Strober Bruce E, Bukhalo Michael, Armstrong April W, Pariser David, Kircik Leon, Johnson Brian, Montgomery Paul, Dickerson Tobin J
Yale University School of Medicine, New Haven, CT, USA.
Central Connecticut Dermatology, Cromwell, CT, USA.
Dermatol Ther (Heidelb). 2025 Jul;15(7):1787-1796. doi: 10.1007/s13555-025-01441-y. Epub 2025 May 11.
This randomized, prospective study (MATCH) was designed to assess the clinical utility of a machine learning-based tool (Mind.Px) that predicts patient response to the biologic drug classes used in the management of psoriasis.
Patients with psoriasis who were biologic naïve or approaching a medication change owing to nonresponse were enrolled into the study (N = 210). At baseline, a dermal biomarker patch was applied to lesional skin, and Mind.Px test results were provided to physicians for patients in the informed arm of the study prior to biologic selection. The choice of biologic was recorded, and, in the case of physician nonconcordance with Mind.Px test results, a questionnaire was completed to determine the reason for nonconcordance. Patients were evaluated at weeks 4 and 12 after baseline using Psoriasis Area and Severity Index (PASI). Statistical analysis between groups was performed using Fisher's exact test.
Physician prescribing behavior was measured with and without the inclusion of Mind.Px test results in the decision-making process (N = 205). Additional comparisons were made to a previously collected data set identical to the Mind.Px-uninformed arm (N = 429). Statistical analysis of concordance between the Mind.Px-informed and Mind.Px-uninformed groups (92.3% versus 62.9%, respectively) showed that when given access to Mind.Px results, physician behavior was significantly altered (p = 8.08 × 10). Furthermore, analysis of patients whose physicians followed Mind.Px results showed that not only did more patients reach the clinical endpoint (PASI75) at 12 weeks (p = 5.4 × 10), but also more patients reached this endpoint by week 4 than those in the treatment-as-usual arm (p = 0.01).
These results provide evidence of the clinical utility of Mind.Px by showing that physicians utilize test results in psoriasis biologic decision-making, leading to improved patient outcomes. These improved patient outcomes can also potentially translate into cost savings for healthcare systems. Mind.Px can minimize the trial-and-error approach to psoriasis treatment, and provide physicians, patients, and payers with an effective tool for re-envisioning the management of patients with psoriasis.
NCT05036889.
这项随机前瞻性研究(MATCH)旨在评估一种基于机器学习的工具(Mind.Px)在预测银屑病患者对用于治疗银屑病的生物药物类别的反应方面的临床效用。
将初治生物制剂或因无反应而即将改变用药方案的银屑病患者纳入研究(N = 210)。在基线时,将皮肤生物标志物贴片应用于皮损处,并在生物制剂选择前,将Mind.Px测试结果提供给研究中知情组的医生。记录生物制剂的选择情况,若医生的选择与Mind.Px测试结果不一致,则填写一份问卷以确定不一致的原因。在基线后第4周和第12周,使用银屑病面积和严重程度指数(PASI)对患者进行评估。组间统计分析采用Fisher精确检验。
在决策过程中纳入和不纳入Mind.Px测试结果来衡量医生的处方行为(N = 205)。还与之前收集的与Mind.Px不知情组相同的数据集进行了额外比较(N = 429)。对Mind.Px知情组和Mind.Px不知情组之间的一致性进行统计分析(分别为92.3%和62.9%),结果表明,当医生能够获取Mind.Px结果时,其行为会发生显著改变(p = 8.08 × 10)。此外,对医生遵循Mind.Px结果的患者进行分析发现,不仅更多患者在12周时达到临床终点(PASI75)(p = 5.4 × 10),而且到第4周时达到该终点的患者也比常规治疗组更多(p = 0.01)。
这些结果通过表明医生在银屑病生物制剂决策中利用测试结果,从而改善患者预后,为Mind.Px的临床效用提供了证据。这些改善的患者预后也可能为医疗保健系统节省成本。Mind.Px可以最大限度地减少银屑病治疗的试错方法,并为医生、患者和支付方提供一种重新规划银屑病患者管理的有效工具。
NCT05036889