Briefing | Antibiotic resistance

The grim prospect

The evolution of pathogens is making many medical problems worse. Time to take drug resistance seriously

FEW, nowadays, would regard gardening as dangerous. But on March 14th 1941 a British policeman called Albert Alexander died of it. Early that year he had been scratched on the face by a rose. The wound became infected by bacteria, probably Staphylococcus aureus with an admixture of various Streptococci, and turned septic. The sepsis spread. First, he lost an eye. Then, he lost his life.

What made Alexander doubly unlucky was that he was almost cured. The hospital treating him, the Radcliffe Infirmary in Oxford, was a few hundred metres from a university laboratory where Howard Florey and Ernst Chain were brewing up extracts of a mould called Penicillium chrysogenum. Repeated injections of this extract came close to abolishing Alexander’s infection, but the two scientists ran out of their home-brewed drug before the bacteria had all been killed. When the treatment stopped the sepsis roared back.

Penicillin is now available in copious amounts, as are other bacteria-killing antibiotics. A thorn scratch today seems a minor irritant, not a potential killer. But that may be too sanguine. A study by America’s Centres for Disease Control (CDC) found that the number of cases of sepsis rose from 621,000 to 1,141,000 between 2000 and 2008, with deaths rising from 154,000 to 207,000. One reason for that is the emergence of MRSA (pictured being attacked by a white blood cell)—a variety of Staphylococcus aureus that cannot be killed with methicillin, one of penicillin’s most effective descendants. This could just be a taste of things to come. Three years ago the CDC produced a list of 18 antibiotic-resistant microbes that threaten the health of Americans (see table). Five of them (including MRSA) cause sepsis.

When people hear about antibiotic resistance creating “superbugs”, they tend to think of new diseases and pandemics spreading out of control. The real threat is less flamboyant, but still serious: existing problems getting worse, sometimes dramatically. Infections acquired in hospital are a prime example. They are already a problem, but with more antibiotic resistance they could become a much worse one. Elective surgery, such as hip replacements, now routine, would come to carry what might be seen as unacceptable risk. So might Caesarean sections. The risks of procedures which suppress the immune system, such as organ transplants and cancer chemotherapies, would increase.

Such worsenings would not be restricted to hospitals. “Multi-drug resistant” and “extensively drug resistant” strains of tuberculosis cause 200,000 deaths a year, mostly in poor countries. Most people who die of tuberculosis at the moment do not die of one of these strains. But they are responsible for more than an eighth of fatal cases, and those cases might otherwise be susceptible to treatment.

Neisseria gonorrhoeae is another bug that has repeatedly developed resistance to antibiotics. When penicillin was first introduced it worked very well against gonorrhoea. When its effectiveness began to fall, it was replaced by tetracyclines. Those gave way to fluoroquinolones, and those, in turn, to cephalosporins. Now, some strains can be tackled only with a combination of ceftriaxone, a cephalosporin, and azythromicin, an azalide. There is nothing else in the locker.

If worries about microbial resistance are cast wider to include not just antibiotics (which attack bacteria) but drugs against parasites, like malaria, and viruses, like HIV, the problem multiplies, particularly in poor countries. In the case of malaria, resistance to drugs that kill the parasite responsible has been a problem for decades. Since the turn of the century deployment of a new medicine, artemisinin, has provided some respite. But now parasites resistant to artemisinin are turning up. And the same is true for first-line drug combinations against HIV, which go back to the 1990s. Such resistance can be dealt with by other medicines, kept in reserve for the purpose. But it still makes things worse, complicating treatment.

This trend is longstanding; Alexander Fleming, who first noticed penicillin’s effects, warned of the dangers of resistance almost as soon as the drug had been shown to be a success. But the fact that these are old worries does not mean that they are not serious ones, nor that they cannot get worse. This week sees the publication of the final recommendations of a review on resistance to antimicrobial drugs led by Jim O’Neill, formerly chief economist at Goldman Sachs, on behalf of the British government and the Wellcome Trust, a medical charity. According to Lord O’Neill and his colleagues 700,000 people die each year from infection by drug-resistant pathogens and parasites. And they say that if things carry on as they are that figure will rise to 10m by 2050, knocking 2-3.5% off global GDP. Already the cost to the American health-care system of dealing with infections resistant to one or more antibiotics is $20 billion a year.

Evolution in action

Drug resistance is a simple-to-understand, yet often misunderstood, phenomenon. Antibiotics mostly kill bugs by either blocking the synthesis of new proteins or interfering with the making of cell walls. Any variation in the bacteria’s genome that makes one of these drugs a less effective killer will be of benefit to the bugs that have it, and as a result it will spread through the population. The genetic variation may change the bug’s physiology, for example increasing the production of proteins that flush harmful molecules out of the bacterium. It may cause the production of an enzyme that destroys the drug. Or it may change the shape of the molecule that the drug is aimed at, making it less susceptible to damage.

