Kim A. Margolin, MD


June 18, 2012

This feature requires the newest version of Flash. You can download it here.


Hello. I am Kim Margolin, Professor of Medicine, Division of Oncology, University of Washington in Seattle and member of the Melanoma Research Alliance Advisory Panel. Welcome to this edition of Medscape Oncology Insights on melanoma.

Today, I would like to highlight for you some of the important data on melanoma presented at the 2012 annual meeting of the American Society of Clinical Oncology (ASCO®). I would like to point out that the new data are almost all on advanced disease or treatment in the adjuvant setting because that is what melanoma medical oncologists focus on at the ASCO® meetings.

Adjuvant Therapy: Challenging the Status Quo

I will start with the adjuvant data. This morning, we heard several abstracts on high-risk patients following surgery who were placed on adjuvant clinical trials. A study from Germany[1] demonstrated the absolute equivalence of intermittent, intravenous (IV) delivery of alpha interferon in comparison with a traditional 1-month initial course of IV therapy followed by 11 months of subcutaneous therapy, 3 times weekly, which has been the standard of care for many years in the United States. The outcomes were identical with respect to all aspects of disease, relapse, and survival, but differences in terms of toxicity and tolerability were marked. These measures were influenced by the amount of interferon that was tolerated by patients by the end of the prescribed interval, as well as the dose-limiting toxicity, namely fatigue. Therefore, one option, if using interferon in the adjuvant setting, is to consider this intermittent IV schedule and route of administration in lieu of the traditional form of delivery, because it is less fatiguing and equally efficacious.

We also heard the results of an extremely mature SWOG intergroup study[2] in which the data from a randomized 1:1 trial comparing biochemotherapy (a 5-drug regimen) with standard alpha interferon (using the traditional regimen that I just described) showed substantial and statistically significant benefit in progression-free survival for biochemotherapy over interferon. In this very mature trial with many years of follow-up, there was absolutely no survival benefit, but subset analysis remains to be done. Although the very great promise of the biochemotherapy over interferon might tempt practitioners to use biochemotherapy, which is much shorter in duration although acutely more toxic, one has to wonder just how clinically useful a regimen with such toxicities, and which does not improve overall survival, could be.

We also heard the very provocative but somewhat preliminary data from a Chinese study[3] in purely mucosal melanomas, which is the most common type of melanoma seen in China. This phase 2 study was a 3-arm randomization (1:1:1). However, despite this phase 2 design, the study suggested a very strong relative hierarchy of outcomes for both relapse-free and overall survival favoring the chemotherapy arm (which consisted of temozolomide and cisplatin) followed by standard alpha interferon adjuvant therapy and, finally, observation in the third arm. These results were very dramatic but preliminary in a relatively small number of patients, and they definitely require confirmation in a larger trial. Hopefully that larger trial will also include molecular characterization of those patients.

Breakthroughs Continue in Advanced Disease

Moving on to advanced disease, we saw tremendous advances even over those of last year and the year before, when we thought we were witnessing some amazing breakthroughs in both targeted therapy and immunotherapy for advanced melanoma. Here are some examples: John Kirkwood[4] presented a trial for a multicenter effort in which dabrafenib, which is a new but still experimental BRAF inhibitor -- very similar to but not identical to the approved agent vemurafenib -- was used in patients with advanced melanoma who also had brain metastases. Half of the patients had previously received some form of radiotherapy to the brain, and the other half were untreated. Patients were generally asymptomatic with limited brain disease. The primary endpoint was brain response, although all of these patients also had measurable advanced disease outside the central nervous system.

The results for the primary endpoint, which was activity in the brain, were virtually identical to those that we heard last year and which were updated again this year by Paul Chapman,[5] in which the dabrafenib results looked identical to those that had been reported for vemurafenib in extracranial disease alone. The extracranial responses to dabrafenib were also very similar to those reported for vemurafenib, with the caveat that these are not head-to-head comparisons. Nevertheless, we generally think of dabrafenib and vemurafenib, activity-wise, as being very similar agents.

