Laboratory assessment of trilostane treatment in dogs with pituitary‐dependent hyperadrenocorticism

Abstract Background Results of ACTH stimulation test (ACTHst), pre‐ and post‐trilostane serum cortisol concentrations (SCCs), urine concentration (urine‐specific gravity [USG]), and urine cortisol : creatinine ratios (UCCRs) are common variables used to monitor trilostane treatment of dogs with pituitary‐dependent hyperadrenocorticism (PDH). However, none has consistently discriminated dogs receiving an adequate dose (A) from those overdosed (O) or underdosed (U). Objectives To assess and compare recommended monitoring variables, including serial SCCs in a cohort of dogs with PDH treated with trilostane. Animals Privately owned dogs with PDH (n = 22) and 3 healthy dogs (controls). Methods Prospective, multicenter, 2‐day study. On day “a” (randomized): ACTHst was completed. Day “b” (>2 to <7 days later): SCCs were assessed −0.5 hours, immediately before, and 1, 2, 2.5, 3, 3.5, 4, 6, 8, and 12 hours after trilostane administration. On the first study day, urine collected at home was assessed for USG, UCCR and owner opinions regarding PDH were categorized as: A (clinical signs resolved), U (remains symptomatic), or ill (possible O). Results At 27 pairs of evaluations, 7 dogs were categorized as A, 19 U, and 1 possible O (excluded from the study). There was overlap in SCC results from the A and U dogs at every time point. Results of USG, UCCR, and ACTHst did not discriminate A from U dogs. Trilostane suppresses SCC within 1 hour of administration and its duration of action in most PDH dogs is <8 hours. Conclusions and Clinical Importance No single variable or group of variables reliably discriminated A dogs from U dogs during trilostane treatment for PDH.

mitotane (o,p'-DDD), a cytotoxic drug that targets adrenocortical cells. [1][2][3][4] Since 2000, the most common medical treatment for dogs with PDH has been trilostane, a competitive inhibitor of the 3β-hydroxysteroid dehydrogenase-isomerase enzyme system. [5][6][7][8][9][10][11][12] Several different mitotane protocols were suggested for dogs with PDH. 3,4 Regardless of protocol, authors consistently considered ACTH stimulation test (ACTHst) results to be a reliable objective indicator of mitotane overdose, underdose, or adequate dosage, regardless of when the test was begun relative to time of previous mitotane administration. [1][2][3][4] Perhaps because results of the ACTHst were considered so reliable in monitoring mitotane treatment, the test began being used for the same purpose in trilostane-treated dogs in initial reports [8][9][10][11] and in the manufacturer's insert. 12 Subsequent studies, however, indicated that ACTHst results were influenced by when the test was started relative to the most recent trilostane administration. [13][14][15][16] The drug manufacturer recommends beginning the ACTHst 4 to 6 hours post-trilostane administration, 12 but results obtained with tests started 4 hours after trilostane administration are likely to be different from those obtained when the test is begun 6 hours after, regardless of dose. 15,17 Although no study has validated the 4 to 6 hour recommendation, it has been reported that results of ACTHst initiated 3 hours after trilostane administration were significantly different from tests started 9 hours after, and those started 2 hours after were significantly different from those started 4 hours after. 14,15 Completely different decisions regarding maintaining, increasing, or decreasing trilostane dose or frequency could be the result of test timing. Adding to the confusion, suggested ACTHst starting times in published studies have varied from as early as 2 hours to as late as 12 or 24 hours after trilostane administration. [13][14][15][16][17][18] Regardless of timing issues, and perhaps more important, concerns persist that ACTHst results do not reliably indicate which dogs are overdosed, underdosed, or dosed adequately. 13,14,16,19,20 With so many issues raised regarding the ACTHst, it is not surprising that alternative monitoring variables have been investigated, including endogenous ACTH (eACTH), cortisol : eACTH ratios, urine cortisol : creatinine ratios (UCCRs), and circulating baseline serum cortisol concentrations (SCCs) before or after trilostane administration or both. 17,[20][21][22][23][24][25][26] None of these variables has provided consistently sensitive and specific indications for discriminating dogs adequately dosed from those receiving too little or too much medication. With failure to identify a gold standard for objective monitoring of trilostane treatment, it is also fair to suggest that veterinary clinicians using trilostane to treat PDH either do not have confidence in any proposed monitoring variable or have placed their confidence in a variable that may not reliably provide the information they seek. Our aim was to assess several previously reported and promoted objective variables as well as serial circulating SCCs in a group of dogs with confirmed PDH given trilostane, and categorized subjectively by their owners as receiving an "adequate dose" (A) or being "underdosed" (U). Our hypothesis was that only urinespecific gravity (USG) from trilostane-treated dogs would consistently discriminate those dosed adequately from underdosed dogs.

