Robenacoxib for Feline Musculoskeletal Pain: COX-2 Selectivity, DJD Evidence, Safety, and the ISFM/AAFP Framework

Bottom line

• Robenacoxib (Onsior) is the only licensed coxib-class NSAID for cats. Its COX-2 tissue selectivity (feline whole-blood in vitro IC₅₀ COX-1:COX-2 ratio 32:1, Schmid assay), short blood half-life (<2 h), and selective persistence in inflamed tissue together underpin its feline safety profile and rationalize once-daily dosing.[1]
• For chronic degenerative joint disease-associated pain (DJD-pain), the pivotal evidence is a randomized, blinded pilot RCT in 109 cats (Adrian 2021): robenacoxib produced a 49% reduction in owner-assessed disability (p = 0.01, effect size ~0.3) and a higher categorical success rate at 6 weeks (p = 0.018, NNT 3.8) versus placebo. Objective accelerometry favored robenacoxib for the proportion of cats gaining >10% activity (p = 0.046 at 3 weeks; p = 0.026 at 6 weeks) but was underpowered as a continuous endpoint.[2]
• On chronic safety, a pooled analysis of 4 RCTs (King 2021; n = 449, 4–12 weeks) found the adverse-event relative risk was 1.15 (95% CI 0.93–1.43) versus placebo overall, and 1.09 (95% CI 0.78–1.52) in a 126-cat CKD subgroup — neither significant. A small but statistically significant creatinine rise (+4.36 µmol/L) occurred without clinical consequence.[3]
• Comparator and perioperative data are accruing: grapiprant (EP4 antagonist) was non-inferior to robenacoxib for post-ovariohysterectomy analgesia (Pisack 2024), and an NSAID-timing RCT (Krekis 2024) found no renal-marker difference between robenacoxib and meloxicam across dosing windows.[5][6]
• The 2024 ISFM/AAFP consensus endorses individualized risk–benefit assessment with serial renal monitoring — not categorical avoidance — for NSAIDs in cats, naming robenacoxib and meloxicam as the most studied agents and supporting cautious use in stable IRIS Stage 1–2 CKD.[4]
• This is a clinician-facing evidence summary, not a dosing protocol. Confirm regimen, monitoring, and contraindications against current product labeling and a veterinary formulary.

Drug facts

• Class: Coxib-class NSAID (selective COX-2 inhibitor); diphenylamine derivative.[1]
• Mechanism / COX selectivity: Preferential COX-2 inhibition via a diphenylamine scaffold. Feline whole-blood in vitro IC₅₀ COX-1:COX-2 ratios are assay-dependent — 32:1 (Schmid) and higher in other assays — but all confirm significant COX-2 inhibition with relative COX-1 sparing at registered doses. Anchor clinical framing to the conservative 32:1 figure rather than lower secondary estimates.[1]
• Pharmacokinetics (cats): Plasma protein binding >98%; terminal blood half-life <2 h (geometric mean ~1.1–1.5 h); low body clearance; biliary excretion predominates (~72% in cats), with minimal renal elimination. The drug partitions into and persists in inflamed tissue, enabling once-daily efficacy despite the short blood half-life.[1]
• Licensed dosages (cats): Oral 1–2.4 mg/kg once daily; 2 mg/kg SC once (or up to 2 consecutive days perioperatively). Always confirm the dose range and approved treatment duration with current product labeling.[1]
• Approved indications (cats, EU/EMA): Acute pain and inflammation associated with musculoskeletal disorders (oral tablets, up to 6 days in many markets); perioperative pain (injection). Confirm local-market authorization before prescribing for chronic pain beyond the labeled duration.[1][4]
• Contraindications / cautions: Concurrent NSAID or corticosteroid use is contraindicated regardless of renal status. Avoid in dehydration, active vomiting/anorexia, uncontrolled hypertension, decompensated (IRIS Stage 3–4) CKD, and known hypersensitivity.[4]
• Key species note: Cats have limited glucuronyl transferase capacity, making NSAID selection and dosing interval particularly important. Robenacoxib's short blood half-life and biliary elimination partially mitigate — but do not eliminate — NSAID risk in this species.[1][4]

Efficacy: what robenacoxib does for feline DJD pain

The pivotal efficacy evidence is the Adrian et al. (2021) randomized, blinded pilot clinical trial in 109 client-owned cats with chronic DJD-pain, using a parallel-group, three-arm design (placebo–placebo [PP], robenacoxib–robenacoxib [RR], robenacoxib–placebo [RP]; each period 3 weeks) with both objective actimetry and owner-assessed endpoints.[2]

