Researchers shed light on how to predict which skin cancer patients are most likely to respond to immunotherapy

An international team of UK and US scientists has discovered that the activity of macrophages – a type of white blood cell that engulf pathogens and cancer cells – can be used to predict whether or not a melanoma patient will respond to immunotherapy. Their findings, published in a landmark paper in JCO Oncology Advances, will help clinicians select treatments that are most likely to be effective for their patients.

Immunotherapy is a powerful treatment for a range of cancers including skin and kidney cancer, but unfortunately only around half of patients respond to this type of treatment.

Therefore, selecting the best treatment is often a trial-and-error process, leaving patients who don’t respond suffering side effects whilst their cancer remains untouched and potentially their condition worsens.

Now researchers, from the Universities of Bath (UK) and Stanford University (CA, USA), have examined novel biomarkers – indicators of the body’s immune system – that may identify melanoma patients more likely to respond to an immunotherapy treatment called TVEC.

TVEC is a modified oncolytic virus that is injected into melanoma directly to stimulate an immune response. It previously has been used in advanced melanoma, however this study was the first to examine its potential to treat high-risk stage II melanoma patients upfront.

It was conventionally thought that TVEC worked by activating T cells – a type of white blood cell – causing them to attack the cancer cells and shrink the melanoma.

However, the team found that pre-existing and post-treatment T cell populations did not have association with treatment responses. Instead, they found that changes in the macrophages   correlated to which patients responded to the treatment and which did not.

Additionally, previous research has monitored amounts of protein indicators such as PD-L1 and at the genes involved in T cells to assess whether immunotherapy is effective.

However, this latest study shows these techniques do not accurately predict which patients will respond to treatment.

Measuring activation, not just amount

In their study, the researchers used a method called iFRET, which monitors protein activation instead of simply measuring the amounts of protein present.

They found that T cell presence showed no consistent trends to viral stimulation or tumour response before and after the treatment, but there was a heavy infiltration of macrophages after treatment in responding patients, associated with very high activation across immune checkpoint regulators – proteins that help regulate the immune system, so that immune system does not attack healthy cells.

The researchers will use the findings to develop clinically predictive tests of which patients will respond to the therapy and enable clinicians to tailor a personalised treatment, saving time and reducing side effects for the patient, as well as reducing the use of costly treatments that do not work.

Professor Banafshé Larijani, Department of Life Sciences and Director of the Centre for Therapeutic Innovation at the University of Bath, co-led the study. She said: “We know that people respond to immunotherapy very differently – in some cases the tumours shrink, and in others, sadly the patients do not survive.

“Our findings show that it’s not enough to simply look at T cell activity, instead it’s imperative to look at the whole immune response environment in detail to predict how a patient will respond to different treatments.

“Our results suggest that, in non-responding patients, we should be targeting these macrophages to reprogramme the tumour immune environment.

“We hope our research will enable clinicians to make important decisions over which patients would be better served by surgery or immune checkpoint blockade by immunotherapy.”

Dr Amanda Kirane, Director of Cutaneous Surgical Oncology Department at Stanford University School of Medicine, who led the clinical part of the study, said:

“This study is highly informative in establishing a connection between pre-existing innate immune functions and ability to respond to immune-stimulating drugs.

“It also strongly supports emerging evidence that there may be biological differences in patients more likely to respond to this kind of immunotherapy - oncolytic viruses - versus other types that target immune checkpoint regulators. 

“Lastly, it extends new and important context to the disconnect between measuring PD-L1 protein values as a clinical biomarker and protein activity in the tumour.

“The added information of iFRET-based immune activity measurements may offer the critical missing link of why current biomarkers have failed to yield a usable test to aid patients in treatment decision-making.”

Next, the team aims to characterise all cells contributing to the immune checkpoint interaction, which will further improve patient stratification and therefore tailoring of personalised medicine.