Efficacy and safety of the Ca_v2.2 blocker Ziconotide in pain: a meta-analysis and systematic review

Supplemental Digital Content is Available in the Text.

Evidence suggests Ziconotide's analgesic promise, but adverse effects and study limitations temper confidence, underscoring the need for rigorous trials before broad clinical adoption.

Abstract

Chronic pain remains a significant clinical challenge, especially in refractory cases. Ziconotide, a selective Ca_v2.2 channel blocker, offers an intrathecal approach, but concerns about its safety, potential biases, and impact on opioid consumption persist. Following Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, we conducted a systematic review and meta-analysis of 23 studies (1,531 participants) to evaluate the efficacy, safety, and secondary outcomes of Ziconotide. Key outcomes were pain intensity reduction, adverse events, and serious adverse events. As secondary outcomes, we assessed changes in opioid consumption. The risk of bias and the certainty of evidence were assessed using risk of bias 2 and Grading of Recommendations, Assessment, Development and Evaluation for randomized controlled trials, ROBINS-I and the Newcastle-Ottawa Scale for observational studies, and the Joanna Briggs Institute checklist for case reports. Publication bias was explored using funnel plot analysis. Ziconotide demonstrated significant pain reduction compared with placebo (mean difference: −22.54, 95% confidence interval [CI]: −36.70 to −8.38, P = 0.002), although heterogeneity was high. Adverse events were frequent (94.9% vs 76.9% in placebo; risk ratio 1.24, 95% CI: 1.09–1.41, P = 0.0008), with serious adverse events reported in 17.85% patients (risk ratio 2.63, 95% CI: 1.52–4.57, P = 0.0006). Secondary outcomes suggested potential reductions in opioid consumption in observational studies, with decreases ranging from 6.4% to 91.5%, though randomized trials showed inconsistent results. Certainty of the evidence was rated as low to moderate. Although Ziconotide shows promise as an intrathecal treatment for refractory pain, its frequent adverse effects, the availability of high-certainty evidence, and inconclusive impact on opioid consumption highlight the need for cautious use. This meta-analysis underscores the need for future research.

1. Introduction

The management of chronic pain represents one of the numerous challenges in present-day medical practice.¹¹ This field is characterized by a paradoxical scenario where diverse treatments exhibit both efficacy and ineffectiveness in the literature, highlighting the complex nature of pain management and the consequential implications associated with the use of prevalent drugs.^11,40 In particular, opioids have become the subject of substantial debate, with criticisms extending to nonopioid analgesics. Consequently, there is an emphasis on the critical need to explore new horizons that should aim not only to alleviate pain but also to address key issues such as drug tolerance and addiction features.^11,47,61 The use of biologically derived toxins, specifically those found in the venoms of various organisms, has gained traction in biomedical research due to their specificity.⁹ An example of this is Ziconotide (Prialt), a peptide toxin derived from the venom of Conus magnus (a marine snail that preys on fish) that has been approved for clinical use in the management of severe pain by the US Food and Drug Administration (2004) and the European Medicines Agency (2005).⁴⁸

Ziconotide is a synthetic version of the ω-conotoxin MVIIA (ω-MVIIA) peptide, which belongs to a broader family of ω-conotoxins, distinguished by peptides comprising 24 to 29 amino acids, including 6 cysteine residues that structurally arrange into 4 loops.^36,49 Within this diverse family, a range of conotoxins shows selective affinity for various ion channels. MVIIA is one of the most extensively studied, known for its specific targeting of certain voltage-gated calcium channels (VGCCs), specifically the N-type Ca_v 2.2 channel. Voltage-gated calcium channels are expressed in different muscular and nervous cells and regulate calcium signaling in response to voltage fluctuations. They serve as transducers of electrical activity, playing a significant role in neurotransmission at the presynaptic level.⁸ The activation of Ca_v channels upon membrane depolarization hinges on the presence of α1 subunits in their composition. These subunits are crucial for pore formation and confer Ca_v channels with distinct biophysical and pharmacological properties, leading to a broad spectrum of VGCCs. Consequently, they represent a diverse family, with various types of channels distinguished by the thresholds required for activation (high- and low-threshold VGCCs) and their sensitivity to pharmacological inhibition. This includes L-type channels (Ca_v1.1–1.4; DHPs), P/Q-type (Ca_v2.1; ω-Agatoxin), N-type (Ca_v2.2; ω-CTx-GVIA), R-type (Ca_v2.3; SNX-482), and T-type channels (Ca_v3.1, 3.2, 3.3).⁷

N-type Ca_v 2.2 channels are crucial players in neurotransmission. Positioned presynaptically, they facilitate calcium influx upon neuronal action potential arrival and depolarization, which in turn triggers vesicular fusion with the membrane mediated by the SNARE complex (syntaxin, SNAP-25, & VAMP/synaptobrevin).⁵⁷ Beyond this, there exists a direct interaction between some Ca_v and SNARE proteins, mediated by specific intracellular loops.⁵² Hence, they are vital in neurotransmitter exocytosis, where most synapses utilize Ca_v channels to orchestrate the synchronous release of vesicles in response to incoming action potentials.²⁴ In the context of pain, these channels are notably expressed at key locations associated with somatosensory processing and nociception. This includes areas such as the dorsal horn of the spinal cord—laminas I and II—and the dorsal root ganglia, where there is a marked colocalization between N-type Ca_v 2.2 channels and glutamatergic neurotransmission, substance P, and calcitonin gene-related peptide.³⁹ Such findings underscore their significance as a therapeutic target in the pharmacological treatment of pain.³¹ As the ω-conotoxin MVIIA Ziconotide positions itself just above the receptor's selectivity filter, it effectively blocks the pore, thereby preventing the influx of calcium from the extracellular space.¹⁴ As a result, Ziconotide can reduce the release of presynaptic vesicles and lessen nociceptive neurotransmission, thus influencing the modulation of pain perception. Importantly, the underlying structural interaction between Ca_v 2.2 channels and Ziconotide has been recently elucidated through cryo-electron microscopy.^14,16 This represents a pivotal discovery, providing mechanistic insight into both our understanding and the future development of pain management pharmaceuticals.

