Doxycycline is a commonly used antimicrobial prescribed by veterinarians to treat a wide array of bacterial infections. Compounded formulations of doxycycline, which are not approved by the US FDA, may be prescribed for veterinary patients when FDA-approved formulations are not appropriate because of tablet or capsule size limitations, compliance problems that prevent successful administration of liquid or solid dose formulations, or a lack of FDA-approved product availability.
Administration of FDA-approved medications is considered the treatment of choice because the content and stability of these drugs have been established. The AVMA provides a policy to veterinarians for appropriate use of compounded medications.1 Although compounding is an important tool in select circumstances to allow veterinarians to optimally treat patients, compounded products may not have pharmacokinetic characteristics in companion animals that are similar to those of FDA-approved products. Compounded formulations may also vary in content, bioavailability, stability, safety, and efficacy when compared with FDA-approved products, and research has shown that some drugs are not ideal for compounding. The USP sets standards that both FDA-approved and compounded drugs must meet to be considered accepted drug products. Random testing of compounded drugs intended for use in people and animals by the Missouri State Board of Pharmacy from 2006 to 2016 indicated product failure rates between 28 of 242 (12%) and 69 of 274 (25%) for drug content out of USP acceptable ranges (ie, 90% to 110% of stated content).2 In a bioequivalence study,3 dogs that received compounded itraconazole had 5.5% of the total exposure (area under the plasma concentration-vs-time curve) and 4.1% of the maximum plasma concentration of the drug, compared with that achieved in the same dogs when the FDA-approved formulation of itraconazole was administered. The lower itraconazole drug exposure achieved with the compounded product was considered unlikely to be effective,3 and as such, its use could markedly increase the risk for therapeutic failure and selection for resistant fungal organisms. Other studies4,5 have investigated the chemotherapeutic drug lomustine, with results revealing low drug content (potency) in various compounded formulations that could lead to treatment failure and selection of lomustine-resistant neoplasia. However, anecdotally, clinical experience with some compounded drugs has suggested that some compounded drug formulations might be clinically efficacious. With this information, further research is warranted to determine whether commonly used medications such as doxycycline are good candidates for compounding and to determine whether there are specific formulations of such drugs that might be more reliable when compounded than others. This information could ultimately improve the potential for therapeutic success when compounding is necessary.
Doxycycline hyclate tablets that are approved by the FDA for use in people have been compounded into an oral suspension with concentrations comparable to reference standards and reliable stability through day 7; however, by day 14 and thereafter, doxycycline concentrations were significantly lower than that of the reference standard and no longer considered appropriate for therapeutic use.6 To the authors’ knowledge, no similar research has been published to evaluate the drug content of compounded doxycycline liquid formulations, tablets, or chews that are advertised by compounding pharmacies and available for veterinary practitioners to prescribe. Doxycycline might not be the best candidate for drug compounding because of its interactions with multivalent cations (eg, calcium, iron, magnesium, or aluminum), which could alter the content of free drug and decrease its oral bioavailability.7 Fluoroquinolones have similar interactions with cations, and compounded orbifloxacin has been reported to be unstable when compounded in the presence of multivalent cations.a
The purpose of the study reported here was to evaluate the drug content of compounded doxycycline products available from national veterinary compounding pharmacies that routinely advertise to veterinarians at national meetings, through direct email marketing, and in trade magazines. The study was performed to assess various formulations (tablets, chews, and liquids) for doxycycline content upon receipt and at the end of a typical treatment time (21 days) and to compare the results with those obtained for FDA-approved formulations (capsules, tablets, and liquid) evaluated in the same manner. We hypothesized that all FDA-approved formulations would be within the acceptable content range at both time periods and that compounded tablets, chews, and liquid formulations would have content that was within the acceptable range on receipt from the source, with chews having greater content variability (owing to potential interactions with additives). We also hypothesized that liquid formulations would be less stable than other formulations, causing drug content to decrease over the 21-day storage period.
