My memories of metastatic melanoma are all unpleasant ones. The very first melanoma patient I took care of was a musician from Austin who, as a veteran, showed up at the San Antonio VA hospital. We treated him with DTIC, then a relatively new drug, in an era before modern antiemetics. We made him miserable, his tumor grew anyway, and then he died even more miserably with progressive liver and subcutaneous metastases. Several decades later, my partner (and cancer center director) Steve Williams, one of the nicer human beings I have ever known, developed a superficial melanoma, which was resected. Then he developed an axillary lymph node metastasis, which was removed. He got interferon. Then he developed lung metastases. He was treated with IL-2. Both interferon and IL-2 beat him up. Late one day I ran into him and his wife leaving the cancer center: his arm had begun twitching for no good reason, a brain MRI had shown metastases, and he was starting radiation. I gave him a hug, but it was melanoma, and we both knew what this meant. He died a few weeks later. As far as I could tell, no therapy had ever done him the least bit of good. My colleagues and I still mourn him.
And that, in a paragraph, is the modern history of treatment for recurrent melanoma: therapeutic failure with toxic drugs.
2011 is just slightly more than half through, yet it is already clear that this is the year of melanoma. Within just a few weeks we have seen the publication of three Phase III trials in the New England Journal of Medicine (two of them presented more or less simultaneously at the ASCO Annual Meeting) showing overall survival advantages in metastatic melanoma.
3 Papers, 3 Separate Agents
That was amazing enough, but even more amazing was the fact that the three papers employed three separate agents: an immune modulator (ipilimumab), a vaccine (gp100), and a small molecule kinase inhibitor (vemurafenib). Ipilimumab is already on the market, and vemurafenib will likely not be far behind. The vaccine (so the study's author, and my IU colleague, Doug Schwartzendruber tells me) is not quite ready for prime time, due to the vagaries of the adjuvants used with the polypeptide in the vaccine, but the establishment of the principle of vaccine therapy in melanoma is certainly intriguing, if not cause for immediate celebration.
A melanoma specialist could be forgiven for joyful public outbursts. It is rather as if one had been wandering through a desert, becoming increasingly desperate, only to suddenly find oneself in an oasis amidst three life-giving fountains. And the desert wanderings were long, full of false hopes never realized.
Chicago Cubs fans had nothing on melanoma docs. Perhaps we should expect the Cubbies to win the World Series this fall: it would be no more remarkable than what we have seen in recent weeks with melanoma.
I am not a tumor immunologist, so I have found the BRAF kinase inhibitor vemurafenib the easiest of the three to wrap my head around. Helen Davies and colleagues first identified BRAF as a potential target in 2002 with a publication in Nature. Part of the RAS–RAF–MEK–ERK–MAP kinase pathway, it was discovered to be mutated at the V600E site in over half of melanoma cell lines tested, an observation that was quickly confirmed in clinical samples, where roughly half of melanomas harbor the mutation. Early attempts at inhibition in the clinic failed (think sorafenib), and for a few years one heard little about BRAF.
When assaults on a therapeutic target fail, either the hypothesis is wrong or the drug is inappropriate to the target for some reason. My laboratory colleagues sometimes labor under the impression that once they have demonstrated the ability of siRNA to inhibit cell growth in vitro, everything else is just trivial piecework, capable of being performed by not very intelligent hominids, called “clinicians” to distinguish them from “real scientists.”
The BRAF story suggests the importance of effective translation to the clinic, of the importance of getting the drug right (medicinal chemistry, preclinical toxicology, PK, PD, ADME, etc.) and of the need for persistence when the biology is telling you that the target just must be right.
One is reminded of a comment that Einstein made when asked how he would have felt had the experiments testing general relativity proved it wrong. “Then I would feel sorry for the good Lord. The theory is correct.”
