Originally published June 6.
By Turna Ray
CHICAGO — In order to improve the chances that next-generation MEK inhibitors are commercialized for the treatment of melanoma and other solid tumors, researchers should take a more molecularly targeted approach by genotyping patients to stratify best responders to the drugs, a reviewer asserted at the American Society of Clinical Oncology's annual meeting here this week.
"Patients don't go to Phase I trials just to complete cohort recruitment ... Patients go to Phase I trials because they want to live, they want response, they want improvement in quality of life," Patricia LoRusso, an oncologist at the Karmanos Cancer Institute at Wayne State University, said during a clinical science symposium on MEK inhibitors at ASCO.
"Seeing that we have had other agents in this class, there should be some sort of selection of tumor types, or some type of genomic profiling, or at least retainment of archival tissue, so that we can go on to profile these patients at a later date," added LoRusso, who reviewed early dosing trials from three MEK inhibitors being developed by GlaxoSmithKline, Merck Serono, and Eisai. LoRusso has also participated in pharmacogenomic analysis of a MEK inhibitor being developed by AstraZeneca employing genotyping, for which data was also presented at a separate session at the meeting.
BRAF activates MEK, which activates the extracellular signal-regulated kinase or ERK, which is an upregulator of the transcriptional pathways that promote cellular proliferation and survival. As such, MEK inhibitors have shown efficacy in cancer patients, particularly in those with melanoma who harbor BRAF mutations. Published research suggests that BRAF mutations occur in more than 50 percent of melanoma patients.
At the clinical symposium, researchers presented data on three early-stage MEK-inhibitors: GlaxoSmithKline's GSK1120212, Merck Serono's AS703026, and Eisai's E6201. These three trials are all first-in-human dosing studies, and as such the data are preliminary. GSK conducted dosing analysis based on genotypes, but the latter two companies did not include such analysis in their abstracts.
Despite the early nature of the studies, LoRusso's review suggested that genotyping perhaps should have played a greater role in determining the maximum tolerated dose of these agents. In reviewing the data, LoRusso questioned whether these agents were viable as monotherapies and advised sponsors to genotype patients to personalize these treatments based on patients' molecular profiles.
"It's not just important to know why these patients respond, but it's equally, if not more, important to know why they don't. And you can't do that by just supposing; you have to look at the tissue," LoRusso said.
A recommendation for genotyping colorectal cancer patients for KRAS mutations before treating them with EGFR-inhibiting monoclonal antibodies such as Vectibix and Erbitux was issued at ASCO's annual meeting two years ago. The impact of that recommendation — although not based on results from early studies but on Phase III results — resulted in driving adoption of KRAS testing in the metastatic colorectal cancer population.
Since genomic profiling is being considered by drug firms much earlier in drug development, LoRusso's recommendation did not fall on deaf ears. Although limited genomic analysis of first-generation MEK inhibitors didn't bear too much fruit, "the refinement of assays and molecular profiling have come a long way," and can guide targeted development of next-generation drugs in this class, LoRusso suggested.
The Studies
GSK1120212
At the symposium, GlaxoSmithKline presented safety and efficacy data from the first-in-human study of an oral MEK 1/2 inhibitor, GSK1120212. The objective of the study was to define the maximum tolerated dose of the drug and to evaluate the pharmacokinetics, pharmacodynamics, and response rate of the agent in advanced solid tumors and lymphoma.
In the study, researchers led by Jeffrey Infante from Sarah Cannon Research Institute and Tennessee Oncology, reported that the drug's long half-life of four-and-a-half days and its small trough:peak ratio of 2 "allows constant target inhibition within a narrow range of exposure." Additionally, the recommended Phase II dose of the drug "is well tolerated and associated with durable clinical activity in specific tumor types."
Researchers enrolled 84 patients, 29 of whom were melanoma patients and 15 had pancreatic cancer. The maximum tolerated dose was 3 mg once daily and the recommended Phase II dose was 2 mg once daily.
Those "specific patients" who saw some benefit in the trial were 5 of 20 melanoma patients with known BRAF mutations, who saw partial responses to GSK1120212 and had 50 percent or greater reduction in their tumors. Three of these patients stayed 30 weeks or more in the study, while the other two patients are still being studied.
