Copper Deficiency as Wilson’s Disease Overtreatment: A Systematic Review

Background: Treatment of Wilson’s disease (WD), an inherited disease characterized by copper overload, is lifelong and there is the possibility that copper deficiency (CD) may occur. We systematically reviewed the literature to describe treatment patterns, symptoms and outcomes associated with CD. Methods: Using preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines, the PubMed database was searched up to 6 April 2023. Results: Across 17 articles, 20 cases of CD were described, most commonly (15 cases) in WD patients treated with zinc salts (ZS), less often on combined chelator and ZS therapy (3 cases), molybdate salts plus ZS (1), or molybdate alone (1). CD symptoms occurred insidiously, including sideroblastic anemia, neutropenia, axonal sensory neuropathy, posterior cord myelopathy and increased ratio of epileptic seizures (or epilepsy). CD diagnosis was based on symptoms and severely reduced urinary copper excretion (<20 µg/24 h [<0.3 µmol/24 h] on ZS, or <100 µg/24 h [<1.6 µmol/24 h] on chelators) with low total serum copper and ceruloplasmin. Conclusions: Awareness of CD and regular monitoring of copper metabolism is needed during WD treatment. Temporary cessation of anti-copper treatment usually reverses serum copper reductions as well as pancytopenia; however, some symptoms, especially neuropathy and myelopathy, may persist.


Introduction
Wilson's disease (WD) is an inherited disorder of copper metabolism with systemic copper overload, leading to clinical symptoms in affected organs, mainly hepatic and neuropsychiatric [1][2][3]. The pathogenesis of WD includes dysfunction of trans-membrane copper transporting type P ATPase 7B, involved in copper excretion from hepatocytes into the bile and copper incorporation into ceruloplasmin (Cp) [1][2][3]. Failure of this transporting protein leads to progressive copper accumulation in hepatocytes, their progressive damage and release of bioavailable so-called "free" non-Cp-bound copper (NCC) into the blood with secondary copper accumulation, and damage to other organs and tissues. In the course of WD, high serum NCC is observed, usually above 15 µg/dL (normal range: 5-15 µg/dL), in contrast to total serum copper which decreases below 70 µg/dL (normal range: 70-140 µg/dL) in line with decreased serum Cp level [1,2].
NCC has not been validated as a reliable biomarker of WD treatment due to the fact that the current method for NCC calculation is indirect (calculated from the difference between the copper concentrations in µg/dL and 3.15 times the serum Cp concentration in mg/dL) and may give non-diagnostic false negative results in up to 20% of patients, particularly when immunological methods of Cp assessment are used due to overestimation of holoceruloplasmin. Direct methods to measure NCC are under investigation [2,4]. The most promising is direct quantification of copper pools from human plasma by combining Cp immunocapture, chelation and filtered inductively coupled plasma mass spectrometry (ICP-MS) to permit quantitative analysis of Cp-copper and directly measured NCC; however, this method currently requires validation before introduction into WD management [2,4,5].
Correct anti-copper treatment should lead to the normalization of calculated, indirect NCC (5-15 µg/dL); the stabilization and increase of copper urinary copper excretion between 200-500 µg/day in the case of chelators; and a decrease of <100 µg/day in the case of ZS use, as well as clinical symptoms regression [1,2]. However, in some cases, especially following long-term treatment, copper deficiency (CD) may occur [1,2,9].
Copper is an essential important cofactor for many enzymes (e.g., cytochrome-C oxidase, dopamine-beta-hydroxylase, copper-zinc superoxide dismutase, proline hydroxylase, lysine oxidase and ascorbic oxidase) involved in the functioning of the bone marrow (erythropoiesis), the central nervous system (CNS; myelinization, dopamine synthesis), skin and vessel construction (collagen synthesis and bridging), as well as respiratory chain enzymes (mitochondrial and other energetic processes). Hence, CD due to WD overtreatment may lead to severe multisystem symptoms including hematological (bone marrow dysfunction leading to anemia and neutropenia) and neurological disorders (e.g., axonal neuropathy, myelopathy, posterior spinal column demyelination and myopathy) .
As the frequency of CD in WD, its association with this type of anti-copper treatment, its risk factors and clinical outcomes have not been analyzed, we performed a systematic review of published studies and case reports on CD in WD.

