兽研局部纳米包裹植物制剂单独或联合oclacitinib治疗犬特应性皮炎：一项探索性临床研究 Sara Daniela Risse 巴拉那天主教大学 Vanessa Cunningham Gmyterco 巴拉那天主教大学 Michel Fleith Otuki 巴拉那联邦大学 Fernanda Gabriela Menegon 巴拉那天主教大学 Maitiani Festinalli 巴拉那天主教大学 Marconi Rodrigues de Farias 巴拉那天主教大学 Fernando Bittencourt Luciano 巴拉那天主教大学 简报 关键词：迷迭香、金盏花、洋甘菊、瘙痒、CADESI-4、多模式治疗 发布日期：2026年10月23日 DOI：https://doi.org/10.21203/rs.3.rs-11310195/v1 许可：本作品根据知识共享署名4.0国际许可协议授权。阅读完整许可 附加声明：未报告利益冲突。第1页/共11页
摘要 犬特应性皮炎（cAD）通常需要多模式管理，一种能够在无需全身免疫调节的情况下实现有意义的临床控制的局部治疗将具有临床价值。这项探索性三臂研究在30只客户拥有的轻至中度cAD犬（每组10只）中比较了一种新型局部纳米包裹植物制剂（Nanobalm）、口服oclacitinib及其联合治疗。在第0、7、15、30和45天记录CADESI-4和瘙痒视觉模拟量表（pVAS）评分。纵向混合效应模型显示CADESI-4（似然比χ²=48.79，df = 8，p < 0.001）和pVAS（χ²=23.57，df = 8，p = 0.0027）存在显著的组别×时间交互作用。在第45天的事后探索性非劣效性分析中，局部单药治疗相对于基线的百分比降低符合对oclacitinib的非劣效性，使用基于文献的15个百分点界值，CADESI-4（调整后G1-G2差异 + 3.3个百分点；95% CI -9.9至16.5；pNI = 0.0047）和pVAS（+ 15.0；95% CI -4.5至34.4；pNI = 0.0023）均如此；使用20个百分点界值时结论不变。联合治疗产生最一致的应答，10/10只犬的CADESI-4降低≥ 50%，9/10只犬的pVAS降低≥ 50%。这些发现支持该局部制剂作为一种有前景的独立选择，并提示与oclacitinib联合时具有额外临床获益，但仍需在前瞻性把握度充分的试验中加以确认。引言 犬特应性皮炎（cAD）是一种慢性瘙痒性和炎症性皮肤病，通过相容的临床特征并排除其他瘙痒性疾病来诊断（Hensel等，2015）。管理通常为多模式，结合发作因素控制与止痒或抗炎治疗，并在适当时进行局部皮肤屏障支持（Nuttall等，2019；Mueller等，2026）。Oclacitinib可在许多患病犬中快速控制瘙痒和皮肤病变，但长期管理常需要个体化的干预组合（Cosgrove等，2013；Mueller等，2026）。局部植物产品在兽医皮肤病学中有所应用，尽管临床证据异质且具有制剂特异性（Tresch等，2019）。一种含有洋甘菊、金盏花和迷迭香提取物的纳米包裹局部制剂可能提供局部抗炎和皮肤支持活性，而无需全身JAK抑制。这一区别具有临床相关性：一种局部单药治疗如果能够实现病变严重程度和瘙痒的持续改善，接近已建立的全身免疫调节剂，则可拓宽长期管理选择，尤其是作为多模式cAD护理的一部分。本探索性研究的目的是比较接受局部制剂单药、oclacitinib单药或两种治疗的轻至中度cAD犬的瘙痒和病变严重程度变化，并评估短期耐受性。由于每组包括10只犬，该研究被解释为产生假设，而非 superiority 或协同作用的确定性检验。第2页/共11页
材料与方法 共纳入30只先前诊断为轻至中度犬异位性皮炎（cAD）的客户拥有犬。所有犬在纳入前均已进行至少八周的消除饮食，并已排除感染性和寄生虫性疾病。排除继发细菌或真菌感染，或在前一个月内接受过口服或外用糖皮质激素、全身性抗生素或抗菌洗剂的犬。本研究经巴西库里蒂巴巴拉那天主教大学动物使用伦理委员会（CEUA）批准（方案编号2469200524）。犬只通过抽签随机分配至三组（每组n=10）。第1组接受外用纳米包裹洋甘菊、金盏花和迷迭香提取物（Nanobalm；Wesen Green Solutions Ltda），每12小时涂抹于指间、腋下、腹股沟区域及耳廓，持续45天。第2组接受口服奥拉替尼（Apoquel；Zoetis）0.4–0.6 mg/kg，每12小时一次，给药14天，此后每日一次至第45天。第3组接受两种方案联合治疗。外用产品轻轻按摩入指定区域，不予冲洗。允许使用美容性非药物洗发剂，但禁止使用保湿或其他外用屏障产品。未进行正式的先验样本量计算。在研究规划时，尚无关于所研究外用制剂（单用或与奥拉替尼联用）预期效应量的可靠估计，且招募仅限于符合预定纳入标准的客户拥有犬。因此，本研究设计为探索性研究，每组10只犬。对最终样本的敏感性分析表明，在双侧α=0.05时，每组10只犬仅对大的配对标准化效应（Cohen's d≈1.32）提供80%把握度；对于总体三组比较（N=30），80%把握度对应Cohen's f≈0.60。因此，本研究检测或排除小至中等程度治疗差异的能力有限。下文描述的非劣效性分析为事后添加，用于量化最终外用单药治疗反应与奥拉替尼的比较，并解释为探索性而非确证性。临床评估在第0、7、15、30和45天进行。犬主使用经验证瘙痒视觉模拟量表（pVAS）对瘙痒进行评分（Rybníček等，2009）。皮肤病变由兽医使用CADESI-4评分（Olivry等，2014）。安全性监测包括红斑或皮肤体征加重、涂抹时不适或抗拒、过度舔舐、流涎、黏膜病变、呕吐及随访期间报告的任何其他不良事件。主要纵向分析采用线性混合效应模型，以犬为随机截距，固定效应包括治疗组、天（分类变量）及组别×天交互作用。CADESI-4因未转换模型残差非正态，经log(评分+1)转换后分析；pVAS按原始尺度分析。总体交互作用通过全交互模型与相应加性模型的似然比检验进行评估。交互作用显著后，在每个时间点估计双侧组间配对对比，并在每个时间点内使用Holm法进行调整。第3/11页
除非另有说明，数据以中位数[四分位距]报告。由于组样本量小且评分分布特殊，采用非参数分析（Kruskal-Wallis检验及双侧Mann-Whitney U事后比较）作为敏感性分析。探索性应答者分析将临床有意义改善定义为第45天时个体基线评分较基线降低≥50%；应答者比例采用双侧Fisher精确检验进行比较，两两比较采用Holm校正。此外，一项事后探索性非劣效性分析在第45天比较了局部Nanobalm单药治疗（G1）与奥拉替尼单药治疗（G2），采用CADESI-4和pVAS较基线的百分比降低值。百分比降低值采用以中心化基线评分作为协变量的ANCOVA进行分析。15个百分点的非劣效性界值选自既往一项比较奥拉替尼与环孢素的犬特应性皮炎试验（Little等，2015），而非来自观察到的治疗差异；20个百分点的界值（如近期一项JAK抑制剂与奥拉替尼比较的研究所采用，Forster等，2025）作为敏感性分析进行评估。由于评估了两项临床结局，采用保守的单侧α=0.025。当校正后G1-G2百分比降低值差异的95%双侧置信区间下限大于−15个百分点时，认为非劣效性成立。此外，还探索性地采用三组第45天百分比降低值的ANCOVA（事后比较采用Holm校正）来描述联合治疗的应答特征。图1和图2中，中心点代表每次访视时的组中位数，误差线代表四分位距（IQR）。分析采用Python 3.13、SciPy 1.17.0和statsmodels 0.14.6完成。在适用处报告效应估计值、精确p值和多重性校正p值。结果 入组的30只犬包括16只雌性（53.3%）和14只雄性（46.7%），年龄1-13岁（平均5.7岁）。混种犬最常见（33.3%），其次为西施犬（16.7%）、约克夏梗（13.3%）和法国斗牛犬（10.0%）。第2组和第3组性别均衡，而第1组70%为雄性。纵向pVAS轨迹在各组间存在显著差异（组×时间交互作用：似然比χ²=23.57，df=8，p=0.0027）。第1、2、3组的基线pVAS中位数分别为6.0、6.0和6.5。第45天时，第1组pVAS中位数[IQR]为3.0[2.25-3.0]，第2组为3.5[3.0-5.75]，第3组为2.0[2.0-2.0]（图1）。第45天混合模型对比中差异最大者为第2组与第3组之间（估计平均差异1.9个pVAS点；名义p=0.020），但该对比经Holm校正后未保持在0.05阈值以下（校正p=0.060）。从基线至第45天pVAS降低≥50%者，第1组为6/10（60%），第2组为2/10（20%），第3组为9/10（90%）。第3组应答者比例高于第2组（双侧Fisher精确检验p=0.0055；Holm校正p=0.016）。当采用基线校正ANCOVA分析全部三组第45天百分比降低值时，第1、2组校正后降低值分别为51.7%、36.5%，第3组为62.5%，第4页/共11页
