The new “guardrails” in the AMNOG process

The AMNOG procedure has been legally adjusted several times in recent years. A significant change recently occurred with the GKV Financial Stabilization Act (GKV-FinStG), which was passed at the end of 2022. The legal package of measures includes, among other things, new requirements for reimbursement amount negotiations, which change the basic principle of AMNOG. According to these new requirements (so-called “guardrails”), in certain cases a new drug may not cost more than the comparative therapy despite the proven additional benefit. If there is no proven additional benefit, the costs must even be lower. The details of these new requirements are subjected to a critical analysis below.

>> The AMNOG requirements previously limited the contracting parties' scope for negotiation if the drug could not demonstrate any additional benefit. Section 130b SGB V stipulated that for new drugs without proven additional benefit, a reimbursement amount should be agreed that does not lead to higher annual therapy costs than the appropriate comparative therapy (zVT) or the most economical alternative in the case of several zVT alternatives. If there was an additional benefit, correspondingly higher prices could be agreed (principle of so-called “value-based pricing”).
The newly implemented regulations, euphemistically referred to as “guardrails” in the legal materials, come into play at this point and tighten the negotiation requirements in Section 130b SGB V in a complex way. Their validity depends on two factors: Firstly, on the type of appropriate comparative therapy – the guidelines apply in cases in which the G-BA, as zVT, has determined a medicinal product with an active ingredient for which patent or document protection exists. Secondly, the extent of the additional benefit - they apply if the additional benefit has been classified as unproven, low or unquantifiable. Combining both factors, three negotiation requirements are now essentially set out in Section 130b SGB V:
a) If an additional benefit is not proven and the zVT is a medicinal product with an active ingredient for which patent and document protection has been lost, a reimbursement amount should be agreed that does not lead to higher annual therapy costs than the zVT.
b) If an additional benefit is not proven and the zVT is a medicinal product with an active ingredient for which patent or document protection exists, a reimbursement amount must be agreed that leads to annual therapy costs that are at least 10 percent below those of zVT.
c) If an additional benefit cannot be quantified or is classified as low and the zVT is a medicinal product with an active ingredient for which patent or document protection exists, a reimbursement amount must be agreed that does not lead to higher annual therapy costs than the zVT.

