Delivery options for sublingual immunotherapy for allergic rhinoconjunctivitis: clinical considerations for North America*

Background : Sublingual immunotherapy (SLIT) can be delivered via tablets (SLIT-T) or aqueous drops (SLIT-D). SLIT-D dosing recommendations using North American extracts were published in 2015. We review the 2015 recommendations in the context of recent research, and compare and contrast dosing, efficacy, safety, adherence, and cost of SLIT-T and SLIT-D for allergic rhinoconjunctivitis (ARC) in North America. Methods : Randomized controlled trials (RCT) of SLIT-D and SLIT-T trials were identified by a systematic PubMed search through March 1, 2022. Results : Dose-finding studies have been conducted for all approved SLIT-T; efficacy in North American populations was demonstrated in 11 RCTs. Approved SLIT-T are uniform internationally. Few dose-finding studies for SLIT-D have been conducted using North American extracts; efficacy was demonstrated in 2 RCTs. Extrapolation of dosing from SLIT-D studies conducted with extracts from other geographic regions is unreliable. Since the 2015 SLIT-D dosing recommendations, no new RCTs of SLIT-D have been conducted with North American extracts, whereas 6 SLIT-T RCTs have since been conducted in North America. Local allergic reactions are the most common adverse events with SLIT-T and SLIT-D, but both can induce systemic allergic reactions. Adherence to SLIT-D and SLIT-T remains a challenge. Patients must pay for SLIT-D directly, whereas SLIT-T is usually covered by insurance. Conclusion : As part of shared decision-making, patients should be informed about the scientific evidence supporting the use of SLIT-T and SLIT-D for ARC.


Introduction
Allergy immunotherapy (AIT) is one of the many treatment options for allergic rhinoconjunctivitis (ARC) and asthma (1) . AIT improves ARC symptoms and reduces the need for symptomrelieving pharmacotherapies by inducing tolerance to the specific allergen(s) (2) . The developed tolerance can be long-term, lasting years after AIT has been stopped (3)(4)(5) . Two methods of AIT administration are in common use, subcutaneous immunotherapy (SCIT) and sublingual immunotherapy (SLIT). SLIT can be delivered via tablets (SLIT-T) or aqueous drops (SLIT-D) (6) . SLIT can be administered daily at home after a medically supervised first dose, which affords the patient a convenient alternative to office visits required for SCIT. SLIT also generally has a better safety profile than SCIT (7,8) . There have been no deaths attributed to SLIT, in contrast to SCIT, which very rarely can induce nearfatal and fatal reactions (9) .
SLIT-T products are standardized and approved for the treatment of ARC caused by grass, ragweed, house dust mite (HDM), birch (and related trees), and Japanese cedar allergens, depending on the geographic region ( Table 1). The HDM SLIT-T is also approved as an add-on treatment for asthma in Europe (10) . There are a few SLIT-D products approved outside of North America, but none are approved in the US (Table 1). Despite the lack of Food and Drug Administration (FDA) approval, SLIT-D in the US are formulated with allergen extracts that are intended for SCIT administration. Surveys indicate that otolaryngologists in North America tend to use SLIT-D over SLIT-T, in contrast to allergists who tend to use SLIT-T over SLIT-D (11,12) .
A recognized limitation for the use of SLIT-D in North America is a lack of optimized doses with established efficacy and safety based on randomized controlled trials (RCTs) (13,14) . To address this issue, Leatherman et al. published recommendations for SLIT-D dosing in 2015 based on both available RCTs and expert opinion (13) . The purpose of this review is to critically review the published 2015 recommendations in the context of recent research, and compare and contrast dosing, efficacy, safety, adherence, and cost of SLIT-T and SLIT-D for ARC in North America.