分别为3；第3组的改善程度大于第2组（差异26.0个百分点；Holm校正p = 0.0118），而第3组与第1组相比无显著性差异（校正p = 0.199）。纵向CADESI-4轨迹在各组间也差异显著（对数转换CADESI-4的组×时间交互作用：似然比χ²=48.79，df = 8，p < 0.001）。第1、2、3组的基线中位数分别为29.0、28.5和32.5。第15天时，CADESI-4评分中位数分别为18.0、10.0和20.0，事后混合模型对比显示第2组病变评分低于第1组和第3组（Holm校正p = 0.010和p = 0.006）。第45天时，中位数[IQR]评分分别为16.0 [11.0–18.0]、15.5 [11.75–17.75]和11.5 [9.25–14.0]；经多重性校正后，第45天原始评分的两两对比均无显著性差异（图2）。从基线至第45天CADESI-4降低≥50%的犬只分别为第1组4/10（40%）、第2组3/10（30%）和第3组10/10（100%）。第3组的应答者比例高于第1组（双侧Fisher精确检验p = 0.0108；Holm校正p = 0.022）和第2组（p = 0.0031；Holm校正p = 0.009）。在三组基线校正的百分比降低分析中，第1、2、3组第45天的校正改善分别为50.8%、45.8%和66.3%。第3组的百分比改善大于第2组（差异20.5个百分点；Holm校正p = 0.0096）和第1组（差异15.5个百分点；校正p = 0.0345）。代表性临床进展见图3。第45天外用Nanobalm与oclacitinib的探索性非劣效性分析对两项疗效结局支持相同的临床解读。对于CADESI-4，第1组和第2组从基线的未校正平均百分比降低分别为50.2%和43.6%；经基线严重程度校正后，估计降低分别为48.6%和45.2%。校正后G1-G2差异为+3.3个百分点（95% CI -9.9至16.5），置信下限仍高于−15个百分点的界值（pNI = 0.0047）。对于pVAS，未校正平均降低分别为51.3%和36.3%，相应的基线校正估计值分别为51.3%和36.3%。校正后差异为+15.0个百分点（95% CI -4.5至34.4；pNI = 0.0023）。因此，两项结局均符合在15个百分点界值下外用治疗对oclacitinib的非劣效性；使用20个百分点界值时结论不变（两项pNI均< 0.001）。由于这些分析未在研究启动前预先设定，应将其解释为支持性探索性证据，而非非劣效性的正式证明。45天随访期间未记录到治疗中出现的不良事件。讨论 最具临床相关性的发现是，尽管该新型外用制剂是作为局部单药治疗而非全身性JAK抑制剂使用，但其在病变严重程度和瘙痒两方面均产生了持续改善。Oclacitinib产生了最强的早期病变应答，尤其是在第15天，但至第45天时，外用Nanobalm的改善幅度与oclacitinib观察到的改善相当。事后非劣效性分析为此提供了定量框架 第5/11页
观察：对于 CADESI-4 和 pVAS，基线校正后百分比降低的差异其置信下限均保持在既往犬 cAD 非劣效性试验（Little 等，2015）所使用的 15 个百分点界值以内，并且使用近期一项 oclacitinib 对照试验（Forster 等，2025）所使用的 20 个百分点界值也得出相同结论。这一点值得注意，因为对照药是一种已确立的全身性免疫调节治疗，而 Nanobalm 仅局部给药。然而，由于非劣效性假设和界值未在研究启动时预先设定，且样本量小，这些发现应描述为与非劣效性相容，而非治疗等效性的确证性证明。联合用药组在随访结束时表现出最强且最一致的整体临床反应。接受 Nanobalm 联合 oclacitinib 的全部 10 只犬均达到 CADESI-4 降低 ≥ 50%，9/10 达到 pVAS 降低 ≥ 50%。与这些应答者数据一致，第 45 天时联合治疗在 CADESI-4 和 pVAS 上的基线校正百分比降低均大于单用 oclacitinib，且在 CADESI-4 上大于局部单药治疗。这些结果支持局部外用治疗联合全身性 JAK 抑制具有辅助临床获益。然而，这些数据并未确立药理学协同作用，确立协同作用需要专门设计以检验交互效应且具有足够把握度的研究。尽管如此，该模式与当前 cAD 的多模式治疗策略一致，即作用于疾病过程不同层面的干预措施可能提供互补获益（Mueller 等，2026）。若干局限性对结果解读有实质性影响。样本量小（每组 10 只犬），且未进行正式的事前样本量计算。敏感性分析表明，本研究主要能够检测大的标准化效应，因此不显著的对比不能解释为等效性的证明。非劣效性分析为事后分析：尽管 15 和 20 个百分点界值均取自已发表的犬 cAD 对照试验而非根据观察到的治疗差异选择，但界值和非劣效性假设均未在入组前设定，且本研究未针对该目标进行前瞻性把握度计算。因此，所得的置信区间提供支持性证据，但需要在更大规模的前瞻性设计的非劣效性试验中确认。应答者和百分比变化分析也属于探索性分析。主人报告的 pVAS 可能因缺乏盲法而受到影响，因为主人知晓治疗分组。oclacitinib 方案在第 14 天后从每日两次改为每日一次给药；该转换后第 2 组评分的描述性升高在临床上值得关注，但在没有预先设定的定义和验证性分析的情况下，不应将其标记为反弹效应。最后，45 天随访不足以评估长期安全性、持续疾病控制或全身性药物节约效应。总之，这种新型局部纳米包裹植物制剂对 cAD 产生了持续且有临床意义的改善，并且在第 45 天时，在采用基于文献的 15% 和 20% 界值的探索性分析中，其在 CADESI-4 和 pVAS 上的降低与全身性 oclacitinib 非劣效性相容。这一发现具有临床相关性，因为可比的晚期治疗控制是通过纯局部干预而非全身性 JAK 抑制实现的。联合治疗产生了第 6/11 页
总体反应最佳，100%的犬CADESI-4改善≥50%，90%的犬pVAS改善≥50%，且两项结局的基线校正百分比降幅均大于单用奥拉替尼。这些结果支持Nanobalm作为cAD有前景的单独及辅助治疗选择，但仍需开展更大规模、设盲、前瞻性定义非劣效和优效假设的试验予以确认。声明 资助：本研究由巴西国家研究委员会（CNPq，项目编号303067/2023-3）和Wesen Green Solutions Ltda.资助。本文发表的文章处理费由高等教育人员改进协调组织——CAPES资助（ROR标识符：00x0ma614）。为开放获取之目的，作者已将本文任何被接受版本的Creative Commons CC BY许可授予。利益冲突：Fernando Bittencourt Luciano是Wesen Green Solutions Ltda.的合伙人。作者贡献：S.D.R.研究实施、统计分析、撰写；V.C.G.审阅/编辑；M.F.O.构思；F.G.M.研究实施、撰写；M.R.F.构思/设计、审阅/编辑、监督；F.B.L.构思/设计、数据管理、统计分析、审阅/编辑、监督。数据可用性：本研究生成和/或分析的数据集可在合理请求下向通讯作者获取。伦理批准：本研究经巴拉那天主教大学动物使用伦理委员会（CEUA）批准，方案编号2469200524。发表同意：已获得所有者书面知情同意，同意发表图3中呈现的临床照片。参考文献 1. Cosgrove SB, Wren JA, Cleaver DM, Walsh KF, Follis SI, King VL, Tena JKS, Stegemann MR (2013) A blinded, randomized, placebo-controlled trial of the efficacy and safety of the Janus kinase inhibitor oclacitinib (Apoquel) in client-owned dogs with atopic dermatitis. Vet Dermatol 24:587–597e141. https://doi.org/10.1111/vde.12088 2. Forster S, Boegel A, Despa S, Trout C, King S (2025) Comparative efficacy and safety of ilunocitinib and oclacitinib for the control of pruritus and associated skin lesions in dogs with atopic dermatitis. Vet Dermatol 36:165–176. https://doi.org/10.1111/vde.13319 3. Hensel P, Santoro D, Favrot C, Hill P, Griffin C (2015) Canine atopic dermatitis: detailed guidelines for diagnosis and allergen identification. BMC Vet Res 11:196. https://doi.org/10.1186/s12917-015- 0515-5 第7/11页
4. Little PR, King VL, Davis KR, Cosgrove SB, Stegemann MR (2015) A blinded, randomized clinical trial comparing the efficacy and safety of oclacitinib and ciclosporin for the control of atopic dermatitis in client-owned dogs. Vet Dermatol 26:23–30. https://doi.org/10.1111/vde.12186 5. Mueller RS, Banovic F, Bensignor E, Santoro D, Nuttall T (2026) Treatment of canine atopic dermatitis: updated guidelines from the International Committee on Allergic Diseases of Animals (ICADA). Vet Dermatol. https://doi.org/10.1111/vde.70128 6. Nuttall TJ, Marsella R, Rosenbaum MR, Gonzales AJ, Fadok VA (2019) Update on pathogenesis, diagnosis, and treatment of atopic dermatitis in dogs. J Am Vet Med Assoc 254:1291–1300. https://doi.org/10.2460/javma.254.11.1291 