| Trilostane treatment
The initial trilostane dosage for each PDH dog enrolled was 0.5 to 1.1 mg/kg PO q12h. Each dog must have been treated for a minimum of 30 days and their doses adjusted at previous visits, if such visits were made, in an attempt to achieve clinical control. Commercially available trilostane (Vetoryl, Dechra, Shrewsbury, UK) was used and, if needed, capsules with the calculated dose were prepared by a compounding pharmacy using licensed trilostane. Each owner of a dog enrolled in the study agreed to bring the dog to the hospital for 2 study days, the second no sooner than 3, and no more than 7 days after the first visit. It also was agreed that no change in trilostane dose or frequency of administration would take place between the first and second study days. Diet and environment were kept as stable as possible by the owner of each dog during this intervening period. Any dog evaluated more than once had a minimum of 60 days between the 2 pairs of assessment days. Repeat assessments were evaluated independently.

| Study protocol
Other than not receiving trilostane, the healthy dogs were evaluated in the same manner as the trilostane-treated PDH dogs. Two different sets of evaluation were completed on all dogs. At each evaluation, a complete physical examination was performed after asking the owner about the dog's general well-being. Owners of PDH dogs were specifically questioned on the first study morning regarding their dog's response to trilostane, understanding that their primary reason for pursuing treatment was resolution of clinical signs. From that ownerveterinarian conversation, each veterinarian confirmed with the owner 1 of 3 subjective conclusions: that their dog's clinical signs had resolved (adequate dose, A), that their dog was symptomatic and likely would benefit from additional medication even if some improvement in PU, PD, PP or some combination of these had been noted (underdose, U), or that their dog was ill (decreased or no appetite, vomiting, diarrhea or some combination of these) and possibly overdosed (O).
Any dog considered ill by the owner or considered ill after the physical examination was to be removed from the study and treated as necessary.
Dogs were randomly assigned to be first evaluated with either the "a" or "b" protocol. Owners were asked to refrain from feeding their dogs on study days and to bring food and trilostane to the hospital with their dogs 60 minutes before trilostane was to be administered. This time might have allowed some dogs to become acclimated to the hospital while the owner history was obtained.
Urine, collected and brought in by the owner on the first of 2 study mornings, was assessed for USG, analytes, and UCCR. On study day "a," 2 hours after trilostane administration, blood was obtained for CBC, serum biochemistry profile, and baseline SCC. Adrenocorticotropic hormone (Nuvacthen. Afasigma S.P.A. Via Ragazzi, Bolonia. Italy) then was administered (5 μg/kg, IM) and blood collected 1 hour later. On study day "b," each dog had an IV catheter placed 15 minutes before approximately 1 mL of blood was obtained for SCC 30 minutes (−0.5 hour) before trilostane administration. Samples also were obtained via the catheter immediately before (0 hour) and 1, 2, 2.5, 3, 3.5, 4, 6, 8, and 12 hours after trilostane was given.
The 12-hour blood sample was obtained immediately before trilostane was again given and also could be considered a 0 hour SCC.
F I G U R E 1 Trilostane dose versus duration of trilostane treatment in dogs considered to be underdosed (U) and to be adequately dosed (A) F I G U R E 2 Results of the pre-ACTH cortisol concentrations for dogs included in the study. First column represents cortisol values for dogs in the underdosed (U) group and the second column the values for the dogs in the adequately dosed (A) group

| Assessments
Owner opinion (A, U) was the standard against which the studied variables were assessed. Those variables included results of USG, UCCR, ACTHst, pretrilostane basal SCCs, and post-trilostane serial SCCs.

