Combining Immune & Targeted Therapies in Melanoma: A Paradigm for Other Tumor Types?

CHICAGO-For the last few years, melanoma has been in the cancer research spotlight, both because of the development of targeted therapies that have yielded dramatically improved responses rates in patients with certain mutations, and as the tumor type in which the greatest efficacy to-date has been seen for new immunotherapy agents.

In a talk at the ASCO 2014 annual meeting, Alexander Eggermont, MD, PhD, Director of the Gustave Roussy Comprehensive Cancer Center in Paris, posited that melanoma might thus be taken as a model for first-class cancer medicine development. The story of how the targeted therapies, immunotherapies, and combinations of these therapies have been developed for use in patients with advanced melanoma holds a number of important lessons for the development of agents for use in other tumor types.

In the first decade of this century, it was discovered that RAS pathway mutations were responsible for a worse prognosis in Western patients with advanced melanoma-especially younger patients. The further discovery that BRAF inhibitors (vemurafenib, dabrafenib) resulted in dramatic improvements in response rates in patients positive for these mutations stirred real excitement in the melanoma community. However, this excitement was somewhat short-lived, as it wasn’t long before researchers realized that these responses were only a transient effect-that resistance to BRAF inhibitors inevitably develops, such that these agents have little impact on overall survival. Moreover, this modest benefit came at the cost of a worse-than-expected adverse event profile. Patients treated with BRAF inhibitors developed squamous cell carcinomas, new melanomas, colon polyps, and gastric polyps.

It was postulated that a different class of targeted agents-MEK inhibitors-might improve this situation if used in combination with BRAF inhibitors. This proved to be the case: patients with BRAF-mutated tumors who were treated with a BRAF inhibitor and a MEK inhibitor (e.g., dabrafenib + trametinib) did demonstrate improved overall response rates compared with a BRAF inhibitor alone. Also, MEK inhibitors alone showed some efficacy in patients with NRAS-mutated melanomas, a group that had traditionally had the worst prognosis of all, and for which, until then, no effective new options had been available. However, these responses, too, were short-lived.

So what lessons have been learned from the trials of targeted therapies in advanced melanoma? Chiefly, Dr. Eggermont suggested, that cross-talk between pathways introduces complexity, and that we need to move away from mono-dimensional thinking about pathways. An important reason why shutting down the RAS region of melanoma has proved so difficult is that p53 and TGF signaling remain active, leading to increased epidermal growth factor receptor (EGFR) and platelet-derived growth factor receptor B (PDGFRB) expression.¹ Of note, it has been established that the reason BRAF-mutated colon cancers do not respond to BRAF inhibitors involves feedback activation of EGFR-and that giving the EGFR inhibitor cetuximab along with vemurafenib to patients with BRAF-mutated colon cancer does result in a response. As an example of a strategy that takes into account and works with this complexity, Dr. Eggermont cited a recent Nature study by Sun et al.,¹ which described how the EGFR expression found in melanoma patients in whom resistance to BRAF or MEK inhibitors has developed declined rapidly when their BRAF inhibitor was stopped; the researchers showed that intermittent drug dosing is a strategy that can restore the efficacy of BRAF inhibitors in these patients.

Despite the exciting results seen with new targeted therapies, Dr. Eggermont noted that the big challenge of cancer drug development remained finding a way to break tolerance. In oncology, he said, the goal is to reawaken the T-cell response. This has begun to be accomplished in melanoma through an innovative approach to immunomodulation-an approach focused on “inhibiting the inhibitor” rather than “activating the activator.” T-cells have both activating receptors (e.g., CD26, OX40, CD137) and inhibitory receptors (e.g., CTLA-4, PD-1, LAG-3). Both agonistic antibodies that bind to activating receptors and blocking antibodies that bind to inhibitory receptors can, in theory, result in a reawakening of the T-cell response.

The agents that are central to this new approach to breaking tolerance are the immune checkpoint blockers. Anti-CTLA-4 antibodies were the first class of immune checkpoint blockers to demonstrate the efficacy of the “inhibiting the inhibitor” strategy.  Although the first of these, ipilimumab, resulted in lower response rates in melanoma patients than were seen with BRAF inhibitors, they achieved significantly more durable responses. In a pooled analysis of data from trials of melanoma patients treated with ipilimumab at a dose of 3 mg/kg, the 3-year overall survival was 22%. In fact, Dr. Eggermont pointed out; these drugs may provide a mechanism for long-term control in a fraction of patients.

Another class of agents that “inhibit the inhibitor” are the anti-PD-1/PD1-L antibodies nivolumab and pembrolizumab (formerly MK-3475)-which Dr. Eggermont termed the “drugs of the year” for 2013. Unlike anti-CTLA-4 antibodies, which are centrally active, the anti-PD-1/PD1-L antibodies work more peripherally, in the tumor microenvironment. These agents have shown response rates in melanoma patients that are 3 times those achieved with anti-CTLA-4 agents (a 55% overall response rate in ipilimumab-naive patients, and a 41% overall response rate in ipilimumab-pretreated patients). Moreover, they cause significantly less toxicity.

Given all the exciting data for both classes of immune checkpoint inhibitors, as well as for targeted agents, melanoma researchers were naturally drawn to the prospect of combining agents from these various classes.

The prospect of combining a targeted agent-with its high rate of initial response-with an immune checkpoint blocker-with its promise of a more durable response-was especially enticing. However, the combination of vemurafenib and ipilimumab led to severe hepatotoxicity in the first small trial in which this was tried; several patients died and the study had to be stopped. And although researchers were able to complete the first small trial of dabrafenib and ipilimumab, another early study of dabrafenib plus trametinib plus ipilimumab also had to be stopped early because of toxicity.

The lesson to be learned here, said Dr. Eggermont, is that we are very bad at predicting toxicities. These must be sorted out by carefully planned and conducted clinical trials.

Combining immune checkpoint blockers from the two different classes (anti-CTLA-4 antibodies and anti-PD-1/PDL-1 antibodies) has proved more promising. While not wholly unanticipated, high levels of immune-related toxicity have been an issue with the ipilimumab-plus-nivolumab combination-although for the most part the toxicities have been manageable. Results with concurrent ipilimumab plus nivolumab have been very good, demonstrating the same long-term responses in a fraction of patients now regarded as characteristic of immune checkpoint blocking approaches. Still, the waterfall plot for the ipilimumab-plus-nivolumab combination is not that different from the waterfall plot for pembrolizumab alone, and it is not yet settled whether the combination represents an improvement over the single-agent therapy.

Dr. Eggermont noted, however, that once T-cell inhibitory mechanisms have been inhibited, the door is opened to the possibility of using agonists. Before, he said, these didn’t get us very far, but in combination with inhibitors of T-cell inhibition (such as anti-CTLA-4 and anti-PD-1 agents), they may be quite effective. There are no data yet, as trials are just getting started, but he felt this was definitely an area to watch.

In sum, Dr. Eggermont believes immune therapy combinations will dominate melanoma therapy for years to come. Because they go the furthest toward the goal of breaking tolerance, he believes these combinations will be the backbone of melanoma treatment and within the next 5 to 10 years will likely have the potential to produce “clinical cure” in a major fraction of melanoma patients. Moreover, because anti-PD-1 agents have already shown efficacy in other cancers (e.g., renal, lung), such combinations also have real potential to produce similar results in a number of other tumors: lung cancer, renal cancer, bladder cancer, lymphoma, and more.

REFERENCE:

• Sun C, Wang L, Huang S, et al. Reversible and adaptive resistance to BRAF (V600E) inhibition in melanoma. Nature. 2014; 508:118-22.