Ziconotide

Ziconotide is a synthetic copy of ω-conotoxin MVIIA from the fish-hunting snail Conus magus. It binds N-type voltage-gated calcium channels with high selectivity, and those channels sit on the presynaptic terminals of nociceptive C and A-delta fibres in the superficial dorsal horn. Blocking calcium entry there prevents release of substance P, glutamate and CGRP, which stops nociceptive transmission at the first synapse in the spinal cord. Because it works nowhere near the opioid receptor, it produces no tolerance, no physical dependence and no respiratory depression — the trade-off is that N-type channels are also present in brain circuits, which is why cerebellar and psychiatric side effects dominate. It is not orally, subcutaneously or intravenously active for pain; systemic dosing does not get meaningful drug into the CSF, which is why it only exists as an intrathecal infusion.

Targets: N-type voltage-gated calcium channel (Cav2.2 / CACNA1B), Dorsal horn primary afferent terminals, Substance P and glutamate release

• · Route is INTRATHECAL — into the cerebrospinal fluid via a surgically implanted programmable pump. The route enum on this site has no intrathecal entry; nothing about this drug is intravenous or subcutaneous, and systemic injection produces no analgesia. Note the unit: single-digit micrograms per DAY, roughly a thousand-fold below typical peptide dosing.
• · Slower titration is standard practice in pain clinics — many centres go up by 0.5-1 mcg/day per week because psychiatric side effects track the speed of escalation more than the absolute dose. Doses above 19.2 mcg/day are not recommended in the label.

Work out your exact syringe units →

• very commonConfusion, memory impairment and cognitive slowing— The most common reason people come off the drug.
• very commonDizziness, ataxia and nystagmus— Dose-dependent and reversible on dose reduction.
• very commonNausea and vomiting— Prominent during titration.
• commonHallucinations and frank psychosis— Can occur even without a psychiatric history; stop the infusion.
• commonElevated creatine kinase, occasionally with rhabdomyolysis— CK is checked periodically; myalgia plus a rising CK means stopping.
• uncommonNew or worsened depression and suicidal ideation— Carries a boxed warning in the US label.
• uncommonMeningitis or pump-pocket infection— A hazard of the delivery hardware rather than the molecule.

• A history of psychosis — this is an explicit contraindication in the label, not a caution.
• Any infection at the infusion site, spinal canal obstruction, or uncontrolled bleeding risk that makes intrathecal catheter placement unsafe.
• Self-administration of any kind. There is no home, subcutaneous or intravenous version of this drug; attempting it either does nothing or is dangerous.

• cautiondifelikefalin — Both can cause dizziness and mental clouding; combined use compounds the cognitive load without adding a distinct analgesic mechanism.
• cautiondermorphin — Ziconotide is often used precisely because opioids failed; adding a potent mu agonist reintroduces respiratory and dependence risk that ziconotide alone does not carry.

• · Cognitive and psychiatric status at every visit and after every dose increase — a simple orientation and mood check catches most problems early.
• · Serum creatine kinase periodically, especially in the first months.
• · Pump function, reservoir volume and catheter integrity at each refill.
• · Signs of meningitis: fever, headache, neck stiffness, altered mental state.

Approved prescription drug in the US and EU, restricted to intrathecal use in specialist pain services. It is not a controlled substance because it has no abuse liability, but it is not obtainable or usable outside a clinical setting.

• Staats et al. 2004, JAMA — intrathecal ziconotide for refractory cancer and AIDS pain (trial)
• Rauck et al. 2006, slow-titration ziconotide in severe chronic pain (trial)
• Prialt (ziconotide) FDA prescribing information (label)