Owner-assessed disability fell 49% in robenacoxib-treated cats versus placebo after 6 weeks (p = 0.01, effect size ~0.3), with significant improvements in owner-reported temperament (p = 0.0039) and happiness (p = 0.021).[2] The categorical success rate at 6 weeks favored robenacoxib (p = 0.018, NNT 3.8).[2] Objective accelerometry was directionally consistent — mean total activity was numerically 5.7% higher and significantly more robenacoxib cats achieved a >10% activity increase at 3 weeks (p = 0.046) and 6 weeks (p = 0.026) — but the continuous activity analysis did not reach significance (p = 0.24).[2]

The methodological signal is as important as the efficacy signal. Actimetry in cats is technically difficult: 82.4% of all activity values were zero, reflecting the sedentary, burst-activity feline pattern that erodes statistical power for continuous accelerometry.[2] The trial therefore supports owner-reported disability scales and global assessments — not raw activity counts — as the more sensitive, deployable primary endpoints for confirmatory feline pain trials. This matters because feline analgesic trials carry a well-documented caregiver placebo effect, which is precisely why blinded, placebo-controlled designs and validated owner instruments (rather than unblinded impressions) are required to establish true drug effect.[8]

Comparators: robenacoxib versus grapiprant

As of 2024, the head-to-head comparator evidence is perioperative rather than chronic. Pisack et al. (2024) conducted a prospective, randomized, masked, non-inferiority trial comparing grapiprant with robenacoxib for post-ovariohysterectomy analgesia in cats and concluded grapiprant was non-inferior to robenacoxib on the primary pain endpoint.[5] Clinically, this establishes that selective EP4 antagonism (grapiprant) and COX-2 inhibition (robenacoxib) deliver comparable acute perioperative analgesia in this population — it does not extend to chronic DJD, where robenacoxib retains the placebo-controlled RCT evidence (Adrian 2021) and grapiprant does not have an equivalent feline chronic-pain dataset. Treat the two as mechanistically distinct options whose chronic-pain equivalence in cats is not yet established.

Safety and adverse events in chronic use

The most comprehensive safety dataset is King et al. (2021), a pooled analysis of 4 prospective, randomized, blinded trials — 449 client-owned cats with chronic musculoskeletal disease (robenacoxib n = 222, placebo n = 227), dosed once daily for 4 to 12 weeks (individual trials: 12, 4, 6, and 6 weeks; target 1 mg/kg, range 1–2.4 mg/kg orally).[3] Adverse events did not differ significantly: 47.8% (robenacoxib) versus 41.0% (placebo) had at least one AE, RR 1.15 (95% CI 0.93–1.43, P = .15); the number of clinical signs per cat (range 0–9) also did not differ (P = .23).[3]

The one statistically significant biochemical signal was a marginally higher serum creatinine during robenacoxib administration (+4.36 µmol/L, 95% CI 0.21–8.50, ~0.05 mg/dL) with no clinically detectable adverse renal effect — small in magnitude but sufficient to justify periodic creatinine monitoring on long-term therapy.[3] Adverse-event frequencies were likewise similar across arms in the Adrian 2021 DJD trial, consistent with this profile.[2]

Special populations: robenacoxib in cats with CKD

CKD is the comorbidity that most often constrains long-term NSAID use in cats, and it co-occurs heavily with DJD — CKD was present in 68.8% of cats recruited for DJD studies in one retrospective cohort — which makes this the central safety question for the DJD-pain population.[7]

The reassuring evidence. In the King 2021 pooled analysis, the pre-specified 126-cat CKD subgroup showed an adverse-event RR of 1.09 (95% CI 0.78–1.52, P = .61) versus placebo — no significant increase over 4–12 weeks.[3] Krekis et al. (2024), a prospective randomized blinded trial in 60 cats undergoing ovariohysterectomy comparing robenacoxib (2 mg/kg SC) and meloxicam (0.2 mg/kg SC) across three timing windows, found no significant difference in creatinine or other renal markers across groups; plasma renin activity rose at end-of-surgery independently of drug or timing (i.e., anesthesia itself activates the renin–angiotensin system), and thromboxane B2 was suppressed less by robenacoxib than meloxicam (p = 0.001), consistent with robenacoxib's higher COX-2 selectivity and potentially more renal-prostaglandin-sparing profile.[6]

The countervailing caution. These datasets do not license blanket use. King 2021 excluded cats with severe or uncontrolled concomitant disease — IRIS Stage 3–4 CKD, uncontrolled hypertension, and active proteinuria are unrepresented, so the RR ~1.09 cannot be extrapolated to decompensated renal failure.[3] Krekis 2024 recorded one suspected IRIS Grade II AKI (in a meloxicam-at-induction cat), which the authors attributed to individual patient vulnerability rather than a class effect — a reminder that patient selection and administration context can matter more than drug choice alone.[6]

Neutral synthesis. The evidence converges on a monitored, individualized approach rather than either categorical avoidance or unrestricted use: robenacoxib is a defensible option in stable IRIS Stage 1–2 CKD with controlled blood pressure and adequate hydration when musculoskeletal pain is otherwise unmanaged, provided baseline and serial renal monitoring are in place; it should generally be avoided in Stage 3–4 disease unless analgesic options are exhausted. This is exactly the position the 2024 ISFM/AAFP consensus adopts.