The role of Ca_v2.2 channels and Ziconotide in pain management has been evidenced through both preclinical and clinical research. Concerning the channel, the ablation of Ca_v2.2 in murine models leads to reduced sensitivity to neuropathic and inflammatory pain.²³ In addition, inhibiting Ca_v2.2 channels diminishes the onset of thermal, mechanical, and inflammatory pain.^30,37 These observations have been effectively translated into clinical investigations. Over the past 2 decades, both randomized clinical trials and observational studies have underscored its efficacy in treating severe pain, with substantial pain reduction.^45,53,62 Nevertheless, its administration is viable only through the intrathecal route, as it cannot cross the blood-brain barrier. Consequently, its use is primarily reserved for severe chronic pain cases unresponsive to other treatments, in alignment with current guidelines.¹³ In this scenario, intrathecal administration offers distinct advantages regarding pharmacokinetics and dose-effect correlation, with the drug being directly released into the cerebrospinal fluid (CSF).^67,71 In such cases, intrathecal therapy is employed, involving an implanted pump connected through a catheter to the CSF, authorized for morphine, baclofen, and Ziconotide.¹³ However, the widespread adoption of Ziconotide is limited in many countries, and there is a notable gap in high-quality research, despite its potential benefits. These include its synergistic combination with morphine or its use in monotherapy to mitigate issues such as tolerance or the induction of long-term hyperalgesia, among others.^{10,11,47,60,61} In addition, preclinical studies indicate a higher hypoalgesic effect of Ziconotide compared with morphine, an aspect yet to be proven in humans through interventional comparative research.³²

Concerns regarding Ziconotide's safety profile, the limited number of high-quality clinical trials, and the heterogeneity of available evidence have hindered a broader understanding of its clinical potential. Although previous reviews have provided valuable insights, further comprehensive evaluations remain necessary to incorporate evidence and address clinically relevant outcomes. Currently, there is an ongoing controversy among clinicians, with some advocating against its use due to safety concerns, while others support its application in selected cases. In this context, this study offers a synthesis of both efficacy and safety data, extending the scope of previous analyses. In addition, secondary outcomes such as reductions in oral opioid consumption and the relationship between dosage and adverse events have been explored.

2. Methods

2.1. Protocol and registration

This investigation adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and checklist, in conjunction with the directives outlined in the Cochrane Handbook for Systematic Reviews of Interventions.^20,38 Furthermore, it has been duly registered in the International Prospective Register of Systematic Reviews (PROSPERO) (ID: CRD42024497261).

2.2. Eligibility criteria

This systematic review and meta-analysis included studies meeting the following eligibility criteria: (1) Randomized/nonrandomized trials and observational studies; (2) studies involving individuals suffering from severe chronic pain, which includes neuropathic pain, cancer-related pain, and noncancer-related pain; (3) studies evaluating the administration of intrathecal Ziconotide as monotherapy; and (4) studies reporting on any of the key outcomes related to the efficacy, effectiveness, safety, or tolerability of Ziconotide. We excluded studies that (1) involved intrathecal administration of Ziconotide in combination with other intrathecal drugs; (2) did not report on the specified outcomes of interest; (3) were not published in English; or (4) had overlapping patient populations.

In our approach, we prioritized a comprehensive understanding of Ziconotide's impact by including diverse study designs and imposing no restrictions on follow-up durations, enabling an inclusive assessment of both its short-term and long-term effects. This wide inclusion criterion was established based on key considerations during availability assessment: (1) the urgent need for a meta-analysis in this area of research, (2) the relative lack of evidence within this specific domain, and (3) the significance and clinical repercussions that may arise in the management of severe chronic pain. Only studies meeting these eligibility criteria were included in the systematic review. However, inclusion in quantitative meta-analysis was further contingent on the availability of appropriate statistical data (eg, pain scores, standard deviations, event rates, etc).

2.3. Search strategy and study selection

A systematic search was conducted across 3 databases in September 2024: PubMed, Web of Science, and the Cochrane Central Register of Controlled Trials. The search strategy was organized by dividing it into 2 key elements: population and intervention. For the population component, a comprehensive list of keywords pertaining to various pain conditions was included, linked by the Boolean operator “OR.” These terms included: “pain,” “chronic pain,” “severe pain,” “persistent pain,” “neuropathic pain,” “nociceptive pain,” “somatic pain,” “visceral pain,” “cancer pain,” “non-cancer pain,” “acute pain,” “postoperative pain,” “musculoskeletal pain,” “inflammatory pain,” “malignant pain,” and “breakthrough pain.” In the intervention component, terms related to Ziconotide were contemplated, such as “Ziconotide,” “Ziconotida,” “Prialt,” “SNX-111,” “calcium channel blocker,” and “N-type calcium channel antagonist.” Both components of the equation were combined using the Boolean operator “AND” (detailed in Supplementary Material 1, http://links.lww.com/PR9/A357). Databases search strategy and posterior study selection were conducted by both J.P. and A.C. independently using the Reference Manager Zotero v6.0 (George Mason University, 2023). Any disagreements that arose were resolved through consensus. When further deliberation was needed, a third reviewer—who had not participated in the initial screening—was consulted to provide an impartial judgment.

