21 January 2020

10 minutes read time

The Gut

Shit or bust

Without an understanding of the microbiome, the drugs don’t work. Big pharma is finally catching on

By James Kinross

In 2003, Allen Roses, worldwide vice-president of genetics at GlaxoSmithKline (GSK), said the following:

“I would say that most drugs work in 30 to 50 per cent of people. Drugs out there on the market work, but they don’t work in everybody.”

When Gerald Ratner publicly criticised the quality of his company’s jewellery, it folded overnight. But, when Roses said those words almost two decades ago, GSK didn’t. That is because, even when the drugs don’t work optimally or safely they still provide some control of our blood pressure and heart rate, or they slow the furring of our cardiac arteries, or they kill cancer cells. They can make life bearable, and possibly even extend it.

Dr Allen Roses: The drugs don’t (always) work

The global drugs market was worth $1.2 trillion in 2018 and 2.5 million prescriptions are written each day in the UK. It is big business, but it is also a risky business. It costs about $1.3 billion dollars and takes about 12 years to take a drug from laboratory to clinic. And there is a high failure rate: out of every 10 new drug candidates, 9 will fail to reach the market. The biggest problem of all is that the pipeline of new drugs is running dry. We have an epidemic of diseases such as obesity and Alzheimer’s, and no effective treatments for them.

Since 2003, pharma companies have championed the concept of personalised healthcare as a solution to this problem. That is, getting the right drug into the right person at the right time for the right reason. What pharma didn’t realise is that you can’t have personalised drug treatments without the microbiome. After all, the gut bicrobiome is one of the most important sources of human biological variation.

Many drugs are chemically modified by the gut microbiome, either directly or indirectly. It influences how safe or effective a huge number of over-the-counter medicines are – even paracetamol. Not only that, many drugs that are not antimicrobial have antimicrobial effects; drugs like Omeprazole, which was prescribed almost 25 million times in the NHS in 2017. If you change the acidity in the foregut, you change the prevalence of bacteria in both the foregut and the colon, with significant consequences for both the bugs and the host. So massive drug prescribing is probably as bad for our microbiome as a western diet.

You can’t have personalised drug treatment without allowing for the microbiome

It is disingenuous to claim that drug manufacturers knew nothing of the gut microbiome. In the 1970s, scientists started making a number of drugs that took advantage of specific chemical functions found in gut bacteria to improve drug efficacy. Sulfasalazine, for example, reaches the colon and is metabolised by bacteria into two active metabolites that have an anti-inflammatory effect. Sulfasalazine is still used to this day in the treatment of inflammatory bowel disease, and there are lots of other examples of drugs that are designed to work in partnership with bacteria.

And drugs developers haven’t exactly been twiddling their thumbs since the 1970s. Modern, data-driven approaches and massive computing power mean that scientists can now efficiently search for new drug candidates and try to predict how they will work through simulation.

But, even today, animal research remains a cornerstone of the development process. You can’t put a drug into a human until you know a little bit about how it performs in a real biological system. And so there is a big market for genetically engineered animals. Genetic mutations can be ordered from a macabre shopping list to simulate just about any human condition.

During drug experiments, mice or rats are typically kept in cages in small numbers. But these animals are coprophagic (they eat poo) and of course they come into contact with each other. So they share bacteria. Until very recently, no one really thought that this had a big impact on drug function. At least, they thought there would not be a big variation between cages, or even between the animal houses that produce them. Everyone was wrong.

A recent study of a new drug called a “PD-L1 checkpoint inhibitor” illustrates this nicely. This drug is used in the treatment of a skin cancer called melanoma and in some other cancer types. It is a blockbuster drug that may revolutionise survival from this terrible disease, which previously had an awful prognosis.

Rats eat poo too

Thomas Gajewski’s lab in Chicago acquired mice from two different mouse facilities, Jackson Laboratory (JAX) and Taconic Farms (TAC), to test the drug. These are genetically identical animals and they should therefore have an identical response to the drug when a melanoma was induced. But, instead, tumours grew much more aggressively in the TAC mice. This was because the bacteria in the guts of these mice regulated a part of the immune system called “T cells”, which in turn influenced how effective the drug was. Amazingly, when faeces from the JAX mice were transplanted into the TAC mice, this effect was reversed.