The reason ziconotide behaves so differently from every other intrathecal analgesic is that it is not a receptor agonist at all. It is a pore blocker. Cav2.2 (N-type) calcium channels cluster on the presynaptic terminals of nociceptive C and A-delta fibres in laminae I and II of the dorsal horn. When an action potential arrives, Cav2.2 opens, calcium floods the terminal, and that calcium transient triggers vesicular release of glutamate, substance P and CGRP onto the second-order projection neuron. Ziconotide sits in the outer vestibule of the channel and occludes it. No calcium, no vesicle fusion, no transmitter, no pain signal handed upward. Two consequences follow directly and explain the whole clinical picture. First, because none of this involves a G-protein-coupled receptor, there is no receptor desensitisation, no beta-arrestin recruitment, no compensatory upregulation of adenylyl cyclase on withdrawal — hence no tolerance, no dependence and no withdrawal syndrome. You can stop the infusion abruptly and the pain returns, but the patient does not get sick. Second, Cav2.2 is not confined to the dorsal horn. It is expressed throughout the CNS, densely in cerebellum, hippocampus and the monoaminergic nuclei, where it controls noradrenaline and dopamine release. Ziconotide diffusing rostrally in the CSF blocks those channels too, and that — not any exotic mechanism — is the entire origin of the ataxia, nystagmus, confusion, memory loss and psychosis. The neuropsychiatric burden is not a side effect in the usual sense; it is the same mechanism acting on the wrong neurons, which is why it tracks the CSF concentration gradient and therefore the rate of dose escalation. Binding is also slow to reverse, so effects lag dose changes and psychiatric events can keep worsening for a day or more after an increase. That kinetic fact is why waiting longer between steps is the single most effective safety intervention with this drug.

Almost everything that goes wrong with ziconotide goes wrong because someone escalated too fast. The label permits increases two to three times a week; that schedule was used in the early trials and produced the hallucination and psychosis rates that define the drug's reputation. Titrate weekly in half-microgram steps instead and most of that disappears. The second failure mode is misattribution: a patient with an intrathecal pump becomes confused, gets labelled as having a ziconotide side effect, the dose is cut, nobody samples the CSF — and it was meningitis. Fever plus headache plus new confusion is a lumbar puncture, not a pump reprogramming. The third is the CK problem, which is silent until it is not; people skip the CK because the patient feels fine, and rhabdomyolysis announces itself as dark urine. The fourth is expectation. Ziconotide relieves refractory non-cancer pain by a mean of roughly 15%, which is a real effect and also a disappointing one to someone told it is their last option. Patients who go in expecting to be pain-free stop the drug; patients who go in expecting a modest improvement after a hard titration often stay on it for years. And finally the fatal category error that shows up in grey-market contexts: ziconotide is not a subcutaneous peptide. Injecting it under the skin achieves nothing analgesic because it does not cross into CSF, and the vial contents are dosed for a compartment of 140 mL. There is no home version of this drug.

1. 2.4 mcgDay 1 — 2.4 micrograms per day, which is 0.1 mcg per hour on the pump. The label is explicit that you start at no more than this. Note the scale: micrograms per day, not per dose.
2. 4.8 mcgDays 3-7 — Increments of up to 2.4 mcg/day, no more often than two to three times a week. Most pain services move more slowly than the label permits.
3. 9.6 mcgWeek 2 — Two further increments. This is where cognitive and vestibular symptoms typically declare themselves. Assess orientation, memory and gait before every step, not after.
4. 14.4 mcgWeek 3 — Analgesia in the pivotal trials was already partly present well below the ceiling. If pain has responded acceptably here, stopping here is a legitimate endpoint.
5. 19.2 mcgWeek 3-4 — 0.8 mcg/hr, the maximum recommended dose in the US label. Going higher is off-label and buys little except psychiatric risk — the dose-response for side effects is steeper than the dose-response for analgesia above this point.
6. —Alternative slow ladder, weeks 1-12 — The protocol many specialist centres actually use: start at 0.5-1 mcg/day and increase by 0.5-1 mcg/day no more than once a week. It takes two to three months to reach a therapeutic dose instead of three weeks, and it converts ziconotide from a drug most patients cannot tolerate into one most can. Rauck 2006 is the evidence that slow titration preserves efficacy while cutting adverse events.

• N-type voltage-gated calcium channel, Cav2.2 (CACNA1B), presynaptic — Sub-nanomolar at Cav2.2, with selectivity over L-type, P/Q-type and T-type channels large enough that the other subtypes are not clinically relevant. The US label publishes no Kd, and the figures circulating in reviews come from rat brain synaptosome binding rather than human channel, so no precise number is asserted here.