2.4. Data extraction

Data extraction and quality assessment were conducted by 2 independent authors. Discrepancies were resolved by mutual agreement. In cases where consensus could not be reached, an independent third reviewer not involved in the extraction assisted in reaching a final decision.

2.5. Quality assessment and qualitative analysis

Randomized clinical trials were evaluated by J.P. and A.C. using the Cochrane Collaboration's tool for assessing risk of bias 2 (RoB2).¹⁹ Studies were appraised for high, low, or unclear risk of bias across 5 domains: selection, performance, detection, attrition, and reporting biases. In addition, potential publication bias was assessed through funnel plot analysis of primary outcomes. Open-label trials were evaluated using the Risk of Bias in Nonrandomized Studies of Interventions (ROBINS-I) tool, which categorizes risk as low, moderate, serious, critical, or no information across 7 distinct domains, deeming a study as low risk of bias only if all domains are rated as low.⁵⁵ Prospective cohort studies and case series reports were assessed using the Newcastle-Ottawa Scale, where each study was awarded 0 to 9 stars based on methodological quality concerning participant selection, group comparability, and outcome assessment.⁶⁶ Finally, case reports were appraised using the Joanna Briggs Institute (JBI) Critical Appraisal Checklist.³⁵ In addition, all included studies were qualitatively analyzed according to the 5 domains (study design, imprecision, indirectness, heterogeneity, and publication bias) of the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) scale.¹⁸ According to this analysis, the evidence was categorized and rated as (1) high certainty, (2) moderate certainty, (3) low certainty, and (4) very low certainty.

2.6. End points and subgroup analysis

The primary outcome of interest was the mean change in pain, assessed by the Visual Analogue Scale of Pain Intensity or Numerical Rating Scale (NRS), and pain relief as measured by the Categorical Pain Relief Scale (CPRS). In addition, other essential outcomes included the incidence of any adverse events, serious adverse events, and all-cause mortality, as well as the reduction in concomitant non-intrathecal opioid consumption and the number of treatment responders. For a comprehensive analysis of these outcomes, we performed subgroup analyses comparing (1) data from randomized controlled trials (RCTs) and (2) data from non-RCT studies.

2.7. Statistical analysis

The meta-analysis was conducted in accordance with Cochrane methodological recommendations and reported following PRISMA guidelines. Comparisons for binary endpoints were made using odds ratios (OR) or risk ratios (RR) with a 95% confidence interval (CI). Continuous variables were compared using weighted mean differences with a 95% CI. For the observational studies, a proportional meta-analysis of single-arm for dichotomous variables was conducted using R. As there were data points close to 0 and 1, they were transformed according to the Freeman-Tukey Double arcsine transformation. Anticipating heterogeneity among the studies, the random-effects models of DerSimonian and Laird were utilized for all meta-analysis. Heterogeneity was examined using the Cochrane Q test and the inconsistency index (I²). To evaluate publication bias, a funnel plot of primary outcomes was carried.^2,56 The mentioned analysis was performed using both Review Manager 5.1 (Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark) and R Studio (PBC, Boston, MA). Data visualization and figures were refined with Adobe Illustrator (v2021, Adobe Inc., San Jose, CA).

3. Results

3.1. Study selection

The search strategy across PubMed, Cochrane, and Web of Science yielded 1,859 records. After removing duplicates (n = 509) and excluding nonrelevant studies based on title and/or abstract screening (n = 1,296), 54 articles were selected for full-text review, as illustrated in the PRISMA flow diagram (Fig. 1). Of these, 23 studies were ultimately included in the qualitative and quantitative analysis. A total of 31 studies were excluded at the full-text stage for the following reasons: overlapping populations (n = 4), non–peer-reviewed sources (n = 8), intervention not meeting eligibility criteria (n = 10), and other specific reasons (n = 9). The complete list of excluded full-text studies, along with the specific reason for exclusion of each, is provided in Supplementary Material 2, http://links.lww.com/PR9/A357.

Figure 1.

PRISMA flow diagram of study screening and selection process. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

3.2. Study characteristics

A total of 1,531 participants were included in this study, encompassing data from 3 RCTs (583 participants),^45,53,62 3 open-label trials (735 participants),^34,59,63 5 prospective/retrospective cohorts (169 participants),^4,42–44,51 6 case series reports (47 participants),^{17,25,27,41,58,68} and 6 case reports (7 participants) (Table 1).^{6,21,28,29,33,54} All patients involved suffer from severe intractable chronic pain, with various underlying causes (Supplementary Material 3, http://links.lww.com/PR9/A357). Concerning the primary outcome, most of the studies assessed pain intensity using the Visual Analog Scale or the NRS. In addition, there was frequent reporting of adverse events, serious adverse events, and concomitant medication. Given the diverse nature of these studies, not all were directly comparable. Therefore, to minimize the risk of bias, a comprehensive subgroup analysis of the RCTs was performed. In addition, non-RCT studies amenable to single-arm proportional analysis underwent thorough examination.

Table 1.

Baseline characteristics of the included studies.