The scientists managed to identify a specific Bifidobacterium species, which when transplanted alone improved tumour control to the same degree as the PD-L1. Yes, a single strain of bacteria was as effective as the billion-dollar drug. When the same bug was given with the drug, the effect of the drug was further enhanced. This may mean that, if we want our cancer drugs to be more effective, we need to optimise our microbiome first, which is pretty startling. In fact, this is the case in any therapy that targets the immune system. But this is not the really important bit. The significance of this work is that it means the entire basis for drug discovery is potentially flawed.

New experimental models that can more accurately and consistently recreate host-drug-microbiome interactions within our bodies are now needed. One potential solution is to make a gut on a chip; a completely synthetic gut system made from plastic, stem cells, an artificial blood supply, mucus and bacteria. If we can get these to work, we will have a greater understanding of how the microbiome influences drug metabolism. Then drug design becomes cheaper, trials are more efficient, there are fewer side effects, there are fewer failures, and ultimately the cost of drugs comes down.

When Jeffrey I. Gordon and his group performed their famous obesity experiment, in which they demonstrated that they could make a mouse obese through faecal transplant, they changed the world of drug discovery for ever. Because if you can change an animal’s weight by faecal transplant, it means that there is a chemical pathway that regulates it. That, in turn, means you can make a drug to treat obesity without FMT. Imagine a magic pill that makes you thin, and you can eat what you want. Kerching!

Jeffrey I. Gordon changed drug discovery forever

The gut microbiome is therefore an unbelievably rich source for new drug candidates that could be leveraged to resupply the dwindling pharma pipeline. Massive searches of gut bacterial functions are being performed to try and identify new treatments for a huge list of chronic (and profitable) diseases. Pharma companies are hungrily snapping up microbiome start-ups. We even have a catchy new word to describe this new science: pharmacomicrobiomics.

But just because a company is working in the microbiome field it doesn’t meant it is guaranteed a hit. In 2015, a new microbiome therapeutics company called Seres Therapeutic raised $131 million on its IPO and it had a juicy deal with Nestle. It looked great until 2016 when it published data from a Phase II trial (essentially a small trial in humans) of their oral microbiome therapeutic “SER-109” that was designed to treat C diff. It was no better than a placebo and, overnight, the company’s stock price collapsed.

Ultimately, however, drugs will be designed to target specific bacteria or microbiome functions for human benefit. Already, ingenious ways of doing this are being found. We are starting to leverage synthetic biology to build bacteria that will manufacture drugs in your gut. We are engineering bacteria to produce enzymes, antibiotics, toxins and metabolites that can kill bad bugs, cancer, or even instruct other bacteria and yeasts to start producing molecules that treat chronic human disease. These small molecules can then disseminate to any organ in the body, including the brain. It is very likely that future treatments for neurological conditions will leverage the gut microbiome in some way.

One of the most important biological discoveries in modern times is gene editing. This is known as CRISPR and it allows us to change the genes within a genome. You will not be surprised to hear that this was discovered in bacteria. CRISPR sequences are derived from DNA fragments of phages that have previously infected a bacteria. It “remembers” this and uses these same segments to detect and destroy DNA from similar phages in subsequent infections. CRISPR will be put to good use on the gut microbiome.

Work to do for Craig Venter back on dry land

The probiotics market will have a turnover value of $46.6 billion by 2020. Its growth will be dependent on the development of gene-edited, next-generation probiotics. It won’t just be the gut microbiome that will be targeted. The skin, lung, bladder and vaginal microbiomes will be leveraged. And the microbiome will also be used by the food, beauty and wellness industries.

And let’s not forget Craig Venter, the biological entrepreneur who started much of this off. In 2010, his team claimed to have created the first form of synthetic life. They built, synthesised and assembled a bacterial genome for Mycobacterium mycoides and they have repeated this feat in yeasts. There is a lot of hyperbole in this, but the ultimate goal is to literally create new bacterial life. It is only a matter of time before this technology is used to engineer not only your microbiome but also oceanic and soil-based microbiomes that can solve some of mankind’s biggest, self-imposed problems.

Someone, somewhere is going to make a shit load of money.

Photographs Getty Images, Sandy Huffaker/The New York Times/Redux/eyevine, Dan Gill/Polaris/eyevine 

All our journalism is built to be shared. No walls here - as a member you have unlimited sharing