Materials and methods
To identify SLIT-D and SLIT-T RCTs conducted since the recommendations for SLIT-D dosing were published by Leatherman et al. in 2015 (13) , a systematic search was conducted in PubMed using the terms "allergen immunotherapy AND (rhinoconjunctivitis OR rhinitis) AND (sublingual OR oral OR drops OR aqueous) AND randomized" and were limited to English only articles published between May 1, 2014 and March 1, 2022. Articles that were previously included in the 2015 SLIT-D recommendations (13) , those without standard ARC efficacy outcomes data (i.e., combined symptom and medication scores or other efficacy outcomes), those that did not compare active SLIT vs no SLIT treatment (change from baseline was allowed as a comparator for studies with an active control group instead of a placebo control group), those that were secondary analyses of a previously published trial (long-term follow-up results were allowed), and those that were not RCTs were excluded. The search terms resulted in 185 publications, which after initial title and abstract review, were narrowed down to 75 relevant publications for review of the full publication. After review for exclusion criteria, 13 publications were excluded because they were secondary analyses, 9 were excluded because they did not include efficacy outcomes, 8 were not RCTs, and 3 were not SLIT trials. Therefore, a total of 42 SLIT-D and SLIT-T RCTs based on the selection and exclusion criteria were ultimately identified. Characteristics and outcomes of these 42 trials are described in Supplemental Table   1. Non-systematic literature searches were conducted to obtain supplemental information about SLIT dosing, efficacy, safety, and adherence.

Dosing
The efficacy and safety of AIT is dependent on dose and duration (15)(16)(17)(18) . Repeated doses do not accumulate, meaning repeated ineffective doses do not result in an effective dose over time (15,19,20) . For SLIT-T, freeze-dried and compressed formulations have been shown to differ in the kinetics and quantity of major allergen released, which may impact efficacy and safety (21)(22)(23) .
Therefore, allergen-specific and formulation-specific dosefinding studies are essential to determine the optimal dose of SLIT products. In the 2015 dosing recommendations, the authors recognized that there were few large dose-finding studies of SLIT-D for consideration (13) . Effective SLIT-D doses were difficult to determine because doses that were effective in some studies were not effective in others and because of differences in extract formulations (i.e., aqueous or alum-absorbed). Neither timing nor duration for SLIT-D are clearly defined (24) . Nevertheless, Leatherman et al. provided a range of SLIT-D dosing recommendations using best available evidence from the available RCTs and expert opinion (Table 2) (13) .
One of the limitations of the 2015 SLIT-D dosing recommendati- Table 1. Regulatory-approved SLIT-T and SLIT-D products in the United States, Canada, Germany, Australia, and Japan.

