A study by scientists at New Zealand's Otago University has opened the door to the discovery of new biomarkers that potentially predict whether melanoma patients will respond to treatment with novel checkpoint inhibitor immunotherapies.

Published in the June 29, 2018, edition of iScience, the study provides insights into why new therapies for metastatic melanoma do not always work and why some patients who initially respond develop resistance, paving the way for predicting which patients will benefit.

The team found that in patients who were resistant to PD-1/PD-L1-targeting immunotherapies, the loss of epigenetic methylation tags led to permanent expression of the checkpoint molecule PD-L1 on tumor cells.

The findings will need to be validated in human trials undergoing treatment. But the group's findings suggest epigenetic therapies could be used in clinical trials in combination with immunotherapy in melanoma patients.

Melanoma is particularly prevalent in Australia and New Zealand, which have the highest rates of the disease worldwide. "In 2012, Australia and New Zealand had almost twice the rate of melanoma than the next highest country, Switzerland, with approximately 34-36 versus 18-19 per 100,000 population, respectively," study leader Mike Eccles told BioWorld Asia.

"The main reasons for this are the intense ultraviolet [UV] light levels in those countries, combined with the relatively high proportion of fair-skinned individuals among the population," said Eccles, a professor in the Department of Pathology at Otago University.

Non-immune treatments have limited efficacy. For example, "combination treatment with dabrafenib (Tafinlar) and trametinib (Mekinist) in patients with V600E BRAF mutations, which comprise 40-50 percent of cases, has a median overall survival of approximately 25-27 months," he noted.

The best conventional non-immune treatments are less effective than the new immune checkpoint inhibitor therapies, including Opdivo (nivolumab, Bristol-Myers Squibb Co.), Merck & Co Inc's Keytruda (pembrolizumab) and Yervoy (ipilimumab, Bristol-Myers Squibb Co.).

In two separate studies, for example, "patients who received Opdivo treatment alone had a median overall survival of approximately 37 months, and in melanoma patients with Opdivo plus Yervoy median overall survival has not yet been reached, i.e. it is longer than 37 months," Eccles said.

The findings of the new study cast new light on why immune checkpoint inhibitor therapies are ineffective against metastatic melanoma in many patients, while those who do respond often develop resistance.

"Analysis of clinical trial data identifies three broad patient populations: those who respond initially and continue to respond (responders), those that fail to ever respond (innate resistance), and initial responders but develop disease progression (acquired resistance)," said Eccles.

He noted that the relative percentages of those groups remain uncertain. Nevertheless, in the phase III Checkmate 067 trial of nivolumab or nivolumab plus ipilimumab, "the immunotherapy combination group had a two-year overall survival rate of 64 percent."

One of the key components of the immune checkpoint mechanism is a protein on the surface of cancer cells, PD-L1, which can potentially either be receptive to or block immunotherapy.

"In immune cells such as T cells, the normal role of PD-L1 is to suppress the immune response, with suppression being triggered when PD-L1's binding partner, PD-1, and PD-L1 interact," explained Eccles.

"Cancer cells expressing PD-L1 can therefore induce immune suppression, but PD-L1 is not a robust biomarker of immunotherapy response," he said.

"However, in the melanoma genomes, apart from PD-L1 itself, we showed that an epigenetic modification, specifically DNA methylation, influenced whether PD-L1 was expressed on the cancer cell surface."

Epigenetic modifications do not directly alter DNA sequence, but they change the frequency by which a cell uses specific genes, thereby playing a key role in switching genes on or off and helping determine cell function. Generally, DNA methylation silences gene expression and has been implicated in cancer.

"Our research provides evidence that it is the global loss of DNA methylation that regulates constitutive expression of the immune checkpoint PD-L1 in melanoma," study co-author Aniruddha Chatterjee, a senior research fellow in Eccles' department, told BioWorld Asia.

Currently, there are no reliable biomarkers for predicting benefit from immune therapy in melanoma. Biomarkers would help to identify patients likely to benefit from immunotherapy, so this is an area of active research worldwide.

Further trials would be needed before this could become a possibility. "We need to do a lot more work around testing and identifying suitable epigenetic biomarkers for possible future use in the clinic," noted Eccles.

His group recently received major government funding and plans to develop a DNA methylation marker panel that predicts the likelihood of melanoma patients responding to immunotherapy.

"Our present research indicates that using this new funding could well be fruitful and it's worthwhile exploring this avenue. But it's likely to be at least three to five years before the results could be used clinically," noted Eccles.

"This work will contribute to selecting the best treatment option for patients, and also for developing new targets for epigenetic therapies," he said.

Looking ahead, "there is evidence that epigenetic therapy such as the DNA methylation inhibitor decitabine (Dacogen, Johnson & Johnson) plus immunotherapy could lead to more favorable outcomes for treating melanoma, warranting further clinical trials," Eccles said.

"Our research suggests tumor cell epigenetics impact the response to immunotherapy. Moreover, tumor cell epigenetic alterations are reversible, so theoretically can be readily modified. Knowing the precise role of tumor cell epigenetics in this could open up an array of new predictive tools for treatment response, so we will next try to identify suitable epigenetic biomarkers for possible clinical use."