Melanoma: Can GSK2118436 overtake PLX-4032 (a.k.a. RG-7204 a.k.a. RO5185426)?

Posted by Jeff Berk, BOLT International;   I have melanoma on the brain this week.  Yervoy (ipilimumab; Bristol-Myers Squibb) was approved, and I shared earlier in the week some of the thoughts on that drug which I took from BOLT’s Melanoma Thought Leader panel; published December 2010.  So, the next melanoma parlor game is guessing who comes in second.  The easy money says PLX-4032, at least here in the States.  But in Europe, if drug development was a race, this is what GlaxoSmithKline’s and Roche’s (Plexxikon’s) melanoma programs would look like:

So the rest of this blog summarizes some of this discussion that BOLT’s melanoma panel had related to the strengths and weaknesses of GSK2118436.  We’re saving the PLX-4032 (a.k.a. look in the blog title) discussion for some other night.  Anyway, here’s how our Thought Leaders see the two in a head-to-head thought exercise:


  • In Europe RG-7204 and GSK2118436 will be available simultaneously; presumably by the end of 2011.  The Europeans will use BRAF inhibitors 1st-line in BRAF mutant V600E + melanoma, regardless of whether EMEA grants 1st-line labeling.  Since both drugs produce response rates of 60%-80% and PFS of 6-8 months, they assume that there is a survival benefit, even if the clinical data has not yet matured enough to show it.


“I believe that the GSK inhibitor and the Roche BRAF inhibitor works as good in first and in second line.  So there will not be a very big surprise about the mode of action and the efficacy rate.  I expect to see the same efficacy in first and second line”.


“I guess it is relatively easy, and I may be not only speaking for my own practice but also for very many of my colleagues in Germany and Austria.  One of the drugs, we have clearly targeted therapies and BRAF is the one they will target for BRAF inhibitors, and this is not only one but also two BRAF inhibitors; the one from Roche and Plexxikon and the other one from GSK which showed almost similar activity.  So if patients have a BRAF mutation and they have a stage IV melanoma disease it is very unlikely that they will not be treated with BRAF inhibitor simply because the response rate is exciting.  It is the in the range of let’s say 60 to 80 percent which depends if you are considering confirmed or unconfirmed responses.  So the PFS is in the range of six to eight months and it is extremely unlikely that these drugs don’t affect the overall survival time.  So therefore patients with BRAF mutations will be treated.  It is just a question of the logistics as these patients will be tested or the tumor material will be tested as routine in whatever, one or two years, or maybe in only half of a year”.



  • Reactivation of wild-type BRAF is responsible for the increase in squamous cell carcinomas seen with RG-7204.  This is managed with close dermatological surveillance and biopsy.  The drug also produced notable phototoxicity.  This is due to the formulation, not due to an intrinsic property of BRAF inhibition.  So this problem is uniquely Roche’s.  Neither of these AEs is a critical setback in the metastatic setting, but they will pose problems in the stage-2/stage-3 adjuvant setting.  Roche is developing a formulation change which will probably reduce the photosensitivity.


“But for sure what is different between the Roche compound and the GSK compound is the formulation.  The formulation is also changing some of the behavior in terms of the side effects.  For example, one of the prominent side effects for the Roche compound is the photosensitivity, which is not a property of the BRAF inhibitor itself but of the pharmacologic formulation.  The photosensitivity seems like it is going to be more of a problem treating in the future using the Plexxikon drug in the adjuvant setting.  I see it as being very difficult to take a younger person and say to them you can’t go outside. I completely agree but from what I know is that the company is working on that to change the formulation because that has been recognized, not only in Europe but also in sunny south parts of the US or Australia to put people on for months or years on such a drug.  I think in terms of quality of life the patient will tell you it is not possible”.


“So the pluses of the agent obviously are the dramatic response rate in the majority of patients treated who are properly molecularly selected and downsides are going to be the lack of durability in the majority of the patients as well as toxicities.  The biggest concern obviously is the squamous cell cancers which have been developing. But we have been able to manage that clinically with close dermatologic surveillance and biopsies and so forth”.





  • The key claim for an efficacy advantage for GSK2118436 will be that it is active in the brain, whereas RG-7204 is supposedly not active against the brain metastases. At autopsy 60%-70% of melanoma patients have brain metastases.  In a sub-study, 9/10 melanoma patients with brain metastases experienced tumor shrinkage on GSK2118436.  The, caveat we offer from a few of our panelists is that the activity of RG-7204 in brain mets hasn’t been fully evaluated, so there may turn out to be less of a difference than initially thought.


“So one thing is the suggestion that it has activity in the brain.  That is something that we don’t have with PLX.  So that would make it a big difference because the relapse in the brain is a big problem”.


“So right now GSK is rapidly approaching a brain metastasis trial which starts very soon.  I think that is one of the advantages with GSK; they are extremely fast.  They are asking good scientific questions and they are addressing these questions in sound clinical trials.  I like this very much.  They would certainly be in the lead in developing this drug also for brain metastasis patients.  But nobody knows yet.  But in theory I cannot see why PLX 4032 should not induce the same response rate as GSK”.