United States
Canada Germany Australia Japan

Pollen
Short ragweed HDM, house dust mite; SLIT-D, sublingual immunotherapy drops; SLIT-T, sublingual immunotherapy tablets. *5-grass mix or timothy grass. † Also approved for the treatment of alder and hazel-related allergic rhinoconjunctivitis. ‡ Mixed with birch and hazel. § Mixed with birch and alder.
with SLIT-D extracts in Europe and in US SCIT extracts can also differ substantially depending on the manufacturer (25,26) . new studies have been conducted in North America, all of which were with SLIT-T (Supplemental Table 1) (18,(27)(28)(29)(30)(31) . No new RCTs of SLIT-D using North American extracts have been conducted to update or expand upon the prior recommendations.
Another limitation of the 2015 SLIT-D dosing recommendations was the overlap in effective and non-effective doses (e.g., micrograms per month) reported in the studies examined. This led to a need to provide a broad range of recommended dosing for some allergens (Table 2) (13) . It is known that a 3-fold difference may separate an effective dose from an ineffective dose (15,32) . A 2018 survey of allergists in practice showed that a large variation in SLIT-D doses are used (33) . Aside from the 2015 SLIT-D dosing recommendations, healthcare providers may also use other methods to determine SCIT-D dosing. In general, higher doses than those used in SCIT are necessary (24) , and a daily SLIT-D dose equivalent to a monthly maintenance SCIT dose has been suggested, at least for grass (34) . One resource for otolaryngic allergy practice simply suggests adding 1 mL of each extract concentrate when preparing a maintenance vial (24) . Because there are no clinical trial data, it is unknown if the doses derived ons was that the majority of the studies examined for guidance were not conducted in North America (13) . Although there have been several successful trials with SLIT-D products in Europe and Asia and data from these trials are important, the allergen extracts used in those trials are unavailable to North American healthcare providers. Comparison of efficacy and safety studies with extracts manufactured in other regions of the world versus those manufactured in North America is unreliable because of differences in extract protein content, major allergen content, and relative potencies (25) . To further confuse the issue, potency,  Table 3. Non-standardized extracts are labeled by weight/volume or protein nitrogen units.
These units do not describe the amount of active ingredient and, therefore, do not provide a measure of potency or allergen content. Moreover, the quantity of the allergens administered using these methods are clinically efficacious and whether the microgram content is an appropriate measure linked to efficacy.
The use of modified quantitative testing (MQT) has been suggested to help plan the starting dose for SLIT-D (24) , although there is little evidence to support this practice. Based on clinical trials with SCIT, intradermal dilutional testing to determine dosing is not recommended (35)(36)(37) .
There is less ambiguity in the dosing for SLIT-T because randomized, double-blind, placebo-controlled (RDBPC) dose-finding studies have been conducted for all of the SLIT-T approved in North America (15,16,19,38) . The approved SLIT-T are uniform internationally. For the five-grass SLIT-T, the daily maintenance dose in adults and children is 300 IR (39) . The five-grass SLIT-T is typically given pre-and co-seasonally. For the Timothy grass SLIT-T, the daily maintenance dose in North America is 2800 BAU, for the ragweed SLIT-T is 12 Amb a 1-Units, for the HDM SLIT-T is 12 SQ-HDM, and for the tree SLIT-T is 12 SQ-Bet (40)(41)(42)(43) . SQ is a standardization method based on potency, major allergen content, and complexity of the allergen extract (characterization of major and minor allergens) (42) . In North America, the Timothy grass, ragweed, and tree SLIT-T are typically given pre-and co-seasonally, and HDM SLIT-T is typically given year-round. However, all SLIT-T are permitted to be initiated at any time of the year and may be administered year-round. Year-round use of Timothy grass SLIT-T is becoming more common.
The effective allergen content measured in micrograms for SLIT-D should not be assumed to be equivalent across different SLIT-D preparations, or to the effective allergen content in a SLIT-T product and vice versa (14) . This principle is demonstrated by comparing the only successful large RDBPC trial of SLIT-D conducted in North America, where the major allergen dose was approximately 50 mg Amb a 1 in 94% of the subjects (44) , whereas the approved dose for the ragweed SLIT-T is 12 mg Amb a 1 (40) .
Thus, the same microgram of allergen that is effective in a SLIT-T product cannot be assumed to be effective for SLIT-D.
A perceived advantage of SLIT-D is that it allows for mixture of allergens, which is common practice for SCIT in North America.
However, for both SCIT and SLIT-D, there is limited data on efficacy and safety of allergen mixtures (45) and Leatherman et al.
were not able to make an evidenced-based recommendation on multiallergen SLIT-D mixes (13) . for all allergens within the group (46,47) . The AAAAI/ACAAI AIT Task Force does not currently endorse SLIT-D (14) .
In all of the pivotal RDBPC trials of SLIT-T, the majority of the patients were polysensitized to more than one allergen but were only treated with one SLIT-T (15,30,(48)(49)(50)(51)(52) . There have been 2 studies on the safety of co-administering multiple tablets and 2 studies on co-administering SLIT-T with SLIT-D (53)(54)(55)(56) . In the dual SLIT-T studies, a sequential dosing schedule was followed (4 weeks with each tablet alone in 1 study and 2 weeks with each tablet alone in 1 study) (53,55) . Ultimately the tablets were co-administered within 5 minutes of each other. Combinations evaluated were grass + ragweed SLIT-T and HDM + Japanese cedar SLIT-T (53,55) . There do not seem to be any additional safety concerns when the SLIT-T are co-administered. To date, there are no data on the efficacy of co-administered SLIT-T.