Humans are not the first of Earth’s creatures to want to kill bugs. Fungi don’t make penicillin for fun, they do it to protect themselves from certain bacteria. The existence of these natural bug-killers has been a great help to human medicine; many of the 20-odd classes of antibiotic used medically are derived from them. But it also means that today’s bugs are not facing entirely new threats. There are often resistance genes tailored to abiding threats lurking, at a low level, in bacterial populations, waiting for their hour to come.

When it does, the gene can spread to other bacteria quickly. Bacteria keep some of their genes on little loops of DNA called plasmids that can be swapped quite easily; think of them as programs on USB sticks. These plasmids allow resistance to pass not just from individual to individual but from species to species. (Genes that make diseases virulent can spread the same way.)

The genes needed for resistance can thus be quite readily available. But like any biological attribute, resistance is not a free good. Building extra bacterial bilge pumps or special drug-smashing enzymes costs a micro-organism energy and materials; changing the shape of molecules to make them drug-proof is likely to leave them working less well than they did. Simply copying the DNA of the resistance gene imposes a metabolic load. And different antibiotics require different resistance genes; the more a bug needs to use, the greater the costs. So resistance tends to be sustained at a high level only when actively provoked by the presence of the drugs in question. That leads to an important corollary: expose the bacteria to fewer drugs and resistance should abate.

INTERACTIVE: Clinical trial simulator - Run your own trials and play the system in favour of your own drug

This is where the aforementioned misunderstanding comes in. The public tends to think that it is the person taking the drugs who becomes resistant to their effects, not the microbes. Last year research published by the World Health Organisation showed that three-quarters of people in poor and middle-income countries misunderstood the problem that way. A survey carried out earlier in 2015 by the Wellcome Trust suggested a similar prevalence of misunderstanding in Britain.

Such ignorance has consequences. If you know that resistance is an attribute of the bacteria, then using drugs rarely but definitively makes sense. Do not use them when not needed; when you do use them, use them in such a way as to kill off all the bacteria, rather than leaving behind a small resistant rump. If you mistakenly think resistance is an attribute of people, on the other hand, you will have no compunction about using antibiotics, provided they seem to have some effect. And you will not think twice about stopping the course when the symptoms subside, rather than carrying it on until all the bacteria are gone. These problems are at their worst in places where antibiotics are easily bought over the counter.

Lord O’Neill argues that public-awareness campaigns might put things right. But this seems a little optimistic. Even when prescriptions are needed and experts are involved things still go wrong. In America some 40m people are prescribed antibiotics for respiratory problems every year. In 2013 a paper published in the Journal of Antimicrobial Chemotherapy estimated that two-thirds of those people may well not have needed the antibiotics they got.

Some of this is down to “pester power”: having gone to the doctor, a patient wants something tangible to show for it, even if his sore throat is probably viral and antibiotics will do him no good. Sometimes, though, it is the other way around. If a doctor cannot be sure of the cause, prescribing an antibiotic may help. The real chance of healing a specific person outweighs the imperceptible increase to the threat of bacterial resistance.

Know your enemy

This suggests one way to reduce the development of resistance would be to have diagnostic kits that ruled bacterial infection in or out on the spot. Such kits would have to be very quick, cheap and convenient indeed to supplant pre-emptive antibiotics. But if they could also tell to which antibiotics an infection was susceptible they would increase their value. If a doctor were to know for sure whether a dose of gonorrhoea could be dealt with by penicillin—which is the case for as many as 80% in England and Wales—he would not have to prescribe more expensive antibiotics just in case; good for the purse and good for public health.

Resistance is not only encouraged and spread in medical settings. In many places, more antibiotics are given to farm animals than to people. In America 70% of those sold end up in beasts and fowl. Some of this is to treat disease; most is not. For reasons only dimly understood, many animals put on weight faster when fed these drugs. A lot of these drugs pass into the soil and watercourses, where they further encourage resistance. The bacteria that become resistant this way are unlikely to be human pathogens. But their resistance genes can quite easily get into bugs that are.

Some of the antibiotics farmers use are those that doctors hold in reserve for the most difficult cases. Colistin is not much used in people because it can damage their kidneys, but it is a vital last line of defence against Acinetobacter, Pseudomonas aeruginosa, Klebsiella and Enterobacter, two of which are specifically mentioned on the CDC watch list. Last year bacteria with plasmids bearing colistin-resistant genes were discovered, to general horror, in hospital patients in China. Agricultural use of colistin is thought to be the culprit.

The cost of banning antibiotics as growth enhancers would not be great: an American government study suggests it might reduce the bottom line of those who currently use them by less than 1%. The European Union has already enacted such a ban. Despite practical difficulties—the difference between a growth-enhancing dose and a veterinarially defensible prophylaxis may often be in the eye of the beholder—more should follow.