Materials and Methods
Doxycycline formulations
Doxycycline hyclate tablets (100 mg)b and capsules (100 mg)c and doxycycline calcium suspension (10 mg/mL)d approved by the FDA for use in human patients were obtained for the study; there was no FDA-approved chew formulation of doxycycline. Compounded doxycycline was obtained from 3 national compounding pharmacies; the formulations chosen differed from FDA formulations because the intent was to test products that a practicing veterinarian might select on the basis of a patient's size or palatability preference and to comply with the AVMA policy on veterinary compounding.1 Three compounding pharmaciese were arbitrarily selected from those used by clinicians at the authors’ veterinary teaching hospital and that advertised compounded doxycycline at national meetings, by paper mailings to veterinary hospitals, and by email sent directly to veterinarians. Compounded doxycycline tablets were obtained as 25-mg unflavored tiny tablets, 100-mg flavored scored tablets, and 150-mg scored tablets. Compounded 100-mg chews were also obtained from all 3 pharmacies. One chew formulation arrived in a blister pack, and 2 arrived in plastic cartons, 1 of which was inside a plastic food-storage type of bag. One chew was labeled to be stored refrigerated, and 2 were labeled to be stored at room temperature. The labeled doxycycline concentration in 2 liquid formulations was 6 mg/mL, and that for the remaining liquid was 50 mg/mL; 1 was an aqueous solution, 1 was an aqueous suspension, and 1 was an oil-based suspension. Although the intent of compounding medications is to create a product specifically for administration to a particular patient, no animals were involved in the study; thus, the compounded formulations were purchased with the submitted prescription labeled for office use. Samples were stored according to the storage directions stated on each product, and all doxycycline content testing was completed prior to the labeled expiration or beyond-use dates. For reporting purposes, the compounded tablets were arbitrarily designated as tablets A, B, and C; the compounded chews as K, L, and M; and the compounded liquid formulations as X, Y, and Z.
Tablets (n = 30) and capsules (30) were selected arbitrarily from their respective containers, and all chews from each product (n = 21, 30, and 30) were weighed on a calibrated laboratory balance on the day of receipt (day 1) to evaluate the consistency of weights within formulations. Arbitrarily selected chews were additionally weighed on days 21 and 98. When tablets were scored so that they could be broken into halves (2/3 formulations), thirds (1/3 formulations), or quarters (1/3 formulations), they were broken along the scored lines with a pill cutter, and each portion was weighed. The FDA-approved tablets were not scored and therefore were not split.
Doxycycline content testing
Each formulation was tested for doxycycline content within 24 hours after receipt (day 1) and on day 21 after shipping by a commercial carrier from the compounding pharmacies (compounded formulations) or from the distributor (FDA-approved formulations). For each testing day and for each tested formulation, 5 tablets, capsules, or chews were arbitrarily selected and analyzed for doxycycline content by UHPLC testing as described. For each liquid formulation, 5 replicates/bottle (1 bottle from each of the 3 compounding pharmacies and 1 bottle of an FDA-approved formulation) were tested for content after thoroughly mixing. When tablets were scored, 5 tablets/formulation were divided as directed, and the doxycycline content of each portion was tested within 24 hours of receipt. In addition to whole-capsule content testing as previously described, 5 arbitrarily selected capsules were also emptied, and both the capsule content and the capsule casing were tested for doxycycline content. All UHPLC analyses were conducted by the same investigator (MW), who was not blinded to the formulation.
Preparation of capsules, tablets, chews, and liquid formulations
Owing to a lack of USP methods for doxycycline content determination in compounded tablet and chew formulations, methods were developed to optimize content determination for these formulations and then applied uniformly to all solid formulations to minimize variability in processing. As a result, the method for content assessment of the FDA-approved capsules and tablets was not the USP method. Similarly, a method was developed to optimize the content determination for the aqueous solution and oil-based suspension because USP methods were unavailable for these formulations. The newly developed methods were then applied to all the liquid formulations to maintain uniformity and to minimize variability in processing. Therefore, the method was different from the published USP method for the FDA-approved suspension of doxycycline.
