Can a set of blood proteins predict colon, breast, and other cancers?

Original source (on modern site) | Article images: [1] [2]

Share on PinterestResearchers have found a link between over 1,000 blood proteins and 19 different forms of cancer. Image credit: Yoshiyoshi Hirokawa/Getty Images.

• A pair of new studies investigate possible associations between thousands of blood plasma proteins and a range of cancers.
• A significant number of proteins appeared linked to cancers that were not diagnosed for 7 years after blood samples were taken, raising the possibility of a cancer early warning system.
• With so many proteins, so many cancers, and the complex pathogenesis of those cancers, the study represents a first step toward a greater understanding of plasma proteins and cancers.

Two new studies from Oxford Population Health at the University of Oxford in the United Kingdom have investigated associations between thousands of blood plasma proteins and an array of cancers.

The first study — which appeared in Nature Communications on May 15 — found 371 plasma protein markers of cancer risk. Of these, 107 were associated with cancers that were not formally diagnosed until 7 years later, suggesting a potential opportunity for early detection and treatment.

The second study — published in Nature Communications a little earlier, on April 29 — observed associations between 40 plasma proteins and various commonly occurring cancers.

The authors' hope is that this may help lead the way to detecting and treating cancers early in their development, and perhaps preventing them from occurring at all.

With data from the U.K. Biobank, statistical links between 1,463 plasma proteins and 19 types of cancer in 503,317 adults aged 39 to 73 were the target of the first study. The second sought associations between 2,047 proteins and nine cancer types in 300,000 people from the U.K. Biobank.

The researchers also explored the possible reasons that proteins were not associated with cancer.

The researchers used a discipline called proteomics, which is the study of proteins, wherever they are in the body — in this case, in blood plasma. Proteomics involves physical and biochemistry, computer science, genetics, and bioinformatics.

Proteins are ubiquitous in our bodies, in blood serum, in muscle, skin, bone, hair, urine, and elsewhere. We each carry at least 10,000 different proteins.

The study represents an early step in understanding the relationship between plasma proteins and cancer. Definitively determining specific plasma protein levels that may signify, or reflect, cancer is beyond the scope of the current research.

The first study found potential links between plasma proteins and increased risk of cancers of the liver, digestive and gastrointestinal tracts, and non-Hodgkin lymphoma, as well as colorectal, lung, kidney, brain, stomach, esophagus, endometrium, and blood cancers.

The second study observed ties to triple-negative breast cancer, bladder cancer, lung cancer, and pancreatic cancer.

"Some of the other links are quite interesting, too," said one of the studies' co-authors, Joshua Atkins, PhD, BBmedSci, a senior genomic epidemiologist at the University of Oxford.

"Proteins that are not causal for cancer development but are a consequence of cancer growth can provide avenues for detecting cancers at an earlier stage when treatment can be more successful," noted Atkins.

Richard Reitherman, MD, medical director of Breast Imaging at the MemorialCare Breast Center at Orange Coast Medical Center, CA, who was not involved in the studies, explained to Medical News Today that:

"These publications demonstrate the association — not cause or existence — of specific proteins and their correlation with known common cancers. This level of basic research is designed to further understand how specific proteins are related to human cancers."

Nevertheless, "[d]isruption of these processes can result in diseases," Atkins said, "including cancer. For some proteins, higher blood levels are linked to higher cancer risk, while others may be protective, so higher levels are linked to lower risk."

Atkins also noted that his team is working now to understand the levels of proteins that should be of concern. This may take some time.

David S. B. Hoon, PhD, professor and director of the Genomic Sequencing Center at St. John's Cancer Institute, CA — who was also not involved in the studies — pointed out, by way of example, that determining healthy levels of lipoprotein cholesterol "took thousands and thousands of tests, male, female, before finally the clinical chemistry made cutoffs of what is truly positive, what is dangerous, and what is basically your norm?"

Hoon also expressed his concern that the research may not represent adults of all ages. "In the paper," he said, "most of these patients are in their sixth, seventh, decade. So you are going to have a lot of proteins that are what we call high-risk for cancers because they just start occurring as you age."

"These are proteins that we all have," said Atkins.

Reitherman explained that "[o]ur blood system is the obligatory middleman, a complex subway system that connects all of the body's organs and functions with each other."

"Like a subway system, passengers — e.g., proteins, carbohydrates, lipids, DNA, RNA, intact cells and cellular sub-particles, oxygen, carbon dioxide, minerals — enter the blood system," said Reitherman. "They are transported to other locations in the body to be used in a myriad of complex processes of metabolism, respiration, cellular growth regulation, or on their way to be eliminated."

He cited insulin, which is secreted by the pancreas into the blood and subsequently "diffuses into every body tissue we have, ultimately regulating numerous cellular functions including our blood glucose levels."

The argument against aggressively manipulating levels of proteins of concern goes beyond the earliness of this research, said Atkins.

"Because proteins have important roles in many of the body's key processes, disrupting levels or the function of the proteins may well have adverse effects," he cautioned.

He noted that the protein FGFR3 is linked to an increase risk of bladder cancer, but that lowering its levels is tied to an increased osteoarthritis risk.

Atkins said:

"Most current drugs target proteins in some form or shape, and we already have good hazard profiles on approved drugs. For novel proteins that we may want to target, this new research using genetic methods enables us to predict what hazards or side effects might occur, and to prioritize targets to move forward with."