| Hormone assays
Serum and urine cortisol concentrations were measured using of a commercial cortisol assay (Immulite 2000, Siemens Healthcare Diagnostics, Cornellà del Llobregat, Barcelona, Spain) that has been validated for use in dogs. The sensitivity of this assay was 0.5 μg/dL (13.8 nM/L).  Except for the ill dog, the CBC, serum biochemistry, and urinalysis results from each dog were within reference limits or had alterations expected in dogs with HAC.

| Classification and trilostane doses
Based on owner observations, the 1 ill dog was removed from the study but classified as a possible overdose, 7 were placed in the A Results of the serial SCCs (μg/dL) throughout the day    The mean prepill SCC at −0.5 hour for all PDH dogs was significantly lower than at 0 hour (P = .01) and no significant difference (P > .05) was found between their prepill SCC at 0 hour and their prepill SCC at 12 hours. The pretrilostane SCC means of A dogs and U dogs were not significantly different (P > .05; Figure 8). There was overlap in results of the SCC from both groups (Figures 6, 7, and 8).
As seen in Table 1 and Figures 6 and 7, SCCs decreased approximately 2 hours post-trilostane administration in all treated dogs, followed by increasing SCCs thereafter ( Figure 6).

| DISCUSSION
Our aim was to assess the reliability of several commonly used objec-  (7), however, suggests that such low starting doses may extend the time needed to resolve clinical signs. As seen in Figure 1, of the 11 studies on dogs treated <10 months, 9 were described by owners as U, whereas 3 of 5 studies on dogs treated >20 months were described as A. Low initial doses and the additional time required to achieve satisfactory control could lead to owner disappointment, frustration, and dissatisfaction.
The issue of starting dosage remains to be better understood.
Some authors have evaluated pretrilostane-administration SCCs to assess HAC control and need for dose adjustment. 21,33,34 This approach suggests that SCCs after trilostane effects have waned will provide objective information regarding the dose and frequency needs of the dog. Results of 1 study suggested that pre-trilostane and 3 hour post-trilostane SCCs potentially were better monitoring variables than ACTHst results, 21 but many studies have suggested ACTHst results to be of questionable monitoring value. Furthermore, there was no clear indication of when the pretrilostane SCCs were measured relative to timing of the next dose or relative to the timing of arrival at the hospital. In a subsequent study, pretrilostane SCCs were found to be more consistent and reliable than 3 hour post-trilostane SCCs. 33  Retrospectively, an additional time point SCC assessment at 11.5 hours of the serial cortisol concentration study also would have provided additional data regarding sampling "30 minutes before" trilostane administration. As discussed, a study of overdosed dogs will be important as would a study of dogs with adrenaldependent hyperadrenocorticism.
Many of the tests most commonly used in monitoring dogs being treated with trilostane for PDH were included in our study. Furthermore, serial SCCs also were obtained beginning before and extending 12 hours after trilostane administration. No variable consistently discriminated dogs receiving an adequate dose from dogs that continued to exhibit clinical signs. As discussed, these results support the importance of knowing the owner's and veterinarian's goals for treatment, and these should be aligned before beginning treatment. Based on results of our study and many others, veterinarians treating dogs with trilostane for PDH are encouraged to develop an effective team approach. Valuable members of the team include the owner in addition to the veterinarian. Adequate time must be available for both team members when reevaluating these dogs so as to fully understand owner observations.

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 hospital board of the Veterinary Teaching Hospital Complutense, Las Palmas de Gran Canaria and from the Hospital Auna Especialidades Veterinarias.

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