  Physical occlusion of the channel pore. Calcium entry into the nociceptive terminal is prevented, so the calcium-dependent vesicle release machinery never fires.

• Substance P, glutamate and CGRP release from primary afferent terminals — Not a binding target — a downstream consequence.

  Release is suppressed. CSF substance P falls during ziconotide infusion, which is the closest thing this drug has to a pharmacodynamic biomarker.

• Cav2.2 in cerebellum, hippocampus and monoaminergic nuclei — Same channel, same affinity — there is no subtype selectivity to exploit.

  Off-target in the therapeutic sense and the direct cause of ataxia, nystagmus, confusion, hallucination and psychosis. Rostral CSF spread rather than receptor promiscuity determines how much of this a given patient gets.

• Mu, delta and kappa opioid receptors — None. Ziconotide has no measurable opioid receptor activity.

  Nothing — and this absence is the point. It is why naloxone does not reverse ziconotide, why there is no respiratory depression, and why it still works in patients whose opioid receptors are thoroughly tolerant.

• Staats et al. — intrathecal ziconotide for refractory pain in cancer and AIDS Phase 3, randomised, double-blind, placebo-controlled (2:1) · n=111 · 2 weeks · 2004

  Mean percentage change in visual analogue pain intensity: 53.1% improvement with ziconotide versus 18.1% with placebo (p<0.001). This is the trial the FDA approval rests on, and the effect size is genuinely large — but it was run with fast titration, which is why the adverse-event rate looks so bad in the same paper.

• Rauck et al. — slow-titration intrathecal ziconotide in severe chronic pain Phase 3, randomised, double-blind, placebo-controlled · n=220 · 3 weeks · 2006

  Mean percentage change in VASPI at week 3: 14.7% with ziconotide versus 7.2% with placebo (p=0.036). A far more modest effect than the cancer trial, obtained with slower titration and a lower mean dose — and it is the more honest picture of what ziconotide does for chronic non-cancer pain. Adverse events were substantially lower than in the earlier fast-titration studies.

Hours after a dose increase some patients notice a shift in pain — the CSF half-life is 4.6 hours, so a rate change reaches its new steady state within about a day. But that first-day change is small, because the starting dose is deliberately sub-therapeutic. Meaningful analgesia appears over the first two to four weeks of titration on the label schedule, or six to twelve weeks on the slow schedule most specialist centres use. Side effects run on a different and slightly delayed clock: dizziness and nausea appear within a day or two of an increase, while cognitive and psychiatric symptoms often build over the following three to seven days, which is exactly why increasing every two days is a trap. Stopping works immediately in the other direction — abrupt discontinuation produces no withdrawal syndrome at all, just the return of pain over roughly 24 hours, and cognitive side effects typically clear within a few days to two weeks of reducing or stopping.

Ziconotide is one of the few compounds on this site where combination therapy is standard clinical practice rather than a forum idea, and it is done inside the pump. Intrathecal morphine or hydromorphone plus ziconotide is a common admixture: the mechanisms are completely non-overlapping, and adding ziconotide often lets the intrathecal opioid dose come down, which matters because intrathecal opioid escalation is what drives catheter-tip granuloma formation. Clonidine and bupivacaine are also combined with it. What none of that licenses is stacking ziconotide with anything you would inject at home. Adding systemic opioids, gabapentinoids, benzodiazepines or alcohol compounds the cognitive load without adding a mechanism the spinal cord is missing, and confusion is already the single most common reason people abandon this drug. If you are on ziconotide and something else is making you foggy, the something else is what to remove first — you have far more headroom to cut a gabapentinoid than to cut a drug that took three months to titrate.