FDA-approved capsules—Each capsule was opened, the contents were emptied into a 50-mL tube, and the capsule casing was added. The tube contained a volume of 50% methanol (with 2 mg of oxytetracycline/mL added as an internal standard) that would result in a 2-mg/mL concentration of doxycycline when the capsule was added. The tube was sealed, vortexed for 30 seconds, sonicated for 20 minutes, vortexed for another 30 seconds, and then allowed to settle for 15 minutes. The supernatant was diluted to a nominal doxycycline concentration of 20 μg/mL by serial dilution in 50% methanol. The final dilution was performed in the injection vial with deionized water added to create a nominal doxycycline concentration of 2 μg/mL, on the basis of the stated drug content, in 5% methanol (with the internal standard diluted to 2 μg/mL), and the mixture was vortexed for 10 seconds.
Additionally, the residual doxycycline content on capsule casings was assessed by emptying the capsule filling into a plastic tube as described above, but then placing capsule casings in a separate tube. Five capsules were used for the determination of doxycycline content in the capsule casing and filling. The doxycycline content of the capsule casings and filling from the same capsule were determined separately in the same manner as described for whole capsules.
FDA-approved and compounded tablets—Whole tablets were assessed intact and compounded tablets that were scored were assessed as whole tablets and after splitting along score marks. Tablets (or pieces) were added to 50% methanol (with 2 mg of oxytetracycline/mL added as an internal standard) to yield an initial nominal doxycycline concentration of 2 mg/mL on the basis of the stated content for each tablet or fraction of tablet, except that the nominal concentration for whole 150-mg tablets was 3 mg/mL. The tubes were then processed as described for the capsules to yield a nominal doxycycline concentration of 2 μg/mL in 5% methanol, except that the nominal concentration for whole 150-mg tablets was 3 μg/mL.
Chews—Each chew was added to 10 mL of 50% methanol containing 2 mg of oxytetracycline/mL as an internal standard. A homogenizerf was used to disintegrate the chews. The chews were homogenized until the pieces were ≤ 1 mm3 in approximate size, and then 40 mL of the same methanol-based diluent was added to achieve a nominal concentration of 2 mg/mL on the basis of the stated content. The tubes were processed as previously described to yield a final nominal doxycycline concentration of 2 μg/mL in 5% methanol in each injection vial, and the mixture was then vortexed for 10 seconds.
Liquid formulations—Five replicates from each bottle of compounded and FDA-approved liquid doxycycline formulations were tested for content and content variability. Immediately prior to liquid formulation testing, EDTA-McIlvaine buffer was freshly prepared by dissolving 27.6 g of sodium hydrogen phosphate dodecahydrate, 12.9 g of citric acid, and 37.2 g of disodium EDTA dihydrate in approximately 800 mL of deionized water, then further diluting the solution with water to a total volume of 1 L. Hydrochloric acid was added as needed to adjust the pH to 4.0 ± 0.05.
For each formulation, EDTA-McIlvaine buffer containing 2 mg of oxytetracycline/mL was added to 5 tubes to achieve nominal concentrations of doxycycline ranging from 0.5 to 1 mg/mL on the basis of the stated concentration, depending on the initial concentration of the formulation.8 Each sample tube was sonicated for 10 minutes and then vortexed for 30 seconds. The contents were then serially diluted by addition of EDTA-McIlvaine buffer. The final dilution was made in the injection vial with deionized water to yield nominal concentrations ranging from 5 to 6 μg of doxycycline/mL in 5% EDTA-McIlvaine buffer with 2 μg/mL of oxytetracycline.
Reference and internal standard solutions
All reference and internal standard solutions were protected from light and refrigerated at 4°C until used. Buffer solutions containing internal standard were protected from light and refrigerated at 4°C for up to 24 hours until used.
Reference standard doxycycline hydrochloride was weighed to the nearest 0.1 mg and dissolved in a solution of 50% methanol in water to achieve a concentration of 10 mg of doxycycline free base/mL. Oxytetracycline hydrochloride was similarly dissolved in 50% methanol in water or EDTA-McIlvaine buffer to create an internal standard stock solution containing 2 mg of oxytetracycline hydrochloride/mL. The standard curve for doxycycline included 50, 25, 10, 5, 2.5, 1, and 0.5 pg/mL, each with oxytetracycline (2.0 μg/mL), and was prepared daily.
