Pain researchers must learn from the opioid crisis

    0
    Pain researchers must learn from the opioid crisis


    In 2022, more than 80,000 people died of an opioid overdose in the United States. To reduce this toll, we need to abandon analgesic drugs that act through the µ-opioid receptor and identify painkillers that cannot be so easily abused. Fortunately, developments in how we can evaluate the efficacy and adverse effects of drugs are making this possible.

    In 1985, the World Health Organization (WHO) proposed an ‘analgesic ladder’ to provide pain relief for people with cancer, it recommended non-steroidal anti-inflammatory drugs (NSAIDs) for mild pain; a combination of an NSAID and a weak opioid for moderate pain; and strong opioids for severe pain1. Although well-intentioned, this framework unfortunately promoted the use of opioids by providing access to these addictive substances.

    There were three fundamental flaws to the WHO strategy. The first was the assumption that the severity of pain is the only important factor to consider when selecting a suitable treatment. The second was the presumption that µ-opioid receptor agonists would be effective for all types of severe acute or chronic pain. And the third problem was the expectation that prescription opioids could be distributed safely.

    After almost four decades, the WHO ladder remains widely used by clinicians — even though we now appreciate that opioids are highly dangerous. Scientists also recognize that there are several distinct types of pain, each driven by different mechanisms and requiring specific distinct interventions. These include nociceptive pain, which is initiated by tissue damaging stimuli; inflammatory pain that is associated with the activation and recruitment of immune cells; neuropathic pain owing to damage to the nervous system; and nociplastic pain, which reflects pathological dysfunction of the nervous system.

    Researchers have now identified sensory neurons in the peripheral nervous system that trigger pain. They have also discovered immune drivers of pain2, as well as the specific circuits in the central nervous system that generate the sensation of pain and the modulatory circuits that suppress pain3. Scientists have also worked out how these circuits can change to drive persistent clinical pain, which is pathological and quite distinct from the protective aspects of acute nociceptive pain that warns of danger in the environment. We now have sufficient understanding of pain mechanisms to identify analgesic drugs that can reduce pathological pain without eliminating the protective function of acute physiological pain.

    With these pieces in place, the key question becomes: what is the best way to set up an effective and efficient analgesic discovery platform that will identify the sort of drugs that are so urgently needed?

    The standard industry approach is to identify a single target that might contribute to a disease state; validate it by conducting gain- and loss-of-function manipulations in animal models and in vitro assays, and by using human genetics; and then run screens for compounds that act only on the target. (A target can be a receptor, ion channel or enzyme.) Unfortunately, this has not worked well; there have been few approvals of analgesics acting on new targets over the past decade. One reason for this is that compounds that act on multiple targets might be required.

    Clearly, an alternative strategy is needed. One option with great promise is a technique called phenotypic screening, which identifies drugs with cell-selective or disease-modifying activity. This method can detect a particular observable characteristic of a cell, such as its excitability. This is possible because human induced pluripotent stem cells can be differentiated into any cell type and used in diverse ways to model and measure disease states.

    Scientists can generate human nociceptors — the sensory neurons that trigger pain in response to noxious stimuli — to study pain in a dish4. In unpublished work, researchers have been able to screen for compounds that selectively reduce activity only in these neurons without affecting other excitable cells, thereby silencing peripheral pain triggers. Furthermore, researchers can run phenotypic screens to identify analgesics that reduce the heightened sensitivity of nociceptors. An important element of these phenotypic screens is that, because they use human cells, there is no risk of wasting development resources on compounds that won’t work in people.

    In conventional drug discovery for medicinal chemistry, compound profiles (or hits) are designed on the basis of the interaction between the compound and the active side of the target protein. By contrast, phenotypic screens will need to be developed using different methods because of the lack of a single target in which the structure and interaction with drugs can be directly determined. Among the most promising technologies is artificial-intelligence-tool-based drug design.

    Machine learning is also assisting efforts to identify analgesics by helping to assess behaviour in rodents, enabling identification of reliable surrogates both of clinical pain states and adverse effects, as well as identifying drugs that are likely to be misused5.

    This strategy will provide us with the means to identify effective and safe analgesics, and in turn permit us to at last abandon prescription opioids.

    Competing Interests

    The author declares no competing interests.



    Source link