It was a 1:1 randomized, double-blind, placebo-controlled study in parallel groups (PMBL vs placebo). The study received a favorable opinion from the Bioethics Committee of the Medical University of Lublin (Resolution No KE-0254/251/2020, 26 November 2020) and was prospectively registered with ClinicalTrials.gov (Trial Registration No NCT04802616, 17 March 2021). The study was conducted in accordance with Good Clinical Practice standards, and the ethical principles that have their origin in the Declaration of Helsinki. The project was implemented in 3 centers in eastern Poland from 22 March 2021 to 29 October 2021.

The primary objectives of this study were to evaluate the effectiveness of 3-month PMBL therapy in improving the clinical course of grass pollen-induced AR in children, and above all its effect on changes in the blood level of the γδT cell subsets: Th1-like, Th2-like, Th10-like, Th17-like, Treg-like; iNKT cell subsets: iNKT1, iNKT2, iNKT10, iNKT17, iNKTreg; cytotoxic T (Tc) cell subsets: Tc1, Tc2, Tc10, Tc17, Treg-like and to assess the relationship between the level of these lymphocytes and the severity of SAR symptoms assessed using the total nasal symptom score (TNSS).

The secondary study objectives were to assess the effect of PMBL therapy on the need for oral H1-antihistamines and intranasal corticosteroids during the grass pollen season and to evaluate the safety and tolerability of this therapy.

Eligible for the study were children aged 5 to 17 years with SAR diagnosed, as per current ARIA recommendations (4), with predominant sensitization to grass pollen allergens determined by skin prick testing (wheal diameter ≥ 3 mm and greater than negative control; Allergopharma-Nexter Ltd, Poland) or based on serum allergen-specific immunoglobulin E (asIgE) levels (≥ class 2; Polycheck, Biocheck GmbH, Germany), with SAR symptoms present in at least two previous grass pollen seasons. Exclusion criteria were as follows: allergy to tree pollens, receipt of PMBL within 12 months or PCBL within 6 months before the randomization visit, oral or subcutaneous allergen-specific immunotherapy within 3 years before the start of the study, treatment with systemic corticosteroids within the last 6 months before the start of the study, history of respiratory infection within 2 weeks before the randomization visit, systemic immunologic disorders, history of transfusion of blood, blood components or blood products.

Children were recruited for the study in the second half of March 2021, before the start of the grass pollen season in Poland. For all patients, written informed consent was obtained from a parent or legal guardian and assent was obtained from the patient.

The study used PMBL obtained by sonification, whereby ultrasound is used to inactivate the bacterial cell wall (Ismigen, Lallemand Pharma AG, Switzerland). The product is available in the form of sublingual tablets containing 7 mg of bacterial lysate of the following bacteria: Streptococcus pneumoniae, Streptococcus pyogenes, Streptococcus viridans, Staphylococcus aureus, Neisseria catarrhalis, Haemophilus influenzae, Klebsiella pneumoniae and Klebsiella ozaenae. PMBL was taken by patients according to its dosage regimen in recurrent respiratory infections. Thus, children in the PMBL group received 3-cycles of PMBL treatment, each consisting of sublingual administration of one PMBL tablet on an empty stomach for 10 consecutive days followed by a 20-day break (25). In the control group, the patients received placebo tablets indistinguishable from the PMBL tablets in shape, color, smell/taste and dissolution time under the tongue.

The randomization list was generated by hospital pharmacy staff using Random Allocation Software version 2.0. Pharmacists also prepared identical packs containing 30 PMBL or placebo tablets each and labelled them with a unique code consistent with the randomization list. The individuals responsible for preparing the randomization list and tablet packs were not involved in the implementation of the study. Children attending the first visit and meeting all inclusion criteria and none of the exclusion criteria drew a card with a unique package number. Patients and investigators, including those responsible for immunoassays, were blind to allocation. Unblinding was performed after receiving patient diaries from all participants, that was in early November 2021.

The dates of visits to the centers and the periods of assessment of SAR symptoms subjected to statistical analysis were determined based on measurements of grass pollen grain concentrations in ambient air.

A Hirst-type apparatus (Lanzoni VPPS 2000) (35) placed on the flat roof of the building of the University of Life Sciences in Lublin (22032’25’’ E and 51014’37’’ N; 197 m a.s.l.) at 18 m above ground level was used to monitor the concentration of grass pollen grains in the air. The apparatus operated in continuous mode, sucking in air along with bioaerosol elements. The pollen grains stuck to a sticky-coated tape placed inside the apparatus on a moving drum. After a week’s exposure, the tape was cut into sections corresponding to 24 hours and microscope slides were prepared. The results were expressed as the number of pollen grains in 1 m³ of air per day. Pollen monitoring was conducted in accordance with the recommendations of the International Association for Aerobiology and Quality Control Working Group (36). The beginning and end of the grass pollen season in eastern Poland were determined using the 95% method ( Figure 1 ) (37).