UHPLC method
Published USP HPLC methods9–11 required different approaches for doxycycline tablets and capsules, liquid suspension, and compounded veterinary aqueous suspension, and no published USP methods were available for evaluation of doxycycline chew formulations, oil-based suspensions, or aqueous solutions. Also, the available system in the authors’ laboratory was a technologically advanced UHPLC system, rather than HPLC as described in USP requirements. Therefore, a single analytic UHPLC method was developed and validated in the authors’ laboratory and applied uniformly for assessment of all doxycycline formulations studied as in a previous study.6
Blank (water and methanol) and internal standard injections were made prior to all standard curve injections to monitor for carryover and shifts in chromatographic quality. Thermal equilibration of the mobile phase and column was monitored by assessment of pressure fluctuation and was stabilized before any injections were performed. Analysis of all samples was carried out by UHPLCg with a C18 column.h Samples were analyzed at 350 nm with the photodiode array detector by use of the ratio of the peak area of doxycycline to the peak area of the internal standard oxytetracycline.
A gradient elution with a flow rate of 0.5 mL/min was used. Mobile phase A contained water with 0.2% formic acid, and mobile phase B contained acetonitrile with 0.2% formic acid. Conditions were 98% mobile phase (A) and 2% mobile phase (B) from 0 to 0.1 minutes, followed by a linear gradient to 50% phase A at 0.7 minutes and then a linear gradient to 20% phase A at 1.1 minutes, holding at 20% phase A until 2 minutes, and a stepped return to 98% phase A at 2.01 minutes with a total run time of 3 minutes. The interday accuracy of the assay was 103% at 0.5 μg/mL, 102% at 5 μg/mL, and 99% at 25 μg/mL, with 12 quality control replicates evaluated at each concentration. The interday precision (coefficient of variation) of the assay was 3% at 0.5 μg/mL, 1% at 5 μg/mL, and 3% at 25 μg/mL, with 12 quality control replicates evaluated at each concentration.
Statistical analysis
Measured weights of doxycycline tablets (whole and divided along designated score lines), capsules, and chews on day 1 were reported as the percentage range around the mean. Actual weights of chews on days 1, 21, and 98 were compared within each formulation by 1-way ANOVA with values of P < 0.05 considered significant. Analyses were conducted with statistical software.i Mean and range of drug content results were reported as the percentage of the stated doxycycline drug content for each formulation. For content testing of scored tablets, expected content per piece was adjusted as appropriate (ie, half of a 100-mg tablet would be expected to have 50 mg of doxycycline content and if measured at 50 mg would be expressed as having 100% of the stated content). The a priori acceptance criteria for compounded formulations was measured doxycycline content between 90% and 110% of the stated content, determined on the basis of USP standards for compounded drugs.12 The USP standards acceptance range for FDA-approved doxycycline tablet and capsule (90% to 120%)9,10 or liquid formulations (90°% to 125°%)11 was applied for those products.
Results
Weights were consistent within all tested formulations of whole tablets, capsules, and chews, with a maximum product weight variability of 9% from the mean weight on receipt. When tablets were split once along a single score, weight variability ranged from 10% to 24% from the mean weight of the pieces, depending on the product. When tablets were scored more than once and were divided into > 2 pieces, weight variability ranged from 28% to 35% from the mean weight of the pieces.
Shrinkage of all formulations of chews was visually evident after 98 days of storage (98 days after the prescription was received), although on day 98, 1 of the 3 chew formulations was past the stated beyond-use date. The mean weight of chew K on day 1 was 0.940 g, that on day 21 was 0.901 g (a mean 96% of original mean weight; P = 0.011), and that on day 98 was 0.846 g (a mean 90% of original mean weight; P < 0.001). Mean weights of chew L on days 1, 21, and 98 were 1.060 g, 0.989 g (mean 93% of original mean weight; P < 0.001), and 0.743 g (mean 70°% of original mean weight; P < 0.001), respectively. Mean weights of chew M did not significantly (P = 0.132) change between day 1 (1.020 g) and day 21 (1.041 g), and the labeled beyond-use date had been exceeded for this product by day 98, so it was excluded from further analysis.