Against intrathecal morphine, the direct comparator, ziconotide wins on tolerance (none versus substantial), dependence (none versus certain), respiratory depression (none versus dose-limiting) and catheter-tip granuloma (a known intrathecal opioid complication that ziconotide does not cause). It loses badly on cognitive and psychiatric tolerability, on speed of onset, and on the sheer practicality of titration. That trade is why consensus guidance places ziconotide as a first-line intrathecal agent for neuropathic and mixed pain despite everyone knowing how hard it is to run. Against everything else in this class on the site it is in a different category entirely: ziconotide has FDA and EMA approval and two positive randomised controlled trials, while dermorphin, DALDA, the enkephalins and the endomorphins have no human dosing data at all. The comparison worth internalising is that this is what a peptide analgesic looks like when it is developed properly — narrow indication, hospital-only delivery, modest effect size, well-characterised toxicity — and it is a useful yardstick for how far the research-chemical opioid peptides are from that standard.

Not priced here because I could not verify current figures in this session, and the honest answer is that the drug cost is the small part. Ziconotide is delivered by a surgically implanted programmable pump: an implantation procedure, periodic clinic refills, and pump replacement every several years when the battery expires. In any real-world setting this is a hospital-billed therapy in the tens of thousands of dollars for the first year, dominated by hardware and procedures rather than by the peptide. There is no self-pay grey-market version and no meaningful monthly figure to quote.

• ?The disulfide connectivity given here (Cys1-Cys16, Cys8-Cys20, Cys15-Cys25) is the standard inhibitor cystine knot arrangement reported for omega-conotoxin MVIIA. I confirmed the primary sequence and C-terminal amide in this session but did not independently re-verify the bond pairings.
• ?No Kd or IC50 for ziconotide at human Cav2.2 is quoted, because the figures in circulation derive from rat synaptosomal binding studies I did not resolve in this session.
• ?Time to steady state is given as approximately one day, derived from the 4.6-hour CSF half-life on a fixed infusion rate, not from a stated label value.
• ?Reported incidence of CK elevation and of rhabdomyolysis varies widely between the label, the trials and post-marketing series, so no single percentage is given.
• ?The claim that CSF substance P falls during ziconotide infusion comes from the mechanistic literature rather than a source I resolved in this session.
• ?Cost figures are omitted entirely rather than estimated, because pump hardware and procedure costs dominate and I could not verify current pricing.

• Intrathecal ziconotide in the treatment of refractory pain in patients with cancer or AIDS: a randomized controlled trial Staats PS, Yearwood T, Charapata SG, et al., JAMA, 2004 · PMID 14709577

  The registration trial. 111 patients, 2:1 randomisation, 53.1% versus 18.1% pain improvement. Read the adverse-event table alongside the efficacy result — this study established both that ziconotide works and that it is hard to tolerate.

• A randomized, double-blind, placebo-controlled study of intrathecal ziconotide in adults with severe chronic pain Rauck RL, Wallace MS, Leong MS, et al., Journal of Pain and Symptom Management, 2006 · PMID 16716870

  220 patients, slower titration, smaller effect, far better tolerability. The single most useful paper for understanding why titration speed is the whole argument with this drug.

• Pharmacokinetics and pharmacodynamics of intrathecal ziconotide in chronic pain patients Wermeling D, Drass M, Ellis D, et al., Journal of Clinical Pharmacology, 2003 · PMID 12817525

  The human CSF pharmacokinetic study — 1 to 10 mcg intrathecal doses in 22 chronic pain patients. Source of the observation that higher CSF concentrations track analgesia, and of the CSF half-life figures the label quotes.

• Pharmacokinetic analysis of ziconotide (SNX-111), an intrathecal N-type calcium channel blocking analgesic, delivered by bolus and infusion in the dog Yaksh TL, de Kater A, Dean R, et al., Neuromodulation, 2012 · PMID 22748108

  The cleanest description of intrathecal distribution kinetics — bolus versus infusion, rostral spread, and why the drug behaves as though CSF bulk flow is its clearance mechanism.

• PRIALT (ziconotide acetate) injection, solution — US prescribing information TerSera Therapeutics LLC, DailyMed, US National Library of Medicine

  Source of every pharmacokinetic and titration number quoted here: CSF volume of distribution approximately 140 mL, CSF clearance 0.3-0.4 mL/min, CSF terminal half-life 4.6 hours (range 2.9-6.5), serum half-life 1.3 hours, approximately 50% plasma protein binding, under 1% renal recovery, and the 2.4 to 19.2 mcg/day dose range. Also carries the boxed warning about psychosis.