Stallergenes Greer (France)

Proton Exchange Membrane Fuel Cells (PEMFCs) are increasingly deployed in clean energy systems, such as GEH2 hydrogen generators, where they operate under highly dynamic and unpredictable load conditions. Accurate prediction of their Remaining Useful Life (RUL) is essential for ensuring reliable, cost-effective, and proactive maintenance strategies. However, conventional voltage-based Health Indicators (HIs) are highly sensitive to power fluctuations and fail to provide consistent degradation trends in real-world industrial scenarios, particularly when system usage varies significantly across different clients, as in the GEH2 case. In this paper, we propose a scalable two-stage framework for RUL prediction of PEMFCs operating under such conditions. First, we introduce a machine learning-based method to extract a degradation-specific Health Indicator directly from voltage measurements, effectively filtering out transient operational effects. Second, we develop a hybrid deep learning architecture that combines Transformer networks and Gated Recurrent Units (GRUs) to model temporal dependencies and provide accurate RUL predictions under dynamic conditions. The proposed approach is validated on a real-world industrial dataset collected from three PEMFC stacks deployed in GEH2 systems operating under highly variable conditions. Comparative results show that our method consistently outperforms baseline machine learning and deep learning models, achieving superior accuracy, robustness, and generalization across diverse mission profiles.

BackgroundAllergic rhinitis (AR) and asthma may impact health-related quality-of-life. However, national estimates on the quality-of-life of patients with AR or asthma are lacking. ObjectiveTo provide estimates for utility scores and EQ-5D Visual Analog Scale (VAS) for patients with AR or asthma. MethodsWe conducted a cross-sectional study using direct patient data from the MASK-air® app on European MASK-air® users with self-reported AR or asthma. We used a multiattribute instrument (EQ-5D) to measure quality-of-life (as utility scores and EQ-5D VAS values). Mean scores were calculated per country and disease control level using multilevel regression models with post-stratification, accounting for age and sex biases. ResultsWe assessed data from up to 7905 MASK-air® users reporting a total of up to 82,737 days. For AR, utilities ranged from 0.86-0.99 for good control versus 0.72-0.85 for poor control; EQ-5D VAS levels ranged from 78.9-87.9 for good control versus 55.3-64.2 for poor control. For asthma, utilities ranged from 0.84-0.97 for good control versus 0.73-0.87 for poor control; EQ-5D VAS levels ranged from 68.4-81.5 for good control versus 51.4-64.2 for poor control. Poor disease control was associated with a mean loss of 0.14 utilities for both AR and asthma. For the same control levels, AR and asthma were associated with similar utilities and EQ-5D VAS levels. However, lower values were observed for asthma+AR when compared to AR alone. ConclusionPoor AR or asthma control are associated with reduced quality-of-life. The estimates obtained from mHealth data may provide valuable insights for health technology assessment studies.

Like in many fields of medicine, the concept of precision dosing has re-emerged in routine practice in allergology. Only one retrospective study on French physicians' practice has addressed this topic so far and generated preliminary data supporting dose adaptation, mainly based on experience, patient profile understanding and response to treatment. Both intrinsic and extrinsic factors shape the individual immune system response to allergen immunotherapy (AIT). Herein, we focus on key immune cells (i.e., dendritic cells, innate lymphoid cells, B and T cells, basophils and mast cells) involved in allergic disease and its resolution to further understand the effect of AIT on the phenotype, frequency or polarization of these cells. We strive to discriminate differences in immune responses between responders and non-responders to AIT, and discuss the eligibility of a non/low-responder subset for dose adaptation. A differential behavior in immune cells is clearly observed in responders, highlighting the importance of conducting clinical trials with large cohorts of well-characterized subjects to decipher the immune mechanism of AIT. We conclude that there is a need for designing new clinical and mechanistic studies to support the scientific rationale of dose adaptation in the interest of patients who do not properly respond to AIT.