7. Olivry T, Bensignor E, Favrot C, Griffin CE, Hill PB, Mueller RS, Plant JD, Williams HC (2018) Development of a core outcome set for therapeutic clinical trials enrolling dogs with atopic dermatitis (COSCAD’18). BMC Vet Res 14:238. https://doi.org/10.1186/s12917-018-1569-y 8. Olivry T, Saridomichelakis M, Nuttall T, Bensignor E, Griffin CE, Hill PB (2014) Validation of the Canine Atopic Dermatitis Extent and Severity Index (CADESI)-4, a simplified severity scale for assessing skin lesions of atopic dermatitis in dogs. Vet Dermatol 25:77–85e25. https://doi.org/10.1111/vde.12107 9. Rybníček J, Lau-Gillard PJ, Harvey R, Hill PB (2009) Further validation of a pruritus severity scale for use in dogs. Vet Dermatol 20:115–122. https://doi.org/10.1111/j.1365-3164.2008.00728.x 10. Tresch M, Mevissen M, Ayrle H, Melzig M, Roosje P, Walkenhorst M (2019) Medicinal plants as therapeutic options for topical treatment in canine dermatology? A systematic review. BMC Vet Res 15:174. https://doi.org/10.1186/s12917-019-1854-4 图 第8/11页
图1 在45天研究期内，接受单用局部Nanobalm（G1）、单用口服奥拉替尼（G2）或两种治疗联合（G3）的犬瘙痒视觉模拟量表（pVAS）评分的时间变化。点代表中位数，误差线代表四分位距。线性混合效应模型中总体组别×时间交互作用显著（似然比χ² = 23.57，df = 8，p = 0.0027）。第45天时，经Holm校正后无任何两两对比仍具有统计学显著性，尽管G3显示最低的中位pVAS评分。第9/11页
图2 在45天研究期间，单独使用局部Nanobalm（G1）、单独口服oclacitinib（G2）或两者联合治疗（G3）的犬中，犬特应性皮炎范围和严重程度指数第4版（CADESI-4）评分的时间进展。点代表中位数，误差线代表四分位距。线性混合效应模型中的总体组别×时间交互作用显著（似然比χ² = 48.79，df = 8，p < 0.001）。在第15天，经Holm校正后，G2的CADESI-4评分低于G1和G3（校正p分别为0.010和0.006）。第45天经多重性校正后，未有任何两两差异保持显著。第10/11页
图3 第1组（G1，上排）、第2组（G2，中排）和第3组（G3，下排）的代表性临床照片。每排内，a = 第0天，b = 第15天，c = 第45天。第1组接受局部Nanobalm，第2组接受口服oclacitinib，第3组接受联合方案。已获得所有者书面知情同意以发表临床照片。第11/11页
查看英文原文▾
Topical nanoencapsulated botanical formulation alone or with oclacitinib in canine atopic dermatitis: an exploratory clinical study Sara Daniela Risse Pontifícia Universidade Católica do Paraná Vanessa Cunningham Gmyterco Pontifícia Universidade Católica do Paraná Michel Fleith Otuki Universidade Federal do Paraná Fernanda Gabriela Menegon Pontifícia Universidade Católica do Paraná Maitiani Festinalli Pontifícia Universidade Católica do Paraná Marconi Rodrigues de Farias Pontifícia Universidade Católica do Paraná Fernando Bittencourt Luciano Pontifícia Universidade Católica do Paraná Short Report Keywords: rosemary, marigold, chamomille, pruritus, CADESI-4, multimodal therapy Posted Date: October 23rd, 2026 DOI: https://doi.org/10.21203/rs.3.rs-11310195/v1 License: This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Additional Declarations: No competing interests reported. Page 1/11
Abstract Canine atopic dermatitis (cAD) commonly requires multimodal management, and a topical treatment capable of achieving meaningful clinical control without systemic immunomodulation would be clinically valuable. This exploratory three-arm study compared a novel topical nanoencapsulated botanical formulation (Nanobalm), oral oclacitinib, and their combination in 30 client-owned dogs with mild-tomoderate cAD (10 dogs per group). CADESI-4 and pruritus visual analog scale (pVAS) scores were recorded on days 0, 7, 15, 30, and 45. Longitudinal mixed-effects models showed significant group-bytime interactions for CADESI-4 (likelihood-ratio χ²=48.79, df = 8, p < 0.001) and pVAS (χ²=23.57, df = 8, p = 0.0027). In a post hoc exploratory non-inferiority analysis at day 45, the percentage reduction from baseline with topical monotherapy was compatible with non-inferiority to oclacitinib using a literaturebased 15-percentage-point margin for both CADESI-4 (adjusted G1-G2 difference + 3.3 percentage points; 95% CI -9.9 to 16.5; pNI = 0.0047) and pVAS (+ 15.0; 95% CI -4.5 to 34.4; pNI = 0.0023); conclusions were unchanged using a 20-point margin. Combination therapy produced the most consistent response, with ≥ 50% reductions in CADESI-4 in 10/10 dogs and pVAS in 9/10 dogs. These findings support the topical formulation as a promising stand-alone option and suggest additional clinical benefit when combined with oclacitinib, while requiring confirmation in prospectively powered trials. INTRODUCTION Canine atopic dermatitis (cAD) is a chronic pruritic and inflammatory skin disease diagnosed through compatible clinical features and exclusion of other pruritic disorders (Hensel et al. 