Efficacy of intramuscular hydroxocobalamin supplementation in cats with cobalamin deficiency and gastrointestinal disease

Abstract Background In humans, absorption and tissue retention rates of intramuscularly administered hydroxocobalamin (OH‐Cbl) are superior compared to cyanocobalamin (CN‐Cbl). Supplementation with OH‐Cbl has not been described in cats. Objectives To evaluate effects of parenteral OH‐Cbl supplementation on clinical signs, serum Cbl and methylmalonic acid (MMA) concentrations in hypocobalaminemic cats with gastrointestinal disease. Animals Twenty‐three client‐owned cats. Methods Prospective study. Serum Cbl and MMA concentrations were determined at enrollment (t0), immediately before the 4th OH‐Cbl IM injection (300 μg, given q2 weeks) (t1), and 4 weeks after the 4th injection (t2). Severity of clinical signs (activity, appetite, vomiting, diarrhea, body weight) was graded at each time point and expressed as clinical disease activity score. Results Median clinical disease activity score decreased significantly from t0 (6; range, 2‐10) to t1 (1; range, 0‐6) and t2 (1; range, 0‐9). Median serum Cbl concentration increased significantly from 111 pmol/L (range, 111‐218; reference range, 225‐1451 pmol/L) at t0 to 1612 pmol/L (range, 526‐14 756) (P < .001) at t1, and decreased again significantly to 712 pmol/L (range, 205‐4265) (P < .01) at t2. Median baseline serum MMA concentration at t0 (802 nmol/L; range, 238‐151 000; reference range, 120‐420 nmol/L) decreased significantly (P < .001) to 199 nmol/L (range, 29‐478) at t1, and was 205 nmol/L (range, 88‐734) at t2. Serum MMA concentrations normalized in 22/23 cats at t1, and were not significantly higher at t2 compared to t1. Conclusions and Clinical Importance The herein described OH‐Cbl injection scheme appears efficacious for normalization of cellular Cbl deficiency in cats with gastrointestinal disease.


| INTRODUCTION
Cobalamin (Cbl), also known as vitamin B 12 , is a water-soluble B-group vitamin and an essential cofactor for nucleic acid synthesis and hematopoiesis. Naturally occurring forms of B 12 are methyl-Cbl, adenosyl-Cbl, and hydroxo-Cbl (OH-Cbl), whereas cyano-Cbl (CN-Cbl) is a synthetic B 12 compound commonly used in supplements. 1 Cbl absorption requires binding proteins and specific receptors along various parts of the gastrointestinal tract, 1 and gastrointestinal disease can therefore lead to Cbl deficiency. In cell metabolism, adenosyl-Cbl is needed as a cofactor for the conversion of methylmalonyl-CoA to succinyl-CoA via methylmalonyl-CoA mutase and for remethylation of homocysteine via methionine synthase. In cats, Cbl deficiency causes a reduction in the activity of methylmalonyl-CoA mutase, resulting in increases in serum methylmalonic acid (MMA) concentrations, but cats do not have increased homocysteine concentrations. 2 Measurement of MMA allows assessment of availability of Cbl for cells and is considered the test of choice to detect cobalamin deficiency in people. 3 Cats with undetectable or subnormal Cbl concentrations also have significantly increased MMA concentrations. [4][5][6][7] Although it is not yet known when depletion of cellular Cbl stores and increases in MMA concentrations begin in the course of subnormal serum Cbl concentrations, MMA is currently considered the best indicator of Cbl status in cats. 2,[5][6][7][8][9] Administration of parenteral Cbl to cats with subnormal Cbl values is currently considered a routine therapeutic procedure, 5-8 especially when daily oral Cbl supplementation is difficult to achieve. 10 A widely followed recommendation suggests to administer 250 μg Cbl once weekly, for 6 weeks, followed by a dose 30 days later and determination of Cbl concentration 30 days after the last injection. 11 At present, CN-Cbl is generally recommended for routine usage in veterinary medicine, most probably because it is widely available and inexpensive. 11 In cats, only the use of CN-Cbl has been investigated so far, 4,7,8 and it was recently shown that a Cbl supplementation scheme consisting of 6 weekly IM (250 μg) injections of CN-Cbl failed to fully normalize serum and urine MMA concentrations in hypocobalaminemic cats with enteropathy. 7 In people with various types of Cbl deficiencies, the natural-occurring OH-Cbl is considered the mainstay treatment and numerous studies consistently showed superior tissue retention rates of Cbl after supplementation with OH-Cbl rather than CN-Cbl together with increased urinary excretion of CN-Cbl. [12][13][14][15][16][17][18][19][20] In addition, IM injections of OH-Cbl are thought to cause less pain than CN-Cbl in people. 15 For these reasons, we replaced CN-Cbl with OH-Cbl in our hospital in 2016 and this approach seemed to work well in our patients.
Based on clinical experience as well as sporadic assessments of Cbl and MMA values, we decided to simplify our supplementation scheme from 6 weekly IM (250 μg) injections to 4 IM (300 μg) injections given every 2 weeks. The aim of the present study was therefore to evaluate the efficacy of intramuscularly administered OH-Cbl on serum Cbl and MMA concentrations in cats with hypocobalaminemia and gastrointestinal disease before and after IM injections of OH-Cbl given every 2 weeks for a total of 4 doses. A secondary aim was to assess the clinical benefits during OH-Cbl supplementation.