Sublingual allergen immunotherapy (SLIT) is a safe, effective, disease-modifying treatment for moderate-to-severe respiratory allergies. The function and responsiveness of the immune system components underlying the effects of allergen immunotherapy may vary from one patient to another. Furthermore, the severity of the symptoms of allergic disease can fluctuate over time, due to changes in environmental allergen exposure, effector cell responsiveness, and cell signaling. Hence, the allergen dose provided through SLIT can be fine-tuned to establish an optimal balance between effectiveness and tolerability. The objective of the MaDo study was to describe and understand dose adjustments of SLIT liquid formulations in France. We performed a retrospective, observational, cross-sectional, real-life study of allergists and other specialist physicians. Physicians described their patients via an anonymous case report form (CRF). The main patient inclusion criteria were age 5 years or over, at least one physician-confirmed IgE-driven respiratory allergy, and treatment for at least 2 years with one or more SLIT liquid preparations. A nationally representative sample of 33 specialist physicians participated in the study. The physicians' main stated reasons for dose adjustment were adverse events (according to 90.9% of the physicians), treatment effectiveness (60.6%), sensitivity to the allergen (42.4%) and other characteristics (30.3%: mainly symptom severity, type of allergen, and asthma). 392 CRFs (mean ± standard deviation patient age: 27.8 ± 17.5; under-18s: 42.1%; polyallergy: 30.9%) were analyzed. Respectively 53.6%, 25.8%, 15.3%, and 8.7% of the patients received house dust mite, grass pollen, birch pollen and cypress pollen SLIT. Dose adjustments were noted in 258 (65.8%) patients (at the start of the maintenance phase for 101 patients (39.2%) and later for 247 (95.7%)). Dose adjustment was not linked to sex, age, or the number of allergens administered. All measures of disease severity (including symptom severity noted on a 0-to-10 visual analogue scale by the physician) decreased significantly during SLIT. Notably, the mean AR symptom severity score decreased to a clinically relevant extent from 7.6 at SLIT initiation to 2.4 at last follow-up, and the mean asthma symptom severity score decreased from 5.0 to 1.3. The few differences in effectiveness between patients with vs. without dose adjustment were not major. For about one patient in five, a specialist physician decided to reduce or increase the SLIT liquid dose at the start of maintenance treatment and/or during maintenance treatment. This decision was influenced by a broad range of patient and treatment factors, mainly to improve tolerability to treatment and/or enhance effectiveness. In France, dose adjustment of SLIT liquid preparations as a function of the patient profile and/or treatment response is anchored in clinical practice. Precision dosing might optimize the overall benefit-risk profile of AIT for individual patients throughout their entire treatment course, enabling them to achieve both short- and long-term treatment goals, whilst maximizing the safety and tolerability.

BackgroundAlthough respiratory allergies to house dust mites (HDMs) can often be controlled with symptomatic medications, some patients do not achieve satisfactory disease control. ObjectiveTo assess fortnightly fluctuations (notably worsening and/or exacerbations) in disease parameters among patients taking only symptomatic medications for HDM allergy. MethodsIn a 13-month, observational, multicenter survey of adults with a self-reported history of poorly controlled, moderate-to-severe, physician-diagnosed HDM respiratory allergy in France, Italy, and Spain, fortnightly telephone interviews were used to gather information on medication use, symptoms, the disease burden, and medical consultations from late May 2012 to early July 2013. ResultsA total of 313 patients completed the study (n = 114 in Italy, 92 in France, and 107 in Spain). Although most participants reported improvements in symptoms, a substantial minority (ranging from 12% to 44% per fortnightly telephone interview in 2012 and from 16% to 37% in 2013) complained of worsening. A few study participants did not improve at any time in the study: 4% overall, and 2%, 2%, and 7% in Italy, France and Spain, respectively. A change in the weather and/or contact with other allergens were the most frequent self-reported reasons for worsening, although the answer “I don't know” was also prominent. ConclusionIn a 13-month survey of patients with HDM allergy in Italy, France, and Spain, the participants’ symptom status fluctuated significantly — illustrating the complexity of this condition. Although most participants reported improvements, the “never-improver” profile warrants further investigation. More prominence could be given to symptom control and a low exacerbation risk as treatment goals in allergic rhinitis.

It has been demonstrated that immunoglobulin (Ig)E specific for cross-reactive carbohydrate determinants (CCD) is present in the serum of sensitized humans, dogs and cats, and that these CCD-specific antibodies might confound serological testing. The objective was to determine whether or not CCD-reactive antibodies occur in horses and to investigate the prevalence of CCD-reactive IgE antibodies in equine sera using a monoclonal cocktail-based enzyme-linked immunosorbent assay designed to detect allergen-specific IgE in horses, and to evaluate a means for successful inhibition of these CCD. Sera from 28 horses suspected of clinical allergy were evaluated, with and without a proprietary inhibitor which contains carbohydrates derived from bromelain (BROM-CCD), using a panel of 72 allergens that include 15 grasses, 17 trees, nine weeds, five mites, 12 fungi, 12 insects and two environmental allergens. Twenty-five samples were shown to be reactive to at least one of the allergens, and 15 were reactive to 10 allergens or more. BROM-CCD had minimal effect on the mite reactivity in any of the positive samples; however, substantial inhibition for pollen allergens (trees, grasses and weeds) was demonstrable. Reduction in signal to pollens ranged from 20% to 100% for samples that were inhibited by CCD-BROM. These results demonstrate that CCD-reactive IgE antibodies are evident in horses and that BROM-CCD can be effective in reducing reactions with these irrelevant carbohydrates and will likely yield a more accurate invitro allergen reactivity profile for selection of allergens included in an immunotherapeutic regime.