The study included two site visits: visit 1 (V1) – a screening/randomization visit, before the start of the grass pollen season (22–31 March 2021), visit 2 (V2) – at the peak of the grass pollen season (23 June – 2 July 2021), and two telephone contacts (TC1, TC2; the study protocol registered at ClinicalTrials.gov provided for 4 telephone contacts, but two contacts were dropped due to the late start of the grass pollen season) ( Figure 2 ). For 7 days after each visit, parents rated the severity of four SAR symptoms (nasal congestion, rhinorrhea, itching of the nose and sneezing) in the patient’s diary, each scoring from 0 to 3 (0 – no symptom, 3 – severe symptom), and then summed the scores to obtain the TNSS (weekly averages were analyzed statistically) (38). Patients started the first cycle of sublingual tablets on 1 April 2021. In addition, the use of an oral antihistamine (desloratadine) or an intranasal corticosteroid (mometasone furoate) was allowed in case of severe allergy symptoms, and the intake of these drugs was recorded daily in the diary (the number of days of the grass pollen season in which patients took the above drugs was analyzed statistically).

During the visits to the center (V1, V2), blood was taken for immunological tests, the severity of four SAR symptoms was assessed cumulatively on a visual analogue scale (VAS), and peak nasal inspiratory flow (PNIF) was measured using the Youlten Peak Flow Meter (Clement Clarke International, UK) (all subjects received appropriate instructions on how to use the Peak Flow Meter correctly; three measurements were recorded for each subject and the highest flow was used for analysis) (39). During telephone contacts, patients were asked to rate the severity of their SAR symptoms cumulatively on a VAS scale using an appropriately prepared tool in a patient diary (instructions, 10 cm line).

The peripheral blood taken from patients was collected into 4.9 ml tubes: K2-EDTA S-Monovette (Sarstedt, Germany). Next, peripheral blood mononuclear cells (PBMC) were isolated by density gradient centrifugation in Gradisol L (Aqua-Med, Poland) at 700 rcf for 20 minutes at room temperature. PBMC were counted in the Neubauer chamber. PBMC at a density of 1 × 10⁶ per ml were distributed into cell culture plates and stimulated at 37°C in a CO₂ incubator for 4 hours in the presence of monensin (2 µM/ml), ionomycin (1 µg/ml) and phorbol 12-myristate 13-acetate (50 ng/ml). After stimulation, the cells were incubated with the combination of monoclonal antibodies. The following monoclonal antibodies were used in the current study: anti-TCR V alpha 24 J alpha 18 FITC (cat. no 342906, clone: 6B11, BioLegend), anti-TCR V alpha 24 J alpha 18 PE-Cy7 (cat. no 342912, clone: 6B11, BioLegend), anti-CD3 PE (cat. no 555333, BD Biosciences), anti-CD3 Krome Orange (cat. no B00068, clone: UCHT1, Beckman Coulter), anti-CD4V450 (cat. no 560345, clone: RPA-T4, BD Biosciences), anti-CD8 Alexa Fluor 700 (cat. no B76279, clone: B9.11, Beckman Coulter), anti-CD8 PE-Cy5 (cat. no 300909, clone: HIT8a, BioLegend), anti-TCR PAN delta/gamma PE-Cy5 (cat. no IM2662U, clone: IMMU510, Beckaman Coulter), anti-TCR PAN delta/gamma FITC (cat. no 347903, clone: 11F2, BD Biosciences), anti-CD45BUV395 (cat. no 563792, clone: HI30, BD Biosciences) and ViaKrome 808 (cat. no C36628, Beckman Coulter). After 20 minutes of incubation, cells were permeabilized with methanol at 4–8°C for 30 minutes in the darkness.

