The search for the perfect painkiller

Only the word “epidemic” really does justice to the scale of human tragedy caused by opioid drug use in America.

How else can we describe 145 largely avoidable deaths a day? Opioids – mostly prescribed as painkillers or obtained illicitly by those hooked through previous treatment – killed 53,000 people in the United States in 2016 – more than guns or road accidents. That same year saw deaths caused by synthetic opioids, mainly fentanyl, double to 20,145.

A commission set up by Donald Trump has been praised for 56 relatively progressive recommendations issued in October. Yet with a 90-day public health emergency declared by Trump having run out on Tuesday 23 January, some have accused him of failing to turn words into funds and action.

Opioid drugs are important for acute pain management following trauma and surgery, as well as end-of-life settings. Their effects vary by type and for individuals but taking them for extended periods can lead to tolerance – meaning larger doses are needed – along with withdrawal symptoms, physical dependence and addiction. Opioid painkillers and related drugs like heroin sedate users and slow their breathing. At higher doses, especially with alcohol or sleeping pills, users can fall asleep and die when their automatic breathing mechanisms shut down.

While there have been concerns about increasing opioid abuse in other countries, the US has by far the greatest death toll – largely the result of pharmaceutical companies aggressively marketing them there for a wide range of conditions, including for long-term use, from the mid-1990s. Millions of Americans are still hooked.

Reversing the tide is likely to be beyond the reach of politicians alone. Doctors and drug companies have roles to play, though recent research suggests that scientists may hold the key. Evidence emerging from labs in the UK, US, Germany and elsewhere suggests a new generation of safer painkilling opioids could soon be available to help end the epidemic.

The human body produces its own opioids. Compounds such as endorphins lock on to receptors embedded in the membranes of cells in the brain and spinal cord, acting as cellular gatekeepers that help modulate things like pain, stress, mood, motivation, hunger and thirst. Synthetic opioids and those derived from poppies work in the same way.

Modern research in this field has its origins in the 1970s, when scientists began to isolate and understand the roles of opioid receptors and their subtypes, called mu, delta and kappa, as well as what happens on the inside of cells when these are activated. They all belong to a larger group called G protein-coupled receptors (GPCRs), and were originally seen as molecular on-off switches, which once flipped triggered a cascade of chemical changes beginning with molecules called G proteins binding to the inside of the receptor. From the 1980s, scientists realised they were more like switchboards.

From the 1970s to the 1990s, Prof Robert Lefkowitz of Duke University, North Carolina, made a series of discoveries about GPCRs that would eventually earn him the 2012 Nobel prize for chemistry. He revealed, for example, that proteins called arrestins, previously seen as “dimmer switches” that acted on the inside of cells to turn down the effects of G proteins, also acted as separate signalling pathways with their own physiological effects.

During the late 1990s, Laura Bohn was working in a lab close to Prof Lefkowitz’s at Duke, genetically manipulating mice so that they no longer produced a protein called beta-arrestin 2. When she gave morphine to these “knockout” mice, it had improved painkilling powers and triggered fewer breathing problems. They also suffered less constipation – another side-effect of opioid drugs.

Prof Bohn’s work helped open the door to new ways of making potentially safer opioid drugs. Like traditional opioids, they target mu opioid receptors and reduce pain through G protein signalling. Unlike them, however, they are designed to avoid attracting beta-arrestin 2 to the receptor, thereby avoiding the risk of users building up tolerance and taking potentially fatal overdoses.

In 2007, several researchers who worked with Lefkowitz set up Trevena, a biotech company based near Philadelphia, to develop treatments based on this concept, which has become known as a “biased” opioid. Last year, Trevena published partial results from two phase-three trials of Olinvo, a biased opioid designed for use as an intravenous analgesic to manage acute pain in hospital settings. A low dose caused fewer breathing problems, nausea and vomiting than morphine in patients following bunion removal and tummy tuck procedures. However improvements were not as good as had been hoped for higher doses with the greatest painkilling effects.

The US Food and Drugs Administration (FDA) is due to rule on whether to approve Olinvo in November. “Our hope is that Olinvo can help physicians manage patients’ pain in hospitals with the potential for better safety and tolerability, and can therefore play a role in solving the US’s pain management crisis,” says Dr Jonathan Violin, Trevena co-founder.

Some scientists wonder whether the devastating side-effects can be further reduced with an even greater “bias ratio” in favour of G protein signalling and against the beta arrestin-2 pathway. “It might be that a ‘super Olinvo’ with a greater bias factor would work better than Olinvo,” says Lefkowitz.

Research published in November by Bohn – now at the Scripps Research Institute in Jupiter, Florida – suggests he may be right. Her group identified existing favourably biased compounds and tweaked them to increase their bias. Those with larger biases produced fewer breathing problems.

While Bohn and others seek a more targeted approach to maximise opioid benefits and minimise harm, others have gone the other way. Most drugs on the market are designed to target just one receptor, but actually interact with others as well. Although today’s painkillers largely target the mu opioid receptor because of its influence on pain, research suggests delta opioid receptors help mediate tolerance and dependence.

Prof Andrew Coop is a Yorkshireman working at the University of Maryland School of Pharmacy in Baltimore – a city that saw nearly 700 drug overdose deaths in 2016. His search for a compound to activate the mu receptor while blocking the delta receptor began in 1999. Research he published in 2013 showed that mice given a compound he called UMB425 achieved similar pain relief to those given morphine but developed much less tolerance. If such results held true for humans, this could mean users are less likely to become dependent.

