Pain relief: designing better opioids

Opioid overdoses in the United States were responsible for at least 42,000 deaths in 2016. This represents two-thirds of all US drug deaths, and more deaths than caused by car accidents or even guns. The crisis stems from an epidemic of prescription drug misuse; opioid prescribing rates increased threefold from 1999 to a 2015 peak. Today, 12 million Americans are using prescription painkillers without a prescription and an estimated 2 million have become dependent. The US situation has focused attention sharply on the problem of pain relief. The hunt for new, safer opioids that carry a lower risk of side effects is now gaining momentum with several strategies showing promise and new drugs getting nearer to the clinic.

Historical solution

Derived originally from the opium poppy, morphine has been known almost as long as civilisation itself. There are references to the cultivation of “the joy plant” in ancient Mesopotamia more than 5,000 years ago, and opium use was widespread in ancient Egypt and was also mentioned by the Greek author Homer. The move from plant to pharmaceutical started in 1803 when German chemist Friedrich Sertürner first isolated morphine from opium, but synthetic analogues were not developed until the mid-20th century.

It is often argued that today’s crisis stems from the ever-present imperative to make money from opioids. It shouldn’t be forgotten that in the mid-19th century two wars were fought over opium itself — the Opium Wars (1839–1842 and 1856–1860) followed China’s attempts to stop Great Britain trafficking the drug. However, the US epidemic can also be traced back to a shift in physician attitudes to pain relief, which has had an impact on opioid prescribing rates in Europe, too.

I put my hand up, like the [rest of the] pain community, [in] really buying into [opioids] in the early 2000s and promoting, and thinking that we had a cure for, persistent pain

“I put my hand up, like the [rest of the] pain community, [in] really buying into [opioids] in the early 2000s and promoting, and thinking that we had a cure for, persistent pain,” explains Roger Knaggs, a pain management specialist pharmacist at the University of Nottingham. “I think this increased willingness to prescribe [opioids] then began to diffuse into primary care.”

As a result of this shift in attitude, drugs that were once largely the preserve of hospital medicine became more widely prescribed, so much so that — according to the Public Health Research Consortium, which brings together senior researchers from 11 UK institutions in a new integrated programme of research — the proportion of patients on the UK Clinical Practice Research Datalink prescribed opioids doubled between 2000 and 2012, from 2.6% to 5.4%. These data also show that the length of continuous prescribing periods increased from 64 days to a peak of 102 days in 2013 and 2014.

In the US, this attitude shift was further spurred on by aggressive (and sometimes fraudulent) marketing of drugs, such as OxyContin. In 1996, Purdue Pharmaceuticals introduced this controlled-release formulation of the opioid oxycodone with the promise that its smooth and sustained 12-hour pain control lowered the abuse potential. But there was no scientific evidence to support this claim and, in 2007, Purdue was fined US$600m as a result. But, by then, the metaphorical horse had bolted and although prescription opioid deaths had levelled off by 2011, those now addicted were moving to street drugs, such as heroin or the more potent opioid fentanyl, which has proven particularly dangerous. In fact, since 2015, fentanyl-related deaths in the US have increased by 540%. The risk of overdose stems from the respiratory depression opioids cause. But they cause multiple side effects, including nausea, vomiting, sedation, euphoria and, often most problematically, constipation. As with many medicines, patients eventually become tolerant to opioids, progressively needing larger doses to maintain the same effect. They can also develop physical dependence, causing withdrawal symptoms if the medicine is stopped, and they sometimes become addicted, leading to the sorts of drug-seeking behaviour that causes harm to the patient and, potentially, whole communities.

The quest for safer opioids started long before the current crisis, though. Heroin was developed in 1874 by UK chemist Alder Wright as a safer, less addictive alternative to morphine. In the past 30 years few new opioids have made it to market. OxyContin was only a reformulation of oxycodone, in clinical use since 1916. Thankfully, several new opioids are nearing market, with a dozen or so in development (see Table). In 2017, Pennsylvania-based Trevena Pharmaceuticals submitted Olinvo (oliceridine) — an intravenous opioid with fewer of the usual side effects — for US Food and Drug Administration approval. This and other novel opioids have been designed with a variety of innovative strategies to maximise analgesic properties, while removing the dangerous side effects and risk of addiction.

Biased agonist approach

One of the most widely discussed strategies is the ‘biased agonist’ approach. Laura Bohn, a pharmacologist at the Scripps Research Institute in Jupiter, Florida, came up with this concept in 1999 when looking at analgesic responses to morphine in mice engineered without the intracellular signalling molecule beta (β)-arrestin2. This protein plays a role in dampening cellular responses, including those involving the intracellular G-proteins recruited when opioid receptors are activated. “[Without β-arrestin2, we] found that the analgesic response was actually enhanced and prolonged and [we] realised that the tolerance went away — you can chronically treat [the mice] with morphine and they remain responsive to it,” says Bohn. She also found the mice did not experience respiratory depression, which led to the idea that minimising the β-arrestin2 response to opioids could provide a safer route to pain relief.