Methods
This systematic review was performed in concordance with the internationally accepted criteria of the preferred reporting items for systematic reviews and meta-analyses (PRISMA) statement [32].

Literature Search and Eligibility Criteria
By searching the PubMed database (up to 6 April 2023), we identified eligible original studies (prospective and retrospective), as well as case and series reports describing and analyzing patients with WD and CD. Terms used for searches included: "copper deficiency and Wilson's disease/Wilson disease" and "copper depletion and Wilson's disease/Wilson disease." Eligible studies were defined as: (1) human studies; (2) original studies (prospective or retrospective); (3) case and series reports of CD in WD patients; (4) published in English. The reference lists of extracted publications were also searched.
Titles and abstracts of all retrieved papers were screened independently by all authors and duplicate records were removed. Reviews, editorials, commentaries, and overlapping studies were excluded after assessment and revisions. The full texts of eligible articles were checked according to eligibility. All identified studies were analyzed and verified independently by all authors to confirm the inclusion criteria and analyze the data.

Data Extraction
The following data were extracted from eligible studies: (1) study characteristic: authors, year of publication, study design (prospective, retrospective, series report or case report); (2) patient characteristic-gender, WD phenotype, age at WD diagnosis and CD onset, initial anti-copper treatment, copper metabolism at CD diagnosis; (3) CD clinical symptoms with available examinations which confirmed diagnosis; (4) CD treatment and outcome. The data extraction was conducted by all authors, with any disagreements being resolved by consensus.

Statistical Analysis
The statistical analysis were carried out using Statistica v.10 (Stat Soft Inc. 2011, Tulsa, OK, USA). The mean, range, percentage, and standard deviation (SD) were noted for descriptive summary statistics.

Results
During the systematic review of the literature, we found 725 records from PubMed and reference list searches ( Figure 1). After duplicate paper removals, 120 papers remained. Further analyzing titles, abstracts and full texts, we additionally removed 103 records. Finally, 16 papers [10][11][12][13][14][15][16][17][18][19][20][21][22][23][24]29] and 1 abstract [28] were included in the analysis, describing 20 WD patients with CD as case and series reports ( Table 1). The following data were extracted from eligible studies: 1) study characteristic: authors, year of publication, study design (prospective, retrospective, series report or case report); 2) patient characteristic -gender, WD phenotype, age at WD diagnosis and CD onset, initial anti-copper treatment, copper metabolism at CD diagnosis; 3) CD clinical symptoms with available examinations which confirmed diagnosis; 4) CD treatment and outcome. The data extraction was conducted by all authors, with any disagreements being resolved by consensus.

Results
Management of CD included temporary anti-copper drug cessation in all cases, temporary copper supplementation in 5 cases, blood transfusions in 2 cases (due to severe anemia), and treatment with granulocyte stimulating factor in 1 case (Table 1). Hematological parameters (hemoglobin, red blood cells, white blood cells and neutrophils) returned to normal values in all cases. Neurological deficits resulting from polyneuropathy and myelopathy subsided in only 1 case (16%) and improved partially in a further 5 cases (83%). Reduction of epileptic seizures was observed in both cases.