2015). Management is typically multimodal and combines control of flare factors with antipruritic or anti-inflammatory therapy and, when appropriate, topical skin-barrier support (Nuttall et al. 2019; Mueller et al. 2026). Oclacitinib provides rapid control of pruritus and skin lesions in many affected dogs, but long-term management frequently requires individualized combinations of interventions (Cosgrove et al. 2013; Mueller et al. 2026). Topical botanical products are used in veterinary dermatology, although clinical evidence is heterogeneous and formulation-specific (Tresch et al. 2019). A nanoencapsulated topical formulation containing chamomile, calendula, and rosemary extracts may provide local anti-inflammatory and skinsupportive activity without systemic JAK inhibition. This distinction is clinically relevant: a topical monotherapy that achieves sustained improvement of lesion severity and pruritus approaching that of an established systemic immunomodulator could broaden long-term management options, particularly as part of multimodal cAD care. The objective of this exploratory study was to compare changes in pruritus and lesion severity and to assess short-term tolerability in dogs with mild-to-moderate cAD receiving the topical formulation alone, oclacitinib alone, or both treatments. Because each group included 10 dogs, the study was interpreted as hypothesis-generating rather than as a definitive test of superiority or synergy. Page 2/11
MATERIALS AND METHODS Thirty client-owned dogs with previously diagnosed mild-to-moderate cAD were enrolled. All dogs had undergone an elimination diet for at least eight weeks, and infectious and parasitic diseases had been ruled out before inclusion. Dogs with secondary bacterial or fungal infection or those that had received oral or topical corticosteroids, systemic antibiotics, or antiseptic shampoos within the preceding month were excluded. The study was approved by the Ethics Committee for Animal Use (CEUA) of Pontifícia Universidade Católica do Paraná, Curitiba, Brazil (protocol 2469200524). Dogs were randomly allocated by drawing to three groups (n = 10 each). Group 1 received the topical nanoencapsulated chamomile, marigold and rosemary extracts (Nanobalm; Wesen Green Solutions Ltda) applied every 12 h to the interdigital, axillary, inguinal regions, and ear pinnae for 45 days. Group 2 received oral oclacitinib (Apoquel; Zoetis) at 0.4–0.6 mg/kg every 12 h for 14 days and once daily thereafter through day 45. Group 3 received both protocols. The topical product was gently massaged into the designated areas and was not rinsed. Cosmetic non-medicated shampoos were allowed, whereas moisturizing or other topical barrier products were prohibited. No formal a priori sample-size calculation was performed. At study planning, reliable estimates of the expected effect size of the investigated topical formulation, either alone or in combination with oclacitinib, were unavailable, and recruitment was constrained to client-owned dogs meeting the predefined eligibility criteria. The study was therefore designed as exploratory with 10 dogs per group. A sensitivity analysis of the final sample indicated that, at a two-sided α = 0.05, 10 dogs per group provides 80% power only for a large pairwise standardized effect (Cohen's d ≈ 1.32); for an overall three-group comparison (N = 30), 80% power corresponds to Cohen's f ≈ 0.60. Accordingly, the study had limited ability to detect or exclude small-to-moderate treatment differences. The non-inferiority analyses described below were added post hoc to quantify how the final topical-monotherapy response compared with oclacitinib and are interpreted as exploratory rather than confirmatory. Clinical assessments were performed on days 0, 7, 15, 30, and 45. Owners scored pruritus using the validated pruritus visual analog scale (pVAS) (Rybníček et al. 2009). Skin lesions were scored by a veterinarian using CADESI-4 (Olivry et al. 2014). Safety monitoring included erythema or worsening skin signs, discomfort or reluctance during application, excessive licking, sialorrhea, mucosal lesions, vomiting, and any other adverse event reported during follow-up. The primary longitudinal analysis used linear mixed-effects models with a random intercept for dog and fix ed effects for treatment group, day (categorical), and the group-by-day interaction. CADESI-4 was analyzed after log(score + 1) transformation because residuals from the untransformed model were nonnormal; pVAS was analyzed on the original scale. The overall interaction was evaluated by likelihoodratio testing of the full interaction model against the corresponding additive model. Following a significant interaction, two-sided pairwise group contrasts were estimated at each time point and adjusted using the Holm procedure within each time point. Page 3/11