| Animals and study design
This study was conducted at the Clinic for Small Animal Internal Medicine, Vetsuisse faculty, University of Zurich between September 2017 and November 2019. All cats were prospectively enrolled in the study based on the following criteria: a history of clinical signs compatible with gastrointestinal disease, which included diarrhea, vomiting, weight loss, anorexia, or weight loss with polyphagia; a serum Cbl concentration below the reference interval (225-1451 pmol/L), 22 and informed owner consent. Concurrent hyperthyroidism at inclusion was not con-

| Analyses
Serum Cbl concentration was measured with an automated competitive binding chemiluminescence assay (Immulite 2000, vitamin B 12 , Siemens Healthcare Diagnostics Inc, Newark, DE). The intra-and interassay coefficients of variation (CV) of the Cbl assay were 2.1% and 3.4%, respectively. The reference interval for serum Cbl was 225-1452 pmol/L. 22 Cbl concentrations at inclusion <111 pmol/L (below the detection limit) were truncated to 111 pmol/L for statistical analysis. All serum MMA concentrations were analyzed in the Division of Clinical Chemistry of the University Children's Hospital Zurich according to accredited methods. In brief, the samples were supplemented with an internal standard, precipitated, and analysis was done by ultraperformance liquid chromatography-tandem mass spectrometry (UPLC-TMS) on an Ultimate 3000 XRS UHPLC system (Dionex; Thermo Scientific) with a SCIEX5500 mass spectrometer (SCIEX, Framingham, Massachusetts) using multiple reaction monitoring. The lower limit of quantification for this method was 25 nmol/L. The interassay CV of the analyses were 5.8%. The reference intervals for serum MMA concentration was 120-420 nmol/L. 7

| Clinicopathologic evaluation
After obtaining a clinical history, all cats underwent a physical examination and laboratory testing including a CBC, serum biochemistry panel, and urinalysis. Two cats were hyperthyroid and received medical treatment. Serum thyroxine concentration was measured in all 23 cats at baseline evaluation, and was well within the reference range in all cats.
All cats were evaluated for pancreatitis using the DGGR-lipase assay 25

| Clinical signs
The clinical disease activity scores ( Figure 3) differed significantly over time (χ 2 24.617, P < .001). The median clinical disease activity score was 6 (range, 2-10) at t0 and decreased significantly (P < .001) to 1 (range, 0-6) at t1, and remained at 1 (range, 0-9) 4 weeks later (t2). It has traditionally been recommended that serum Cbl concentration should be above the reference range at the time of retesting, and if supranormal serum cobalamin concentrations have not been achieved to continue supplementation every 2-4 weeks. 11 Our data at t1 and t2 do not support a strict need for supranormal serum Cbl results at retesting as MMA concentrations did not differ between both time points, even though median serum Cbl concentration had decreased again into the reference range. The results of the clinical disease activity score (significant decrease from t0 to t1 and no further change at t2) support this assumption, even if clinical signs must be interpreted with caution as all cats had also received additional treatment.