And persistence was rewarded. In 2010 we saw the first glimmerings of hope,
with positive Phase II trials and tales of rapid and miraculous tumor shrinkage. The Phase III trial presentation at the ASCO plenary session, accompanying the ipilimumab Phase III, was something of a coronation ceremony: impressive increases in response rate, marvelous hazard ratios for progression-free and overall survival. This was a great moment for targeted therapy, and for melanoma patients.
And yet… The Progression Free Survival curves do not plateau, or at least not in the early going (follow-up is short). Melanoma is another smart tumor: before the ASCO presentation there were already six documented mechanisms of resistance to BRAF inhibition that I was able to count in the literature. And these were not just cell line epiphenomena, but rather lurking disasters documented from tumor samples in patients receiving the drug.
Is this yet another false dawn for melanoma patients? I certainly hope not, but the real test will be in the adjuvant setting, with small volume micrometastases in a potentially curative setting. Let's hope this is HER2 in breast cancer, not bevacizumab in colorectal cancer.
As to the other two advances, what fascinates there is the underlying class of these treatments. I have been a long-term skeptic regarding immune-based approaches to cancer. And, delightfully (if not, alas, for the first time), I was wrong, at least as far as melanoma was concerned. I grew up as an oncologist in the period when interferon and IL-2 were going to be the next wonder drugs, actively promoted as the greatest thing since sliced bread. A generation of oncologists was bitterly disappointed (and some perhaps secretly pleased, in a schadenfreude sort of way) by their abject failure, and by the many subsequent failures of immune-based approaches.
That's all past. We now have two T cell immunoregulatory agents that produce overall survival advantages in metastatic melanoma. Ipilimumab blocks cytotoxic T-lymphocyte–associated antigen 4 (CTLA-4), a negative regulator of T cells. Vaccination with the gp100:209-217(210M) peptide generates high levels of circulating T cells capable of recognizing and killing melanomas, particularly when combined with Interleukin-2 (though not, curiously, when combined with ipilimumab).
So Different from BRAF Story
The clinical results are interesting because they are so different from the BRAF story. The toxicities are different (real and significant for ipilimumab, trivial for the vaccine), the patients are different (no BRAF mutation needed, though the peptide study required expression of HLA*A0201), and the PFS curves look subtly different, with a hint of long-term benefit for both immune-based therapies.
We seem to be making a patchwork quilt for melanoma, overlapping pieces of cloth covering different aspects of the disease. This makes clinical trialists happy: non-overlapping toxicities and different mechanisms of action offer up combinatorial possibilities that will be exploited in the clinic.
Overall this has been a good year for melanoma patients and for the physicians caring for them, and for the principles underlying the biologic revolution we are experiencing. That isn't to say that all our challenges are gone. We will not be able to afford to treat everyone (ipilimumab and vemurafenib will both be “break the bank” drugs from a pricing standpoint), and we will undoubtedly face another round of the depressing “How much is a year of life worth?” arguments that increasingly dominate healthcare.
We will face the long hard slog of trying to overcome a priori drug resistance, something we have not been very good at. We also should devote more energy as a society to preventing this largely self-inflicted wound (sunscreen, yes; tanning salons, no). But for the moment we can be grateful for the times we live in, a glorious period in the history of our profession.
Before I leave this topic, let me mention one fascinating spin-off of the melanoma story: hairy cell leukemia. Hairy cell, a rare and fascinating cancer, with its absolutely distinctive morphology. It is not a big killer, nor a common cancer, even among leukemias, and it already has effective therapy in the form of purine nucleoside analogues. But a recent publication in the New England Journal of Medicine says that 100% of hairy cell leukemias have the V600E BRAF mutation seen in melanoma.
This strikes me as golden. When 100% of a particular type of cancer has a specific mutation, it is a safe bet that it is a driving mutation crucial to that disease, even defining the existence of the disease. I would be willing to bet that this will end up being something special for hairy cell, something akin to BCR-ABL for CML. But that is a story yet to be told.