**4. Discussion**

AIT is the only disease-modifying and potentially resolving treatment available for patients with IgE-mediated allergic diseases, and its efficacy has been proven with a high degree of evidence [34–36]. However, one of the major problems of AIT management in clinical practice is that current guidelines give no clear indication about the algorithm to be used for choosing patients eligible for this treatment [37–40]. This could be one of the reasons for the low patient compliance with AIT reported in many clinical studies [41–43]. The introduction of personalized medicine, envisioned as a patient-tailored diagnostic and therapeutic approach, is currently influencing all fields of medicine; therefore, dedicated tools for identifying patients eligible for AIT are strongly needed [13].

According to current guidelines, AIT is indicated in patients with allergic rhinitis, with or without co-existing asthma [37–40]. Identification of the allergen(s) driving symptoms is the first level of patient stratification to ensure that the correct allergen solution is used for AIT. However, this treatment is preferentially used in patients with few sensitizations or polysensitized patients with one to three dominant allergens in clinical practice. In addition, to take advantage of AIT long-term effects, younger patients with few allergic diseases and a recent onset of allergic rhinitis are preferred for AIT. Previous studies have tried to correlate AIT response with a single marker. For instance, changes in cytokine pattern, such as an increase in Th2-dependent cytokines IL-4 and IL-13, have been associated with AIT response [9]. In addition, IL-10 mRNA levels have been suggested to be predictive of clinical responses to AIT [14]. IL-10 producing regulatory B- and T-cells specific for allergens were reported to increase during AIT or following the natural allergen exposure [15]. However, no cytokine has been clearly established as a marker for AIT efficacy to be used in the clinical routine. In addition, the specific and total IgE ratio has been formerly proposed as a biomarker of AIT efficacy [11,44]. A study by Di Lorenzo et al. [11] analyzing 279 monosensitized patients treated with both SCIT and SLIT immunotherapy found that specific IgE/total IgE ratio >16.2 (i.e., specific IgE/total IgE ratio × 100) was associated with an effective response to AIT. On the contrary, a randomized, double-blind, placebo-controlled clinical trial by Fujimura et al. [44] reported that patients with specific IgE/total IgE ratio <0.19 achieved better AIT outcomes. Other authors have suggested considering allergen-specific IgE level rather than the specific IgE/total IgE ratio, describing higher specific IgE levels in AIT responders than in non-responder adults [45] and children with allergic rhinitis [46]. The same research group has proposed a cut-off value of allergenspecific IgE levels (>9.74 kUA/L) that could predict a successful response to AIT [10]. Nonetheless, these observations were based on a small number of studied patients, with a consistent discrepancy between the sample size of the responders and non-responders, which should be considered a limitation of these results [9]. We chose to use the specific IgE/total IgE ratio rather than allergen-specific IgE level, because we would need a normalized specificity index that could be easily stratified into categories. In addition, the linear regression analysis of PRIS parameters (Table 4) shows that the ratio was a significant PRIS predictor (*p* < 0.001). CRD can help to differentiate patients with genuine sensitization from those with cross-reactive sensitization to other allergen sources [1,17,47]. This could help avoid administering irrelevant allergens in AIT, improving its clinical efficacy and cost effectiveness [1,17,47]. A pilot study by di Coste et al. [1] including 36 children with allergic rhinoconjunctivitis monosensitized to grass pollen aimed to evaluate the correlation between the sensitization to different molecular Phleum pratense (Phl p) allergens and clinical efficacy of SLIT. The authors performed serum analysis of specific IgE to Phl p 1, 2, 4, 5, 6, 7, 11, and 12, and showed that SLIT was effective irrespective of the patients' baseline sensitization to either single or multiple grass pollen allergens [1]. However, a direct correlation between IgE sensitization for other major allergens detected at the molecular component level and AIT outcome has currently not been found. Finally, other potential biomarkers that have been suggested for the assessment of AIT efficacy are the assessment of IgG1 and IgG4 levels [12] or the basophil activation test [48,49], but they did not show real reliability, and there is no consensus in their usage in patients undergoing SLIT [50]. Indeed, numerous studies indicate IgG1 and IgG4 levels increase during SLIT, but they may reflect compliance instead of clinical efficacy [12,50].

To approach the heterogeneity of allergic patients, we developed a multi-parameter score, namely PRIS, potentially able to predict AIT effectiveness and identify eligible patients. PRIS includes clinical and laboratory parameters (Table 2) chosen based on clinical practice, literature review, and previous work evaluating AIT responsiveness and possible predictive factors [1,10–13,44,46,48,49,51–53]. We rated each PRIS parameter to reach a maximum of 100 points to mimic the odds of achieving a clinical improvement, thereby making it easy to be used by clinicians and intelligible by the patients. We included age, number of allergic diseases, and disease onset as clinical parameters. In addition, we tried to increase the awareness of the major allergen(s) responsible for clinical symptoms by including in the PRIS the evaluation of the occurrence of symptoms following the exposure to one or more allergens [51]. As diagnostic parameters, we started by evaluating the sensitization profiles of patients, and we included the number of sensitizations, CRD, and the specific/total IgE ratio. This last parameter was preferred to specific IgE level because we would need a normalized specificity index. Finally, we included the number of dominant allergens used to calculate PRIS and decide the allergen(s) to be used with SLIT.