All FDA-approved capsules and tablets had appropriate doxycycline content (percentage of the labeled dose within the USP specifications) on day 1 and day 21 (Table 1). When capsule contents alone were tested (without casings), doxycycline content was appropriate, and there was no measurable doxycycline content in the capsule casings alone.
Comparison of doxycycline content in FDA-approved tablets and capsules for human use and compounded tablets and chews for veterinary use.
| Drug content (%) | Proportion within USP specifications | |||
|---|---|---|---|---|
| Product | Day 1 | Day 21 | Day 1 | Day 21 |
| FDA-approved formulations | ||||
| Tablet | 107 (106–109) | 103 (103–104) | 5/5 | 5/5 |
| Capsule | 107 (106–110) | 105 (103–107) | 5/5 | 5/5 |
| Capsule contents only | 108 (106–109) | NA | 5/5 | NA |
| Compounded formulations | ||||
| Tablet A | 89 (88–91) | 88 (86–91) | 3/5 | 1/5 |
| Tablet B | 98 (97–99) | 99 (96–100) | 5/5 | 5/5 |
| Tablet C | 116 (110–123) | 109 (108–112) | 1/5 | 4/5 |
| Chew K | 81 (75–88) | 67 (64–69) | 0/5 | 0/5 |
| Chew L | 78 (74–81) | 82 (74–86) | 0/5 | 0/5 |
| Chew M | 98 (91–106) | 69 (60–76) | 5/5 | 0/5 |
All products were tested by UHPLC. Results are expressed as mean (range) percentage of drug content relative to labeled amount and the proportion of samples tested for each formulation within passing specifications determined by the USP. Acceptable content was 90% to 120% of the labeled amount for FDA-approved products and 90% to 110% of the labeled amount for compounded products. The date of receipt of each product was day 1.
NA = Not applicable.
The only compounded whole tablet or chew with appropriate content for all samples on both days 1 (5/5 tablets) and 21 (5/5) was compounded tablet B (Table 1). Compounded tablet C had 1 of 5 tablets with appropriate content on day 1 with the remainder higher than stated content, but on day 21, content in 4 of 5 was acceptable. Compounded chew M had appropriate content in all samples (5/5) on day 1, but by day 21, none had appropriate content.
Splitting of compounded tablets along the score lines, when applicable, into halves (2 formulations), thirds (1 formulation), or quarters (1 formulation) resulted in fragments of varied doxycycline content, calculated on the basis of stated drug content per fragment. One of the singly-scored tablet formulations had a 100% (10/10 halves) passing rate when split, but the other formulation only had a 60% (6/10) passing rate. The mean doxycycline content of tablets split on 1 score was 101% (range, 96% to 106%) of stated content for one formulation and 89% (range, 70% to 94%) for the other. The mean doxycycline content of tablets cut on 2 score lines was 96% (range, 83% to 111%) and 108% (range, 88% to 123%), with passing rates of 53% (8/15) for one formulation and 65% (13/20) for the other.
The 5 tested replicates of the FDA-approved liquid doxycycline had appropriate doxycycline content on day 1 (mean, 109%; range, 107% to 112%) and on day 21 (mean, 99%; range, 89% to 103%). None of the 3 compounded liquid formulations had doxycycline content within the acceptable content range for any replicate at either testing time. Mean doxycycline content for 5 replicates of liquid X was 50% (range, 46% to 54%) on day 1 and 50% (range, 43% to 69%) on day 21. For liquid Y, mean content of 5 replicates was 85% (range, 83% to 87%) and 59% (range, 52% to 62%) on days 1 and 21, respectively. Finally, mean doxycycline content for 5 replicates of liquid Z was 52% (range, 47% to 58%) and 39% (range, 36% to 43%) on days 1 and 21, respectively.
