Consider immunotherapy as first-line treatment.

Application of immunotherapy is becoming increasingly more sophisticated as we learn to control the many signals that activate and deactivate T cells. We now have two approved immunotherapies for metastatic melanoma – high-dose interleukin-2 (Proleukin) and ipilimumab (Yervoy) – and at least two investigational ones – adoptive T-cell therapy and a novel antibody to PD-1 (MDX-1106; ASCO 2010 annual meeting. Abstract 2506) – that are especially promising.

There is no question that if you give only a single agent and compare agents head to head, ipilimumab will lose to vemurafenib (Zelboraf), a targeted BRAF inhibitor, in terms of response rate, progression-free survival, and probably median survival. But our goal is the best overall outcome in this population, and in that regard, the sequencing of agents comes into play, and there may be advantages to giving immunotherapy first.

The first of several factors to consider when selecting initial therapy is that immunotherapy can produce long-term durable remissions; for example, about 5% of patients given high-dose interleukin-2 have a remission lasting more than 2 years, in some cases more than 10 years, often without maintenance therapy (Clin. Cancer Res. 2006;12:2353s-8s). In contrast, virtually all patients given targeted agents will require some sort of second therapy when they have progression.

Second, the actual treatment duration of immunotherapy is quite limited, and patients having a response may never need a second therapy. But for targeted therapy, as far as we can tell, patients will have to remain on that therapy forever.

A third factor is the speed of response. Reponses to immunotherapy are slow, in some cases taking at least 12 weeks, whereas responses to targeted agents often occur within days. So it may be difficult to give immunotherapy if you wait until a patient has progression on the targeted agent, whereas the targeted agent works so quickly that it may still be an option after failure of immunotherapy.

It is also noteworthy that targeted therapy can be associated with fulminant relapse. Our limited experience has been that a subset of patients has incredibly rapid progression when this therapy begins to fail and the targeted agent is discontinued, so there’s no opportunity to give immunotherapy at that point.

To be sure, patients with high tumor burden and very rapid tumor progression at initial metastatic presentation are generally not good candidates for receiving immunotherapy first. So you really have to look at the overall patient and consider potential predictive markers like low baseline lymphocyte count (which would disfavor immunotherapy) and preexisting autoimmune disease (which would exclude patients from some immunotherapies and favor targeted therapy).

Fourth, it has been proposed that giving immunotherapy first and targeted therapy second may be synergistic. Admittedly, this is purely theoretical. But the idea is that ipilimumab, for instance, remains in the body for up to 3 months, so if a patient starts to develop progression on this agent and a targeted agent is given at that time, you are essentially giving combination therapy.

Also, antitumor immunity may play an important role in the efficacy of targeted therapy. Very interesting preclinical data suggest that some of the activity of targeted agents requires existing immune responses (Cancer Cell. 2010;18:485-98), so inducing or expanding an immune response first may produce a better overall effect from the targeted agent. Clinically, evidence was presented that vemurafenib induces T-cell infiltrates in the tumor.