| DISCUSSION
To our knowledge, supplementation with OH-Cbl has not been reported before in cats. In a previous study assessing a supplementa- Although varied concurrent treatments could have also influenced our results, we presume these differences in outcome might relate to the Cbl formulation. It has been shown experimentally in dogs that only a minor part of the injected dose of CN-Cbl is eventually utilized, whereas a large percentage is being lost in the urine during the first hours after injection. 18 When the same dogs received OH-Cbl, higher serum binding and more prolonged increases in serum Cbl concentrations and lower urinary excretion was noted compared with CN-Cbl. 18 It was speculated that the difference in serum binding may lie in the chemical stability of CN-Cbl, which limits the sites at which it may be bound as compared to OH-Cbl which readily dissociated and may equilibrate with a variety of serum constituents. 17 Possible differences in protein binding likely have also implications for cellular markers of Cbl availability as only Cbl bound to transcobalamin is available to cells. 30 Superior bioavailability of OH-Cbl over CN-Cbl has also been proven in humans. Slower disappearance of OH-Cbl from site of injection, increased liver uptake, and less rapid urinary excretion when compared with CN-Cbl were demonstrated by various authors. 14,[15][16][17][18][19][20]28 Hertz et al reported that after IM injection of 1 mg CN-Cbl and OH-Cbl, healthy individuals excreted within 24 hours about 80% and about 25%, respectively, in the urine. 15 Further dialysis experiments showed that OH-Cbl passed more slowly through membranes than does CN-Cbl, and that OH-Cbl is bound to serum proteins in far greater quantities than is CN-Cbl. 15 These observations were confirmed in long-term studies when humans with pernicious anemia were given equal amounts of either OH-Cbl or CN-Cbl IM injections, 14,17 and similar results have been published when comparing oral OH-Cbl to CN-Cbl. 29 Against this background, it is interesting to note that it has also been mentioned in small animal medicine that CN-Cbl might fail to increase serum Cbl concentrations for reasons currently not understood, and it has been speculated that OH-Cbl might instead be more effective in these patients. 11 Superior bioavailability of OH-Cbl in humans may be because of different affinities for the blood-transport binding proteins, cell receptors for Cbl uptake, and intracellular enzymes involved in their conversion to intracellular cobalamin. 13 The latter could have been the reason that CN-Cbl injections in humans with an inherited disorder of intracellular Cbl metabolism were inadequate, and biochemical (ie, MMA, homocysteine) as well as clinical parameters normalized only after patients were switched to OH-Cbl injections. 21 Unfortunately, very little is known about Cbl-binding proteins in cats, 31  Our study had several limitations. One limitation was the inclusion of cats with more than 1 clearly defined disease process as well as with different stages and severity of disease. However, we believe this was unavoidable. Even if only cats with 1 diagnosis (eg, intestinal SCL) had been included, it would have been impossible to guarantee uniform spatial distribution of disease among patients. That means we could not control for individual interference with renewed absorption of biliary Cbl during enterohepatic circulation. Another limitation was that our modified injection scheme made direct comparisons of OH-Cbl versus CN-Cbl supplementation difficult. Furthermore, concurrent individual treatments made it difficult to accurately assess the magnitude of the beneficial clinical effect of OH-Cbl supplementation, but we could not justify withholding necessary medical treatments. The magnitude of remission of clinical signs at t1 (1; range, 0-6) was the same compared to the identical score used in the previous study with CN-Cbl, 7 but also in the previous study most cats received additional treatments.
In conclusion, our findings show that 4 injections of OH-Cbl are efficacious for normalization of serum Cbl and cellular Cbl deficiency in cats with hypocobalaminemia and gastrointestinal disease. The beneficial metabolic effects appear to be greater when compared to previously published results with CN-Cbl. 7 Comparative studies on the optimal Cbl formulation (OH-Cbl versus CN-Cbl) are needed in cats with Cbl deficiency.

CONFLICT OF INTEREST DECLARATION
Authors declare no conflict of interest.

OFF-LABEL ANTIMICROBIAL DECLARATION
Authors declare no off-label use of antimicrobials.

INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE (IACUC) OR OTHER APPROVAL DECLARATION
Approved by the Cantonal Veterinary Office of Zurich and conducted in accordance with guidelines established by the Animal Welfare Act of Switzerland.

HUMAN ETHICS APPROVAL DECLARATION
Authors declare human ethics approval was not needed for this study.