Data are reported as median [interquartile range] unless otherwise specified. Because of the small group size and score distributions, nonparametric analyses (Kruskal-Wallis with two-sided Mann-Whitney U post hoc comparisons) were used as sensitivity analyses. Exploratory responder analyses defined clinically meaningful improvement as a ≥ 50% reduction from the individual baseline score at day 45; responder proportions were compared using two-sided Fisher exact tests with Holm adjustment across pairwise comparisons. In addition, a post hoc exploratory non-inferiority analysis compared topical Nanobalm monotherapy (G1) with oclacitinib monotherapy (G2) at day 45 using percentage reduction from baseline in CADESI-4 and pVAS. Percentage reduction was analyzed by ANCOVA with centered baseline score as a covariate. A 15-percentage-point non-inferiority margin was selected from a previous canine atopic dermatitis trial comparing oclacitinib with ciclosporin (Little et al. 2015), rather than from the observed treatment difference; a 20-point margin, as used in a more recent JAK-inhibitor comparison with oclacitinib (Forster et al. 2025), was evaluated as a sensitivity analysis. Because two clinical outcomes were assessed, a conservative one-sided α = 0.025 was used. Non-inferiority was considered supported when the lower bound of the 95% two-sided confidence interval for the adjusted G1-G2 difference in percentage reduction was greater than − 15 percentage points. Three-group ANCOVA of percentage reduction at day 45, with Holm-adjusted post hoc contrasts, was also used exploratorily to characterize the combination-treatment response. In Figs. 1 and 2, central points represent the group median at each visit and error bars represent the interquartile range (IQR). The analysis was performed in Python 3.13 using SciPy 1.17.0 and statsmodels 0.14.6. Effect estimates, exact p values, and multiplicity-adjusted p values are reported where applicable. RESULTS The 30 enrolled dogs comprised 16 females (53.3%) and 14 males (46.7%), aged 1–13 years (mean 5.7 years). Mixed-breed dogs were most frequent (33.3%), followed by Shih Tzu (16.7%), Yorkshire Terrier (13.3%), and French Bulldog (10.0%). Groups 2 and 3 were balanced by sex, whereas 70% of Group 1 were male. The longitudinal pVAS trajectory differed significantly among groups (group-by-time interaction: likelihood-ratio χ²=23.57, df = 8, p = 0.0027). Baseline pVAS medians were 6.0, 6.0, and 6.5 in Groups 1, 2, and 3, respectively. At day 45, median [IQR] pVAS was 3.0 [2.25-3.0] in Group 1, 3.5 [3.0-5.75] in Group 2, and 2.0 [2.0–2.0] in Group 3 (Fig. 1). The largest day-45 mixed-model contrast was between Groups 2 and 3 (estimated mean difference 1.9 pVAS points; nominal p = 0.020), but this contrast did not remain below the 0.05 threshold after Holm adjustment (adjusted p = 0.060). A ≥ 50% reduction in pVAS from baseline to day 45 occurred in 6/10 (60%) dogs in Group 1, 2/10 (20%) in Group 2, and 9/10 (90%) in Group 3. Group 3 had a higher responder proportion than Group 2 (two-sided Fisher exact p = 0.0055; Holm-adjusted p = 0.016). When percentage reduction at day 45 was analyzed by baseline-adjusted ANCOVA across all three groups, the adjusted reductions were 51.7%, 36.5%, and 62.5% for Groups 1, 2, Page 4/11
