Evidence vs. efficacy in allergen-specific immunotherapy: Considerations using the example of tradable products in Germany

The recently published S2k-guideline on (allergen-) specific immunotherapy (AIT) provides an excellent overview of the evidence on allergen preparations available for AIT in Germany based on the published efficacy studies. Publications based on the guideline are currently being used by the German associations of statutory health insurance physicians and German health insurance funds to open a discussion on the reimbursement status of allergen preparations. In our view, calling the reimbursement status of perscribable and tradable AIT preparations into question on the basis of an assessment of the current body of evidence in the guideline is to be rigidly opposed. In Germany the Paul Ehrlich Institute (PEI) is the only authority empowered to decide on the marketability of AIT preparations, and decisions on the reimbursement status of AIT products need to be based on cost-benefit analyses and not solely on an evaluation of the evidence. The present article aims to examine the relationship between the evidence, efficacy, tradability, and reimbursability of AIT preparations.


Evidence and e cacy
(Allergen-)Speci c immunotherapy (AIT) is currently the only treatment available that is capable of inducing speci c tolerance to individual allergens, particularly in allergies to inhalant allergens [1]. In principle, due to its disease-modifying e ects that may manifest as asthma prevention and the prevention of new sensitizations, AIT is superior to purely symptomatic treatment [2,3]. However, studies have proven these e ects for only a scant number of products. A variety of preparations are available for subcutaneous and sublingual use, whereby native or chemically modi ed allergens (allergoids) can be used.
e practitioner needs to establish which product should be used, taking into consideration e cacy, safety, time expenditure for the patient, as wells as costs. e revised S2k-guideline [4] published at the end of 2014 is based on the current state of knowledge and provides physicians with excellent support when deciding for or against AIT, as well as in the selection of allergen preparations. One of the guideline's core statements is a call to evaluate each individual preparation on the basis of the study results available for that preparation, irrespective of its mode of administration. In the opinion of the authors, approved allergen preparations with documented e cacy and safety, or preparations mar-Abstract e recently published S2k-guideline on (allergen-) speci c immunotherapy (AIT) provides an excellent overview of the evidence on allergen preparations available for AIT in Germany based on the published e cacy studies. Publications based on the guideline are currently being used by the German associations of statutory health insurance physicians and German health insurance funds to open a discussion on the reimbursement status of allergen preparations. In our view, calling the reimbursement status of perscribable and tradable AIT preparations into question on the basis of an assessment of the current body of evidence in the guideline is to be rigidly opposed. In Germany the Paul Ehrlich Institute (PEI) is the only authority empowered to decide on the marketability of AIT preparations, and decisions on the reimbursement status of AIT products need to be based on costbene t analyses and not solely on an evaluation of the evidence. e present article aims to examine the relationship between the evidence, e cacy, tradability, and reimbursability of AIT preparations.
ketable under the erapy Allergen Ordinance (TAO) for which e cacy and safety have already been documented in clinical trials meeting World Allergy Organization (WAO) or European Medicines Agency (EMA) standards, "should be preferentially used" [4].
AIT products are essentially challenging to compare due to their heterogeneous composition. To date, the characterization and standardization of allergen extracts has been performed according to manufacturer-speci c standards and, even when major allergen concentrations are given, these do not permit conclusions to be drawn on possible e cacy, since methods for the analysis of major allergens are not consistent, and most allergen extracts contain several major allergens. is has lead to the indispensable requirement (as with other drugs) for reliable studies to analyze each individual AIT preparation for e cacy and safety. is presents the prescribing physician with the dilemma of needing to be familiar with, and capable of evaluating, the study data.
By way of support here, the guideline refers readers to tables listing allergen preparations that fall under the German TOA. ese tables are not an integral part of the guideline. ey provide information on the following: -Current marketing authorization status of individual products -e number of full publications evidencing the e cacy of each individual preparation listed -−An overview of trials that have been conducted with o cial approval (according to the EMA register www.clinicaltrialsregister.eu [www.dgaki. de/leitlinien/s2k-leitlinie-sit/clinical-trials-register-eu-nachzulassungen und www.dgaki.de/ leitlinien/s2k-leitlinie-sit/clinical-trials-register-eu-neuzulassungen]) e tables are updated bi-annually and brought in line with the current status. Only the results of clinical trials that ful ll conditions de ned on the basis of WAO criteria are included in the list (www. dgaki.de/leitlinien/s2k-leitlinie-sit/sit-produkte-studien-zulassung) [5]. e conditions to be ful lled for the guideline are the following: 1. Standardized allergen extract with dose information, 2. randomized, double-blind, placebo-controlled study design, 3. information relating to a combined symptom-medication score and/or the two individual scores, 4. information relating to statistical analysis and statistically signi cant results, 5. a threshold in e cacy of 20 % above placebo. e rst point represents a minimum requirement in terms of product characterization stipulated by the Paul Ehrlich Institute for the marketing authorization of an AIT product. Requirements 2-4 emphasize the evidence documenting that the product is e ective. e term evidence (see de nition in Tab. 1) stems from the Latin "evidentia" and is the colloquial term for "obviousness" meaning that something does not need to be questioned [6]. us, evidence does not assess the magnitude of e cacy, but concerns rather the reliability with which something is proven to be e ective. e h criterion speci es a minimum level of e cacy. is criterion is not relevant to the marketing authorization of an AIT product at the Paul Ehrlich Institute. e requirement is based on the concept that speci c immunotherapy that is far more costly than pharmacotherapy needs to demonstrate at least an e cacy comparable to that of (here in the sense of antihistamine and/or topical corticosteroids treatment). However, it must be borne in mind that clinical studies to assess the e ect of speci c immunotherapy or of pharmacotherapy are designed di erently, and that the possible disease-modifying e ect of AIT is not taken into account in this comparison. According to published studies, a 20 % improvement in the symptom and medication score (generally measured in the rst year following AIT initiation) results in a relevant improvement in quality of life. When considering the various AIT studies, one should bear in mind that the resulting strength of treatment can be a ected by a number of factors that, to a certain extent, cannot be in uenced and which are independent of the investigated preparation: -Symptom severity in the study population -Extent of sensitization among subjects -Degree of pollination during the study season -De nition of the endpoint -Statistical analysis A further crucial factor is whether the intent-totreat (ITT) or the per-protocol (PP) population (see de nitions in Tab. 1) is analyzed. While e ect sizes may be identical in the PP and ITT, they can also di er widely.