Study*	Study design	Location	Chronic pain population†	N	Female, %	Mean age	VASPI/NRS score‡	Follow-up
Staats et al.53	RCT	USA, AU, NL	AIDS or cancer	108	50	55.95	78.5	4 wk
Rauck et al.45	RCT	USA	Severe pain	220	50.9	53.65	80.7	3 wk
Wallace et al.62	RCT	USA, UK	Severe pain	255	43.92	52.2	78.5	2 wk
Ver Donck et al.59	Open-label trial	USA, SE, BE	Non-/malignant pain	71	45.1	52.8	79.8	4 wk
Wallace et al.63	Open-label trial	USA	Severe pain	644	51.1	52.1	76	42 mo
Mohammed et al.34	Open-label pilot trial	UK	Severe pain	20	60	50.8	69	6 mo
Raffaeli et al.44	Retrospective cohort	Italy	Intractable pain	104	58.65	63.76	85.6	53 d
Prusik et al.43	Retrospective cohort	USA	Intractable pain	15	60	57.3	81.7	15.5 mo
Brinzeu et al.4	Prospective cohort	France	Spinal cord injury-related	20	25	49.79	61	3.59 y
Shao et al.51	Prospective cohort	USA	Intractable pain	14	27.27	55.36	NA	11 mo
Pope et al.42	Prospective cohort	USA	Intractable pain	16	56.25	58.13	NA	6 mo
Lindley27	Case series report	USA	Spine-related	17	58.82	69.12	83	4.7 mo
Poli et al.41	Case series report	Italy	Chronic pain	18	NA	NA	NA	1 y
Thompson et al.58	Case series report	USA	Lumbar spondylolysis	3	66.67	53.33	80	4 wk
Kapural et al.25	Case series report	USA	CRPS	3/7	66.67	20.67	93.33	3.2 y
Grajny et al.17	Case series report	NA	Spine-related and DM1	3	33.33	59	NA	1.58–2 y
Wermeling et al.68	Case series report	USA	Neuropathic pain	3	66.67	49.33	73.33	1–180 d
Staub et al.54	Case report	USA	Idiopathic facial pain	1	100	37	70	22 mo
William et al.6	Case report	USA	Phantom limb pain	1	0	43	85	8 d
Eberhard28	Case report	Germany	Trigeminal neuropathy	1	100	50	90	8 mo
Maier et al.29	Case report	Germany	Polyneuropathy and PTP	2	50	52.5	80	10 wk
Holden et al.21	Case report	USA	Chronic migraine	1	100	46	70	1 y
Mohammed et al.33	Case report	UK	Breast cancer-related pain	1	100	63	90	3–4 mo

^*Detailed additional information about the pain duration and specific outcomes reported in this study is outlined in Supplementary Material 3, http://links.lww.com/PR9/A357.

^†Not specified population or categorized as experiencing “severe pain” or “chronic pain” encompass a variety of distinct etiologies, which are also detailed in Supplementary Material 3, http://links.lww.com/PR9/A357.

^‡Both the VASPI and NRS scales are expressed in a 100-point scale.

AIDS, acquired immune deficiency syndrome; AU, Australia; BE, Belgium; CRPS, complex regional pain syndrome; DM1, diabetes mellitus type I; NA, not available; NL, Netherlands; NRS, Numeric Rating Scale; PTP, posttraumatic pain; RCT, randomized controlled trial; SE, Sweden; USA, United States of America; VASPI, Visual Analogue Scale for Pain Intensity.

3.3. Efficacy and safety of Ziconotide: a pooled analysis from randomized controlled trials

In the analyzed randomized controlled trials, Ziconotide treatment was associated with a mean pain reduction ranging from 14.7% to 53.1%, in contrast to the placebo groups, which showed a reduction between 3.04% and 6.79%. This yielded a statistically significant mean difference of 22.54 (95% CI: −36.70 to −8.38; P = 0.002; n = 347 for Ziconotide vs n = 234 for placebo), despite considerable heterogeneity (I² = 100%) as depicted in Figure 2A. Concordantly, the CPRS supported these findings, with 40.91% of patients treated with Ziconotide achieving moderate to complete pain relief vs 15.81% in the placebo cohort—a significant intergroup disparity (OR 3.49; 95% CI: 1.99–6.14; P < 0.0001; I² = 42%; n = 352 for Ziconotide vs n = 234 for placebo) (Fig. 2B). Complete pain alleviation is notably challenging in persistent chronic conditions, especially for refractory pain. Of note, 6.25% of patients receiving Ziconotide attained full relief, compared with a minimal 0.43% in the placebo group (OR 5.35; 95% CI: 1.13–25.33; P = 0.03; I² = 0%; n = 352 for Ziconotide vs n = 234 for placebo) (Fig. 2C). Nonetheless, the drug did not elicit a response in all patients, as reflected in the RCTs. When analyzing the total number of nonresponders, a greater proportion was evident in the placebo group (72.22%) relative to the Ziconotide group (57.65%) (OR 0.32; 95% CI: 0.19–0.56; P < 0.0001; I² = 51%; n = 349 for Ziconotide vs n = 234 for placebo) (Fig. 2D). Therefore, Ziconotide demonstrates significant efficacy in average pain reduction, CPRS, nonresponder rates, and complete pain relief when benchmarked against placebo, as collectively presented in Figure 2.

Figure 2.

Pooled analysis of efficacy-related outcomes. (A) Forest plot depicting the mean difference in pain reduction percentage on the VAPSI scale for Ziconotide vs placebo, showcasing the standardized mean differences and 95% confidence intervals. (B) Odds ratios for the number of patients who achieved moderate to complete pain relief as assessed by the CPRS, comparing Ziconotide to placebo, with each study's odds ratio and the overall combined odds ratio. (C) Odds ratios for the number of patients with complete relief of chronic severe pain when treated with Ziconotide vs placebo, including the odds ratios for individual studies and the total combined odds ratio. (D) Odds ratios for the number of nonresponsive patients who did not achieve pain relief with Ziconotide compared with placebo, detailing individual study outcomes and the total combined odds ratio. CPRS, Categorical Pain Relief Scale; VASPI, Visual Analogue Scale of Pain Intensity.

