HLA genes encode for cell surface molecules which can help inform the immune system whether it should attack. Mutations in cancer cells cause HLA molecules to present modified surface proteins, called neoantigens, which can identify the cell as non-self. This raises the alarm to the immune system.

If an HLA gene is mutated or disrupted, neoantigens may not be presented on the cell surface, providing the cancer cell with an invisibility cloak, so it can avoid being detected by the immune system. This is a problem for cancer treatments that involve boosting the immune system. 

In research published today in Nature Genetics, the research team developed a tool called MHC Hammer and used it to evaluate HLA molecules in both healthy tissue and lung and breast cancer samples.

Each HLA gene is in a pair with two copies, one which is inherited maternally and one paternally. Using MHC Hammer, the team evaluated four potential types of HLA disruption:

• Losing one copy in the pair;
• Mutations in one copy in the pair;
• Repression of one copy in the pair, and
• Cutting (splicing) of HLA molecules.

They found a high degree of variability in these events in tissues from a bank of healthy volunteers, which means that some people naturally have more HLA molecules on their cells.

But all four types of HLA disruption were prevalent in the lung and breast cancer samples, often resulting in fewer neoantigens being present on the cell surface. For example, it was more likely for HLA genes to be cut in the wrong places in the tumour cells than in healthy cells from the same person.

The researchers also reported that losing one copy of an HLA gene was associated with cancer spread, also called metastasis. This may be because the cancer cells that ‘seeded’ the metastasis had fewer neoantigens so could travel around the body undetected.

The team believe epigenetics – changes to DNA that don’t involve the genes themselves – may play a role in fewer HLA molecules being present, as they saw an increase in an epigenetic change called methylation in HLA in the cancer tissues.

Charles Swanton, Principal Group Leader of the Crick's Cancer Evolution and Genome Instability lab, medical oncologist at University College London Hospitals, Chair in Personalised Cancer Medicine at the UCL Cancer Institute, and Chief Investigator of the CRUK TRACERx study, said: 

Drugs to stimulate the immune system are often used in lung cancer and breast cancer, but there’s a large group of patients who don’t respond. Our research suggests that disruption to HLA genes is involved in hiding tumours from the immune system. Reversing the epigenetic changes to HLA genes in people with high levels of disruption could reveal the tumours to the immune system and ensure a better response to treatment.

Clare Puttick, first author and postdoc at the Crick, said: 

We’ve shown that there are at least four different ways that the HLA molecules can be disrupted in lung and breast cancer. We think that this disruption may play a role in tumour metastasis and could be important to account for when evaluating what immune therapies may work for patients.

The three patient datasets used in the research were the Cancer Research UK-funded TRACERx study, the Cancer Genome Atlas (TCGA) and the Genotype-Tissue Expression (GTEx) dataset. This study was also funded by the CRUK City of London Centre Award.