In addition, if several zVT alternatives have been determined, the application of the above specifications must be based on the zVT that represents the most economical alternative after the annual therapy costs. If a medicinal product with a patent-protected active ingredient that is not subject to the benefit assessment according to Section 35a was determined as zVT, a fictitious discount of 15 percent must be applied to the costs of this medicinal product.
While the content of the requirement mentioned under (a) corresponds to the status before the GKV-FinStG, the requirements under (b) and (c) represent a significant tightening. In the case of an unproven additional benefit, a target specification now applies instead of a limitation Minimum discount of 10 percent, which acts as a hard price cap. In the case of a non-quantifiable or small additional benefit, instead of the possibility of negotiating a higher reimbursement amount, a hard cap at the level of the zVT now applies (see Table 1).
Initial estimates assume that with these pricing specifications, no restrictions now apply to only one in five subpopulations assessed and that a reimbursement amount above the zVT is possible. Around 3/4 of the new active ingredients would be affected by a restriction due to the guardrails in an initial assessment (9). Overall, with the GKV-FinStG, the retroactive effect of the reimbursement amounts applies from the 7th month instead of the 13th month after the start of the procedure.
Justification of the legislator
In the justification for the cabinet draft of the GKV-FinStG, the new guidelines for non-quantifiable or low additional benefits are described as necessary and appropriate to ensure the financial stability of the GKV, since “the price level of patent-protected drugs is already in a range that is sufficient for research-based drug manufacturers, the additional Surcharges are not justified if the additional benefit is not quantifiable or if the additional benefit is small.” In addition, the incentive is set to “place a greater focus on patient and care-relevant questions for the benefit assessment in order to be able to achieve a considerable or significant additional benefit in the benefit assessment”.
With regard to the minimum discount for an unproven additional benefit, the justification states that this change became necessary “as practice has shown that the opening of the upper cost limit introduced by the Act to Strengthen Drug Supply in the Statutory Health Insurance Fund (AMVSG). For medicines without additional benefit, in justified individual cases, the agreed reimbursement amounts were regularly higher than the most economical alternative to the appropriate comparative therapy. In this way, the GKV's expenditure on medicines without additional benefits was able to rise to a level that was not adequate, from which subsequent active ingredients in particular were able to benefit through a correspondingly higher price level.
The new pricing specifications and the legal justifications listed above have become the subject of critical discussions. The points of criticism are explained in detail below.
Significance of an unproven, minor or unquantifiable additional benefit
First, it must be questioned whether the systematic devaluation of an unquantifiable or minor additional benefit of a new drug is appropriate. The first thing that needs to be pointed out here is what lies behind a small or unquantifiable additional benefit in accordance with the legal requirements. According to the Drug Benefit Assessment Ordinance (AM-NutzenV), there is a slight additional benefit “if a moderate and not just a slight improvement in the therapy-relevant benefit that has not yet been achieved compared to the appropriate comparator therapy [...] is achieved, in particular a reduction in non-serious symptoms the disease or a relevant avoidance of side effects” (see Table 2).
This makes it clear that the small extent already includes a minimum threshold for patient-relevant therapy effects, so that negligible or minor effects cannot be assumed. In the G-BA's assessment practice, this also includes assessments with an extension of overall survival and prevention of recurrences in oncological diseases or more frequent complete remissions of severe plaque psoriasis in children and adolescents (3, 4, 5).
According to AM-NutzenV, if the additional benefit is not quantifiable, there is an additional benefit, but “it cannot be quantified because the scientific data does not allow this” (see Table 2).
In the G-BA's assessment practice, this was always associated with an additional benefit, which can in principle include all positive extent categories (1). A non-quantifiable additional benefit is therefore at least small and at the same time includes the possibility of a considerable or significant additional benefit with significant improvements in the course of therapy up to healing. Non-quantifiable additional benefits also include: To find drug reviews that significantly increase overall survival or cure chronic hepatitis C in pediatric patients (6,7).
Overall, the blanket statement that there is “only” a small or non-quantifiable additional benefit that does not require any reward must be clearly questioned. Rather, the examples mentioned make it clear that such medicines can have a high priority in care, but that this can no longer be adequately rewarded as a rule with the new regulations imposed by the GKV-FinStG.
Regarding the unproven additional benefit, it should be noted that in most cases this determination was not made on the basis of a weighing up of positive and negative effects, but because the available evidence from approval studies did not meet the requirements of the G-BA in the context of its specific question and therefore did not was included in the benefit assessment (8). Furthermore, every “equally good” medicine expands the available therapy options in everyday care. A compulsory punishment for this situation is therefore incomprehensible.