The prevalence of adverse effects was notably high, with adverse effects being observed in 94.90% of patients treated with Ziconotide, compared with 76.92% in those receiving a placebo. This disparity is statistically significant, as indicated by a relative risk (RR) of 1.24 (95% CI: 1.09–1.41; P = 0.0008; I2 = 59%; n = 353 for Ziconotide vs n = 234 for placebo) (Fig. 3A). The RCTs distinguished between general adverse effects and serious adverse effects (SAEs), revealing a substantially lower prevalence (17.85% in the Ziconotide group vs 6.84% in the placebo group). There was a clear tendency favoring the placebo in terms of SAE prevalence, with Ziconotide associated with a significantly higher likelihood of inducing SAEs (RR 2.63; 95% CI: 1.52–4.57; P = 0.0006; I² = 63%; n = 353 for Ziconotide vs n = 234 for placebo) (Fig. 3B). Moreover, the reported all-cause mortality rates were relatively low, with 2.87% in the Ziconotide group and 2.56% in the placebo group, showing no marked differences (RR 1.05; 95% CI: 0.40–2.72; P = 0.92; I² = 0%; n = 349 for Ziconotide vs 234 for placebo) (Fig. 3C). Consequently, despite Ziconotide's efficacy, the prevalence of adverse effects is significantly elevated compared with the placebo, as delineated in Figure 3. These findings must be interpreted with caution given the limited power of the sample.

Figure 3.

Safety outcomes pooled analysis. (A) Risk ratios for the incidence of any adverse effects in patients treated with Ziconotide vs placebo. (B) Forest plot showing risk ratios for the number of patients experiencing serious adverse effects with Ziconotide compared with placebo. (C) All-cause mortality rates between Ziconotide and placebo groups, represented by risk ratios.

3.4. Single-arm proportional analysis and efficacy review of nonrandomized trials and observational studies

Given the limited evidence from RCTs, a single-arm proportional meta-analysis of observational studies was conducted to provide a better perspective on their safety. Across 15 studies involving a total of 906 patients, it was observed that 704 patients experienced some form of adverse event (Fig. 4A), and 88 patients experienced a serious adverse event (Fig. 4B). The aggregated results of the analysis indicate a combined effect estimate with a high adverse event proportion of 0.65 (95% CI: 0.40–0.87; I² = 92%; n = 906). According to the findings, there is a high variability and heterogeneity that could be attributed to both the characteristics of the population and/or the specific protocols or doses of Ziconotide. Despite the heterogeneity, the estimate suggests a considerable prevalence of adverse effects in the analyzed studies (Fig. 4A). The estimated proportion for serious adverse effects was significantly reduced, where the combined effect demonstrated a total proportion of 0.03 (95% CI: 0.00–0.15; I² = 83%; n = 906) (Fig. 4B). Similarly, moderate-to-high heterogeneity was observed, indicating the presence of variability possibly due to (1) study design, (2) population characteristics, or even (3) definitions of “serious” adverse effects across studies.

Figure 4.

Single-arm proportional meta-analysis of observational studies. (A) Forest plot depicting the estimated proportion of any adverse events for each study with a subanalysis according to study design. (B) Estimated proportion of serious adverse events with a subanalysis per study design.

In addition, an attempt was made to evaluate the efficacy of Ziconotide in observational studies using a single-arm meta-analysis. However, this was hindered by insufficient data on the variability of outcomes across most of the studies. Consequently, the percentage observed pain reduction was qualitatively synthesized as part of the systematic review (Supplementary Material 4, http://links.lww.com/PR9/A357). Overall, the reviewed studies reported highly variable pain reductions following the administration of Ziconotide, ranging from minimal relief to complete alleviation of 100% in some case reports, which documented full pain relief.⁶ Interestingly, it should be noted that studies with small sample sizes tend to overestimate the percentage of pain reduction, while observational studies with substantial samples, such as Wallace et al.,⁶³ 2008 (n = 644), report a more mild-to-moderate median pain reduction percentage ranging from 3.95% to 10.53%. These data could suggest both (1) high inter-individual variability in response to Ziconotide and (2) a potential publication bias in case reports with favorable outcomes. See Supplementary Material 4, http://links.lww.com/PR9/A357, for more details.

3.5. Adverse effects: qualitative and quantitative synthesis

A total of 116 adverse effects were identified upon the completion of a systematic review of the 23 studies involving a total of 1,310 participants (Supplementary Material 5, http://links.lww.com/PR9/A357). The 20 most frequently reported AEs were then compiled and synthesized, including the number of studies in which each event was reported. Finally, a quantitative synthesis was conducted, detailing the aggregate number of patients affected by these events across all 23 studies and presenting the data as a percentage of the total cohort. In our synthesis, the most prominent effects identified include dizziness (n = 491, 37.8%), nausea (n = 382, 29.16%), nystagmus (n = 250, 19.08%), confusion (n = 245, 18.7%), memory alterations (n = 176, 13.44%), abnormal gait (n = 168, 12.82%), and speech disorders (n = 154, 11.76%), among others (Table 2). In addition, data from studies facilitating the association of adverse effects with Ziconotide dosages were synthesized. Challenges arose due to the heterogeneity in reporting methods. Nevertheless, both qualitative and quantitative syntheses were conducted, collating and presenting the current information on the correlation between AEs and dosages (Supplementary Material 6, http://links.lww.com/PR9/A357). In this context, doses associated with adverse effects typically varied from 0.11 to 1.44 µg/hour, with a mean of 0.63 µg/hour (SD ± 0.41). However, it should be noted that this represents a broad overview. Each specific adverse effect demonstrates unique characteristics, as detailed in the table provided in Supplementary Material 6, http://links.lww.com/PR9/A357.