Abbreviations
A harmonization of studies is sought on both an international and a national level. However, the efcacy of di erent allergen products can only be compared if di erent products are directly used in the same study. At present, it appears highly unlikely that head-to-head studies of this kind will be conducted in the foreseeable future.
Evidence assesses the reliability with which an intervention is shown to be e ective and does not permit conclusions to be drawn on the size of the intervention's e ect. us, a lack of evidence is not tan-tamount to a lack of e cacy, and the intervention with the highest level of evidence is not necessarily the most e ective. However, a basic requirement should be that e cacy is shown in controlled studies for all preparations containing common allergens currently on the market (see below). e TAO denes the trial requirements that need to be ful lled for marketing authorization.

Marketing authorization
AIT preparations that were granted marketing authorization by the relevant authorities before the TAO came into force [8] have proven their e cacy in studies that satis ed the marketing authorization requirements at that time. According to the TAO, these preparations are not required to be investigated in new studies ful lling current standards; they remain permanently tradable. Although this fact may appear worthy of discussion, it is understandable given that marketing authorization of drugs cannot be withdrawn simply because the requirements made of trials to establish e cacy have changed. Since the trials conducted in the past partially failed to satisfy WAO requirements for inclusion in the guideline tables, the tables contain older authorized products in particular, for which no e cacy trials are available (since the published trials do not ful ll current criteria). However, this lack of data does not imply that no trials exist or even that these preparations are necessarily ine ective. More importantly, this is in no way linked to the assertion that the marketing authorization of these preparations is questionable and that their prescription represents grounds for recourse. us, the guideline does not release allergologists from their duty to assess trial data on the products they use on an individual basis. are virtually impracticable due to the number of patients required. Nevertheless, patients with allergies to rare allergens are equally entitled to cause-oriented treatment using AIT. Preparations produced and marketed according to § 21 paragraph 2, 1b of the Medicinal Products Act (Arzneimittelgesetz, AMG) of the Federal Republic of Germany continue to be available to these patients as individual formulations for AIT. e 14th amendment of the AMG of 2005 regulates the exemption of therapy allergens produced for individual patients from licensing requirements. Speci c immunotherapeutic agents produced on the basis of a prescription formulation are explicitly exempted from the obligation to obtain marketing authorization in § 21 paragraph 2, 1b of the AMG: "A marketing authorisation (Zulassung) shall not be required for medicinal products which ... are prepared on prescription for individual persons." Here again, no grounds for recourse are provided for.