The G-BA's assessment practice also shows that drugs with unproven, minor or unquantifiable additional benefits are also highly relevant in care. If AMNOG-assessed drugs themselves are named as new ZVT in an area of application, around 2/3 of them have previously “only” shown an unproven, minor or unquantifiable additional benefit. These zVT designations cover a whole range of indications such as cancer, skin, musculoskeletal or metabolic diseases. The relevance of such medicines is also reflected in the guidelines of the medical societies, which have a permanent place there. At the same time, even if there is no proven additional benefit, the G-BA repeatedly points out in the underlying reasons for the decisions that such drugs can be a relevant therapy option in individual cases. This is particularly the case if there is a great need for therapy in care or if the new drug is even the first approved therapy option in this indication. Against this background, too, a blanket marginalization of the importance of these medicines through the “guardrails” seems incomprehensible.
Ultimately, the new requirements must be viewed critically in the context of the innovation landscape of the last few decades. They essentially penalize companies for taking the risk of investing in developing improved therapeutic approaches across a variety of diseases across different lines of therapy. The so-called step innovations are the norm in drug research and development. Step innovations are much rarer. The new guidelines reduce incentives to look for patient-relevant improvements even after a drug has been approved for the first time and to show these in the benefit assessment compared to the standard therapy.
Equality of unequal people
facts
The new regulations of the GKV-FinStG also lead to regular unequal treatment in the AMNOG system. Essentially the same facts, i.e. medicines that are comparable in terms of benefit assessment, are treated unequally when determining prices. If e.g. For example, if a second representative of a class of active ingredients is comparable to the first representative, which has now become the therapeutic standard, the second representative is systematically disadvantaged by the minimum discount regulation. At the same time, significantly different situations, i.e. drugs with proven additional benefits compared to their inferior appropriate comparator therapy, are treated equally. In addition, the use of guidelines equates the conduct of comparative studies with the absence of such evidence. Such a blanket penalty regulation therefore appears to be systematically extremely problematic.
Shifting the
Negotiating positions
The draft law includes the introduction of the guardrails, among other things. with the aim of strengthening the position of the National Association of Statutory Health Insurance Funds (GKV-SV). This justification is incomprehensible since the GKV-SV is in a quasi-monopoly situation anyway. He also plays a key role in the entire AMNOG process. As a key bank in the G-BA, it is directly involved in the benefit assessment process and decides in the consultations with the pharmaceutical companies and in the subsequent resolutions about the exact wording of the zVT and the extent of the additional benefit - and thus exactly about the levers for this possible application of the guard rails in the subsequent reimbursement amount negotiation. Intuitively, this suggests a governance problem. Empirically, this is also reflected in the dissenting voting results in the G-BA plenum. In such cases, since 2011, the GKV-SV has always voted for a worse assessment of the respective drug compared to the other G-BA banks. The GKV-SV is also the only party to have a complete overview of all price negotiations that have taken place and are ongoing, which means a clear negotiating advantage.
Reduction of flexibility
in negotiations
The guardrails significantly limit the contractual parties' scope for negotiating reimbursement amounts. The justification that the creation of a target regulation resulted in the agreed reimbursement amounts regularly being above the most economical alternative of the zVT must be questioned. On the one hand, there is no empirical evidence that such an excess occurred not only in individual cases but on a regular basis. On the other hand, a situational application would indicate the existence of a justified individual case. The target regulation was specifically introduced in 2017 with the Drug Supply Strengthening Act (AMVSG) for such individual cases in order to agree, under special circumstances, a reimbursement amount above the most economical comparison therapy, even for drugs without an established additional benefit. This was intended to make narrow, rigid specifications more flexible in order to ensure more room for negotiation for an appropriate price. Such special circumstances may also exist in the case of a patent-protected comparative therapy. However, the guard rails now limit the possibilities for the negotiating partners or the arbitration board to take these issues into account.
effects on certain
Therapy situations
Chronic diseases
The blanket guardrail regulations also show their limitations with regard to certain therapies. So is e.g. For example, in the case of chronic diseases, it should be noted that AMNOG has so far been unable to demonstrate any significant or even significant additional benefit (e.g. for drugs to treat type 2 diabetes mellitus or for mental illnesses). If, according to the justification for the law, a greater incentive is to be directed towards questions other than those previously addressed, the guardrails here could actually have a counterproductive effect, since the threshold of at least considerable additional benefit for such diseases is also determined by the specifics of the therapies and the AMNOG Targets have rarely been achieved in recent years. The incentives would be even lower, especially for the development and introduction of drugs in such indication areas, and market access would be even more difficult (10).