The multimodal / guideline framework

The 2024 ISFM/AAFP consensus guidelines on long-term NSAID use in cats (Taylor et al., J Feline Med Surg 2024) provide the framework within which robenacoxib decisions should be made.[4] Key positions: chronic musculoskeletal pain in cats is undertreated and prolonged pain carries its own physiologic cost; individualized risk–benefit assessment is endorsed over blanket contraindication in stable CKD; robenacoxib and meloxicam are the most widely studied feline NSAIDs; COX-2-selective agents are preferred when NSAID therapy is needed in renally compromised patients; and IRIS Stage 1–2 cats may be candidates for a monitored NSAID trial whereas Stage 3–4 cats require substantially greater caution.[4] The guidelines fold NSAID therapy into a broader multimodal plan (weight optimization, environmental modification, and adjunctive analgesics as indicated) and repeatedly flag the cat's unique metabolism (limited glucuronidation) as a species-specific prescribing consideration.[4]

Contraindications, precautions & PK

• PK summary (cats): >98% protein bound; terminal blood half-life <2 h (~1.1–1.5 h geometric mean); low body clearance; ~72% biliary excretion with minimal renal elimination; tissue-selective accumulation at inflamed sites drives once-daily efficacy.[1] The preclinical margin is wide — healthy cats tolerated 8× the clinical oral dose for 6 weeks — but clinical populations with comorbidities still warrant monitoring.[1]
• Absolute/relative contraindications: concurrent NSAID or corticosteroid; dehydration; active GI signs (vomiting/anorexia); decompensated (IRIS Stage 3–4) CKD; uncontrolled hypertension; hypersensitivity; concurrent nephrotoxic drugs.[4]
• Drug-interaction screen before starting: identify concurrent ACE inhibitors, ARBs, corticosteroids, or diuretics, which compound renal risk.[4]
• Monitoring: baseline biochemistry including creatinine, SDMA, and UPC (with blood pressure) before initiation; recheck at 2–4 weeks, then every 3–6 months for stable patients; monitor more frequently if CKD stage advances or clinical signs change. Suspend and reassess if creatinine rises >25% from baseline or SDMA crosses an IRIS staging threshold.[4]
• Duration caveat: the labeled oral duration in many markets (6 days) reflects the initial regulatory authorization; the 4–12 week safety trials informing King 2021 extend beyond the label. Confirm local-market authorization before prescribing beyond the label.[3]

Practical decision support

• Reasonable candidate: a cat with chronic DJD-pain and no absolute contraindication; for CKD patients, stable IRIS Stage 1–2 with controlled blood pressure, adequate hydration, and pain unresponsive to other modalities.[4]
• Expected benefit: owner-assessed disability improvement on the order of the Adrian 2021 signal (49% reduction; NNT ~3.8 at 6 weeks). Set owner expectations around functional/behavioral endpoints (disability, temperament, happiness) rather than raw activity counts, which are noisy in cats.[2]
• Efficacy monitoring: use a validated owner-reported instrument, not unblinded clinical impression, because of the caregiver placebo effect in feline analgesia trials.[8]
• Perioperative context: robenacoxib and grapiprant are comparable for acute post-OHE analgesia; chronic equivalence in cats is not established.[5]
• Renal-risk mitigation: anesthesia itself activates the renin–angiotensin system, so ensure normovolemia and blood-pressure support perioperatively regardless of which NSAID is chosen; individual patient vulnerability, not drug class alone, drives most renal AEs.[6]
• Do not infer specific mg/kg doses from this summary. Always confirm the exact regimen, interval, and contraindications against current product labeling and a veterinary formulary before prescribing.

Frequently Asked Questions

What is the COX-2 selectivity of robenacoxib in cats? In feline whole-blood assays, robenacoxib's in vitro IC50 COX-1:COX-2 ratio is 32:1 (Schmid assay), meaning it inhibits COX-2 at concentrations roughly 32 times lower than those needed to inhibit COX-1; other assays report higher ratios. At registered oral dosages of 1–2.4 mg/kg it produces significant COX-2 inhibition while relatively sparing COX-1 (Lees et al., J Vet Pharmacol Ther 2022).