Efficacy
The vast majority of the SLIT studies have been conducted outside of North America. As discussed above, extrapolation of data from European (or other regions) SLIT-D products to SLIT-D preparations of US extracts cannot be made based on allergen content alone. To date, there have been 13 SLIT-T (11 were Phase 3 or 4) and 7 SLIT-D (1 was Phase 3) RDBPC efficacy trials conducted in North America (15, 18, 20, 27-31, 44, 48-52, 57-62) . Of these, 9 of the 13 SLIT-T trials were field trials that demonstrated significant improvement in symptoms and a reduction in symptom-relieving medication use compared with placebo (standard of care which allowed for symptom-relieving medication use) as measured by a combined symptom and medication score (15,27,29,30,(48)(49)(50)(51)(52) . One grass SLIT-T field trial did not show a significant improvement from placebo in the primary endpoint (daily symptom score) (60) . The remaining 3 SLIT-T trials were environmental exposure chamber (EEC) trials, two of which were dose-finding trials for the tree SLIT-T and the HDM SLIT-T that demonstrated significant improvement in symptoms after allergen challenge versus placebo (18,31) . The third SLIT-T EEC trial investigated the potential  bystander effect of grass SLIT-T treatment after birch challenge and found no significant bystander effect on birch-induced symptoms (no confirmatory grass challenge was conducted) (28) .
Only 2 of the 7 trials of SLIT-D using US extracts demonstrated a significant improvement in symptom and medication scores compared with placebo (44,62) .
Three meta-analyses have compared the standardized mean difference in symptom and medication scores from placebo between SLIT-T and SLIT-D studies, including studies conducted outside North America (63)(64)(65) . The placebo effect in allergy medication and AIT trials is wellknown. A meta-analysis of AIT trials demonstrated that up to 68% of the total treatment effect of SLIT in adults was attributed to a placebo effect (68) . Such a large placebo effect in ARC trials is partially because the primary outcome typically includes a subjective reporting of symptom scoring by the patient, which is susceptible to psychological mechanisms (69) . Another contributor to the placebo effect in AIT trials is the allowed use of symptom-relieving medications. The use of symptom-relieving medications in the placebo group potentially lowers (improves) the symptom score of the placebo group, which decreases the ability to detect a true difference in scores between active treatment and placebo (basement effect). To help offset the impact of symptom-relieving medication use in AIT trials, the recommended primary outcome is a combined score of symptom and medication use (70) , and results are often presented as percentage improvement relative to placebo. Such an analysis takes into account 'real world' conditions for patients for whom these medications are widely available.

Disease modification and sustained effect
In contrast to symptom-relieving pharmacotherapy, AIT changes the immunologic mechanisms that drive ARC symptoms and therefore has the potential to have sustained effects after treatment is stopped (71) . Two large RDBPC 5-year trials demonstrated that 3 years of year-round timothy grass SLIT-T treatment resulted in significantly improved symptoms and symptom-relieving medication use compared with placebo 2 years after the end of treatment (3,72) . Thus, the timothy grass SLIT-T is recognized by the US FDA as having a sustained effect and met the criteria set by the European Medicines Association for a disease-modifying effect (41,73) . These studies show that 3 years of timothy grass SLIT-T treatment appears to be required to achieve the sustained effect. The five-grass SLIT-T has demonstrated clinical benefits for up to 2 years after 3 years of pre-and co-seasonal treatment in a RDBPC trial (74) ; however, the results were not sufficiently robust for the tablet to be recognized by regulatory authorities as having a sustained effect or as a disease-modifying treatment.
One RDBPC trial of the Japanese cedar SLIT-T has demonstrated a sustained and disease modifying effect for up to 2 years after 3 years of treatment (17,75,76) . Trials evaluating a sustained and disease-modifying effect with the HDM, ragweed, and tree SLIT-T have not yet been conducted.
There have been 2 RDBPC trials of SLIT-D (both used European extracts) that evaluated a sustained effect. One trial demonstrated a significant improvement in symptom score compared with placebo 1 year after completing 3 years of co-seasonal five-grass SLIT-D treatment, but improvement in the combined symptom and medication score did not reach significance for the 1-year follow-up (77) . The second trial demonstrated a sustained effect from placebo 1 year after completing 32 weeks of Artemisia annua SLIT-D (78) . In a prospective, randomized, open-controlled trial of HDM SLIT-D, patients received treatment for 3, 4, or 5 years (79) . A sustained significant improvement in the combined symptom and medication score compared with placebo was Table 4. Cost of monthly supply of SLIT-T and SLIT-D in the United States.
SLIT-D daily doses are those recommended by Leatherman et al. (13) . observed every year for 6 years after treatment discontinuation in patients who received 3 years of treatment and for 7 years for patients who received 4 or 5 years of treatment. Another RCT of HDM SLIT-D found that 1 or 2 years of treatment resulted in a sustained improvement in symptom scores 1 year after treatment discontinuation (80) . In addition, a retrospective study of patients receiving 1, 2, 3, or 4 years of HDM SLIT-D demonstrated that a sustained improvement in symptom scores compared with placebo was observed for up to 8 years after treatment discontinuation in patients who received 4 years of treatment, but the duration of sustained effect decreased in proportion to the other years of treatment duration (81) .

