The efficacy of coenzyme Q10 treatment in alleviating the symptoms of primary coenzyme Q10 deficiency: A systematic review

Abstract Coenzyme Q10 (CoQ10) is necessary for mitochondrial electron transport. Mutations in CoQ10 biosynthetic genes cause primary CoQ10 deficiency (PCoQD) and manifest as mitochondrial disorders. It is often stated that PCoQD patients can be treated by oral CoQ10 supplementation. To test this, we compiled all studies describing PCoQD patients up to May 2022. We excluded studies with no data on CoQ10 treatment, or with insufficient description of effectiveness. Out of 303 PCoQD patients identified, we retained 89 cases, of which 24 reported improvements after CoQ10 treatment (27.0%). In five cases, the patient's condition was reported to deteriorate after halting of CoQ10 treatment. 12 cases reported improvement in the severity of ataxia and 5 cases in the severity of proteinuria. Only a subjective description of improvement was reported for 4 patients described as responding. All reported responses were partial improvements of only some symptoms. For PCoQD patients, CoQ10 supplementation is replacement therapy. Yet, there is only very weak evidence for the efficacy of the treatment. Our findings, thus, suggest a need for caution when seeking to justify the widespread use of CoQ10 for the treatment of any disease or as dietary supplement.

a clinically heterogeneous and rare disorder. 12,13 Symptoms often resemble those of typical inborn mitochondrial respiratory chain disorders ( Figure 1B), including early onset, multi-organ involvement and with prevalent neurological and muscular manifestations. It some cases the symptoms predominantly affect a particular organ or tissue (e.g. kidney-or cerebellum-limited phenotypes). 8,[14][15][16] Secondary CoQ 10 deficiency refers to all the conditions in which the etiology of a CoQ 10 deficiency is not a molecular lesion in the CoQ 10 biosynthetic pathway. 17,18 In fact, a variety of conditions have been found to be associated with CoQ 10 deficiency. Statins were shown to reduce serum and muscle CoQ 10 levels. 19,20 Mutations in the electron transfer flavoprotein dehydrogenase (ETFDH) and mitochondrial DNA (mtDNA) lesions, including low mtDNA copy number, were also shown to lower steady state level of CoQ 10 . [21][22][23][24][25][26] The mechanisms leading to deficiency in these cases are unknown, except for the effect of statins, which inhibit the synthesis of mevalonate, the molecular precursors of the CoQ 10 sidechain.
In tissue samples or cultured cells from patients, CoQ 10 deficiency can be diagnosed by measuring CoQ 10 levels, which can be complemented by the observation of impaired CoQ 10 -dependent respiratory chain activities (Complex I-III and Complex II-III). In the last few decades, with the increasing availability and affordability of genomic sequencing technology, whole genome or exome sequencing is increasingly becoming the first-line diagnostic test for patients suspected of having genetic disorders, including PCoQD. This has accelerated the discovery of novel PCoQD disease variants. 27 Disease-causing mutations have been reported for 9 out of 11 COQ genes required for CoQ biosynthesis. Below we report that at least 303 PCoQD patients have been reported so far. CoQ 10 supplementation is frequently initiated immediately after diagnosis ( Figure 1C), and the majority of the literature on CoQ 10 deficiency states that CoQ 10 deficiency is treatable by supplementation with exogenous CoQ 10 . [28][29][30][31][32][33] However, there is lack of clear evidence for such a claim.
In addition to patients with documented CoQ 10 deficiency and/ or COQ mutations, CoQ 10 is frequently recommended to mitochondrial disease patients. 34,35 In fact, it is a component of the so-called mitochondrial cocktail, which is a collection of high-dose nutraceuticals with the potential to support mitochondrial functioning. 36 Moreover, although there is no consistent scientific evidence for beneficial effects, CoQ 10 is often recommended for treating a wide range of other conditions (e.g. heart failure and neurodegenerative diseases) and it is widely available over the counter as an anti-ageing dietary supplement. 32 By estimation, the global market size of CoQ 10 amounts to close to 600 M USD a year.
This review aims to summarize and evaluate the available evidence for the effectiveness of CoQ 10 supplementation for the treatment of PCoQD. Patients with PCoQD should be the most amenable to CoQ 10 treatment because their CoQ 10 deficiency is the only cause of all their symptoms, and therefore, CoQ 10 treatment is simple replacement therapy. Thus, examining outcomes of CoQ 10 treatment F I G U R E 1 CoQ 10 in the mitochondria, pathology of CoQ 10 deficiency and oral supplementation. (A) The final steps of CoQ 10 biosynthesis are carried out in the inner mitochondrial membrane. The CoQ 10 biosynthetic pathway includes both enzymes (in blue) and structural or regulatory components (in purple). Only the numbers in their names are shown for COQ proteins (COQ2-7, COQ8A, COQ8B and COQ9). They are known to form a large complex, the CoQ biosynthetic complex or CoQ-synthome. COQ10A and COQ10B whose functions are uncertain and not known to be part of the complex are not shown. The most essential function of CoQ 10 is to transport electrons in the mitochondrial respiratory chain. Although CoQ 10 is found in the mitochondrial membrane, in the figure this is not shown for clarity. (B) Primary CoQ 10 deficiency predominantly manifests as mitochondrial disorder, with organs with high energy needs being most often affected. (C) Intestinal absorption of CoQ 10 is thought to occur through the formation of mixed micelles with other dietary lipids. Once inside the enterocytes, CoQ 10 is incorporated into chylomicrons (CM), which are transported via the lymphatics to the blood circulation. Because of its extreme hydrophobicity and its relatively large size, the absorption of orally administered CoQ 10 has been reported to be poor for these patients is the first key step to address the effectiveness of any CoQ 10 therapy and to promote a rational use of CoQ 10 for disease treatment or as a health supplement.