What efficacy did robenacoxib show in the feline DJD pilot RCT? In the Adrian et al. (Sci Rep 2021) randomized blinded pilot RCT (n=109 cats with DJD-pain), robenacoxib produced a 49% reduction in owner-assessed disability (p=0.01, effect size ~0.3) and a higher categorical success rate at 6 weeks (p=0.018, NNT 3.8) versus placebo. Owner-reported temperament (p=0.0039) and happiness (p=0.021) also improved significantly.

Is robenacoxib safe in cats with chronic kidney disease? In the King et al. (JVIM 2021) pooled analysis of 4 RCTs, the 126-cat CKD subgroup had an adverse-event relative risk of 1.09 (95% CI 0.78–1.52; P=.61) versus placebo — not significant — with a small but significant creatinine rise (+4.36 µmol/L) of no clinical consequence. Cats with severe or uncontrolled disease were excluded, so these data support cautious, monitored use in stable IRIS Stage 1–2 CKD but do not cover decompensated (Stage 3–4) disease.

Why does robenacoxib have once-daily dosing despite a short blood half-life? Robenacoxib's terminal blood half-life is under 2 hours, but it selectively distributes into and persists in inflamed tissue, where it continues to exert local COX-2 inhibition after blood concentrations fall. This tissue-selective accumulation, together with ~72% biliary excretion, is the pharmacokinetic basis for effective once-daily dosing (Lees et al., J Vet Pharmacol Ther 2022).

How does robenacoxib compare to grapiprant for pain in cats? In a prospective randomized masked non-inferiority trial in cats undergoing ovariohysterectomy (Pisack et al., J Feline Med Surg 2024), grapiprant was non-inferior to robenacoxib for postoperative pain, indicating comparable acute perioperative analgesia. This equivalence is perioperative only — robenacoxib retains placebo-controlled RCT evidence for chronic DJD pain (Adrian 2021) that grapiprant does not yet have in cats.

Does perioperative NSAID timing affect renal function in cats? In a randomized blinded trial in 60 cats undergoing ovariohysterectomy (Krekis et al., J Vet Pharmacol Ther 2024), timing of robenacoxib or meloxicam (admission, induction, or end of surgery) did not produce significant differences in creatinine or other renal markers. Plasma renin activity rose at end-of-surgery independently of drug or timing, showing anaesthesia itself activates the renin–angiotensin system; one meloxicam cat developed suspected IRIS Grade II AKI, attributed to individual vulnerability rather than a class effect.

What pre-treatment screening do the 2024 ISFM/AAFP guidelines recommend before long-term NSAIDs in cats? The 2024 ISFM/AAFP consensus (Taylor et al., J Feline Med Surg 2024) recommends a baseline evaluation of renal markers (creatinine, SDMA, urine specific gravity, UPC where indicated), hepatic parameters, and blood pressure, plus a medication history to flag interacting drugs (ACE inhibitors, ARBs, corticosteroids, diuretics). Recheck at 2–4 weeks, then every 3–6 months for stable patients.

How long has robenacoxib been studied for chronic use in cats? The longest pooled chronic dataset is King et al. (JVIM 2021): 4 prospective RCTs covering 4 to 12 weeks of once-daily oral robenacoxib (target 1 mg/kg; range 1–2.4 mg/kg) versus placebo in 449 client-owned cats with chronic musculoskeletal disease. Adverse-event risk did not differ significantly from placebo (RR 1.15; 95% CI 0.93–1.43). Note the labeled oral duration in many markets is 6 days, so use beyond that requires confirming local authorization. See the FAQ section rendered from structured metadata.

Changelog

• 2026-06-23: First published (hub).
• 2026-07-06: Consolidated evidence hub. Folded in four dated dispatch spokes — Adrian 2021 pilot RCT (endpoint methodology + caregiver-placebo context via Gruen 2017), King 2021 pooled chronic safety (CKD co-prevalence context via Marino 2014), the June-13 DJD-activity dispatch (added Pisack 2024 grapiprant non-inferiority comparator), and the June-11 CKD-safety dispatch (added Krekis 2024 NSAID-timing renal RCT and ISFM/AAFP IRIS-staging detail). Deduplicated repeated Adrian/King/Taylor evidence to a single appearance each; reconciled the COX-1:COX-2 selectivity ratio to the cited Lees 2022 figures (32:1 conservative, assay-dependent) rather than the looser 10:1 secondary estimate.

---

For licensed veterinary professionals. Evidence summary, not clinical advice. Voyage is decision support, not a substitute for clinical judgment — always confirm dosing and treatment with current formularies and your own clinical judgment.