Discussion
The purpose of the present study was to assess doxycycline content in FDA-approved (for human use) and compounded doxycycline formulations within 24 hours after receipt from the supplier (day clinically relevant question of product potency during the time of intended use. Additionally, the study was not designed to determine the reasons for incorrect potency of any of the formulations. There are many potential reasons for such findings, including a lack of robust beyond-use dates, ingredient effects (eg, added flavorings, suspension agents, or other excipients causing chelation or degradation), incorrect amounts of the active ingredient, degradation of the active ingredient prior to or during compounding, instability of the active ingredient in the compounded formulation, failure to correct for salt formulations of the active ingredient (ie, use of doxycycline hyclate vs doxycycline hydrochloride), and storage time and conditions prior to receipt at the destination.
The FDA-approved doxycycline formulations included in this study have undergone extensive formulation and development testing to assure that the measured drug content relative to the stated content remains within the acceptable range up to the labeled expiration date. As expected, all tested FDA-approved doxycycline formulations had drug content within the acceptable range on day 1 and day 21. The consistency of these findings for the doxycycline capsule, tablet, and liquid formulations approved for use in human patients indicate they are the preferred formulations for use in veterinary patients in the United States at this time.
When compounding is necessary, our results indicated that compounded doxycycline tablets were closest to their stated drug content, whereas compounded liquid or chew formulations were not within acceptable drug content ranges for the duration of the study. One formulation of compounded tablets (tablet B) had acceptable doxycycline content at the beginning and end of the 21-day study. Another compounded tablet formulation (tablet C) tested above the acceptable content range for compounded doxycycline, but this product would have been considered within the wider acceptable content range for an approved doxycycline formulation (acceptability range, 90% to 120% of the stated content10,11). The other compounded tablet tested (tablet A) had doxycycline content slightly below the acceptable range (mean measured doxycycline representing 89% and 88% of the stated content on days 1 and 21, respectively), with 2 of 5 and 4 of 5 tablets having unacceptable values on days 1 and 21, respectively.
It is important to note that appropriate doxycycline content alone should not be interpreted as indicating bioequivalence with an approved product or that the product is efficacious. Information about bioavailability, rate of absorption, and other pharmacokinetic parameters of compounded doxycycline tablets in dogs and cats is lacking. It is also important to realize that batch-to-batch variability in the pharmacokinetics of compounded drugs can occur, and changes in additives (eg, flavoring) or excipients and variability in the compression or strength of the tablet could also affect the pharmacokinetics of compounded tablets. Therefore, a pharmacokinetic study would provide valuable information, but would only be applicable to a specific batch from that specific compounder, and such information should not be extrapolated to other compounded formulations even within the same compounding pharmacy because of a potential lack of formulation homogeneity.
Of the chew formulations, 1 product (chew M) had doxycycline content within the acceptable range on day 1, although this was not the case on day 21. The other tested chews had content considerably lower than the acceptable range on both testing days. All content testing was completed before the beyond-use dates of the chews. There was no FDA-approved chew formulation to test concurrently with similar techniques; however, we did not believe that the low doxycycline content results for chews were attributed to methodological errors because chew M did have drug content within the acceptable range on receipt. These results suggested a loss of potency over time for chew M.
An additional problem that was observed for 2 of the 3 chew formulations was significant decrease in weight on days 21 and 98, compared with that on day 1, despite those products being within the stated beyond-use dates and stored according to directions. The change in weight was considered most likely due to fluid loss. Future formulations of the chews should include greater stability in regard to maintaining weight and content uniformity through the beyond-use date. Alternatively, the beyond-use dates should be markedly decreased. Our results suggested that prescription of doxycycline in a compounded chew formulation should be avoided.