Discussion
Based on our literature searches, theoretical mechanisms and a general review, the clinical symptoms of CD in patients with WD, as in the population as a whole [29,30], include predominantly hematological changes (especially anemia [sideroblastic] with leuko/neutropenia) and neurological-like sensorimotor distal polyneuropathy and spinal cord myelopathy (cervical or thoracic), and rarely, white matter demyelination and epileptic seizures.
The molecular basis of anemia is likely due to the fact that copper is a component of key enzymes involved in iron hemostasis. Cp has ferroxidase activity that converts ferrous iron to ferric iron, to enable iron incorporation into apo-transferrin and transport iron in the blood. Hence, Cp defects, such as aceruloplasminemia, can lead to iron accumulation in macrophages, anemia and neurodegeneration. Hephaestin is another copper-containing protein with ferroxidase activity, which is involved in iron release from intestinal cells. As with Cp, the lack of transformation from ferrous to ferric ions decreases the release of iron from enterocytes, leading to iron deficiency and anemia. Finally, copper-dependent cytochrome C in mitochondria is involved in reducing ferric into ferrous ions, which is required for iron incorporation in hemoglobin synthesis. Animal studies also indicate that copper-deficient rats have decreased erythrocyte survival, probably due to oxidative stress from a lack of antioxidant copper-dependent cytochrome C activity, leading to damaged membranes and increased vulnerability [29][30][31]33,34].
Interactions between copper and iron resulting in hematopoiesis disturbances were first described in 1932 [34]. In animal models, researchers observed reduced iron uptake in rats fed with a copper-deficient diet and impaired utilization of parenterally administered iron for hemoglobin synthesis in copper-deficient swine, which led to anemia with low hemoglobin levels as well as cytopenia. Based on this observation, without knowledge about copper-dependent proteins with ferroxidase activity, copper-related anemia terms were used in the bibliography [30,34].
The etiology of other hematological symptoms includes the destruction of myeloid progenitor cells, copper-dependent impaired maturation of myeloid precursors, as well as the formation of anti-neutrophils, antibodies related to copper depletion. However, in contrast to anemia, the exact mechanism of these phenomenon is not known. Analyzing the hematological profile, only thrombocytopenia is not in the spectrum of CD symptoms, and the finding of normal platelet values with severe anemia and neutropenia in patients where CD is suspected is an argument in favor of a CD diagnosis [30,31].
Neurological symptoms including myeloneuropathy [12,13,[15][16][17][18][19][20]27], CNS demyelination [11] and epileptic seizures [14,28] are also observed in copper-deficient syndromes due to Menkes disease, in which a lack of activity of copper-dependent enzymes involved in CNS maturation lead to these symptoms. The exact mechanism(s) of CD-related myeloneuropathy are unknown [12,13,[15][16][17][18][19][20]27]. Myelopathy affecting the spinal cord (cervical or thoracic) seems to be most common neurological presentation of CD syndrome. Clinically and radiologically, it is not possible to distinguish between subacute cord degeneration due to vitamin B12 deficiency and CD. However, CD is suspected as the cause of cytochrome C (copper-dependent) dysfunction, with increased lactate found in cerebrospinal fluid in contrast to vitamin B12-dependent cord degeneration. Another hypothesis for CD-related myelopathy is the dysfunction of methylation cycles, including myelin proteins (with spinal cord damage) and neuropathy.
CD symptoms appeared to occur after long-term anti-copper treatment, and most commonly in those treated with ZS or ZS plus chelators. The causality of CD is probably caused by differences in monitoring the anti-copper adequacy of treatment in WD. Urinary copper excretion should be less than 100 µg/24 h with ZS and 200-500 µg/24 h with chelators. Hence, it is easier to overlook decreased urinary copper excretion in CD when patients are treated with ZS than with chelators, especially in long-term treated WD patients when decreased Cp and serum copper are observed. Based on the increasing numbers of CD in WD, current American Association for the Study of Liver Diseases (AASLD) recommendations state that CD may be suspected when daily urinary copper excretion is low: <20 µg/24 h (<0.3 µmol/24 h) in patients treated with ZS or <100 µg/24 h (<1.6 µmol/24 h) on chelators and associated with low serum total copper and low Cp; however, cut-offs for serum copper and Cp are not provided [2]. As TH is currently under investigation in clinical trials, this treatment is not discussed in WD guidelines [1][2][3]. However, based on its dual mechanism of action, low daily copper urinary excretion with low Cp and serum copper, as in ZS, may arouse suspicion of CD.
Analyzing CD outcomes, in the course of anti-copper treatment dose reduction or temporary cessation, hematological symptoms disappeared in all cases over a relatively short period of time, although blood transfusion was needed in some severe cases of anemia. However, while there was some improvement in neurological symptoms with WD treatment discontinuation, they persisted in most patients, which underlines the need for correct and timely diagnosis of CD in WD.
In summarizing, the results from our study highlight the significance of monitoring copper metabolism as well as hematological parameters in WD patients to detect CD, especially those treated for a long duration. Furthermore, from the neurologist's point of view, the occurrence of sensory disturbances in WD patients as well as gait deterioration should always include differential diagnosis of CD, especially in ZS-treated patients, as these symptoms could be potentially reversible [35,36]. However, improper increase of anti-copper medications doses due to suspicion of WD progression may lead to irreversible neurological complications.
Our study has several limitations. Firstly, our searches only retrieved case reports and the total number of patients included was small; however, there were no prospective studies and just 20 case reports in the published literature. Furthermore, the lack of data from retrospective studies did not allow us to establish the frequency of CD in WD and follow-up observations were described over a limited period of time. We did not perform a metaanalysis, as study designs and available data came only from case reports (no prospective or retrospective studies in WD patients analyzing these problems with statistical analysis have been published so far). Furthermore, available data were heterogeneous (including copper metabolism biomarkers, as well as clinical data).
In addition, analysis of Cp was conducted using different methods (immunologic and enzymatic assay), which did not allow us to calculate NCC across the whole group of patients [4,5].
Currently, in order to improve the copper metabolism monitoring, additional copper metabolism biomarkers are studied which include directly measured NCC, labile bound copper (LBC), exchangeable copper (CuEXC) and relative exchange copper (REC) (last both used mostly in France, especially in WD diagnosis) [4,5]. They need to be validated in WD treatment monitoring, as well as in centers involved in WD management. The establishment of the objective, direct measurement of copper metabolism biomarkers will allow us to avoid more frequent CD syndromes by using more adequate doses of anti-copper drugs and more detailed copper metabolism monitoring [1]. Finally, another issue that may improve the monitoring of WD treatment (including adverse events) are the prospective international registries of WD patients. This may help to establish the frequency of CD and other adverse drug reactions in the course of WD treatment, by providing more reliable data. Only such data from studies will allow for meta-analyses for risk factors of CD, and therefore better characterize this problem [1,5].