SLIT product
The RDBPC trials of SLIT-T that included North American populations indicate that treatment induces a significant increase from baseline in allergen specific IgG4, specific IgE, and IgE-blocking antibodies compared with placebo (15,27,29,30,48,49,51,52,60) . Significant increases from baseline in allergen specific IgG4 and specific IgE were also demonstrated in 4 of the 7 SLIT-D trials conducted in North America (20,44,57,58) . One trial of HDM SLIT-D noted a significant increase from baseline in allergen-specific IgG4, but not specific IgE (59) . A trial of dual HDM and timothy grass SLIT-D demonstrated a significant increase in allergen-specific IgG4 but a significant decrease in specific IgE compared with placebo (62) .
In the trial of cat SLIT-D, there was no significant increase from baseline in either allergen-specific IgG4 or specific IgE (61) .

Prevention of asthma and new sensitizations
The potential prevention of asthma is a key motivation for use of AIT in children. One large, RDBPC 5-year trial of grass SLIT-T in children found that the SLIT-T significantly reduced the proportion of patients with asthma symptoms or asthma medication use compared with placebo (16% vs 20%, p=0.036) (72) . Four open-label, controlled trials have demonstrated that treatment with either HDM or pollen SLIT-D with European extracts significantly reduces the proportion of patients that develop asthma or prevents decreases in forced expiratory volume in 1 second (FEV1) (82)(83)(84)(85) .

Prevention of development of new allergen sensitizations is
another potential motivator for AIT use, but a meta-analysis of 18 AIT studies published in 2017 found that there is little high quality evidence to support such an outcome (86) . Of the 18 studies, 6 were with SLIT (1 SLIT-T and 5 SLIT-D). Only the SLIT-T trial was a RDBPC trial (3) . After long-term follow-up 2 years after ending a 3-year treatment period with grass SLIT-T, there was no difference in new sensitizations between treatment and placebo, although this trial was conducted in adults and most were already multisensitized at baseline (3) . The 5 SLIT-D trials were all open-label, with conflicting results on the prevention of sensitizations (79,82,84,85,87) . Since the 2017 SLIT meta-analysis was publis-  (88) . Another open-label, non-randomized trial of HDM SLIT-D demonstrated that after 5 years of treatment, 58% of patients in the control group, 13% in a standard SLIT-D group, and 8% in an adjuvanted (e.g., adjuvant added to improve the antigenic effect) SLIT-D group developed new sensitizations (83) .

Safety/tolerability
SLIT in either tablet or drop form is generally considered a safer treatment than SCIT (7,8) . The most common AEs with SLIT-T and SLIT-D are local allergic reactions such as oral pruritus, throat irritation, ear pruritus, and mouth edema (89)(90)(91) . With SLIT-T, the local allergic reactions have been demonstrated to be mostly mild-tomoderate, last approximately 30 minutes to 1 hour, and typically abate or disappear after 2 weeks of treatment (89) . The duration and recurrence of AEs for SLIT-D are less well characterized.

