The Next Level in Tumor Diagnostics

Fraunhofer researchers have developed a highly effective method for the early detection of pancreatic cancer. This approach could bring huge advances to precision medicine.

Kai Sohn
© Fraunhofer Magazin / Simon Koy
What’s next? Biologist Kai Sohn at Fraunhofer IGB is banking on precision diagnostics as a game changer in medicine.

Feathers, rapeseed, lignin: In the “Circular” display in the foyer of the Fraunhofer Institute for Interfacial Engineering and Biotechnology IGB in Stuttgart, various materials hang from the ceiling, fixed between glass panels. But in area, only four narrow rows of black lines and dots can be seen. “A reminder from my doctoral thesis,” explains Kai Sohn, who holds a doctorate in biology – fragments of human DNA, genetic information made visible. Two decades ago, sequencing just one human genome cost 100 million dollars. Now, the latest technologies enable it to be done for as little as 100 dollars. And science has since grasped what options a deep dive into the genome actually holds: This information can save lives!

Tumors that are based on changes to not just one chromosome, but entire chromosome complexes, pose a special diagnostic challenge. Pancreatic cancer is one such ultimate adversary: With a survival rate of less than ten percent within the first five years after diagnosis, pancreatic cancer is considered the most dangerous cancer. This type of tumor has an especially high rate of growth and undergoes early metastasis to neighboring organs in the abdomen.

However, the mortality rate is also so high because pancreatic tumors grow unnoticed: symptoms such as abdominal pain, weight loss or digestive problems only occur in the later stages. “There are currently no screening tests or instruments enabling the early diagnosis of pancreatic cancer,” says Kai Sohn, summarizing the situation. “So far, it can only be diagnosed when it is too late. And this is precisely why we chose this disease for our research.”

In collaboration with Georg Weber from Erlangen University Hospital and Genedata, a Fraunhofer research team led by Kai Sohn has developed an innovative method that can detect pancreatic cancer at an extremely early stage – without any biopsy. All that is required is a blood sample.

Cleverly wrapped – but with a predetermined break point

In his office, Kai Sohn places a white plastic model on the table, representing a DNA double helix that is carefully wrapped around a sponge-like internal structure. “In our cells,” the researcher explains, “the DNA is compressed as best as possible in a packaged unit like this – the nucleosome.” “The core of the nucleosome is made up of histone proteins around which the DNA strand is wrapped. This results in the most compact possible structure.” The individual nucleosome units are in turn connected to each other by so-called linker DNA. These are the predetermined break points of the complex.

Nucleosome complexes can break apart at precisely these weak points. This can occur in natural cell decay, for example, but also in the development of tumors or through inflammatory responses in the body. Fragments like this, also known as cell-free DNA (cfDNA), are present in the blood of every human being. “We search for these DNA fragments in the plasma,” explains Sohn. Or more precisely, they search to determine whether methyl groups have attached themselves to the DNA fragment. This is because tumor-derived DNA has different biochemical signatures, distinguishing it from healthy DNA. And there’s more. “Depending on where the methylation took place, we can tell whether this nucleosome originates from an immune cell, the intestine, the liver or the pancreas.”

An angular device the size of a desktop printer sits in the basement of the Fraunhofer IGB, with a soft pink luminescent band around its midsection and on which someone has glued two cheerful googly eyes. This is where the detailed analysis is done. The Illumina NextSeq2000 performs high-throughput sequencing, also known as next-generation sequencing (NGS). This technology now enables researchers to analyze genomes at high speed, says Sohn. “We can currently sequence 30 million DNA fragments overnight, meaning we can determine the exact sequence of nucleotides and search for methylation patterns.”

Not only do the biomarkers discovered by researchers in the In-vitro Diagnostics department at Fraunhofer IGB make it possible to distinguish between healthy patients and patients with tumors, but also to differentiate very precisely between pancreatitis and a malignant pancreatic tumor, for example – two diseases with nearly identical symptoms. Using AI-based algorithms, the researchers were even able to show in some example cases that their diagnostic procedure can also classify preliminary stages of pancreatic cancer. Kai Sohn is optimistic: “I am convinced that this technology is a true game-changer in the fight against pancreatic cancer.”

The next step is to conduct further clinical tests, followed by technology transfer to routine hospital use. Sohn underscores that Fraunhofer has a special advantage here: “By working closely with future producers and users, we develop technologies that account for user needs right from the start.”

Genome analysis as the key to personalized medicine

NGS is also used at Fraunhofer IGB to identify new biomarkers for diagnosing other cancers as well as certain infectious diseases. For example, the DISQVER diagnostics platform developed at Fraunhofer IGB – now operated by Fraunhofer spin-off Noscendo – can identify more than 16,000 microbes (bacteria, DNA viruses, fungi and parasites) within 24 hours and can thus provide the fastest and best possible support for treatment decisions. Here too, all that is required is a blood sample from the patient.

According to Sohn, the rapid, effective and inexpensive analysis of genomes using biochemical and bioinformatic methods such as NGS are the key to patient-specific medicine – with the potential to herald a paradigm shift from diagnostic to preventive medicine. Development is still in its infancy, but the progress achieved over just a few years has already been enormous. “We are gradually beginning to understand the wealth of information available from fragmentomics, the analysis of DNA fragments,” the researcher explains. “My dream is that one day we will be able to detect the diseases of tomorrow and beyond at an early stage, simply by analyzing tiny DNA fragments.”