Immunotherapy Advances in Melanoma

Yvonne Saenger, MD

Assistant Professor, Medicine, Hematology and Medical Oncology

Assistant Professor, Dermatology

Icahn School of Medicine

Melanoma care has fortunately undergone a whirlwind of changes over the past several years. Novel immunotherapies are perhaps the most exciting recent development in cancer care because patients can enjoy long-term benefit from these treatments, meaning that they are possibly “cured.” Thus, it is now considered possible to rid the body of cancer using immunotherapy in a significant fraction of patients with advanced melanoma. Immunotherapy can be very toxic, but it lacks the characteristic poisoning effects of chemotherapy and has the appeal of offering a more “natural” solution to cancer as it builds on the body’s own innate defense mechanisms to cause tumor rejection. Over the coming decades, the challenge will be to expand the minority currently curable with immunotherapy and also to introduce immunotherapies earlier in the disease course to prevent systemic spread in the first place.

In 1981, interleukin-2 became the first immunotherapy approved for treating patients with melanoma.

The drug reliably produces an approximate 15% response rate and a “cure” rate perhaps slightly below 5%. However, the mechanisms of action of interleukin-2 are largely unknown and skeptics of immunotherapy have remained unimpressed by the data for the drug. The key development that led to the real expansion of immunotherapy for melanoma was the observation that blockade of a negative regulator of T-cell activity, CTLA-4, could result in melanoma regression.

CTLA-4

CTLA-4 is a protein expressed by T cells in response to activation, and it serves to dampen the immune response, thereby preventing autoimmunity. Thus, mice deficient in CTLA-4 develop widespread autoimmune disease. In this way, CTLA-4 serves to protect normal tissues against the immune system; however, it is sometimes used by cancer cells to hide from the immune system. By blocking CTLA-4, the tumor can be revealed as a foreign threat.

While the initial data for CTLA-4 blockade showed efficacy not far superior to interleukin-2, follow-up studies have shown impressive rates of survival following CTLA-4 therapy, with 4-year survival rates of approximately 40% in previously untreated patients treated with high-dose CTLA-4 blockade on clinical studies. These data suggest that, while most patients do not experience immediate tumor shrinkage, tumors may respond later in the treatment course, resulting in improved survival. The CTLA- 4—blocking antibody ipilimumab (Yervoy) was approved for the treatment of melanoma in March 2011.

PD-1/PD-L1

The success of CTLA-4 blockade spurred further research, resulting in the discovery of a second clinical target, PD-1. PD-1, similar to CTLA-4, is expressed by activated T cells, and it appears to be induced by chronic inflammation. Similar to CTLA-4, PD-1 signaling leads to inactivity and sluggishness of T cells. Thus, CTLA-4 and PD-1 are described as “checkpoints” in immune activation. One way to describe the role of PD-1 is that the immune system might decide to tone down a prolonged immune response, as it appears unlikely that the foreign entity will be successfully eliminated and the attack becomes a waste of resources. Thus, PD-1 expression is high on T cells during chronic viral infections and in T cells infiltrating tumors.

Blockade of PD-1 has produced impressive responses in melanoma patients. In a phase I study, the PD-1 inhibitor pembrolizumab (MK-3475) had an overall response rate of 34%, and preliminary evidence suggests that these responses are durable.¹ In another phase I study, combination therapy with the PD-1 inhibitor nivolumab and ipilimumab produced responses in 40% of patients.2 The latest data on pembrolizumab and nivolumab were presented in June at the 2014 ASCO Annual Meeting, and both drugs are expected to eventually be approved by the FDA.

PD-L1, a protein that binds and activates PD-1, is found on many human tumors, including melanoma. By expressing PD-L1, the tumor can suppress T cells that might enter into the tumor and attempt to attack it. Several antibodies against PD-L1 have shown promising results in clinical trials, with response rates above 20%. Thus, interrupting the PD-1/ PD-L1 interaction appears to be more effective than CTLA-4 blockade, and will likely be the cornerstone of future therapy for melanoma.

T-VEC

The immune checkpoint inhibitor approach has yielded amazing results in melanoma, but other strategies are also under active investigation and nearing fruition. For example, oncolytic viruses are receiving increased interest. Talimogene laherparepvec (T-VEC) is the most advanced in clinical trials.

T-VEC is a modified herpes virus that selectively infects tumor cells and is administered by direct injection into the tumor. This therapy is designed to produce an effect called “auto-vaccination,” whereby the tumor becomes its own vaccine. Infected tumor cells are lysed by the virus and in the process, release inflammatory signals that are immune activating and can cause the immune system to attack other tumor deposits that were not inoculated with the virus. Oncolytic viruses are attractive therapeutic options because they are derived from genuine pathogens and therefore are more potent mediators of immune activation than conventional vaccines.

A phase III study of T-VEC presented at the 2014 ASCO Annual Meeting met its primary endpoint of response rate (16% vs 2% in the control arm) and achieved a tantalizing near advantage in terms of overall survival (P = .051).3 It is highly likely that intralesional therapies such as T-VEC will be part of future melanoma therapy and may enhance the activity of immune checkpoint blockade.

Biomarkers

One challenge in immunotherapy is to develop accurate techniques to measure immune activation. In early stage melanomas, it is known that the immune system can infiltrate into the tumors and control their growth. Recently, our group has proposed a biomarker to predict recurrence-free survival and overall survival in patients with stage II/III melanoma at high risk of recurrence. This biomarker is based on the expression of immune genes and patients with higher expression of immune genes have better clinical outcomes. Similarly, we have found that survival in patients treated with CTLA-4 blockade can be predicted based on patterns of gene expression in blood. Genomic biomarkers based on analysis of blood and tumor tissue are likely to play a role in the future for prognostication and for selection of immunotherapy options to be used in specific patients.

In summary, tremendous progress has been made in immunotherapy for melanoma over the past 5 years. A diagnosis of metastatic melanoma is no longer the death sentence it once was. The future of melanoma therapy is likely to include combinations of different immune therapies, including immune checkpoint blockade and intralesional treatments, such as T-VEC.

These immune therapies may also in the future be combined with targeted treatments, such as BRAF inhibitors, and/or conventional therapies designed to break down the tumor mass, such as irradiation. Further new targets for immune activation, including LAG-3, CD40, GITR, and OX-40, are on the horizon as part of combination strategies.

References

1. Ribas A, Hodi FS, Kefford R, et al. Efficacy and safety of the anti-PD-1 monocolonal antibody MK-3475 in 411 patients with melanoma. J Clin Oncol. 2014;32:5s (suppl; abstr LBA9000).
2. Wolchok JD, Kluger HM, Callahan MK, et al. Safety and clinical activity of nivolumab (anti-PD-1, BMS-936558, ONO-4538) in combination with ipilimumab in patients (pts) with advanced melanoma (MEL). J Clin Oncol. 2013;31 (suppl; abstr 9012).
3. Kaufman HL, Andtbacka RH, Collichio FA, et al. Primary overall survival (OS) from OPTiM, a randomized phase III trial of talimogene laherparepvec (T-VEC) versus subcutaneous (SC) granulocyte-macrophage colony-stimulating factor (GM-CSF) for the treatment (tx) of unresected stage IIIB/C and IV melanoma. J Clin Oncol. 2014;32:5s (suppl; abstr 9008a).