Lord O’Neill favours such prohibitions. He also likes the idea of using more vaccination to head off the need for treatment, both in livestock and in people. Hospital hygiene is another focus; there is some evidence that staff are more careless about cleanliness than they were in pre-antibiotic days, when they saw deaths like Albert Alexander’s on a more regular basis.

All these steps would make existing antibiotics more effective; another approach is to create more such drugs, or their functional equivalents. In the decades after penicillin came to market drug companies fell over each other to develop new antibiotic molecules. Since then, interest has waned. The pipeline of potential new products at various stages of clinical trial is barely 40 strong. Only a fraction of them will reach market; each will represent a big investment (see chart 2).

There are reasons for drug firms not to invest in antibiotics. Such companies increasingly prefer treatments for chronic diseases, not acute ones; the customers stick around longer. And despite the growing problem of resistance, most antibiotics still work for most things most of the time. Given that the incumbents are also cheap, because they are off-patent, new drugs cannot earn back their development costs. Even if they could, it would be poor public policy to let them; much better for new drugs to be used only sparingly, to forestall the development of further resistance. That further puts the kibosh on sales.

Some of the gap might be plugged by reviving old drugs that have fallen out of use; drugs bugs have not recently seen are drugs they are less likely to be resistant to. Another possibility is to revamp the incentives, rewarding the development of antibiotics destined to sit behind “use only in emergency” glass. The O’Neill report suggests one-off payments of between $800m and $1.3 billion to firms that develop drugs which meet predefined criteria of unmet need, to be paid on top of sales revenue. At this year’s meeting of the World Economic Forum in Davos, 85 companies said that if governments offered them money with such conditions attached they would do everything they could to earn it.

Whatever happened to the phage way?

It is possible, though, that the problem goes deeper than incentives. Some think the supply of raw materials for research—molecules capable of being turned into antibiotic medicines—may itself be close to exhaustion. The intensive efforts of the mid-20th century may have more or less emptied nature’s store.

In the late 1990s established drug companies and biotech startups ransacked then-revolutionary genome sequences from bacteria for targets against which they could screen chemicals by the zillion in the search for new drugs. According to David Payne, a researcher at GlaxoSmithKline who wrote about this gold rush last year, his firm looked at 70 apparently promising targets this way. On the basis of experience in other therapeutic areas the company expected to find almost as many “lead compounds”—molecules that looked worth the effort of taking further—as it had targets. Instead it came away with six. Given the attrition rate for lead compounds in subsequent development the whole undertaking was pretty-much wasted effort.

Today’s far more powerful genomics might tell a different story, but it seems wise to look at alternative approaches. One possibility would be to use specially formulated antibodies, instead of conventional “small-molecule” drugs. On the plus side this could provide weapons bacteria have never come across before (yeasts and other microbes do not make antibodies). On the minus side, therapeutic antibodies tend to be very expensive.

Another option is to co-opt viruses, known as bacteriophages, that prey on bacteria. Bacteria have defences against phages, just as they have resistance genes against natural antibiotics (indeed, biotech’s hot new genome-editing tool, CRISPR-Cas9, is based on a system bacteria use to slice up the genes of viruses that attack them). Phages have been looked at as therapies for decades; better understanding of bacterial genomes may mean that they can now be used in more cunning ways than was previously possible.

There are also ecological approaches. Some bits of the body—notably the skin and the gut—are permanently and unproblematically inhabited by bacteria. There is some evidence that manipulating these native populations can make them less welcoming to outsiders. This approach has had early success against Clostridium difficile, the bug at the top of the CDC danger list. And there are also drugs that might be aimed at the patient, not the bug, in an attempt to make his immune response more appropriate and effective.

These possibilities show that there is no reason for panic. But there is strong reason for action. Florey and Chain were motivated by a crisis: sepsis took a heavy toll among the wounded of the second world war. Today’s steady worsening is no crisis; this war is a subtle one, to be fought on behavioural, regulatory and economic fronts, as well as medical ones. But war it nevertheless is. An appropriate response is called for.


Correction: The word "plasmids" was rendered as "plastids" on two occasions in the original version of this story. Plastids are plant-cell organelles, not found in bacteria. Be assured that we do know the difference. Blame poor proof-reading.

This article appeared in the Briefing section of the print edition under the headline "The grim prospect"

When the drugs don’t work: The rise of antibiotic resistance

From the May 21st 2016 edition

Discover stories from this section and more in the list of contents

Explore the edition

Discover more

How to predict Donald Trump’s foreign policy

He may be inconsistent, but his advisers offer some clues

The war in Gaza may topple Hamas without making Israel safer

It will end up even more deeply mired in the conflict that is the main threat to its security


America’s economy has escaped a hard landing

But there are still pitfalls ahead