Reimbursement status
ere has been a discussion in recent months on possible restrictions on the reimbursment of allergen extracts used for AIT. One publication in particular [9], which virtually compiles a positive list of products for AIT, forms the basis for this discussion. e publication categorizes allergen products into a pyramid according to " rst choice" to "last choice." While the current tables in the guideline appendix have our full endorsement, we see a real hazard in terms of interpretation posed by the speci c classi cation of products -conveyed as the personal recommendation of the publication's author by virtue of the pyramid structure -into "recommended" and "not recommended", which does not lie in the intention of the guideline. In our view, representing products in such a way bears the risk that healthcare actors could indirectly instrumentalize the guideline as a means to decide on the reimbursability of preparations. e guideline, as well as the tables in the appendix explicitly state that the table is "not suited as a positive or negative list for decision-making on prescribability or reimbursability." It is our opinion that deriving a positive list based on the guideline tables is also particularly unjustied since the issue does not relate to the actual ecacy of AIT products, but rather to merely presenting evidence according to criteria based on WAO recommendations. It is currently not possible to conduct an accurate, scienti cally founded comparative assessment of the e cacy of the various AIT products, meaning that the prescriber of these products must not be limited in his therapeutic freedom via the loophole of decisions on reimbursability. However, for patients' well-being, any prescription of an AIT preparation should be based on e cacy data, and it is the responsibility of the prescribing physician to establish whether studies have proven the e cacy of the preparation that he/she wishes to prescribe.

Cost-bene t analysis
Whereas the Paul Ehrlich Institute decides on the marketability of drugs, guidelines aimed at ensuring medical care -as compiled by the Joint Federal Committe according to § 92 of the German Social Code (Sozialgesetzbuch, SGB) V -should enable the physician to make a cost-e ective and appropriate selection of drug therapy. It has been speci ed that information shall be included that permits the physician to make an appropriate treatment and drug selection that is also based on treatment bene t in relation to treatment costs, thereby serving the cost e ectiveness of the prescription. According to § 35b SGB V, the Joint Committee tasks the German Institute for Quality and E ciency in Health Care (Institut für Qualität und Wirtscha lichkeit im Gesundheitswesen, IQWIG) with drawing up a cost-bene t analysis in which both the bene t to the patient (disease improvement, reduced disease duration, increased life expectancy, reduced side e ects, improved quality of life) and the economic analysis (appropriateness and reasonableness of requesting the insured community to meets the costs) need to be taken into consideration. us, a well-founded selection of reimbursable or non-reimbursable products must be made on the basis of cost-bene t analyses. A valid, preparation-speci c calculation of the bene t conferred by AIT is currently not possible since, in addition to reduced medication use and improved ability to work, as well as the long-term e ects in terms of asthma reduction and the prevention of new sensitizations, treatment with each individual preparation needs to be assessable.
Calculating the cost of AIT is also not unproblematic. Due to varying treatment protocols, calculating de ned daily doses (DDD) is not helpful and is considered obsolete today. Calculating costs over a de ned period of time (the costs of treatment over a 3-year period are o en used) is favored. However, it should be borne in mind here that the standard period for subcutaneous immunotherapy (SCIT) is currently 3 years, while a period of 3-5 years is increasingly favored for preparations used in the context of sublingual immunotherapy (SLIT). ere is a discussion as to whether a single treatment cycle of several weeks for peptide therapy may be adequate. erefore, rightfully speaking, the costs of a "complete treatment cycle" need to be assumed as a basis for calculating costs.
Alternatively, comparative models in which the nancial expense of achieving a prede ned outcome Conclusion e guideline provides an excellent overview of the AIT preparations used in Germany and, with its tables, refers the reader to a current overview of the marketing approval status of these products, the published study results that ful ll today's criteria (according to WAO), as well as the o cial approval to conduct AIT trials. Using the guideline to deduce the reimbursability of allergen products is to be rigorously rejected. Reimbursability needs to be based on cost-bene t calculations, which are currently not practicable in a preparation-speci c manner. e fact that it was not the intention of the guideline authors to in uence the reimbursability of AIT products is clear from the wording in the tables: " e tables are not suitable as a basis for decision-making on prescribability or reimbursability in the sense of a positive or negative list."