and 3, respectively; improvement was greater in Group 3 than Group 2 (difference 26.0 percentage points; Holm-adjusted p = 0.0118), whereas Group 3 versus Group 1 was not significant (adjusted p = 0.199). The longitudinal CADESI-4 trajectory also differed significantly among groups (group-by-time interaction for log-transformed CADESI-4: likelihood-ratio χ²=48.79, df = 8, p < 0.001). Baseline medians were 29.0, 28.5, and 32.5 in Groups 1, 2, and 3, respectively. At day 15, median CADESI-4 scores were 18.0, 10.0, and 20.0, and post hoc mixed-model contrasts showed lower lesion scores in Group 2 than in Groups 1 and 3 (Holm-adjusted p = 0.010 and p = 0.006, respectively). At day 45, median [IQR] scores were 16.0 [11.0–18.0], 15.5 [11.75–17.75], and 11.5 [9.25-14.0], respectively; no pairwise day-45 raw-score contrast remained significant after multiplicity correction (Fig. 2). A ≥ 50% reduction in CADESI-4 from baseline to day 45 was achieved by 4/10 (40%) dogs in Group 1, 3/10 (30%) in Group 2, and 10/10 (100%) in Group 3. Group 3 had a higher responder proportion than Group 1 (two-sided Fisher exact p = 0.0108; Holmadjusted p = 0.022) and Group 2 (p = 0.0031; Holm-adjusted p = 0.009). In the three-group baselineadjusted analysis of percentage reduction, adjusted day-45 improvements were 50.8%, 45.8%, and 66.3% for Groups 1, 2, and 3, respectively. Group 3 showed a greater percentage improvement than Group 2 (difference 20.5 percentage points; Holm-adjusted p = 0.0096) and Group 1 (difference 15.5 percentage points; adjusted p = 0.0345). Representative clinical progression is shown in Fig. 3. Exploratory non-inferiority analysis of topical Nanobalm versus oclacitinib at day 45 supported the same clinical interpretation for both efficacy outcomes. For CADESI-4, the unadjusted mean percentage reduction from baseline was 50.2% in Group 1 and 43.6% in Group 2; after adjustment for baseline severity, the estimated reductions were 48.6% and 45.2%, respectively. The adjusted G1-G2 difference was + 3.3 percentage points (95% CI -9.9 to 16.5), with the lower confidence limit remaining above the − 15-point margin (pNI = 0.0047). For pVAS, the unadjusted mean reductions were 51.3% and 36.3%, and the corresponding baseline-adjusted estimates were 51.3% and 36.3%. The adjusted difference was + 15.0 percentage points (95% CI -4.5 to 34.4; pNI = 0.0023). Thus, both outcomes were compatible with non-inferiority of the topical treatment to oclacitinib under the 15-point margin; the conclusion was unchanged using a 20-point margin (both pNI < 0.001). Because these analyses were not specified before study initiation, they should be interpreted as supportive exploratory evidence rather than formal proof of non-inferiority. No treatment-emergent adverse events were recorded during the 45-day follow-up. DISCUSSION The most clinically relevant finding is that the novel topical formulation produced sustained improvement in both lesion severity and pruritus despite being used as a local monotherapy rather than a systemic JAK inhibitor. Oclacitinib produced the strongest early lesion response, particularly at day 15, but by day 45 the magnitude of improvement with topical Nanobalm was comparable to that observed with oclacitinib. The post hoc non-inferiority analysis provides a quantitative framework for this Page 5/11
observation: for both CADESI-4 and pVAS, the lower confidence bound for the baseline-adjusted difference in percentage reduction remained within the 15-percentage-point margin previously used in a canine cAD non-inferiority trial (Little et al. 2015), and the same conclusion was obtained using the 20- point margin used in a recent oclacitinib comparator trial (Forster et al. 2025). This is notable because the comparator is an established systemic immunomodulatory treatment, whereas Nanobalm is administered exclusively topically. However, because the non-inferiority hypothesis and margin were not prespecified at study initiation and the sample was small, these findings should be described as compatible with non-inferiority rather than as definitive demonstration of therapeutic equivalence. The combination group showed the strongest and most consistent overall clinical response by the end of follow-up. All 10 dogs receiving Nanobalm plus oclacitinib achieved a ≥ 50% reduction in CADESI-4, and 9/10 achieved a ≥ 50% reduction in pVAS. Consistent with these responder data, the baselineadjusted percentage reduction at day 45 was greater with combination therapy than with oclacitinib alone for both CADESI-4 and pVAS, and greater than topical monotherapy for