Protein biomarkers: Here's a view from Sweden

As we report each week in FierceBiomarkers, there are many biomarkers being tagged as promising, but these need mechanisms to validate them. Ulf Landegren, professor of molecular medicine at Uppsala University and vice chair of the Science for Life Laboratory (SciLifeLab) spoke with FierceBiomarkers at the BBMRI.se HandsOn: Biobanks meeting in Uppsala, Sweden, giving a taste of his views and his work in protein biomarkers. Here is a snapshot of the conversation.

Landegren and his group focus on developing advanced molecular tools to identify, measure and analyze proteins and nucleic acids, including for use as biomarkers in research and medical diagnostics, particularly in neurological disease. The techniques include oligonucleotide ligation assays, padlock probing, and proximity ligation, and these have also been licensed to international biopharma companies, or used by startups spun out from the institution.

The conversation started in a rather startling way, with Landegren saying: "A lot of the current work in proteomics screening is doomed!"

He expanded on this further: "There is a wide variety of protein biomarkers that could be useful, including conjugates, splice variants and post-translationally modified proteins, as well as proteins in exosomes released by cells. However, these are generally in low concentrations. If current technologies and methods are not sufficiently sensitive to detect proteins that are already used as clinical markers, then the chance of using these same techniques to uncover new markers would seem to be limited."

Despite this pronouncement, Landegren is actually very positive about the future of proteomic biomarkers, believing that it is a technical problem that needs to be (and can be) solved. As he explains, work of this type needs a good biophysical understanding, as well as technologies that can handle wide concentration gradients. He sees his group's research, particularly in proximity ligation assays, as well as Sweden's sound background in biobanking, as a good start, though there is perhaps still a way to go.

"We need the right technology as well as the right samples. However, in biobanking, people have been influenced by what is needed by the existing technology--for example, researchers and biobankers started by simply preserving plasma, and then focused on DNA. However, to be able to develop protein predictive markers we will need a combination of clinical data with a series of consecutive samples, so that individuals can act as their own control." Landegren concludes: "Researchers need to start thinking about what samples will be needed in the future, rather than what is simply needed now." -- Suzanne Elvidge (email)