Next, the cells were washed twice with phosphate buffered saline (PBS) for 10 minutes, 700 rcf. Afterwards, intracellular labelling was performed using the following antibodies: anti-FoxP3 Pacific Blue (cat. no B90432, clone: 259D, Beckman Coulter), anti-IFN-gamma PE-CF594 (cat. no 562392, clone: B27, BD Biosciences), anti-IL-10 Brilliant Violet 786 (cat. no 564049, clone: JES3-9D7, BD Biosciences), anti-IL-17A APC-Cy7 (cat. no. 512320, clone: BL168, BioLegend), anti-IL-4 Brilliant Violet 605 (cat. no. 500828, clone: MP4-25D2, BioLegend), anti-E4BP4 PE (cat. no. 12-9812-42, clone: MABA223, Invitrogen), anti-T-bet APC (cat. no. 644814, clone: 4B10, BioLegend), anti-GATA-3 PerCP-eFluor 710 (cat. no. 44-9966-42, clone: TWAJ, Invitrogen) and anti-ROR gamma T Brilliant Violet 650 (cat. no. 563424, clone: RPA-T4, BD Biosciences). Incubation was performed at room temperature for 30 minutes in the darkness. After final wash with PBS, samples were analyzed on CytoFlex LX flow cytometer (Beckman Coulter, USA). Kaluza v2.1.1 (Beckman Coulter, USA) was used to analyze and graphically present the collected data ( Figure 3 ).

Sample size was determined using sample size test in Statistica 13.3 software (StatSoft, Poland) based on previous studies on the effect of immunostimulant products on the number of CD3+ lymphocytes in children with allergic asthma, recurrent respiratory tract infections and adults with chronic obstructive pulmonary disease (40–42). It was calculated that 21 patients should be included in the PMBL and placebo groups, respectively (power 85%, alpha value 0.05). The total sample size calculation required to enroll 50 patients in anticipation of a dropout rate of 15%.

Statistical analysis was conducted by an International Statistical Institute Elected Member not involved in the study using IBM SPSS Statistics 25 package. Using a two-way mixed model analysis of variance, checks were made to assess whether there were statistically significant differences between the study periods analyzed in the two groups, as well as whether the groups differed in terms of the variable under analysis in the same study period. Spearman’s rank correlation made it possible to check whether there was a statistically significant relationship between the analyzed variables. The Mann-Whitney U test was used to assess the presence of statistically significant differences between the two independent groups (in terms of age). A two-tailed P value lower than 0.05 was considered as statistically significant. Analyses were based on the intent-to-treat (ITT) population (patients who took ≥ 1 tablet and had ≥ 1 post-randomization assessment).

62 children were enrolled in the study. After excluding 8 patients who did not meet the inclusion criteria or met at least one exclusion criterion and 4 patients who did not agree to participate in the study, 25 children were randomized to the PMBL group and 25 to the placebo group. A total of 41 children completed the trial, 21 in the PMBL group and 20 in the placebo group ( Figure 4 ). Of the 41 children, 29 (71%) were male and 12 (29%) were female. Mean age of participants was 9.1 ± 2.55 years. Demographic characteristics were similar between the groups as shown in Table 1 .

In both groups, during the grass pollen season, a statistically significant increase and then a decrease in the severity of SAR symptoms expressed on the TNSS and VAS was observed ( Table 2 ). In the placebo group there was a statistically significant decrease in PNIF (p = 0.007), while in the PMBL group there was a statistically insignificant (p = 0.07) but clinically significant increase of 11.19 L/min (minimally clinically important difference [MCID] is 5 L/min) ( Table 3 ).

Children taking PMBL during the grass pollen season displayed much less intensity of nasal symptoms of SAR compared to children receiving placebo. The severity of the nasal symptoms of SAR on the TNSS scale was revealed to have a significantly lower impact in the PMBL group vs the placebo group at measuring points TC1 and V2 (p = 0.01, p = 0.009, respectively) ( Figure 5A ). A statistically significantly lower mean severity of nasal symptoms of SAR on the VAS scale was recorded for children in the PMBL group compared to the placebo group at measuring points TC1, V2 and TC2 (p = 0.04, p = 0.04, p = 0.03, respectively) ( Figure 5B ). The compared groups do not show significant differences in terms of PNIF values at individual measuring points, although a p-value on the verge of statistical significance (p = 0.053) was obtained at measuring point V2, indicating less nasal obstruction in the PMBL group versus the placebo group ( Figure 5C ).

There were no statistically significant changes in immune variables between V1 and V2 in both groups, and between the groups at both measurement points ( Table 4 ). For both groups, there was a statistically significant association between the level of Th1-like γδT cells and the severity of SAR symptoms expressed on the TNSS scale (p = 0.03) – the lower the level of Th1-like γδT cells, the higher the TNSS value ( Figure 6 ).

The mean number of days of use of desloratadine and mometasone furoate per patient was respectively 26% (18.62 ± 4.6 vs 25.2 ± 7.4 days, p = 0.005) and 31% (28.57 ± 7.41 vs 41.6 ± 11.17 days, p < 0.001) lower in the PMBL group vs the placebo group. The therapy was associated with a high safety and tolerability profile. In the PMBL group only one adverse event in forms of mild abdominal pain was reported, which resolved spontaneously without any medication. No adverse event was reported in the placebo group.