Systemic allergic reactions
Both SLIT-T and SLIT-D carry the risk of inducing systemic allergic reactions, including anaphylaxis. Surveillance data from a survey in the US found a systemic allergic reaction rate of 1.4% for SLIT-D between the years of 2012 and 2013 (92) . No equivalent data for SLIT-T are yet available, although a review of SLIT-T trials found that epinephrine was administered to 0.2% of subjects for SLIT-T treatment-related events (93) . Rarely, severe systemic allergic reactions can occur, especially after the first dose of administration (92) . The prescribing information for all the SLIT-T state that the first dose should be administered under the supervision of physicians experienced with severe allergic reactions and that the patient should be observed for 30 minutes after the first dose (39)(40)(41)(42) . In the US, the FDA recommends that self-injectable epinephrine be prescribed along with SLIT-T, although this recommendation is controversial because of the low risk of anaphylaxis with SLIT-T in relation to the high cost of autoinjectors (94) . No other country has a similar recommendation.

Asthma
Uncontrolled asthma is a risk factor for life-threatening reactions for SCIT and appears to be a risk factor for severe reactions to SLIT (92) . In early SLIT trials, even mild asthma was an exclusion criteria, but trials in the last decade have included subjects with mild to moderate persistent asthma. Pooled analyses of SLIT-T trials indicate that there is no difference in the safety profile of subjects with controlled mild and moderate persistent asthma compared with no asthma (89,90) . The HDM SLIT-T is also approved as an add-on treatment for asthma in Europe (10) .

Eosinophilic esophagitis
There is a direct relationship between eosinophilic esophagitis (EoE) and food allergy oral immunotherapy (95) . There are also reported cases of EoE associated with SLIT-T and SLIT-D (90,96,97) , but it is unknown if there is a direct relationship between EoE and SLIT for aeroallergens and the mechanism is unclear. Because of differences in contact with the lower esophagus, there may be differences between SLIT-T (immediate absorption in the mouth) and SLIT-D (swallowed) in relation to EoE. A history of EoE is currently a contraindication for SLIT-T.

Cost
Cost is a common reason patients discontinue SLIT (98,99) . Commercial insurance coverage for SLIT-T in the US is approximately 75% and the tablets are covered under Medicare (100) . Manufacturers of SLIT-T offer coupons on their websites to offset the cost to patients and can range from $15 (€14) to $100 (€92) depending on commercial insurance coverage and the pharmacy network (Table 4). In contrast, SLIT-D are not covered by commercial insurance because of lack of FDA approval. The cost to the prescriber for SLIT-D is dependent on the dose used and the specific allergen (Table 4). The out-of-pocket cost to the patient for SLIT-D will vary depending on individual prescriber practices and other factors.

Adherence
Daily treatments, particularly long-term treatments, have notoriously poor adherence rates (101) . As with all daily longterm treatments, adherence to SLIT remains a challenge and many patients do not complete the recommended 3 years of treatment (102,103) . Although adherence to SLIT in general has been evaluated in many studies, adherence to SLIT-T and SLIT-D specifically is not always distinguished and no study has directly compared adherence between the two.

Selection of SLIT Products
The choice of AIT treatments should be based on scientific evi- AIT guidelines for ARC recommend that only AIT treatments with documented efficacy should be prescribed, with the exception of AIT for rare allergens (104) . It is recognized that a "class effect" of SCIT or SLIT treatments for a particular allergen cannot be assumed based on clinical evidence for other individual treatments (105) . To illustrate the current and unsatisfying clinical reality, a suggested algorithm on how to select a SLIT treatment in North America based on an evidence-based approach is shown in Figure 2.  (44) . The trial outcome was positive, but the trial did not lead to FDA approval. A few bulk extracts (e.g., dog, cat, timothy grass, and HDM) have been evaluated in small US SCIT and SLIT-D trials and shown evidence of effect (62,106,107) . However, such evidence has not yet been replicated in Phase 3 trials and have not led to regulatory approval. Most US allergen extracts are not standardized but remain licensed for diagnostic use and have been grandfathered in to be used for SCIT despite the absence of formal studies.

Conclusion
Since

Ethics approval and consent to participate
Not applicable

Consent for publication
Not applicable

Availability of data and material
Data sharing is not applicable to this article as no datasets were generated or analysed during the current study.