Brightening the Forecast for Skin Cancer

Despite many cases being preventable, melanoma remains one of the most common cancers. Patients with resected stage II or III lesions often develop distant metastases – but current methods cannot accurately predict who is most at risk.

Rebecca Lee, a research associate at the University of Manchester, is first author on research that investigates the predictive potential of circulating tumor DNA (1). We spoke with her to find out how genetic testing using ctDNA could improve patient care.

What main advantage does your approach offer?

Current methods use a combination of clinical and pathological features – mainly based on the American Joint Committee on Cancer (AJCC) staging – to predict recurrence. However, they are limited in their ability to accurately predict relapse for each individual patient. It is of great importance to identify those patients at highest risk of relapse, especially in the era of effective adjuvant therapies for melanoma. We believe that genetic testing using ctDNA detects minimal residual disease following surgery and, therefore, can be used in conjunction with AJCC staging to identify high-risk groups.

What could this test mean for skin cancer patient care?

We believe liquid biopsies can improve care in several key areas. The first is early detection of micro-metastatic relapse in patients who have had surgery performed with curative intent as part of secondary prevention strategies. The second is the monitoring of tumor burden; for example, in patients receiving treatment for stage IV disease, it could be used to detect early progression of treatment. Finally, it can be used to identify mutations associated with either a response to therapy (such as BRAF mutation/BRAF inhibitors) or a resistance to therapy.

We are currently developing clinical trials to test whether early treatment based on detection of ctDNA could improve outcomes for patients, so we would expect it to become available in the clinic in about five to eight years’ time.

How did you approach development?

As a proof of principle, we decided to examine BRAF and NRAS mutations, as these account for approximately 80 percent of driver mutations in melanoma. We could also look at other mutations, which would require extra steps – first, sequencing the tumor to identify the mutations, then analyzing the ctDNA in plasma. We decided to use droplet digital PCR (ddPCR) – a clinically relevant technique because it is relatively fast and inexpensive compared with next generation sequencing (NGS) approaches. In addition, bioinformatics support is not required, which makes it easier to translate into the clinic.

How sensitive is the test?

Sensitivity for predicting relapse was 18 percent and specificity was 95 percent; if ctDNA is detected, there is an extremely high chance that the patient will relapse. A lack of detection, however, does not necessarily mean that the patient will not relapse – although it becomes more likely after a longer period of time. We feel that longitudinal monitoring of patients will improve the ability of the test to pick up micro-metastatic relapse.

How would the test fit into the existing pipeline?

The majority of the processing is relatively straightforward and can be automated. The process is already used to detect certain mutations – for example, EGFR mutations in lung cancer – so it would be a question of building the capacity into the health system to perform more tests, and establishing pipelines whereby samples are sent to the laboratories in a timely manner, just like other blood tests. We see it as an additional tool to aid in the diagnosis and monitoring of the disease.