To our knowledge, this study is the first time that a multi-parameter score has been effective in identifying patients eligible for SLIT in a real-life setting. Overall, PRIS strongly correlates with symptom improvement after 24 months of SLIT (Figure 2), and it was effective in patients undergoing single-allergen SLIT (Figure 3A) as well as in patients treated with a two-allergen SLIT (Figure 3B). Furthermore, PRIS was a good predictor in both patients with only rhinitis (Figure 3C) and patients with rhinitis associated with asthma (Figure 3D). Together, these results indicate that PRIS can be used to predict the efficacy of SLIT regardless of the patient's clinical condition, the product, and the number of allergens used for SLIT. Our observations suggest that PRIS also effectively predicts the degree of clinical response patients may expect from SLIT. Indeed, we observed that patients with a higher PRIS value have significantly more chances of achieving a higher symptom improvement (Figure 4). In addition, we reported that most patients with PRIS ≥ 80 experienced a very high improvement, whereas patients with 60 > PRIS ≥ 40 mostly experienced a high improvement (Figure 5). Further studies are needed to confirm these observations on a larger scale.

One of the surprising observations in our study was to find such high patient compliance to AIT. Indeed, we observed that only 10.9% of patients included in our study did not complete the 24-month follow-up. These data are dramatically lower than that reported by most studies (30–40%) [41–43]. We cannot exclude the possibility that the limited number of patients influenced compliance in our study. However, we believe that one of the reasons for the high patient compliance in our cohort is that the knowledge of their PRIS value conferred them a sort of awareness about the goal they could achieve with SLIT.

The clinical efficacy of AIT is measured using various scores as study endpoints. According to EAACI recommendations [37–40], we used a combined symptom and medication score (MSS) to permit the comparison of results with other studies. Our results confirmed that SLIT effectively improves symptoms of rhinitis and/or asthma after 12 months of SLIT treatment (Figure 1A). Although an additional symptom improvement was recorded after 24 months, no significant difference was observable between MSS-12 and MSS-24 (Figure 1A). Another interesting point of reflection is that up to a 30% improvement is achieved with placebo in AIT placebo-controlled studies [54]. This placebo effect is substantially less than >75% in nearly half of the patients in the present study (Table 1). However, as this study was open, there is no way to evaluate a contribution of the placebo effect on the perceived effects of the AIT. We also observed a significant response in both monosensitized and polysensitized patients (Figure 1B) and patients with or without allergic asthma (Figure 1C). In our cohort, a better efficacy was found in patients with only rhinitis than in patients with rhinitis associated with asthma (Figure 1C). However, further studies are needed to confirm this observation.

Our study is subject to some limitations. First, the sample size was small, but our encouraging results showed that a strong direct correlation between PRIS and SLIT outcome (Figures 2 and 3) could be a starting point for multi-center studies, which could validate PRIS on a larger scale. Second, we used different products for SLIT, and we cannot exclude the possibility that data on SLIT efficacy can be influenced by the product used. However, we avoided any product comparison because products for SLIT cannot be compared at present due to their heterogeneous composition [55,56]. From our perspective, combining several parameters routinely used in clinical practice to obtain a disease score rather than relying on a single parameter or a single product for SLIT may help better manage the within-subject variability.

In conclusion, AIT is very demanding for the patients since it is expensive and requires a long period to achieve a sustained response [7,37]. Therefore, a specific tool able to predict SLIT efficacy is worth being used in clinical practice to select eligible patients and improve patients' compliance to complete the course of treatment.

**Author Contributions:** Conceptualization, I.M. and F.G.; Data curation, I.M., F.G., A.D., D.P. and F.D.R.; Formal analysis, I.M., F.G., A.D., D.P., F.D.C., F.D.R. and A.R.; Investigation, I.M., A.D., D.P., F.D.C., F.D.R. and A.R.; Methodology, I.M., A.D., D.P., F.D.C., F.D.R. and A.R.; Project administration, F.G. and A.d.P.; Resources, F.G., A.D., A.d.P. and F.W.R.; Supervision, F.G., A.d.P. and F.W.R.; Validation, I.M., F.G., A.D., D.P., F.D.C., F.D.R., A.R., A.d.P. and F.W.R.; Visualization, I.M. and F.G.; Writing—original draft, I.M., F.G., F.D.C. and F.D.R.; Writing—review and editing, I.M., F.G., A.D., D.P., A.R., A.d.P. and F.W.R. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** The study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of the University of Naples Federico II (protocol code 75/21).

**Informed Consent Statement:** This study was performed in accordance with the principles of the Helsinki Declaration. Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author.

**Conflicts of Interest:** The authors declare no conflict of interest.