Scoring of tablets is designed to improve dosing options for prescribing clinicians and has become an important tool for veterinarians treating animals of various sizes.13 In the present study, when one of the scored compounded tablet formulations was split in half along a single score, the mean drug content and range remained within the acceptable range, but the second tablet formulation did not have acceptable content in the split pieces. When samples of tablets from 2 pharmacies were split along 2 scores, the variability of drug content increased, and a larger proportion of split pieces had doxycycline content outside of the acceptable range. These results suggested that splitting scored compounded doxycycline tablets could lead to large drug content variability and should be avoided. It is interesting that the premise of compounding is to produce patient-specific formulations, and if this is the case, a scored compounded formulation should not be needed. The AVMA defines compounding consistent with the FDA extralabel drug use regulations as the customized manipulation of an approved drug by a veterinarian, or by a pharmacist upon the prescription of a veterinarian, to meet the needs of a particular patient.1 Collaboration between veterinarians and pharmacists is encouraged to ensure that veterinary patients of all sizes receive customized formulations of doxycycline, and as such, there is no need for compounding scored tablets.1
Content analysis of the liquid doxycycline formulations in the present study should be interpreted differently from the data for solid formulations (tablets, capsules, or chews) because, rather than sampling 5 separate tablets (or capsules or chews) from a single prescription on each day, a single bottle of the FDA-approved formulation and 1 bottle of compounded liquid from each pharmacy had 5 replicates tested on days 1 and 21. The authors speculated that if several bottles were ordered from each pharmacy or distributor, they would be all be filled from 1 batch of the same liquid formulation, which would not truly increase the number of samples from each pharmacy (only the number of replicates). The FDA-approved product and 2 compounded liquid formulations were suspensions that were mixed well prior to sampling, and 1 compounded liquid formulation came in solution. Because 5 replicates were taken from the same bottle, the mean is the main value of importance when interpreting the content results, and the range indicates variability within a single bottle and the importance of mixing suspensions well prior to dosing. That said, the FDA-approved product was the only liquid that had doxycycline content within the acceptable range. All 3 compounded liquid formulations (an aqueous solution, an aqueous suspension, and an oil-based suspension) had doxycycline content less than the stated amount on days 1 and 21, with some of the formulations having a loss of potency over time. A previous study6 to assess doxycycline content of liquid formulations compounded from doxycycline hyclate tablets also found that the drug concentration was decreased substantially between 7 and 14 days after preparation. Taken together, the results of these studies suggest that liquid formulations of compounded doxycycline, whether aqueous or oil based, should be avoided and that the FDA-approved product should be used when a liquid formulation is needed.
As with all studies, the present study had limitations. Tablets, chews, and liquid formulations were chosen for evaluation in this study because they are frequently requested by clients at the authors’ institution; however, further research is required to analyze other formulations such as compounded capsules, powders, and pastes to determine whether they provide acceptable drug content when compounded doxycycline is required. Doxycycline products were evaluated from only 3 compounding pharmacies, and it is possible different pharmacies would have formulations with different measured content. Further studies assessing products from more compounding pharmacies, multiple lots from the same pharmacy, and additional compounded drugs could help determine whether the findings of the present study are indicative of a widespread or more limited problem. It was not the authors’ intent to single out specific pharmacies, but to provide information on compounded doxycycline products advertised at national meetings and generally available for prescription by veterinarians. Previous studies4,5 have also identified problems with drug content in compounded products. Considering that larger independent studies documenting acceptable potency of compounded doxycycline formulations are unavailable, these are presently the only data available on which to base recommendations regarding the use of these products.
The analytic method used in the present study was developed and validated in the authors’ research laboratory. State-of-the-art UHPLC was used, and the methods were applied uniformly to the analysis of all of the doxycycline formulations. The authors contend that if > 3 different USP methods were used to test the different formulations in the study, this could have decreased the robustness of the drug content determinations. Additionally, the authors would have had to downgrade from the UHPLC method to a more dated HPLC method. There are presently no USP content determination methods established for compounded tablets, chews, aqueous solutions, or oil-based suspensions; therefore, our methods were developed to optimize content testing for solid and liquid dosing formulations. Although the methods used were not the USP methods for the FDA-approved formulations, all the FDA-approved solid dose forms and the FDA-approved aqueous suspension still met the USP content specifications, suggesting the methods were indeed robust.
