Biomarkers
and a battle of the bases

We’ve come a long way from the early days of using medicinal leeches and mercury treatments as therapeutics. As described in the book Soonish, the path to precision medicine has been a long one, often lacking scientific rigour and sometimes, common sense.

Thankfully, on the diagnostics front, we’ve advanced further than posthumous autopsies and microscopic pathology developed in the nineteenth century. Today, molecular diagnostics have become commonplace in medicine. Looking into the future, how will the innovations of today shape the face of medicine ten or twenty years from now? We may see the biggest change occur in genomics, with a battle of the bases determining which biomolecule will reign supreme for biomarker discovery.

In one corner is DNA, or deoxyribonucleic acid, in the other corner is RNA, or ribonucleic acid. These two contenders are known for their respective and important roles within the central dogma of biology. And, while both are composed primarily of a sugar and base combination, major differences exist. First, the base pairs differ slightly since DNA uses the bases adenine, thymine, cytosine and guanine while RNA uses adenine, uracil, cytosine and guanine. Second, DNA most often exists as a double-stranded molecule, while RNA is usually found in its single-stranded form. RNA’s ‘bonus’ hydroxyl group results in decreased chemical stability relative to DNA. This, coupled with the pervading presence of RNases, results in a much shorter lifetime for RNA, and is a third defining difference between these two biomolecules.

Finally, while most organisms have only two copies of DNA per cell, even low abundance RNA transcripts often have hundreds of copies present per cell (or organism).

So, let’s start off with a warm-up match: infectious diseases.
RNA’s ubiquitous nature affords an advantage here. Diagnostics which are developed to detect RNA species for infectious organisms could require less patient material or detect the infection earlier due to a stronger signal compared to their DNA counterparts. Further, as described by John Brunstein, if a transcript or set of transcripts is selected correctly, an RNA diagnostic could also provide a measurement of viability of the infectious agent. Looks like RNA has taken an early lead.

One more bout to get warmed up: autoimmune disease.
Since the completion of the Human Genome Project and with the introduction of high-throughput sequencing, we’ve been able to collect massive amounts of DNA data from both healthy and diseased patients. These data sets have enabled us to determine genetic changes implicated in many diseases, including autoimmune diseases like rheumatoid arthritis (RA) and multiple sclerosis (MS). However, one challenge that remains is that DNA only reports on the risk or potential for developing one of these diseases. RNA, however, gets us one step closer to the downstream biology, and can greatly help disease diagnosis. By analysing RNA signals, it’s been reported that earlier, more accurate diagnosis can be achieved.

This warm-up round goes to RNA, as well.
And now, for a championship match of great significance: Cancer.

This disease, or rather an assemblage of diseases, still represents the second most prevalent cause of death in the world. Despite immense improvements in understanding cancer biology and using this to develop new and impactful treatments, including targeted therapies, immunotherapies, cell therapies and cancer vaccines, we still see only a small percentage of patients benefiting with robust responses. Cancer, with its combination of hereditary, environmental and sometimes random origins, has required diagnostic approaches covering many fronts.

Round 1: genetic mutations –
For hereditary cancers, there are a number of mutations that have been determined to play a major role in the likelihood of developing the disease or the path of disease progression.

In breast cancer, for example, these include BRCA1 and BRCA2, along with the less commonly discussed PTEN and p53 mutations.

However, for many patients, a genetic analysis delivers a number of mutations which are listed as ‘Variants of Unknown Significance’. These mutations are known to be different from a ‘normal’ genome, but there is not sufficient clinical evidence to use the information to make medical decisions. However, when RNA data is layered on top of this genetic data, additional context may be used to determine whether these variants should be classified as pathogenic or benign. As described above, while we’re still looking at risk factors in this case, the additional clarity provided by RNA can make a big impact in determining whether a patient receives preventative surgery or more rigorous surveillance. Round 1 goes to RNA.

Round 2: gene fusions –
Understanding the mechanisms of cancer that impact progression and pathogenesis are imperative to improving treatment plans.

Further, specific molecular events have also informed therapy selection, more commonly known as targeted therapies. One excellent example is in non-small cell lung cancer, where the EML4-ALK gene fusion, arising from chromosomal rearrangement, has become both a diagnostic marker and a target for ALK inhibitor therapies. In the diagnostic arena, fluorescence in situ hybridisation (FISH) of the two DNA sequences has been used to detect the presence of this gene fusion. However, there is significant genomic heterogeneity in these ALK rearrangements undetectable by FISH, which has been shown to be related to differences in kinase inhibitor treatment response rates. Use of RNA sequencing technologies, which look not only for expression of these fusions but also enable discovery and quantification of novel fusions, has enabled the field to move beyond the limited information provided by DNA diagnostics. RNA wins another point.

Round 3: immune response –
Undeniably, the most promising path to understanding and treating cancer is immuno-oncology.

This approach, which harnesses the body’s immune system to fight the disease, has garnered much attention, resulting in the 2018 Nobel Prize in Medicine. While these therapies show enhanced and lasting responses, a number of challenges remain. DNA technologies, including tumour mutational burden (TMB), showed promise in predicting response to immune checkpoint inhibitors; however, the results in the clinic were disappointing. Understanding the dynamic immune composition at the site of the solid tumour is imperative, and new approaches using RNA models to achieve that are showing the promise that RNA will have in this space. It may be too early to call this round for RNA, but it is beginning to make a case.

Concluding thoughts
As it turns out, the battle of the bases was not really a fair fight. The advances in high-throughput sequencing and analysis that have bolstered our databases of DNA information have in parallel provided us with a plethora of RNA data to mine and interpret. Further contributing to this are enhanced tools for analysing chemically and enzymatically degraded RNA, removing a roadblock that previously existed. Now, instead of seeing RNA’s lability as a weakness, its dynamic nature is now considered a massive advantage. We can measure the impact of therapy and disease progression in the body’s ever-changing RNA. Today, the field is focused on building and validating new tools using RNA to aid physicians in understanding what causes disease, when disease will manifest itself, how we can predict which therapies will benefit individual patients and when disease will recur or progress. From infectious diseases to oncology, and likely in areas we have not yet discovered, RNA diagnostics will certainly be a key chapter in the tome entitled precision medicine that so many of us in the industry are currently writing. And, if successful, may help us knockout some of these malevolent diseases.

Instead of seeing RNA’s lability as a weakness, its dynamic nature is now considered a massive advantage