Since then, Bohn has been looking for a drug that, when bound, would change the receptor’s shape to bias its response in favour of the pathway that recruits G-proteins and leads to pain relief, as opposed to the β-arrestin2 pathway, which seems to control many of the adverse side effects. Opioids, such as fentanyl, show the opposite bias and are more likely to induce respiratory suppression at lower doses than other drugs.

Bohn has since identified six piperidine-based mu (μ) receptor agonists that provided effective pain relief with fewer respiratory problems^[1]
. The addictive nature of these compounds is still unclear, but Bohn is hopeful that they will not lead to tolerance. “We are hoping that if we can develop a compound that doesn’t produce tolerance, doses can be kept low and pain can be managed,” she explains.

We are hoping that if we can develop a compound that doesn’t produce tolerance, doses can be kept low and pain can be managed

Nevertheless, whether or not Bohn’s biased agonist approach really can provide the required clinical advances remains to be seen. “It’s a good step forward but I don’t know if it’s going to be the magic pill that everyone is hoping for,” says Gavril Pasternak, clinician and pharmacologist at the Memorial Sloan Kettering Cancer Center in New York.

In fact, Trevena’s new opioid, Olinvo, is based on the biased agonist principle and trial results so far suggest its advantages over existing opioids may be more limited than hoped. Phase III trial results in post-surgical patients released by the company in February 2017 showed only trends to improvement over morphine with respect to side effects.

Receptor selectivity

Another strategy in the quest for safer opioids has been to selectively target an alternative opioid receptor. “The analgesic and the side effects of the opioids that are on the market at the moment are almost exclusively related to interaction with the μ receptor,” Knaggs explains. But there are others — the kappa (ĸ) and delta (δ) receptors provide different pathways for pain reduction. They are all G-protein coupled receptors (GPCR) — transmembrane proteins that pass through the cell membrane, usually seven times.

Drugs that bind selectively to the ĸ receptor seem to avoid some of the problems of other opioids. They have less effect on the central nervous system and so cause less of the problematic side effects associated with μ receptor agonists, including addiction. The strategy was first explored in the 1980s. “Initial attempts to look at the ĸ story really did not go well,” Pasternak explains. “They produced psychotic medical effects and some other side effects, and those strategies were dropped for many, many years.”

Initial attempts to look at the ĸ story really did not go well, they produced psychotic medical effects and some other side effects, and those strategies were dropped for many, many years

But the strategy is now returning, with a potential ĸ -targeting painkiller, difelikefalin (Korsuva) in phase I trials at Cara therapeutics based in Stamford, Connecticut. Bryan Roth, pharmacologist at the University of North Carolina School of Medicine, is also targeting the ĸ receptor but using a biased agonist that may avoid the dysphoric effects of previous ĸ agonists. Roth started by trying to understand the molecular structure of the ĸ receptor when it binds to an opioid ligand. By comparing the receptor structure in its active and inactive states he was able to model potential drug structures^[2]
. “We identified some early-generation drugs that target the receptor in a selective way and appear, at least in cells in culture, to have the activity that would predict engaging the receptor to induce analgesia without side effects,” he explains. “Of course, we ultimately need to test them in animals and down the line in humans.”

Pasternak has been taking another approach that also seeks to selectively activate receptors, but in this case a sub-type of the μ receptor. Over the past decade, he has discovered that there are variant μ receptors produced along with the expected seven transmembrane GPCR. “It turns out that the μ receptor, which is the main target for most opiates, is complex and, for some reason, Mother Nature really has it undergoing an extraordinary amount of what we refer to as alternative splicing,” he explains. This means that the gene that codes for the μ receptor also produces multiple variants of the protein, including a truncated six transmembrane version.

Pasternak found that he could create opioids that would target only the truncated receptors, the most potent being IBNtxA (3-iodobenzoyl-6β-naltrexamide), a naltrexone derivative^[3]
. “When we looked at drugs whose actions were pretty limited to those truncated variants, they were very effective analgesics. We could give doses that were five or ten times higher than the doses needed to produce pain relief, but we still saw no respiratory depression, which for us was exciting,” he says. To prove the specificity of the drug’s target, Pasternak engineered knockout mice without the truncated receptor and was able to show these mice did not respond to IBNtxA. But when the gene coding for the truncated version was reintroduced through a lentivirus infection, the analgesic effect of IBNtxA was restored.

Targeting these alternative receptors also seemed to prevent physical dependence developing, and tolerance to a lesser extent than morphine. But again, it’s still early days. “The proof of the pudding is in the clinic. Until you put it in [humans] and have tested it deeply, you just don’t know how things are going to work,” Pasternak concludes.