Table 2.

Qualitative and quantitative synthesis of the 20 most frequently reported adverse events across the 23 selected studies, encompassing a total of 1310 patients.

Adverse effects*	Studies documenting the adverse effect (n, %)	No. of patients reporting the AE	%
Dizziness	8 (34.78)	491	37.48
Nausea†	10 (43.48)	382	29.16
Nystagmus	6 (26.09)	250	19.08
Confusion†	10 (43.48)	245	18.7
Memory alterations	4 (17.39)	176	13.44
Abnormal gait	4 (17.39)	168	12.82
Speech disorder†	7 (30.43)	154	11.76
Urinary retention/impairment	9 (39.13)	127	9.70
Hallucinations	8 (34.78)	103	7.86
Ataxia	5 (21.74)	101	7.71
Blurred vision†	4 (17.39)	86	6.56
Increase CK (creatine kinase)	4 (17.39)	83	6.32
Headache	10 (43.48)	75	5.73
Somnolence	6 (26.09)	69	5.27
Asthenia/Myasthenia†	5 (21.74)	67	5.11
Mental slowing	1 (4.35)	65	4.96
Nausea and vomiting	3 (13.04)	63	4.81
Only vomiting	5 (21.74)	57	4.35
Generalized body pain/back pain	5 (21.74)	52	3.98
Constipation	2 (8.70)	40	3.05

^*Comprehensive information regarding all 116 reported AEs, along with each study's respective contribution to these adverse events, is available in Supplementary Material 5, http://links.lww.com/PR9/A357.

^†AEs marked with an asterisk may be slightly underreported as Thompson et al. reported the presence of the event without specifying the exact number of affected patients.

AE, adverse effect; CK, creatine kinase.

3.6. Reduction in opioid consumption

In the studies reviewed, a significant portion of patients were on concurrent medications, including systemic opioids. Of note, in the 3 randomized controlled trials, modifications in opioid usage were documented. However, owing to a lack of data variability information, an appropriate subanalysis using the corresponding forest plot was unfeasible. Consequently, alongside other studies that provided relevant data, an aggregation was made to assess the change in opioid consumption, especially to determine whether Ziconotide monotherapy could contribute to a reduction in opioid use (Table 3). Among the data compiled, only 8 studies were applicable, with 7 indicating a trend towards decreased opioid consumption. In this observed trend, the RCTs conducted by Staats et al.⁵³ and Rauck et al.⁴⁵ did not show significant differences. In addition, Wallace et al. reported no differences in either the Ziconotide or Placebo groups.⁶² However, observational studies, despite their external validity limitations, indicated reductions in opioid use of 6.4%, 42.43%, and 91.5%. Therefore, in a general overview and extending beyond a qualitative analysis, the current evidence is insufficiently clear and robust to conclusively affirm that Ziconotide can effectively reduce opioid consumption. Moreover, only 34.78% of the studies included in this review provided relevant data on this aspect.

Table 3.

Qualitative summary of mean change in opioid consumption.

Study	Study design	Opioid consumption change
Staats et al.53	RCT	9.9% reduction in Ziconotide group vs 5.1% increase in placebo group
Rauck et al.45	RCT	23.7% reduction in Ziconotide group vs 17.3% reduction in placebo group
Wallace et al.62	RCT	0% change in both groups
Ver Donck et al.59	Open-label trial	6.4% reduction at week 3
Prusik et al.43	Retrospective cohort	4 patients evidenced systemic opioid reduction
Pope et al.42	Prospective cohort	Mean reduction of 91.5%
Brinzeu et al.4	Prospective cohort	50% of patients stopped opioids, 50% decreased by an average of 84%
Shao et al.51	Prospective cohort	Mean reduction of 31.82%

RCT, randomized controlled trial.

3.7. Risk of bias and Grading of Recommendations, Assessment, Development and Evaluation assessment

The RCTs were appraised utilizing the Cochrane RoB2. They were overall rated as presenting “some concerns.” This appraisal acknowledged a low risk of bias in both the randomization procedures and the outcome measurements. Nonetheless, “some concerns” were identified pertaining to deviations from intended interventions, the selection of reported results, and incomplete outcome data. The comprehensive risk of bias summary and the corresponding graphs for the RCTs are depicted in Figure 5A. In addition, the funnel plot analyses for the key outcomes related to the safety and efficacy of Ziconotide were performed. Despite the limited number of studies included in the meta-analysis, they exhibited a trend towards symmetric distribution, which may suggest an absence of publication bias (Fig. 6). However, the small dataset resulted in noticeable gaps in the funnel plots for pain reduction, raising the possibility of publication bias or the underrepresentation of smaller studies with less pronounced effects (Fig. 6A). Regarding safety outcomes, the presence of studies with significantly elevated SE (log [RR]) values could indicate potential biases or suggest their categorization as outliers due to other factors (Fig. 6B).

Figure 5.

