in FLT3-mutated AML supports this notion.²⁷ As hypomethylating agents (ie, decitabine and 5-azacitidine) are already approved for frontline treatment of MDS, CMML, and AML (in elderly patients), such combination trials should ideally be conducted in the frontline setting to have maximum impact.

In summary, trametinib has shown single-agent activity in patients with RAS-mutated myeloid malignancies. Treatment combinations based on a better understanding of hierarchical importance of mutational events in myeloid malignancies and use in frontline therapy will likely extend and expand the role of trametinib in myeloid malignancies.

Condensed abstract.

This is the first study to show a link between RAS-mutant myeloid malignancies and clinical response to MEK inhibitor therapy. These data highlight the importance of the RAS/RAF/MAPK pathway in leukemogenesis and support further study of trametinib in patients with RAS-mutant myeloid malignancies.