| Search strategy and selection criteria
A literature search was performed in PubMed for studies that described PCoQD patients, up until May 01, 2022. The PubMed query used is given in Supporting Information. The references cited in the articles identified were manually screened for any additional relevant study. We imposed no publication status or language restrictions. We considered any type of study regardless of research design.
The following information was sought in each paper: descriptive characteristics of PCoQD patients including sex, age of onset, major symptoms, age at the last reported exam or death, molecular lesions in COQ genes or proteins, severity of CoQ 10 deficit, respiratory chain complex (RCC) activities, CoQ 10 treatment received and clinical outcomes and laboratory tests known to be relevant to mitochondrial disease. CoQ 10 levels and RCC activities are most often reported in patient-derived skin fibroblasts or muscle biopsies. Study data were extracted by one reviewer (YW) and verified by another reviewer (SH) for accuracy, narrative summaries and interpretation. When data were reported more than once for the same patients, which was exceedingly rare, the data that were included were those from the most recent comprehensive report. If no data on patient treatment with CoQ 10 were provided in a study, or if patients were treated but outcome data were not reported, or the reported effects were contradictory or ambiguous, the study was excluded from the final data synthesis ( Figure 2).

| Data analysis
We synthesized data using tabulations that include narrative summaries. The effect of CoQ 10 treatment on clinical outcomes is considered as positive (responding) if one of the following criteria is satisfied: a) a positive effect on a quantifiable clinical measure was reported; b) some improvement was noted after CoQ 10 treatment and stopping/halting the treatment resulted in deterioration of a patient's condition; and/or c) no quantifiable clinical evidence was provided but at least two symptoms were described to be improved following CoQ 10 treatment. Fulfilling any one of the first two criteria is defined as responding with an objective description of the response. Whereas if symptom improvement was described without relying on any quantifiable measure, we categorize it as a subjective description of the response to CoQ 10 therapy. Patients counted as not responding include cases where no significant effect was noted after CoQ 10 treatment, or the reported effect(s) were minimal, or when, though some clinical improvement was noted, the patient's condition had deteriorated (e.g. developed new symptoms) while on CoQ 10 therapy. No restriction on CoQ 10 dosage (dose, formulation, dose frequency), time of initial treatment, duration of treatment or concurrent treatments was made. The two authors independently assigned the patient cases to the categories. Disagreements were resolved by discussion and consensus.

| Statistical analysis
Violin graphs were plotted and analysed by using GraphPad Prism 9 (GraphPad Software, Inc.). Differences between groups were tested using Student's t-test.

F I G U R E 2
Flow diagram for identification and selection of primary CoQ 10 deficiency patients