Despite being almost two decades into his search, Prof Coop isn’t ready to make his compound available to patients. He is all too conscious of the role of previous generations of scientists in fuelling today’s opioid epidemic. “UMB425 is a step forward, but it’s only halfway there,” he says. “It is designed to prevent dependence, but not to stop the rewarding effects associated with abuse, so it would be inappropriate to release it as a drug.” He is now working on a new version that can’t be used to get high.

Coop is not alone in his belief that the incalculable damage caused by previous scientific efforts to make safer opioids means the only ethical approach today is to put addictive potential front and centre of work in this field. Prof Stephen Husbands at the University of Bath began his career working under John Lewis, who oversaw the development of buprenorphine, an opioid painkiller launched in the UK in 1978. Buprenorphine is used as a treatment for opioid addiction because while it too acts on the mu opioid receptor, it cannot activate it fully and so has milder effects. However, it too has become a major drug of abuse.

As well as acting on the mu opioid receptor, buprenorphine also binds to the nociceptin opioid peptide receptor (NOP). Prof Husbands set out to produce a less addictive form of buprenorphine by making a compound with greater NOP activity and reduced affinity to the mu receptor. The result, in 2008, was BU08028. US collaborators found it blocked pain at low doses without suppressing breathing in monkeys. What’s more, the animals did not self-administer it when able to or suffer withdrawal symptoms when no longer given it.

“How things will translate into humans we don’t yet know, but all the evidence we have suggests BU08028 is less open to addiction than buprenorphine, which itself is less addictive than morphine or heroin,” says Husbands. He has licensed BU08028 to California-based Orexigen alongside other compounds, on two of which it hopes to conduct clinical trials.

Trevena may head the race for safer opioid painkillers with Olinvo, but even if the FDA approves it for patients later this year, it is expected to be classified as of high abuse potential. Others have used the biased opioid approach to develop a compound that they say shows early promise of easing pain without triggering addiction.

In 2012, a group at Stanford University, California, led by Nobel laureate Prof Brian Kobilka, used X-ray crystallography to produce a detailed picture of the structure of the mu opioid receptor. Kobilka’s long-time collaborator, Prof Brian Shoichet, at the University of California, San Francisco, modelled vast numbers of potential interactions between this structure and a virtual library of more than 3.5m compounds. He selected the best two fits that were also biased towards G protein pathway and against beta-arrestin. Tests on molecules similar to these led them to the compound PZM21.

Tests in mice showed that PZM21 provided potent analgesic protection but was only effective in blocking the affective or emotional aspect of pain and not the automatic, reflexive part. It did not slow breathing, and mice showed no preference for compartments where they had been trained to expect doses of PZM21, as they did with morphine.

“We don’t know how this will play out in humans, but these experiments give us hope that there may be a way to get analgesia without getting the reinforcement effects that are often connected with addiction,” says Prof Shoichet. He and colleagues have started a company called Epiodyne to optimise PZM21 further.

Others are taking different approaches. German scientists last year published details of a modified version of fentanyl that only acts on the mu opioid receptor in the more acidic environment of inflamed tissue. Animals tests suggested it did not slow breathing or trigger addictive behaviours. Cara Therapeutics, based in Stamford, Connecticut, is carrying out phase two trials of CR845, which targets the kappa opioid receptor with potential to target pain while minimising side-effects.

While all this work is certainly adding greatly to knowledge about precisely how opioids act on the body, some are sceptical at the idea that their potential for abuse and dependence can be entirely designed out. If the systems we have evolved to reward us for advantageous behaviour with feelings of wellbeing, and those that suppress pain when needed, are intertwined, there may be a limit to how much they can be separated.

“Pain is complex,” says Prof Bohn. “I think there are distinct regions of the brain that deal with rewards and the modulation and perception of pain, but there are also overlaps. When it comes to addictive potential, I think we’ve had enough overselling of compounds that people say won’t cause addiction, and troubles as a result of that, so I won’t speculate on that.”

Given how heroin was marketed as a safer, nonaddictive form of morphine that could treat coughs, and the huge toll of related drugs today, scientists are well aware that the burden is on them to prove any claims they make for novel opioids. This does not, however, mean that only entirely side-effect-free new drugs have merit.

“People have a right to be sceptical, given the horrendous consequences of some of the claims made in the past,” says Prof Husbands. “Those of us in the field have to show that things we are developing now are improvements. That doesn’t necessarily mean the holy grail of strong pain relief with absolutely no side-effects, but if we can achieve strong pain relief with significantly reduced side-effects, that’s certainly a worthwhile goal.”

Poppy love

A concise history of human opiate use and abuse

Opium is an extract of the opium poppy plant, Papaver somniferum. The earliest written evidence of its use by humans dates back more than 5,000 years. According to a clay tablet excavated in what is modern-day Iraq, the Sumerians were cultivating the plant by 3,400BC. They called it hul gil, or “the joy plant”.

The Assyrians also used opium, as did the ancient Egyptians who, for example, used it to stop children crying. Poppies appear in Greek artefacts dating back to 1,500BC, and Roman doctors used opium as an anaesthetic during amputations. Arabs used it to treat eye problems and diarrhoea, and introduced it to China through trade in the seventh century.

The British surgeon James Moore is widely credited as the first doctor to record using opium to treat postoperative pain in 1784. German pharmacist Friedrich Sertürner isolated the active ingredient in opium and called it morphine, after Morpheus, the Greek god of dreams.

Morphine use for chronic and postoperative pain and during surgery grew following the invention of the hypodermic needle in 1853. Concerns about its abuse led to the development, in 1895, of heroin, which German chemical company Bayer marketed as a cough remedy and “non-addictive morphine substitute for medical use”.

Widespread abuse led to the drug being banned in the US in 1924 in the US and controlled in the UK from 1921. Scientists have been trying to develop safer alternative opioid painkillers for more than a century.