[ pagebreak ]
Additionally, in the 11 BRAF mutant melanoma patients, three had partial responses, 5 had stable disease, and three had progressive disease, mostly due to new brain lesions seen in two of these patients. In nine BRAF wild-type melanoma patients, three patients reached stable disease, and two had partial responses, including one patient who had a GNAQ mutation.
Dose-limiting toxicities with GSK1120212 were rash, diarrhea, central serous retinopathy, and were reversible. At doses greater than or equal to the recommended Phase II dose in 77 patients, the most common adverse events were rash (77 percent) and diarrhea (45 percent).
In addition to the dosing study with GSK1120212, GSK separately presented data from an ongoing dosing study on a BRAF inhibitor, called GSK2118436, in melanoma patients. According to the abstract, the drug displays "dose-dependent inhibition of MEK and ERK phosphorylation in mutant BRAF cell lines and tumor regression in xenograft models."
In this study, 61 patients (52 of whom had BRAF-mutated melanoma) received 12-400 mg daily. "In patients with mutated BRAF melanoma (without brain mets) 18 out of 30 (60 percent) patients have a greater than 20 percent tumor decrease by RECIST at first restaging (8-9 weeks)," researchers reported, adding that maximum-tolerated dose has not yet been reached in this study.
"Clinical activity with minimal toxicity was observed at multiple dose levels in mutated BRAF tumors," they added. This abstract was not part of the MEK inhibitor symposium.
AS703026
The second Phase I human dosing study presented during the symposium was Merck Serono's AS703026, also an inhibitor of MEK1/2, conducted by Delord et al. In the study, patients with solid tumors received oral, fixed doses of AS703026 once a day, on days 1-5, 8-12, and 15-19 (schedule 1) or days 1-15 (schedule 2), of 21-day cycles.
As of Nov. 1, 2009, 68 patients (36 patients on schedule 1; 32 patients on schedule 2) received AS703026 across 10 dose levels. The most frequent tumors seen in the study were from colorectal and melanoma patients.
With this structure, researchers found that AS703026 "showed a good safety and [pharmacokinetic] profile with complete target inhibition in peripheral blood mononuclear cells at doses ≥28 mg/day." The most common adverse events seen with AS703026 were asthenia, diarrhea, acne-like skin reactions, nausea, constipation, and vomiting. At 28 mg/day, one out of six patients experienced a dose-limiting toxicity, of grade three liver function test elevation. Some patients experienced visual disturbances at 94 mg/day.
In this study, the maximum tolerated dose has not been reached and dose escalation is ongoing. But "antitumor activity has been observed prior to reaching the maximum tolerated dose." While genomic mutation analysis may be planned for later studies, this study did not analyze the impact of AS703026 doses based on patients' genotypes.
E6201
The third presentation, by Borad et al., was a Phase I, open-label study evaluating the maximum tolerated dose of Esai's E6201, which has shown to inhibit MEK1 in solid tumors. Sequential cohorts of three to six subjects were given 30-min IV infusion of E6201 on Days 1, 8, and 15 of a 28-day cycle starting at 20 mg/m2 and increasing in 100 percent increments up to 320 mg/m2 until two grade 2 toxicities or one dose-limiting toxicity was observed. At that point, doses were then increased by 50 percent or less up to 720 mg/m2 or until the maximum tolerated dose was determined.
"The E6201 MTD of 320 mg/m2 IV once weekly for three out of a four-week cycle was well tolerated in patients with advanced solid tumors," the researchers concluded, reporting one dose-limiting toxicity of grade 2 QTc prolongation and grade 2 central nervous system toxicities, such as dizziness, dysarthria, andataxia.
According to the researchers, an expansion of this study will be done at the maximum tolerated dose "to assess efficacy and determine the optimal schedule using biomarker-driven endpoints and tumor response."
The Problems
Following the presentation of E6201, one audience member commented that ASCO should recommend that sponsors do not advance MEK inhibitors in clinical trials without genotyping and stratifying patients based on molecular markers.
This sentiment was echoed by LoRusso in her review of the three abstracts, in which she estimated 272 patients were evaluated in total, of which 50 patients had mutation status available, with 25 exhibiting BRAF mutations. Of these, 21 patients were evaluable, and 19 BRAF-mutated patients had a clinical benefit from treatment.
This further supports the view that researchers should molecularly profile patients to predict whether they are likely to benefit from MEK inhibitors.