| RE SULTS
The literature search yielded 78 published studies, from which a total of 303 patients with PCoQD were identified. Their characteristics are summarized in Table 1, and details are available in Table S1.
Of the 303 PCoQD patients, 142 [46.7%] were reported to receive oral supplement of CoQ 10 . The dosage was reported as mg/day in some studies and as mg/kg/day in others. Doses ranged from 60 mg/ day to 2100 mg/day or from 5 mg/kg/day to 100 mg/kg/day, and the reported duration of treatment was from 1 month to 8 years. [37][38][39][40] Following the exclusion criteria, 53 treated patients were removed from the final analysis (Table S2). Among the excluded patients, 16 were excluded because the reported follow-up findings were judged to be ambiguous or inadequate to judge treatment efficiency, for example reports that mention symptom stabilization or CoQ 10 treatment combined with other simultaneous treatments. All other exclusions were because no treatment outcome was reported.
In the final analysis, we included and assessed a total of 89 patients. The results are shown in Table 2. Details, including total count of patients treated and numbers of exclusions for each gene, can be found in Table S3. We classified 65 out of the 89 patients (73.0%) as not responding to CoQ 10 treatment according to the evaluation criteria (Table S4). Among those, there are nine cases in which patients showed infantile onset and multisystem involvement. Such cases may be more challenging to treat, but this is only speculation. Of the 24 cases (27.0%) that were identified as responders, 20 were found to provide objective descriptions of responses and four are considered to be responders because they meet the criterion of having a subjective description of responses to CoQ 10 therapy (Table S5).
Note, however, that all responses were partial, and responses are frequently only observed with a single symptom. Table 3 highlights the five cases in which a worsening of patient conditions after stopping/halting of CoQ 10 treatment or regimen change was reported.
Four out of the five also reported recovery to some extent following treatment resumption. These cases potentially provide the most tantalizing evidence for a partial efficacy of CoQ 10 treatment for CoQ 10 deficiency. We should note, however, that the possibility of placebo effects cannot be excluded. Furthermore, in one of the cases the patient's condition was reported to worsen after replacement of CoQ 10 with idebenone and thus it is impossible to distinguish between the effects of stopping CoQ 10 and potential idebenone toxicity. Of the other 15 cases of responses with objective description, four cases reported a decrease of proteinuria after CoQ 10 treatment as an indication of kidney function improvement and ten reported a reduction in a severity score of ataxia or another motor performance test at a follow-up. However, five of the patients classified as responders because of an amelioration of proteinuria had only kidney symptoms and in two cases only proteinuria.
As shown in Figure 3 and S1, there is no significant differences in treatment dosage and duration of treatment between the nonresponding and responding patients. The highest reported dosage is 2100 mg/day. No substantial adverse effects have been reported for the CoQ 10 -treated PCoQD patients. However, an adverse reaction has been reported in one case of treatment with the synthetic CoQ analogue idebenone, which has a hydroxydecyl instead of a decaprenyl side chain and higher solubility than CoQ 10 . 41