Dabrafenib and Vemurafenib Differ in Toxicity

There are some very important differences in toxicities, which favor vemurafenib with respect to a systemic "pyrexia reaction," which is almost never seen with vemurafenib but is seen in a substantial percentage of patients taking dabrafenib. On the other hand, the important skin toxicities, particularly the emergence of low-grade squamous cancers and keratoacanthomas, which occurred in a substantial number of patients taking vemurafenib, appear quite unusual with dabrafenib. We don't know yet about the off-target mechanisms of these differences and how much may be due to the vehicle or the formulation for each of these oral agents. Furthermore, there are significant challenges for these patients after failure of these drugs, which tend to work for an average of 5-6 months. In addition to the short duration of activity, we don't know whether there is partial control of some of the lesions, raising the questions of whether to continue therapy after failure in only some of the patients' disease, whether to combine it with radiotherapy, and what to do after failure. There is also the exciting area of learning pertaining to combinations of targeted agents, as well as combinations of targeted therapies with some of the newer immunotherapies.

Proof of Concept for BRAF Inhibition

We also heard about a comparison between dabrafenib and dacarbazine in a study that was similar to, but not identical in design, to the BRIM-3 study that came out last year and updated today. This was presented by Axel Hauschild[6] for an international group, and this study was weighted in a 3:1 randomization favoring dabrafenib over dacarbazine. Of course, that does not mean, statistically, that the study requires fewer patients or has higher power, but ethically it's very important to have the maximum amount of what you know to be a highly active drug available to patients with a devastating disease. The trial and the report that we heard today confirmed what we know to be the case for vemurafenib, further proof of the concept that treating with a drug that targets in a very specific fashion a driver-mutated pathway in melanoma meets the proof of concept and has promise for the future.

Understanding Drug Resistance in Melanoma

One of the most important and immediate challenges, given the very brief term of benefit of these agents, is to understand some of the mechanisms of resistance and to be able to intervene in some of those mechanisms as quickly as possible with new agents. Until we have new agents to manage all of the different mechanisms of resistance, we need to at least study them, if possible, and use them not only to predict possible therapeutic failure before it becomes a big problem for the patient (assuming we have second-line therapy agents available) but also to pick out the small fraction of patients who may not benefit at all and for whom other clinical trials or other forms of novel therapy should be developed.

It is important to study baseline and on-therapy mechanisms of resistance, as very elegantly demonstrated to us by Jeff Sosman,[7] on behalf of a large multi-institution group, who presented a phase 2 trial of vemurafenib in previously treated patients. These patients are ideal because the response rate is so high that it gives the investigator the opportunity to have tissue available not only from pretreatment patients but also from on-treatment patients whose tumors are regressing, and from patients who go on to develop progressive disease. This allows for the investigations of the molecular mechanisms for progression, which do vary quite a bit. It seems as though every month we identify another molecular mechanism of failure on BRAF inhibitors that will require further study and intervention, as well as drug development.

Big News: PD-1 Blockade

Finally, the biggest news of the meeting for the melanoma folks was the PD-1 presentations, which may turn out to change the New England Journal of Medicine to the "New England Journal of Melanoma," as Dr. Antoni Ribas quipped in his discussion at the end of the session. This was the group of presentations that we heard on blockade of PD-1, the PD-L1 axis.[8,9,10] This is one of the checkpoint axes that controls the interaction between T-lymphocytes of the immune system and tumors as they grow. At this point we have more mature and promising data for PD-1 blockade, although we do have some for PD-L1 that we heard earlier in the meeting. We saw today the activity in the melanoma cohort of about 100 patients from the large phase 1 study[10] of 300 patients who had various solid tumors in which the overall response rate was approximately 30%. The responses were early and they were very durable. That is something that may contrast with ipilimumab, particularly the early onset of responses, which has been a challenge with our previously approved checkpoint blocker. There also appears to be much less of the immune-related adverse-event profile compared with ipilimumab, although with a caveat: no direct head-to-head comparisons have been done. Furthermore, there may be a biomarker of the effect of PD-L1 axis blockade -- the presence of PD-L1 on tumors. That would be a proof of concept that blockade of a pathway, which you can stain for immunohistochemically, is the precise mechanism by which this form of therapy works.

Thank you for joining me in this edition of Medscape Oncology Insights. This is Kim Margolin reporting from the 2012 ASCO® meeting in Chicago.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.