New method finds cancer's attack points

Expectations of new forms of cancer treatment, which enable the immune system to attack the cancer directly, have grown over the last 10-20 years. This has happened as the treatments—grouped under the label 'immunotherapy'—have demonstrated that they can completely cure otherwise terminal patients of their illness. The number of different cancers which can be treated using immunotherapy has also increased dramatically, and now includes melanoma, lung cancer, kidney cancer and bone marrow cancer, with many more on the way.

But the treatments do not work for everyone. Far from it, unfortunately. This is because cancer is not just cancer, and an immune system is not just an immune system. All people are different, and so are their cancer cells. It therefore requires both luck and skill to find the right attack points at present. For example, the proportion of melanoma patients completely cured using immunotherapy is around 5-7 per cent in Denmark.

Cancer is personal

For immunotherapy to work, you have to know certain peptides ('peptide MHC complexes', comprising small chains of amino acids) on the surface of cancer cells, which the immune system's T-cells (a type of white blood cells) can recognise. Although T-cells should kill cancer cells automatically, as they respond when we are attacked by viruses etc., they are deactivated in the body by the cancer cells' defense mechanisms. But in combination with medicine which inhibits the cancer cells' defense mechanisms, it is possible to activate the T-cells so they attack the cancer cells.

The problem is that we are not very good at finding attack points on the surface of the cancer cells. With the methods being used today, it is possible to find somewhere between 10 and 100 in a single test, but that is not nearly enough. Over 1000 must be found to have an effective screening of whether a patient's specific cancer can be attacked using immunotherapy. And this is a major challenge.

"The attack points are not the same for different people. There are several thousand of these (MHC) molecules that we have genetically encoded in our genome and which are responsible for showing our immune system what is happening inside a cancer cell or the like. We each have six different MHC molecules—which vary from person to person. But some are more common than others. What we can work with today is typically the 30 most common ones. This allows us to usually look at a few of the molecules the patient might have. Each of these molecules carries protein fragments, which enable the immune system to read what is happening inside the cell. Our challenge is to find out which of the thousands of protein fragments the immune system can recognise and respond to," says Reker Hadrup from DTU Vet, who was responsible for the new study recently published in the Nature Biotechnology journal.

"In relation to cancer, the key is to work out whether the immune system can recognise something in the cancer cells which is not present in healthy cells. The mutations that have accompanied the cancer cell's development are an obvious attack point. The other problem with them is that they are also personal. So we actually have to sequence tumor DNA from the given patient and make a library of peptides. The problem has been that there was no screening tool with a matching level of complexity."

Effective screening using new method

Sine Reker Hadrup from DTU Vet has therefore developed a new method to this end. Instead of using colour codes to distinguish between the various peptides, it uses small DNA sequences which serve as unique bar codes. This makes it possible to measure immune cells' recognition of up to 1000 peptides in a single test, vastly increasing the level of complexity of the analysis of what the immune system recognises from the current level," explains Sine Reker Hadrup:

"With the help of DNA tagging, we are able to screen for this large library. There might be 800-1000 peptides for each patient. We can even screen for these in a single sample using this technology—either in blood or a biopsy from the patient's tumor," she says, noting that without the technology it would require large quantities of blood to analyse the entire library of peptides. With biopsies the task is impossible, as there are not enough cells.

"People have mostly searched for the most likely attack points from a theoretical perspective. We are now better able to perform 'genome-wide' analyses, and look at what has changed in the cancer cells' genome compared to the normal genome for a healthy cell. This means that we don't have to make a lot of predictions, but can screen very broadly."

Put briefly, instead of conducting immunotherapy treatment partially blindfolded, the new method means that many more of the key attack points can be found. Sine Reker Hadrup explains that they conducted experiments using samples from lung cancer patients previously screened using the 'old' method. The new method clearly showed more responses than before. There were more attack points that had not been discovered for technical reasons.

Sine Reker Hadrup hopes that the technology will provide biomarkers which can tell doctors what is required for an immunotherapy treatment to work, so patients can be identified beforehand who are likely to benefit from the treatment. And conversely, to identify those who will not benefit and who should be offered another treatment. Finally, the more effective screening also allows patients to be monitored during the treatment, to see whether things are moving in the right direction or if the treatment should be adjusted.

"I hope and believe that this technology can help move immunotherapy in a direction where we become better at selecting patients for various forms of treatment based on their personal characteristics. And where our treatment is more closely tuned to the given patient's cancer cells, leading to a far more targeted treatment. We can already see the effect of immunotherapy on all major cancers, even in the more unrefined form we currently have."