Risk of bias assessment and case reports evaluation. (A) Summary of the risk of bias assessment for RCTs and the proportion of included studies with low, high, or moderate risk of bias using the RoB-2 Tool. (B) Risk of bias assessment and the proportion analysis for open-label trials using the ROBIN-I Tool. (C) Risk bias assessment of observational studies using the Newcastle-Ottawa Scale. (D) Case Reports Assessment using the JBI Critical Appraisal Checklist. JBI, Joanna Briggs Institute; RCT, randomized controlled trials; RoB2, risk of bias 2 tool; ROBINS-I, Risk of Bias in Nonrandomized Studies of Interventions.

Figure 6.

Funnel plot analysis of main outcomes. (A) Efficacy funnel plot analysis for VASPI Mean Reduction and Pain Relief Responders. (B) Safety funnel plot analysis depicting adverse events and serious adverse events of randomized controlled trials. (C) Safety funnel plot analysis of single-arm proportion meta-analysis of observational studies. VASPI, Visual Analogue Scale for Pain Intensity.

The nonrandomized open trials were assessed using the ROBINS-I tool. Each study demonstrated a significant risk of bias in 2 or more evaluated domains. Despite a low-risk categorization in the intervention domain, they were overall classified as having a “serious risk of bias.” This classification is reflected in a high percentage of serious or moderate risk across most domains, as illustrated in Figure 5B. Cohort studies and case series were appraised using the Newcastle-Ottawa Scale. Among the 11 studies, 4 achieved an optimal overall rating of 7 to 8, attributed to robust performance in the subdomains of selection, comparability, and outcome assessment. By contrast, the remaining 7 studies were assigned a low-to-moderate score, approximately 3 points, predominantly due to a lack of efficacy in the comparability subdomain (Fig. 5C).

Given their nature, the case reports were evaluated using the JBI Critical Appraisal Checklist tool based on 8 key questions. All the case reports complied with a “yes” for the 8 domains (Fig. 5D), being suitable to qualitatively review the information for the systematic review, despite the external validity limitations.

The outcomes from the randomized clinical trials were assessed using the GRADE methodology (Supplementary Material 7, http://links.lww.com/PR9/A357). Overall, the certainty of the evidence was categorized as low to moderate. This grading was primarily influenced by 2 critical issues: (1) the heterogeneity and the wide range of the confidence intervals, and (2) a potential publication bias arising from existing conflicts of interest. Of note, all 3 randomized clinical trials were conducted in collaboration with a private sponsor, and some of the researchers involved were affiliated with the sponsor at the time. Although such arrangements are common in clinical research, the GRADE methodology encourages consideration of potential conflicts of interest and the risk of publication bias when appraising the certainty of evidence. This mention follows that guideline and does not imply any judgment beyond the standard evaluation criteria.

4. Discussion

Other narrative and systematic reviews have been previously published but addressing a different PICO framework (Population, Intervention, Comparison, and Outcome).^5,46,48 This systematic review and meta-analysis acknowledge the previous evidence on Ziconotide and extend its scope by incorporating single-arm proportional analyses, conducting a meta-analysis of adverse events, and applying an evidence certainty assessment, thereby providing an updated evaluation of this selective Cav2.2 channel blocker in the management of severe chronic pain. In total, 1,531 participants from 23 different studies were included (Table 1). According to its location, physiological role, and mechanism of action, modulating Ca_v2.2 channels may reduce nociceptive neurotransmission, as those channels are already the target of pregabalin and gabapentin in the therapeutic use for pain. Preclinical evidence was promising with the observation of hypoalgesic effects more significant than those mediated by morphine itself,³² and although clinical research is more complex, the effects capable of mediating the painful experience of patients can be corroborated. Through the subanalysis of 3 randomized clinical trials comparing Ziconotide with placebo (with a total of approximately 350 patients), its effectiveness was corroborated (Fig. 2). In a limited manner, we also observed these variable trends in observational studies (Supplementary Material 4, http://links.lww.com/PR9/A357). In intractable chronic pains where intrathecal therapy might be indicated, expert panels suggest Ziconotide, alongside morphine and hydromorphone, as a first-line treatment in multimodal analgesic therapies.⁴⁸ From this meta-analysis, we observe a moderate risk of bias in the RCTs and, through GRADE assessment, it is considered that the recommendations are based on evidence of low-to-moderate certainty for some outcomes (Supplementary Material 7, http://links.lww.com/PR9/A357). In this regard, clinical practice is complex and must be framed in the conceptualization of evidence-based medicine, yet it is emphasized that further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Therefore, given the current expert consensus, we should not be satisfied with the current low-to-moderate certainty of evidence. In addition, in most included studies, pain is conceptualized as a unidimensional variable, which differs from what we know about pain today. Only a single study, with a modest sample size and no placebo group, attempts to delve into inseparable variables such as catastrophizing or function, observing favorable effects.⁵¹ Furthermore, none of them have focused on somatosensory variables using Quantitative Sensory Testing. These aspects would be worth delving into in case of subsequent clinical trials.