Conclusions
Physicians treating WD patients should be aware of and familiar with CD symptoms, which include mostly anemia with neutropenia and myeloneuropathy. The diagnosis of CD is based mainly on clinical symptoms and copper metabolism, especially low daily copper urinary excretion. CD usually occurs in patients treated long-term, particularly with ZS. Early detection of CD and temporary anti-copper treatment cessation (or dose decrease) leads to rapid improvement of hematological parameters. Neurological symptoms may subside over a longer time window, but may persist in the case of long-lasting CD.
Author Contributions: T.L.: conception, planning and supervising the work, planning the literature search, selecting the abstracts, extracting data from original papers, writing, revising, approving the manuscript; A.C.: study design, selecting the abstracts, extracting data from original papers, writing, revising, approving the manuscript; J.B.: study design, interpretation, selecting the abstracts, extracting data from original papers, writing, revising, approving the manuscript; A.A.: study design, selecting the abstracts, extracting data from original papers, interpretation of the results, writing, revising, approving the manuscript; A.P.: study design, selecting the abstracts, extracting data from original papers, interpretation of the results, writing, revising, approving the manuscript; M.S.: selecting the abstracts, extracting data from original papers, interpretation of the results, writing, revising, approving the manuscript; I.K.-J.: selecting the abstracts, extracting data from original papers, interpretation of the results writing, approving the manuscript. The corresponding author attests that all listed authors meet the authorship criteria and that no others meeting the criteria have been omitted. All authors have read and agreed to the published version of the manuscript.