“There was some claim from GSK that they have better activity on brain metastasis. Well, they have treated, and it was in the presentation at ASCO, one or two patients with brain metastasis, small brain metastasis, and they have seen some shrinkage.  So they are going to explore that by actually doing a trial for brain metastasis.  Kind of like the trial we did with ipilimumab with Kim Margolin which was reported where you basically had small brain metastasis and then instead of giving them steroids you just gave them ipilimumab.  We have seen 28 percent response rate and stabilization or something like that, and if you added steroids it was lower.  So this will be a trial again with GSK BRAF, if you are positive and you have small brain metastasis you go on trial.  We will see if we can actually treat brain metastasis with this.  So that could be another opening there.  I know that if anybody failed on the Plexxikon, and again it is all published, the brain metastasis were one of the failure sites.  So it could be a difference.  Maybe the one gets into the CNS and the one doesn’t.  Just because you use 150 mg b.i.d. and the other one 960 b.i.d. I don’t think that makes one drug inherently better than the other, just the numbers, the milligrams.  That doesn’t tell me anything.  The sun sensitivity maybe a real differentiating factor, and the penetration to the brain might.  But, again, that is going to be hard to discriminate between those two drugs.  I think they are very similar in my mind”. 


“I guess that is one of the points of differentiation I have heard from some of the guys working specifically with the GSK molecule and that they have shown efficacy in the brain metastasis. Right.  And that is where I am trying to understand from your perspective how common is that as a secondary tumor site. Yeah, brain metastasis are very, very common and I feel like at autopsy series certainly more than 60 or 70 percent of patients will have brain metastasis that weren’t seen on imaging studies.  Right so then that strikes me as a big deal. Yeah, it is a big deal.  It’s a big deal.  If you are looking at patients with metastatic disease I would say at the time of diagnosis 10 percent will probably have clinically radiographically observable brain metastasis, and then the majority will develop them throughout the course at some point.  The presence of brain metastasis is a big, big deal.  I am looking up and seeing that the phase II study excluded patients with brain metastasis and this is for the Plexxikon drug.  The phase I allowed stable brain mets or brain mets were not progressing for at least three months at a time the patients were put on.  The patients that were treated I don’t think we have a good sense of whether this agent is active against progressing brain metastasis or not.  But again, because of the lead that this drug has, I think we are going to have this information.  Maybe we will find that it doesn’t work in brain metastasis.  And if that is the case then the GSK drug may really have a meaningful difference”. 



  • The toxicity profile of GSK2118436 may have less skin involvement than RO5185426 in terms of rates of photosensitivity.  As for squamous cell carcinoma, the further the drug gets in development, the higher the rate of keratoacanthoma is getting.  What was 9% is now 16%-17%.  The reason the GSK rate doesn’t approach the 30% frequency seen with RG-7204 is that GSK2118436 is more purely inhibiting mutant BRAF, and not hitting CRAF as well.  Neither of these adverse events are considered relevant in the metastatic setting, but in younger Stage-II / Stage III patients who are being treated after surgery for the prevention of recurrence, the ability to have at least minimal sun exposure is a definite quality of life advantage.


“The fact is the photosensitivity may be a little bit less with the GSK agent.  That has been a profound problem for a couple of our patients and I know others that have been treated and had the sensitivity.  But why some people are photosensitive and why others are not and whether those differences are going to be in properly done head-to-head comparisons we don’t know”.


“Is the principle on why you are going to see lower rates with the GSK compound because they are more targeted on the mutant BRAF? I think what you mean is that the Plexxikon agent may hit to some degree the CRAF and the GSK is said to be tighter focused upon BRAF, and to an extent I guess that is at least part of the potential explanation”.

“They may or may not try to look at safety profile and claim they are cleaner or easier to administer, although that is hard to achieve because the Plexxikon doesn’t have that horrible safety profile.  As we know from ASCO if you were at that presentation they basically said, well, we initially started with squamous cell carcinoma, keratoacanthoma 9% but as we are going forward with our trial as it is more mature they were quoting like 16% to 17% percent so pretty much getting closer to the Plexxikon rate anyway.  So they cannot claim that and that is the most bothersome one.  The only thing I think they haven’t seen the skin sun sensitivity. That may be the only real difference, the sun sensitivity, which one has to a certain degree and the other one claims not to have it.  But remember it is early.  The toxicity is subject to changes because you treat 50 patients, you see something you treat 150 patients, you see more in 500 and other things”.


“The toxicities of the two drugs, granted it is not a clear picture, but if there are different incidences of squamous cells – something I don’t care about –  but may it have a practical clinical importance, even though squamous cells are a small part of the problem.  The big problem is relapse, not squamous cells.  But also other skin toxicities and photosensitivity.  If the GSK drug has a different profile it may make it more appealing for the adjuvant setting.  Where the PLX drug, if you start having squamous cells and young people having to live indoors because they cannot be exposed to even a minimal amount of sun, that is a big problem.  My understanding is that from a survival standpoint the squamous cell carcinomas have no impact.  Is that why you are saying you don’t care about that? I don’t care about them in the metastatic diseases but in the adjuvant setting, so in stage III melanoma that is BRAF mutant and you want to run a big study to see if you treat BRAF up front do you get less patients developing metastatic cancer, and then do you cure some patients?  That will be the premise.  Then having less toxicities because you are treating a population that is healthy, that has not active melanoma, so you are trying to prevent melanoma then the smaller amount of toxicity is the better.  You will keep more patients on drug and you will see more effect.


“So I guess the differences between the Roche and the GSK compound may obviously hit other areas; the incidence of the squamous carcinoma or dyskeratotic acanthoma, kind of process in one (RG-7204) may be over 30 percent and the other may be very, very small.  Had these two been tested in the same institutions with the same surveillance for those squamous or intermediate level skin tumors?  No.  So I think we really don’t know that they are categorically different”.

Categories: CANCER, melanoma
Tags: , , , , , , , , , , , , , , , , ,
Bookmark the permalink.

Post a comment

Comments on are monitored.

You may use these HTML tags and attributes:
<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>