Importantly, although Ca_v2.2 receptor expression is higher in the peripheral nervous system, they are also widely expressed throughout the central nervous system, including in areas such as the hippocampus, hypothalamus, locus coeruleus, dorsal raphe, thalamic nuclei, and the granular layer of the cortex, among others.³ This distribution suggests that Ziconotide may influence not only nociceptive glutamatergic signaling but also modulate serotonergic, GABAergic, cholinergic, and dopaminergic pathways.^15,22,69,70 Observations from animal models lacking the Ca_v2.2 channel have shown behavioral changes, including increased aggression.^3,26 Thus, it is reasonable to anticipate adverse effects extending beyond nociception reduction. This hypothesis is supported by clinical evidence and the findings of this meta-analysis, which identified up to 116 AEs, 50 of which are psychiatric or neurological (43.10%). In addition, this study aims to provide a detailed overview of the relationship between AE occurrence and dosing (Supplementary Material 6, http://links.lww.com/PR9/A357). Furthermore, while not highly frequent, severe AEs, including cases of depression and suicidal ideation, were identified. Unfortunately, some of these were accompanied by the act of suicide.²⁹ Psychiatric conditions with suicidal ideation emerged rapidly following Ziconotide administration and were reversible upon discontinuation, a critical factor in clinical decision-making and adjustments of the titration phase, where faster rates correlate with a higher incidence of AEs. These findings should be carefully considered in clinical practice, given that behavioral alterations have been previously published in basic research,^3,26 emphasizing their role in monoamine regulation. This underlines the need for deeper exploration in current research that may not yet fully address these aspects.

However, many of the studies included in this review were conducted over a decade ago, when clinical protocols for Ziconotide administration often relied on higher starting doses and faster titration schedules. Current clinical use emphasizes slower titration and more cautious initiation, which may reduce the frequency and severity of adverse events. Moreover, the use of Ziconotide in combination with other agents—particularly intrathecal morphine—is becoming more common in modern pain management, potentially improving both efficacy and tolerability.⁶⁴ These evolving clinical practices suggest that the safety and effectiveness profiles observed in this review may not fully capture the outcomes of polymodal strategies.

Building on this, synergistic effects with opioids have been suggested, where the combination of both intrathecally could offer interesting benefits.^1,64 Indeed, opioids share a similar mechanism of action: the activity of μ-opioid receptors is coupled to Ca_v2.2 through G proteins. Therefore, the activity of morphine mediates the inhibition of N-type calcium currents.^12,50,65 In this regard, preclinical evidence shows synergistic analgesic effects and interactions in terms of morphine tolerance.⁶⁵ In this review, we focus on intrathecal monotherapy analysing a possible reduction of opioid intake, based on the aforementioned. The evidence is controversial and inconclusive (Table 3), but it remains a question to be deepened since only 8 studies of the 23 included have been focused on this question.

Altogether, this meta-analysis highlights the therapeutic potential of intrathecal Ziconotide and exposes a substantial burden of adverse events that must be carefully weighed in clinical decision-making. Given the overall low-to-moderate certainty of the current evidence, further high-quality, prospective randomized trials remain essential, particularly those incorporating patient-centered outcomes and reflecting contemporary multimodal treatment strategies.

4.1. Limitations

This systematic review and meta-analysis are subject to several limitations that may affect the interpretation of its results. (1) There is a limited number of randomized controlled trials available, and only a few were eligible for quantitative meta-analysis. As a result, part of the evidence base derives from observational studies, which were analyzed separately using single-arm proportional meta-analyses that inherently yield less robust evidence. (2) Owing to the scarcity of high-quality controlled studies in this field, we included observational studies and very small case series in a separate, stratified analysis. These data were not pooled with higher-level designs and were incorporated strictly for exploratory purposes, aiming to illustrate the broader spectrum of reported adverse events. Although this approach helps capture some outcomes, the uncontrolled nature and small sample sizes of certain studies inherently limit the interpretability of their findings. (3) Several outcomes exhibited substantial heterogeneity, which may be partly explained by differences in Ziconotide dosage, titration protocols, and the clinical diversity of the populations studied. As detailed in Supplementary Material 3, http://links.lww.com/PR9/A357, studies covered a broad range of pain etiologies—including cancer-related, neuropathic, nociceptive, postsurgical, posttraumatic, and AIDS-related conditions—as well as mixed or poorly classified syndromes. This etiological heterogeneity likely impacted both efficacy and safety outcomes, limiting direct comparability across studies. (4) The timing of outcome assessments also varied across studies, with differences in baseline and endpoint definitions potentially affecting effect size estimation. (5) The underrepresentation of certain chronic pain populations may compromise the external validity of our findings. (6) Finally, although most studies reported pain intensity and adverse events, patient-centered outcomes—such as quality of life, functional status, or satisfaction—were largely underreported, limiting the broader assessment of clinical benefit. Taken together, these limitations, along with the issues raised throughout the discussion, should be considered when interpreting the conclusions and judging their relevance across different clinical scenarios.

5. Conclusions

This meta-analysis and systematic review represent a comprehensive evaluation to date of Ziconotide, a Ca_v2.2 channel blocker, for managing severe intractable chronic pain. Despite demonstrating significant efficacy in pain reduction compared with placebo, Ziconotide's use is accompanied by frequent adverse effects, including serious events such as psychiatric and neurological complications. These findings underscore the need for cautious clinical application, emphasizing appropriate patient selection and dose titration. The evidence on Ziconotide's impact on opioid consumption remains inconsistent, with observational studies suggesting potential reductions but limited support from randomized trials. The variability and heterogeneity of the data, along with methodological limitations in the included studies, highlight the need for robust, well-designed clinical trials to address these gaps. From a clinical perspective, Ziconotide is a valuable option for intrathecal therapy in refractory pain, particularly as a nonopioid alternative. However, the low-to-moderate certainty of the current evidence necessitates further research to establish its safety profile, optimize its use in multimodal pain management, and evaluate its long-term outcomes. These insights are critical for guiding future therapeutic strategies and enhancing the quality of care for patients with severe chronic pain.

Disclosures

The authors have no conflict of interest to declare.