Mellar P. Davis, a palliative oncologist at the Levine Cancer Institute of Atrium Health in Charlotte, North Carolina, does not call for abandoning scheduled dosing. Instead, his analysis weaves together experimental psychology, addiction neuroscience, and clinical prescribing data to advance a single candidate principle for opioid stewardship in cancer survivorship: whenever clinically feasible, opioid exposure should remain linked to meaningful symptom relief. The idea is that pain-contingent use, in which a dose follows the arrival of pain rather than the arrival of a scheduled hour, may help preserve the distinction between therapeutic use and the beginnings of problematic use.

The biological logic behind the proposal comes from a well-documented phenomenon in behavioral pharmacology known as negative reinforcement. In people who are not addicted, opioid self-administration tracks the presence of pain: the drug is taken because it relieves something aversive, and the relief itself reinforces the behavior. Classic laboratory work supports this coupling. In studies of arthritic rats, opiate self-administration rose and fell with the intensity of chronic nociceptive pain, functioning almost as a readout of the animals’ discomfort. Human experiments point the same way: in healthy volunteers given fentanyl, a cold-water immersion stressor changed the drug’s reinforcing and subjective effects, suggesting that distress and pain modulate how rewarding an opioid feels.

Clinical studies in people with chronic pain echo the pattern. Research by Von Korff and colleagues compared time-scheduled with pain-contingent opioid dosing in chronic opioid therapy and found no clear advantage for fixed schedules, while daily-life monitoring studies have shown that prescription opioid use tends to co-occur with, and lag behind, spikes in pain and negative affect. In other words, in the therapeutic setting, taking the drug when it hurts is the normal, expected behavior, and it is the behavior that addiction neuroscience would predict in people whose brains have not been reshaped by dependence.

The trouble, Davis argues, is that this coupling is not permanent. With chronic use, the link between pain and opioid-taking weakens. Neuroadaptations described by Volkow, Koob, and other addiction researchers, including opponent-process mechanisms, escalating negative affect between doses, and the recruitment of stress and anti-reward circuitry, emerge with regular exposure itself, independent of whether pain is present. Over time, opioid-taking can become progressively pain-independent, driven instead by relief of withdrawal-like states or by conditioned cues. In the framework of Koob’s negative reinforcement model, the drug that once relieved pain now relieves the discomfort of its own absence, a transition that blurs the line between analgesia and addiction.

This matters for oncology because the patient population has changed. Around-the-clock dosing was designed for continuous, severe pain in people with limited life expectancy, when the dominant clinical failure was undertreatment and the risk of long-term dependence was largely moot. Contemporary cancer care produces prolonged survival and chronic treatment-related pain: neuropathy from chemotherapy, postsurgical syndromes, radiation fibrosis, and musculoskeletal sequelae of endocrine therapy. Studies of opioid prescribing among older cancer survivors have raised safety concerns, and the concept of complex persistent opioid dependence, described by Manhapra, Ballantyne, and colleagues, captures the gray zone in which long-term opioid therapy creates dependence that is neither classic addiction nor straightforward analgesia.

The evidence base for scheduled versus as-needed dosing in cancer pain itself is thinner than the doctrine suggests. A systematic review and meta-analysis by Edler-Buggy and colleagues comparing regular with as-needed opioid dosing for chronic cancer pain found no clear superiority of regular dosing on the outcomes examined. Meanwhile, current guidelines from the CDC, the American Society of Clinical Oncology, and the National Comprehensive Cancer Network already endorse principles that align with Davis’s proposal: careful patient selection, risk stratification, monitoring, and attention to whether opioids continue to provide functional benefit. The commentary is therefore largely guideline-concordant, offering a unifying rationale rather than a radical departure.

Davis is explicit about the limits of the framework. Pain-contingency is difficult to measure directly in clinical practice; patients cannot always distinguish pain relief from relief of anxiety or dysphoria, and electronic prescribing records rarely capture the temporal relationship between symptoms and doses. Moreover, because relevant neuroadaptation begins with regular use itself, preserving pain-contingency cannot eliminate risk; it can only serve as a marker and a moderating strategy. The framework also does not oppose around-the-clock dosing for continuous severe cancer pain, where scheduled administration remains appropriate and humane. What it offers is a candidate principle, not a proven intervention, for keeping opioid exposure tethered to symptom relief in the growing population of survivors.

The implications reach beyond oncology. If the coupling between pain and opioid-taking is a behavioral signature that distinguishes therapeutic use from escalation toward addiction, then clinical monitoring could, in principle, watch for its erosion: doses requested well before scheduled intervals without reported pain, use that persists as pain resolves, or escalating consumption unaccompanied by functional improvement. Recent work on avoidance learning in chronic opioid users, reported by Geissler and colleagues in JAMA Psychiatry, adds to the picture that long-term exposure changes decision-making circuitry in measurable ways. Consensus-based guidance for patients with advanced cancer and opioid misuse or use disorder, published by Fitzgerald Jones and colleagues in JAMA Oncology, already emphasizes structured assessment; a pain-contingency lens could sharpen those tools.

None of this diminishes the moral core of the original cancer pain movement, which correctly identified untreated pain as a scourge and scheduled opioids as its remedy. Davis’s reappraisal is best read as an update for a new era: the same drugs that freed dying patients from agony are now dispensed for years to survivors whose pain is chronic, fluctuating, and entangled with the vulnerabilities of long-term exposure. In that setting, the timing of a dose may carry information that the dose itself does not. Keeping opioids linked to pain, whenever clinically feasible, is a modest, testable idea, and one that could help clinicians honor both halves of their duty, relieving suffering while protecting the brain’s capacity to tell the difference between relief and craving.

Subject of Research: Pain-contingent opioid dosing and addiction risk in cancer pain management

Article Title: Pain-contingent opioid use and addiction risk in cancer pain: a reappraisal of around-the-clock dosing

Article References: Davis, M. P. (2026). Pain-contingent opioid use and addiction risk in cancer pain: a reappraisal of around-the-clock dosing. Supportive Care in Cancer, 34(11), Article 1086. https://doi.org/10.1007/s00520-026-11321-6

Image Credits: AI Generated

DOI: 10.1007/s00520-026-11321-6

Keywords: cancer pain, opioid stewardship, around-the-clock dosing, pain-contingent opioid use, addiction, negative reinforcement, cancer survivorship, opioid dependence, palliative care, chronic pain, behavioral pharmacology, clinical guidelines