Advanced therapy medicinal products
The justification for the cabinet draft of the GKV-FinStG also states that the price level of patent-protected medicines is in an adequate range for the
new medicines. These should therefore be able to achieve a comparable price. This general statement also needs to be questioned. Patent-protected medicines can also have significantly different price levels. Especially with drugs for advanced therapies (Advanced Therapy Medicinal Products, ATMPs), constellations are conceivable where an additional benefit can only be classified as unquantifiable, but the price level of the previous therapy standard is not sufficient. In these cases, the hard price cap does not allow any flexibility to take the special circumstances of the individual case into account.

Add-on combinations
Other constellations also appear not to have been thought through in the guardrail system. An example is drug B, which is used as an add-on to A in combination therapy. Despite survival advantages, the G-BA classifies the additional benefit for B as only low or unquantifiable. According to the guidelines, the drug should not lead to higher costs overall than the annual therapy costs of basic therapy A. This would mean that the new therapy may cost a maximum of 0 euros despite the additional benefit. If there is no proven additional benefit, the reimbursement amount would have to be less than EUR 0 even for at least 10%. Taking into account further regulatory measures, such as the combination discount, this is likely to lead to the demand for even lower prices (see Fig. 1).
Initial effects in assessment practice and care

Estimates of the effects are only possible to a limited extent in the short term, as a meaningful number of cases within the scope of the new guidelines must first be reached. Nevertheless, the first undesirable developments began to appear just a few months after the law came into force. The new set of rules obviously also applies where it was probably not intended at all - for example in innovations with significant increases in overall survival, which are already establishing themselves as the new standard of therapy. This became clear early on using the example of the active ingredient trastuzumab deruxtecan for the treatment of patients with advanced breast cancer. Overall, the drug shows very clear advantages in terms of overall survival and serious side effects in second-line treatment - a clear case of considerable additional benefit in G-BA practice to date. However, unexpectedly, the G-BA classified the additional benefit for this area of application as non-quantifiable (6).
Due to this classification, the second-line drug falls under the new, rigid price regulations of the GKV-FinStG. The G-BA decision is justified in particular by the formal uncertainty with regard to quantification, since a subgroup analysis of the approval study shows survival advantages (of considerable magnitude according to IQWiG) for the larger group of people under 65 years of age, while for people who are 65 years or older, the effect is not significant. According to the DGHO (German Society for Hematology and Medical Oncology), however, there is no biological rationale for the subgroup findings. Based on the study results in Germany, the drug even replaces the previous standard of care in this indication (11). According to the current assessment by the European specialist society ESMO (European Society For Medical Oncology), the degree of innovation is 4 out of 5 possible points on the ESMO-MCBS (Magnitude of Clinical Benefit Scale) (12). In the ESMO-MCBS system, this corresponds to a “substantial magnitude of clinical benefit”, which overall calls into question the G-BA’s classification as non-quantifiable and the scope of application of the guidelines here.
The validity of the guardrails in such cases is questionable from a regulatory perspective, as innovations relevant to healthcare provision are not rewarded for additional benefits. An unquantifiable extent of the additional benefit in the system of new guard rails has significantly more restrictive effects on the subsequent negotiations. The question also arises as to what level of methodological uncertainty it is possible to quantify the extent (as at least considerable) and the guard rails should not be effective. This also increases the question of when residual methodological uncertainties are not reflected in the probability of additional benefit (e.g. downgrading from indication to indication) but instead or additionally in the G-BA's classification of the extent of additional benefit.
Eptinezumab for use in migraine prophylaxis shows, for example, that the guard rails lead to unequal treatment and inconsistency. The drugs in this drug class that were previously assessed by AMNOG were each compared to Best Supportive Care (BSC) as zVT and were each able to demonstrate considerable additional benefit. In the case of eptinezumab, however, the specification of zVT was changed in the consultation to include drugs that had already been evaluated. Proof of additional benefit compared to BSC was therefore no longer possible here. As a result, there is even a requirement for a minimum discount of 10 percent. In this situation, the guardrails exacerbate inconsistent assessment. In this case, the question also arises as to what significance the requirement under Section 6 Paragraph 3 of the AM-NutzenV has in the G-BA, according to which the same appropriate comparative therapy must be used for drugs of the same active ingredient class in order to ensure a uniform assessment. What makes matters worse is that the G-BA's decision on the need for a change at a certain point in time does not appear to be sufficiently transparent (14). The example also illustrates that changing a ZVT can also develop into a strategic lever for the application of the guardrails.
Another example that should be pointed out is the lack of introduction of the fixed combination of nivolumab and relatlimab for the first-line treatment of advanced melanomas in adults and adolescents aged 12 and over as a result of the new regulation. Although dual immunotherapy shows clear advantages in improving progression-free survival (PFS), in this country PFS is not recognized as a patient-relevant endpoint or as an additional benefit. The current assessment by the European specialist society gives this combination therapy 3 out of 5 points on the ESMO-MCBS (13). Together with the new guidelines, the added value of this therapeutic approach can no longer be reflected, according to the company concerned, so that the drug will not be placed on the market for the time being.
Further direct consequences for supply cannot be estimated at the moment, as the guardrails will have their effect primarily in the medium and long term. In addition to recording unintended use cases, other effects on the availability and speed of market access in Germany must also be further monitored.
conclusion
It should be noted that the new “guardrails” have restricted the basic principle of AMNOG (value-based pricing with higher prices with higher additional benefits) in a number of cases. The resulting blanket approach to the importance of unproven, minor or unquantifiable additional benefits appears questionable. A further shift in negotiating positions in the case of existing governance problems as well as restricting flexibility in negotiations also appear to be debatable. The consequences for certain therapeutic situations such as chronic diseases, drugs for advanced therapies or combination therapies appear to be partly unthought through. In terms of regulatory policy, a counterproductive effect of the guard rails also appears possible, since research and development for certain therapies does not generally result in at least significant additional benefit according to current AMNOG practice. The lack of remuneration for the research risk in step-innovation as the backbone of drug research also appears to be debatable. Even if meaningful assessments of the effects in practice are still hardly possible, initial examples of controversial applications of the guardrails and effects on the availability of medicines are emerging. The upcoming evaluation of this new set of regulations should therefore take the existing problems into account and lead to a revision. <<

from:
Dr. PH Andrej Rasch
Senior Manager Benefit Assessment/
HTA coordination, Association of Research-Based Drug Manufacturers (e. V.)
ORCID: 0000-0003-4098-4502

background
Since 2011, medicines with new active ingredients have been subject to price regulation, which is linked to the AMNOG (Medicine Market Reorganization
law) was introduced. The drugs are first assessed by the G-BA with regard to their additional benefit compared to the current therapeutic standard (so-called appropriate comparator therapy). On the basis of this additional benefit assessment, a reimbursement amount is then negotiated between the pharmaceutical company concerned and the National Association of Statutory Health Insurance Funds. The core principle of this regulation has so far been that prices follow the additional benefit. It should be possible to negotiate a reimbursement amount above the costs of the appropriate comparative therapy if the new drug was able to demonstrate an additional benefit. However, if there is no proven additional benefit, the costs of the comparator therapy should not be exceeded.
Since 2011, around 400 medicines have been subjected to such a process in over 850 procedures. In around 9 out of 10 cases, the National Association of Statutory Health Insurance Funds and pharmaceutical manufacturers were able to agree on a reimbursement amount. If certain issues remained controversial, the AMNOG arbitration board was involved in order to determine a reimbursement amount to balance the interests. The framework for benefit-based pricing was therefore basically in place for all sides. Thanks to the positive incentives, Germany has so far been a leader in Europe in terms of the speed of market launch and availability of innovative medicines.