CADESI-4. These results support an adjunctive clinical benefit of combining local topical therapy with systemic JAK inhibition. The data do not, however, establish pharmacological synergy, which would require a study specifically designed to test an interaction effect with adequate power. The pattern is nevertheless consistent with the current multimodal approach to cAD, in which interventions acting at different levels of the disease process may provide complementary benefit (Mueller et al. 2026). Several limitations materially affect interpretation. The sample was small (10 dogs per group), and no formal a priori sample-size calculation was performed. Sensitivity analysis indicates that the study was primarily capable of detecting large standardized effects, so nonsignificant contrasts cannot be interpreted as proof of equivalence. The non-inferiority analysis was post hoc: although both the 15- and 20-percentage-point margins were taken from published canine cAD comparator trials rather than chosen from the observed treatment difference, neither the margin nor the non-inferiority hypothesis was specified before enrollment, and the study was not prospectively powered for this objective. The resulting confidence intervals therefore provide supportive evidence but require confirmation in a larger prospectively designed non-inferiority trial. Responder and percentage-change analyses were also exploratory. Owner-reported pVAS may be influenced by lack of masking as owners were aware of treatment assignment. The oclacitinib regimen changed from twice-daily to once-daily administration after day 14; the descriptive increase in Group 2 scores after this transition is clinically interesting but should not be labeled a rebound effect without a prespecified definition and confirmatory analysis. Finally, the 45-day follow-up is insufficient to assess long-term safety, sustained disease control, or a systemic drug-sparing effect. In conclusion, the novel topical nanoencapsulated botanical formulation produced sustained clinically meaningful improvement in cAD and, at day 45, its reduction in both CADESI-4 and pVAS was compatible with non-inferiority to systemic oclacitinib under exploratory analyses using literature-based 15% and 20% margins. This finding is clinically relevant because comparable late-treatment control was achieved with a purely topical intervention rather than systemic JAK inhibition. Combination therapy produced the Page 6/11
greatest overall response, with ≥ 50% CADESI-4 improvement in 100% of dogs and ≥ 50% pVAS improvement in 90%, together with greater baseline-adjusted percentage reductions than oclacitinib alone for both outcomes. These results support Nanobalm as a promising stand-alone and adjunctive option for cAD, while larger blinded trials with prospectively defined non-inferiority and superiority hypotheses are required for confirmation. Declarations Funding: This study was supported by the National Research Council of Brazil (CNPq), process number 303067/2023-3 and Wesen Green Solutions Ltda. The Article Processing Charge for the publication of this research was funded by the Coordination for the Improvement of Higher Education Personnel – CAPES (ROR identifier: 00x0ma614). For the purposes of open access, the authors have assigned the Creative Commons CC BY license to any accepted version of the article. Competing interests: Fernando Bittencourt Luciano is a partner of Wesen Green Solutions Ltda. Author contributions: S.D.R. investigation, statistical analysis, writing; V.C.G. review/editing; M.F.O. conception; F.G.M. investigation, writing; M.R.F. conception/design, review/editing, supervision; F.B.L. conception/design, data curation, statistical analysis, review/editing, supervision. Data availability: The datasets generated and/or analysed during the current study are available from the corresponding author on reasonable request Ethics approval: The study was approved by the Ethics Committee for Animal Use (CEUA) of Pontifícia Universidade Católica do Paraná under protocol 2469200524. Consent for publication: Written informed owner consent was obtained for publication of the clinical photographs presented in Fig. 3. References 1. Cosgrove SB, Wren JA, Cleaver DM, Walsh