Acknowledgments
Supported in part by the Departments of Clinical Sciences and Anatomy and Physiology and the Institute for Comparative and Computational Medicine (ICCM), College of Veterinary Medicine, Kansas State University, the Morris Animal Foundation Veterinary Student Scholars Program, the Veterinary Research Scholars Program at Kansas State University, and the Dean's Fund at Kansas State University.
The authors declare there were no conflicts of interest.
Presented in abstract form at the 2015 Merial-NIH National Veterinary Scholars Symposium, Davis, California, July–August 2015.
ABBREVIATIONS
| HPLC | High-performance liquid chromatography |
| UHPLC | Ultrahigh-performance liquid chromatography |
| USP | US Pharmacopeia |
Footnotes
KuKanich B, Papich M. Fluoroquinolone stability in vehicles for oral administration (abstr). J Vet Intern Med 2003;17:449.
Dava Pharmaceuticals Inc, Fort Lee, NJ.
Harris Pharmaceutical, Fort Meyers, Fla.
Vibramycin, Pfizer Labs, New York, NY.
Pharmacy information on file with the authors.
Fisher PowerGen 35, Fisher Scientific, Pittsburgh, Penn.
Acquity UPLC H-Class System, Waters, Milford, Mass.
CSH C18, 50 mm × 2.1 mm, 17-μm pore size, Waters, Milford, Mass.
SigmaPlot, version 12.5, Systat Software Inc, Chicago, Ill.
References
1. AVMA. Veterinary compounding. Available at: www.avma.org/KB/Policies/Pages/Compounding.aspx. Accessed Feb 28, 2017.
2. Missouri Division of Professional Registration. Board of Pharmacy annual reports, 2006–2016. Available at: pr.mo.gov/pharmacists-annual-reports.asp. Accessed Mar 16, 2017.
3. Mawby DI, Whittemore JC, Genger S, et al. Bioequivalence of orally administered generic, compounded, and innovator-formulated itraconazole in healthy dogs. J Vet Intern Med 2014; 28: 72–77.
4. Burton JH, Stanley SD, Knych HK, et al. Frequency and severity of neutropenia associated with Food and Drug Administration approved and compounded formulations of lomustine in dogs with cancer. J Vet Intern Med 2016; 30: 242–246.
5. KuKanich B, Warner M, Hahn K. Analysis of lomustine content in FDA-approved and compounded lomustine capsules. J Am Vet Med Assoc 2017; 250: 322–326.
6. Papich MG, Davidson GS, Fortier LA. Doxycycline concentration over time after storage in a compounded veterinary preparation. J Am Vet Med Assoc 2013; 242: 1674–1678.
7. KuKanich K, KuKanich B. The effect of sucralfate tablets vs. suspension on oral doxycycline absorption in dogs. J Vet Pharmacol Ther 2015; 38: 169–173.
8. Guo L, Chen Y, Zhang L, et al. Development and validation of a liquid chromatographic/tandem mass spectrometric method for determination of chlortetracycline, oxytetracycline, tetracycline, and doxycycline in animal feeds. J AOAC Int 2012; 95: 1010–1015.
9. USP. Doxycycline tablets. In: United States pharmacopeia. 38th revision. Rockville, Md: US Pharmacopeial Convention, 2015.
10. USP. Doxycycline capsules. In: United States pharmacopeia. 38th revision. Rockville, Md: US Pharmacopeial Convention, 2015.
11. USP. Doxycycline calcium oral suspension. In: United States pharmacopeia. 38th revision. Rockville, Md: US Pharmacopeial Convention, 2015.
12. USP. Chapter 795: pharmaceutical compounding—nonsterile preparations. In: USP/NF official compendia of standards. 34th revision. Rockville, Md: US Pharmacopeial Convention, 2010.
13. Margiocco ML, Warren J, Borgarelli M, et al. Analysis of weight uniformity, content uniformity and 30-day stability in halves and quarters of routinely prescribed cardiovascular medications. J Vet Cardiol 2009; 11: 31–39.
