Search for a command to run...
The Minimal Clinically Important Difference (MCID) is the smallest change in a treatment outcome that an individual patient would identify as important and which would indicate a change in the patient's management [1]. A plausible MCID is necessary to evaluate if the treatment effect of AIT compared to placebo is clinically relevant and thus meaningful (and not only noticeable). This is a prerequisite for a marketing authorization (MA). In a Position Paper of the European Academy of Allergy and Clinical Immunology (EAACI), an international group of experts in AIT trial methodology developed a consensus on standardized and globally harmonized methods for analyzing the clinical efficacy of AIT products in RCTs [2]. The authors recommend the use of a homogeneous combined symptom and medication score (CSMS) as a simple, standardized endpoint that equally balances symptoms and the need for antiallergic medication in clinical trials. However, this endpoint has not yet undergone clinical validation, including the definition of its MCID. Although several initiatives have addressed this gap [3-5], the definition of a robust and evidence-based MCID for CSMS is long overdue. In a recent article published in Allergy [6], the authors aimed to further advance this discussion and make an initial contribution to an evidence-based approach toward an MCID. Based on this analysis, they suggest a −0.22 point difference in the CSMS of AIT compared to placebo as MCID in a pivotal clinical study. This objective to determine a suitable difference in the CSMS, that reflects a clinically relevant effect on patients is strongly supported from a regulatory point of view. However, the data obtained from this analysis are not suitable for the MA process due to the following reasons: (i) the design of the patient survey (available at https://www.polleninformation.at/news/umfrage-graeserpollen-allergikerin-gesucht, last accessed 31.08.2025), and (ii) the respective RQLQ value used as an anchor: (i) At the beginning of the survey, a list of 6 symptoms is presented to the participant out of which she/he is asked to select the symptom which is the most important for her/him. Thereafter, the patient survey [6] focuses on this most important symptom only, which is an interesting approach; however, the results retrieved in this survey are not transferable as MCID for the CSMS in AIT-trials since the defined primary endpoint is not identical: the effect in AIT-trials is measured as average difference (of all symptoms) observed in the study population [5], which is not equivalent to the minimum expected improvement in one (the most important) symptom individual for each patient as used in the survey. Yet, the data obtained from the survey might serve to design future clinical trials with the aim of investigating the effect on the patient's most important symptom. In such a clinical trial, at inclusion, each patient would have to define her/his most important symptom. Then, the clinical trial would have to prove that AIT is able to reduce the symptom severity of one class on average over the defined most important symptoms evaluated by the difference between verum and placebo, not in average over all symptoms. Such a study design does not comply with current EMA guidance (CHMP/EWP/18504/2006). Moreover, to investigate individual improvements, a baseline season would be necessary. According to the competent authority's experience, it would be more challenging, and this new primary endpoint might be harder to achieve than the current one [5]. Additionally, in its current form, the survey appears prone to bias: Concerning the minimal improvement in the most important symptom the patient would be satisfied with, questions Q015-Q032 (table S1 [6]) are binary “yes/no” selections starting with the smallest effect. Answering a question with “yes” leads to the immediate end of the survey without the possibility to go back to revise the answer or review further options. The acquiescence bias (respondent tendency to answer “yes” to a question) is a well-known factor that often occurs with surveys that include only binary response options, like “yes/no”; this may give rise to systematic distortions [7]. Only after having answered 4 times “no” (not satisfied), a second symptom is offered for which relief might be desirable for the patients. A third symptom is not offered. Since the participants were not informed about the survey structure and the fact that the answer “yes” will be understood in a way that the participant would not care about other symptoms, the design bears the risk of premature ending where patients' expectations were only partially captured. Consequently, the survey structure bears a risk of underestimating the MCID. To minimize this bias, alternative survey structures are worth further exploring in the future, for example, it might be feasible to use questionnaires that display all possible response options at once, from which the patients can choose the minimal value they consider beneficial. (ii) The RQLQ score might be interesting to explore further for an anchor-based approach [1] to determine the MCID for CSMS0–6 since it is a validated score for symptomatic treatment of rhinoconjunctivitis. According to the WAO, a higher effect is expected for AIT-products than that achieved by antihistamines (symptomatic treatment) [3]. However, it is noteworthy to point out that the MCID of 0.5 for RQLQ [8] and of 0.7 for Mini-RQLQ [9] was described by Juniper in the context of a within-participant difference, whereas the MCID for RQLQ or Mini-RQLQ between treatment arms is yet to be determined and it may not be appropriate to use 0.5 or 0.7, respectively, as a gold standard for the difference between treatment arms without validation. Several research approaches have been applied to translate individual patient perspectives into between-group post hoc comparisons in AIT trials [4, 10], to gain information on clinically relevant differences in the RQLQ. These include evaluations of data from authorized grass pollen AIT products [4, 10]. However, these RQLQ-values retrieved from pivotal clinical AIT trials represent only secondary endpoints [11] and thus were not decisive for the regulatory MA approval. Consequently, from the MA, it cannot be deduced that the reported RQLQ values were clinically relevant or meaningful, and their suitability as an anchor at the given time is therefore not established. In conclusion, the recent report [6] is recognized as an important step toward further optimizing methodology in AIT-trial design and analysis (based on clinical endpoints and MCID) and is expected to foster fruitful discussions on the future development of a valid, evidence-based MCID. Despite the reported MCID of −0.22 in the CSMS, which cannot be regarded as suitable for regulatory assessment, as outlined above, a patient-centric approach and further activities toward an evidence-based MCID for the primary endpoints of clinical AIT-trials are strongly supported. This intention will be further advanced through a Task Force initiative of the EAACI (EAACI Task force “Minimal important difference (MID) in AIT” (chair: Oliver Pfaar)), supported by scientific contributions from regulatory authorities. All authors made substantial contributions to the conception and design of the article. D.H. drafted the article; while S.K., S.S., and V.M. revised it critically for important intellectual content. All authors gave their final approval of the version to be published. The authors have nothing to report. The authors declare no conflicts of interest. This article is linked to Pfaar O. et al. paper. To view these articles, visit https://doi.org/10.1111/all.16654. The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.