Dose-limiting toxicities for the GSK drugs were central serous retinopathy, diarrhea, and rash. These adverse reactions were characterized by LoRusso as being typical of this class of drugs. The Merck Serono compound induced in some patients liver function abnormalities and retinal vein occlusions. According to LoRusso, past experience with MEK inhibitors suggests that retinal vein occlusions may be a class effect.
[ pagebreak ]
"The Esai compound had none of the standard dose-limiting toxicities for MEK inhibition, but rather had grade 3 QTc prolongation," LoRusso pointed out.
Additionally, typical MEK toxicities of skin rash, peripheral edema, and visual disturbance, such as blurred vision, were not seen in the Esai compound. "But rather we see diarrhea, which isn't a class effect, but something we see with a lot of small-molecule inhibitors, as well as nausea, and vomiting."
Based on this data, LoRusso characterized E6201 as "not a selective MEK1/2 inhibitor, as the other two compounds were," but as a "dirty inhibitor" that has inhibition against multiple kinases. "And based on this, maybe the toxicities that ensued were a result of not being able to achieve the doses necessary for MEK inhibition, thus giving a different toxicity profile than we would have expected to see with a classic MEK inhibitor."
The half-lives of the three drugs were also scrutinized with regard to the doses administered to patients. The half-lives of the Merck Serono compound and the Eisai compound are around five hours. In particular, the short half-life of Eisai's drug was seen as problematic by LoRusso and audience members. Meanwhile, the half-life of GSK's MEK inhibitor is approximately four-and-a-half days, which was seen as more favorable from a disease-targeting standpoint.
Next-Gen Successes?
In the past, companies have failed to advance MEK inhibitors in solid tumors. For example, two early-stage MEK inhibitors developed by Pfizer, CI-1040 and PD325901, were terminated due to low efficacy and safety issues, respectively, several years ago.
CI-1040 was aborted due to low bioavailability. After "significant toxicity" was seen in clinical trials of PD325901, the agent was "temporarily shelved" and the "fate of this drug currently rests in the hands of Pfizer," said LoRusso.
At the time, these early attempts did not benefit from advanced molecular profiling techniques more readily applied in drug development today. By comparison, LoRusso highlighted the benefit of genotyping in a study she was involved in that looked at the impact of a MEK inhibitor being developed by AstraZeneca in melanoma patients.
At the meeting, separate from the clinical symposium on MEK inhibitors, LoRusso and colleagues presented data from a Phase I trial of the MEK inhibitor AZD6244, in which patients received the investigational drug in combination with one of three chemotherapies: dacarbazine, docetaxel, or temsirolimus. Upon genotyping 19 out of 25 patients with available archival tissue samples for BRAF and NRAS using pyrosequencing, researchers highlighted that five patients that responded to an AZD6244-combination regimen had BRAF mutations, and none of the patients with BRAF mutations had early disease progression.
"Presence of BRAF mutations was significantly associated with both clinical responses and increased time to progression," the researchers concluded in the abstract. In BRAF-mutant melanoma patients there was a 56 percent response rate with a time-to-tumor progression of 31 weeks; both findings were statistically significant. "While the number of patients analyzed is small, the trend toward clinical benefit in patients with BRAF mutations is inferred," the researchers added.
This research looking at the MEK inhibitor in combination with chemotherapy further supports LoRusso's hypothesis that this class of drugs may not be viable as a single agent. The overall response rate in the AstraZeneca-funded study of AZD6244 resulted in an overall response rate of 28 percent in melanoma patients, "which is higher than any of the ... drugs alone as monotherapy have ever been reported in terms of efficacy," LoRusso said.
"The third generation of MEK inhibitors are not novel first-in-class ... they are just the first in human trials of a specific drug. So, could [the researchers] have explored a more tailored trial design to try to accelerate recruitment and minimize patients being treated at low exposure levels? And could we have also incorporated molecular signatures into our patient cohorts and populations?" she asked.
Additionally, Plexxicon and Roche are developing a treatment in melanoma in patients harboring the V600E mutation of the BRAF kinase gene. Preliminary data in 27 evaluable patients showed 18 patients to have partial responses with greater than 30 percent tumor regression, and 15 patients had responses with more than 50 percent tumor regression. Minor responses of between 10 percent and 30 percent were seen in six patients.
PLX4032 is currently being studied in a Phase II trial. Roche and Plexxicon are developing a companion diagnostic to select melanoma patients with BRAF V600E mutations for treatment with the agent.