The proof of the pudding is in the clinic. Until you put it in [humans] and have tested it deeply, you just don’t know how things are going to work

Acidic trigger

Christoph Stein, an anaesthesiologist at the Charité University hospital in Berlin, has been working on a more local approach to opioid analgesia, which could also avoid the serious side effects linked to the central nervous system. Stein and collaborators have designed a molecule that will be active only in acidic conditions. “We know that many different types of tissue injuries that cause pain are characterised by acidotic tissue,” says Stein, “[Therefore,] the idea was that it would not activate the receptors in the brain [which is not acidic] and cause side effects”.

An opioid ligand changes its behaviour with pH owing to the inclusion of a tertiary amine, which has to be protonated for activation of the receptor. Stein started with the fentanyl molecule and, using computer modelling, found that by replacing a number of hydrogens with fluorine atoms he could create an analogue that would bind to the receptor only in acidic conditions^[4]
. He tested the molecule, N-(3-fluoro-1-phenethylpiperidin-4-yl)-N-phenylpropionamide (NFEPP), in rats with injured paws, but even at high doses, observed no overt sedation, respiratory depression or constipation, while fentanyl proved lethal.

Several other strategies are also in the mix, including polypharmacological approaches, where the drug binds to an additional target to reverse some of the unwanted effects. For example, Steven Husbands, a medicinal chemist at the University of Bath, UK, has developed a μ-receptor agonist, related to buprenorphine, BU08028^[5]
. It is also an agonist to the nociceptin receptor, another member of the opioid receptor family, found in the brain and involved in regulating emotion.

Another avenue is allosteric modulation, which uses molecules that can bind to alternative sites on the μ receptor to enhance and modify its activation. John Traynor, a pharmacologist at the University of Michigan, is working on this approach and has identified the first such molecule that could reduce the level of opioid needed, or even work in tandem with the body’s own endogenous opioid peptides^[6]
.

Are opioids always the answer?

While there are plenty of ideas for opioid-based medicines out there, there is no perfect solution. As Stein notes: “At the moment, none of these strategies has produced any compounds that have no side effects.” But even if we do find the perfect opioid, will it provide the pain relief patients are looking for? A study of 240 patients, published in March 2018, showed that opioids were no better at reducing pain in patients with chronic back, knee and hip problems than non-opioids^[7]
.

At the moment, none of these strategies has produced any compounds that have no side effects

“We know that [opioids] can be very useful for acute pain, in other words pain that is associated with either trauma or surgery or at the end of life, [but there is] a real lack of efficacy in longer term conditions,” says Knaggs. “Most of the [opioid] trials that have been conducted have been for no more than three to four months and, on that basis, effectiveness has been extrapolated over a very long period, which of course is not necessarily [always] the case.”

“There is no high quality evidence that opioids actually help [with chronic pain], and really that is the main issue,” Knaggs adds. “Chronic pain is a very complex beast and most medicines that we have only interact with one single receptor, or one part of that very complex neurological pathway, so it’s perhaps not surprising that this is the case.”

“I think we need a much bigger toolbox for managing pain rather than every person who comes in with an ache gets an opiate — that’s ridiculous, but that’s where we have been,” says Bohn. But new pain solutions, such as selective COX-2 inhibitors, have also proved to be problematic with serious cardiovascular side effects and many have been removed from the market.

I think we need a much bigger toolbox for managing pain rather than every person who comes in with an ache gets an opiate — that’s ridiculous, but that’s where we have been

Pasternak agrees that new targets do need to be considered, such as dopamine, serotonin and cannabinoid receptors. “For chronic pain, people are looking at nerve growth factor (NGF) blockers, either [targeting] receptors directly or [finding] antibody drugs that can sequester the NGF. They have shown that this works.”

But Knaggs suggests that pharmaceuticals may not be the answer for chronic pain at all. “When you experience pain for longer periods of time, the emotional and affective components become much more established and important, so actually dealing with those is probably equally important to providing pain relief.” Knaggs also suggests that a lack of physical activity often exacerbates a pain problem. “From what we know about persistent pain, it’s that sensitisation within the nervous system that is a problem and not ongoing tissue damage. The best way to deal with that sensitisation is to be as active as possible, but that is a very difficult message for a lot of patients.”

The extent of the opioid epidemic in the United States may not reach UK shores, but we certainly cannot be complacent. NHS Digital figures show that, in England, the number of patients admitted to hospital for overdosing on prescription opioid painkillers more than doubled between 2005–2006 and 2016–2017, from 4,891 to 11,660. The pharmacological challenge to create safer and less addictive opioid drugs remains important. “I’m pretty optimistic that there will be a way to find better analgesic drugs, maybe within the next five to ten years,” says Stein.

Until then, says Pasternak, we should remember that opioids “can be extraordinarily effective and provide pain relief in situations that cannot be treated any other way”.

“We have to be careful who we give [pain relief] to and how we give it to them; these drugs need to be used but need to be used appropriately and wisely.”