KF, Follis SI, King VL, Tena JKS, Stegemann MR (2013) A blinded, randomized, placebo-controlled trial of the efficacy and safety of the Janus kinase inhibitor oclacitinib (Apoquel) in client-owned dogs with atopic dermatitis. Vet Dermatol 24:587–597e141. https://doi.org/10.1111/vde.12088 2. Forster S, Boegel A, Despa S, Trout C, King S (2025) Comparative efficacy and safety of ilunocitinib and oclacitinib for the control of pruritus and associated skin lesions in dogs with atopic dermatitis. Vet Dermatol 36:165–176. https://doi.org/10.1111/vde.13319 3. Hensel P, Santoro D, Favrot C, Hill P, Griffin C (2015) Canine atopic dermatitis: detailed guidelines for diagnosis and allergen identification. BMC Vet Res 11:196. https://doi.org/10.1186/s12917-015- 0515-5 Page 7/11
4. Little PR, King VL, Davis KR, Cosgrove SB, Stegemann MR (2015) A blinded, randomized clinical trial comparing the efficacy and safety of oclacitinib and ciclosporin for the control of atopic dermatitis in client-owned dogs. Vet Dermatol 26:23–30. https://doi.org/10.1111/vde.12186 5. Mueller RS, Banovic F, Bensignor E, Santoro D, Nuttall T (2026) Treatment of canine atopic dermatitis: updated guidelines from the International Committee on Allergic Diseases of Animals (ICADA). Vet Dermatol. https://doi.org/10.1111/vde.70128 6. Nuttall TJ, Marsella R, Rosenbaum MR, Gonzales AJ, Fadok VA (2019) Update on pathogenesis, diagnosis, and treatment of atopic dermatitis in dogs. J Am Vet Med Assoc 254:1291–1300. https://doi.org/10.2460/javma.254.11.1291 7. Olivry T, Bensignor E, Favrot C, Griffin CE, Hill PB, Mueller RS, Plant JD, Williams HC (2018) Development of a core outcome set for therapeutic clinical trials enrolling dogs with atopic dermatitis (COSCAD’18). BMC Vet Res 14:238. https://doi.org/10.1186/s12917-018-1569-y 8. Olivry T, Saridomichelakis M, Nuttall T, Bensignor E, Griffin CE, Hill PB (2014) Validation of the Canine Atopic Dermatitis Extent and Severity Index (CADESI)-4, a simplified severity scale for assessing skin lesions of atopic dermatitis in dogs. Vet Dermatol 25:77–85e25. https://doi.org/10.1111/vde.12107 9. Rybníček J, Lau-Gillard PJ, Harvey R, Hill PB (2009) Further validation of a pruritus severity scale for use in dogs. Vet Dermatol 20:115–122. https://doi.org/10.1111/j.1365-3164.2008.00728.x 10. Tresch M, Mevissen M, Ayrle H, Melzig M, Roosje P, Walkenhorst M (2019) Medicinal plants as therapeutic options for topical treatment in canine dermatology? A systematic review. BMC Vet Res 15:174. https://doi.org/10.1186/s12917-019-1854-4 Figures Page 8/11
Figure 1 Temporal progression of pruritus visual analog scale (pVAS) scores over the 45-day study period in dogs treated with topical Nanobalm alone (G1), oral oclacitinib alone (G2), or the combination of both treatments (G3). Points represent medians and error bars represent the interquartile range. The overall group-by-time interaction in the linear mixed-effects model was significant (likelihood-ratio χ² = 23.57, df = 8, p = 0.0027). At day 45, no pairwise contrast remained statistically significant after Holm correction, although G3 showed the lowest median pVAS score. Page 9/11
Figure 2 Temporal progression of Canine Atopic Dermatitis Extent and Severity Index version 4 (CADESI-4) scores over the 45-day study period in dogs treated with topical Nanobalm alone (G1), oral oclacitinib alone (G2), or the combination of both treatments (G3). Points represent medians and error bars represent the interquartile range. The overall group-by-time interaction in the linear mixed-effects model was significant (likelihood-ratio χ² = 48.79, df = 8, p < 0.001). At day 15, G2 had lower CADESI-4 scores than G1 and G3 after Holm adjustment (adjusted p = 0.010 and 0.006, respectively). No pairwise difference remained significant at day 45 after multiplicity correction. Page 10/11
Figure 3 Representative clinical photographs from Group 1 (G1, top row), Group 2 (G2, middle row), and Group 3 (G3, bottom row). Within each row, a = day 0, b = day 15, and c = day 45. Group 1 received topical Nanobalm, Group 2 received oral oclacitinib, and Group 3 received the combination protocol. Written informed owner consent was obtained for publication of the clinical photographs. Page 11/11