| DISCUSS ION
In humans, mutations have so far been reported in all the genes required for CoQ 10 biosynthesis, except COQ3. COQ3 is an Omethyltransferase and it is the only COQ protein that is required for more than one step in the CoQ biosynthetic pathway. 42 Thus, one possible explanation for the lack of reports of COQ3 patients is that, because it is required for two enzymatic steps, pathogenic mutations in COQ3 are more detrimental to CoQ production and thus are more likely to be lethal. Among the reported PCoQD patients, 37.0% (112/303) carry a mutation in the COQ8A gene and 29.0%  57 The scientific literature as well as the general media mostly state that oral CoQ 10 supplementation is effective and thus that CoQ 10 deficiency is treatable. 33   or kidney symptoms were reported to show no response or the condition continued to deteriorate after CoQ 10 treatment (Table S4). Overall, most descriptions of the effects of CoQ 10 treatment have incomplete information and lack a complete clinical picture.
Doctors and patients are aware of the treatments (i.e. no blinding).
There can of course be no 'no-treatment' control group of patients.
For these reasons, we consider the cases where a minimal effect only was reported as not responding to treatment. It has been hypothesized that CoQ 10 treatment cannot reverse severe tissue damage due to PCoQD when the disease has already progressed too far before therapy is initiated. 30,58 However, animal studies with an unnatural CoQ biosynthetic precursor suggest that most phenotypes due to severe CoQ deficiency can be completely rescued by a partial replenishment of CoQ levels. [59][60][61] It remains to be seen how various disease symptoms due to CoQ 10 deficiency can be effectively However, the overall conclusion remained that CoQ 10 is ineffective for the treatment of patients with mitochondrial disorder, or at least there is no solid evidence to suggest otherwise. 63,64 CoQ 10 is extremely lipophilic and practically insoluble in water; therefore, to develop pharmaceutical CoQ 10 preparations, a number Note: Treatment effects established by quantitative or semi-quantitative measures to describe the response to CoQ 10 treatment were counted as responding with objective description, while descriptions of positive effects but without relying on quantitative or semi-quantitative measures were counted as responding with subjective description. 'Not responding' include the patients who were reported not to respond to CoQ 10 treatment or whose responses we consider lacking a convincing demonstration of a response to CoQ 10 supplementation.  Figure 1C) where CoQ 10 is mostly packaged into lipoproteins. 66 In humans, the level of total plasma CoQ 10 is less than 2 μg/ml. Increases severalfold above normal plasma level has been reported after CoQ 10 Weak evidence: small benefits and a possible placebo effect and/or observer bias [78] COQ6 A353D (HOM) [A1072 -22] The patient was diagnosed with SRNS at the age of 2.5 years and SND at 4 years old. CoQ 10 treatment was started at age 5.5 years when the subject was in partial remission from cyclosporine treatment, which was discontinued at 5.8 years. A decrease of proteinuria was observed (from 117 to 76 mg/ day) at 2 months into treatment and remission was maintained at the end of the study period.
No hearing improvement was observed. Proteinuria reoccurred at a level of 1100 mg/ day after temporary cessation of CoQ 10 treatment and decreased again to 188 mg/day following reinstitution of CoQ 10 treatment.
Weak evidence: the effect observed after the first 2 months of treatment is confounded by the presence of another intervention; unclear cause for the surge of proteinuria after interruption of CoQ 10 CoQ 10 and L-carnitine were initiated at age 5, improved exercise tolerance and fewer vomiting episodes were noted after 3 months of therapy. Blood lactate level also decreased but without totally normalizing. The patient continued to develop new neurologic symptoms, for example cerebellar syndrome with tremors, with increasing age. CoQ 10 was replaced with idebenone at the age of 9 years, and within the following 4 months, severe exercise intolerance reappeared with numerous episodes of vomiting. The clinical deterioration was accompanied by an elevation of lactatemia. Reverting back to the initial CoQ 10 treatment resulted in returns to the previous clinical status within 3 months.
Weak evidence: partial improvement of only a few symptoms; confounding effects from another intervention; worsening of the patient's condition after stopping CoQ 10 is also consistent with idebenone toxicity [79] COQ8A T584delACC/P502R (CH) CoQ 10 was initiated at 5 years of age with partial improvement in motor skills, balance and strength. After 6 years, the patient gradually stopped taking CoQ 10 and her condition deteriorated including severe psychiatric involvement.
Weak evidence: partial improvement; vague description of effects; possible placebo effect and/or observer bias [80] COQ8A S616LfsX114/R301Q (CH) Self-reported fatigue and exercise tolerance improvement after 2 weeks of therapy. After 2 years of therapy, ataxia and head tremor diminished. SARA total score improved from 13 to 8. When the treatment was stopped for a month, the patient's condition deteriorated, rendering him to resume taking CoQ 10 .
Weak evidence: placebo effect and improvements as a result of the natural course of the illness, could not be ruled out [81] Abbreviations: CH, compound heterozygous; HET, heterozygous; HOM, homozygous; SARA, Scale for the Assessment and Rating of Ataxia; SND, sensorineural deafness; SRNS, steroid-resistant nephrotic syndrome. a Patient IDs are provided when more than one individual was described in the original patient reports.
treatment. [66][67][68] However, it is not known how blood CoQ 10 concentration is related to effectiveness in relieving symptoms. Moreover, the mechanism of tissue uptake of CoQ 10 is still poorly understood.
In rodents, after oral CoQ 10 supplementation high concentrations of CoQ 10 were reported for several tissues including the liver, ovaries, brown adipocytes and spleen, but not for the heart, kidney, muscle and brain, the main affected tissues in PCoQD. 59,[69][70][71][72][73] Key factors that influence the tissue or cellular uptake of CoQ 10 await future studies.
There have been discussions on the possible merits of using the reduced form of CoQ 10 , also known as ubiquinol, to enhance the bioavailability of CoQ 10 . 74 However, this is not yet strongly supported by all studies, and ubiquinol's claimed to superior bioavailability is still in question. 66 Out of the 89 cases included in our final analysis, 6 were reported to be treated with ubiquinol (Table S4 and S5). Two met our criteria of responding and 4 did not. Thus, these data also do not point to better bioavailability of ubiquinol over regular CoQ 10 in PCoQD patients.
In sum, the results of the present review suggest the need to develop alternative strategies of providing CoQ 10 for treating PCoQD. For example, our recent study suggests the possibility of intravenously administering CoQ 10 solubilized with the fungicide caspofungin to achieve much higher plasma concentration and thus more effective CoQ 10 therapy. 75 Moreover, modified precursors of the quinone ring of CoQ 10 , for example, DHB, have been considered as potential alternative treatment option for some types of PCoQD. [59][60][61]76,77 Future work is warranted to further explore these possibilities and unleash the full potential of CoQ 10 therapy.
Another implication of our study is that better empirical and clinical documentation of the effects of CoQ 10 treatments is needed.
Our study also stresses the need for caution when seeking to justify the widespread use of CoQ 10

SH and YW have received royalty payment from Clarus Therapeutics
Holdings. SH also consults for Clarus Therapeutics Holdings.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.