>> How compatible are innovations with our health?
security system?
As chairman of the AMNOG arbitration board, I can actually only speak for drug innovations, but you can actually see them pars pro toto for everyone - including systemic and procedural ones.

In what way?
We go into the so-called benefit assessment - it doesn't really matter whether an innovation falls under the AMNOG according to Section 35a SGB V
or new examination or treatment methods
§ 136 assumes that, shortly after being brought onto the market, the positive effects that occurred in the very narrow study population with mostly very high inclusion and exclusion criteria will be evident in the population to be treated. That is the difference between the effectiveness, which is shown in the clinical study setting, and the real patient benefit, the effectiveness, under the relatively open real conditions in the healthcare environment. It doesn't actually matter whether it's surgical procedures, medical devices, medicines or other services to be provided in the healthcare system - this difference will always be there. This is also why innovations, the way we design benefit assessments today, often fail. Of course, exceptions always confirm the rule, especially when it comes to oncological drugs that should not be withheld from those affected, even if they may only provide survival for a few months.

However, the transformation of really hard endpoints can actually only be shown in oncology, and in chronic diseases usually only using surrogates.
There are a large number of surrogate parameters. These include those that are recognized and validated, but also, let’s say, those that need to be discussed. The problem here is that the HTA authorities - above all the G-BA and its evaluating institutions - would have to move on to setting up clear criteria, widely agreed upon by specialist societies, on how to deal with which surrogate parameters, instead of relying on evaluation as is currently the case Rating to hang. Unless different rules are practiced in the future, innovations that only lead to changes in patient-relevant endpoints at a relatively late stage will continue to fail mercilessly. It must also be made clear: not everything that seems to glitter is gold.

What if a true breakthrough innovation came onto the market?
If it is a drug with a clear, significant or perhaps even significant additional benefit, there is actually no problem in the early benefit assessment. This will only become apparent later, when the federal association has to agree on a reimbursement price that can be accepted by both parties with the pharmaceutical company that brings it to the market. As is well known, if this does not happen, the case will end up on the table of the AMNOG arbitration board.

Are we talking about a meta-innovation with the power of a “game changer”?
If you look at the method paper of the IQWiG and the decision-making practice of the G-BA, it would be a stroke of luck if a meta-innovation that starts at a very early, asympotmatic stage of the disease or even earlier, when the disease develops, were to survive . As a rule, the change in an actual, patient-relevant endpoint must be shown. And this even at the early benefit assessment stage with the mostly insufficient data available up to that point; especially when you think about hard endpoints.

Let's assume an innovation that is based on certain genes
marker wants to use a “window of opportunity” to treat a disease so early with a suitable therapeutic response that its progression can be delayed or even prevented from breaking out.
You talk about the vision of Disease Interception, a truly extremely complex system innovation that could come to us in five, ten, fifteen years.

How complex?
The approach must not only be understood as a therapeutic and systemic innovation, but will – if it becomes reality at some point – trigger a serious paradigm shift in our healthcare system, which focuses on the treatment of diseases.

Is that bad?
Not at all. It can even be a great opportunity, because in the face of such a breakthrough innovation it will no longer be possible to avoid a discussion that Germany has been spiraling around for years and that neither society nor politics have really dealt with. Above all, this is the question of all questions, which is: What is a life worth and how much should quality of life and time cost?

The QALY discussion that has been going on for a long time in Anglo-Saxon, but has now become somewhat watered down.
Which is understandable. If you have to transfer reimbursement decisions from the system level to the individual case level, there will always be distortions that you have to get under control somehow. Nevertheless, a related discussion makes sense because, in addition to a maximum reimbursement limit, important ethical questions must be addressed and socially agreed upon. This will change at the latest with the disease interception model.

Does such an innovation even fit into our current system?
Not so right. This is because the approach aims at biomarker-based early intervention, even before the first symptoms appear.
tome occur at all. This clashes with the concept of illness, which is introduced in some paragraphs of SGB V but has not been defined in more detail by the legislature.

While the WHO understands illness as the absence of health and as a disturbance of physical, mental and social well-being...
... the Social Security Code is more about a condition that only becomes an illness when symptoms or disorders are diagnosed and as such becomes worthy of treatment and reimbursement. But that is exactly what is not the case with the disease interception approach, which is based on the drug response to a probable risk of the occurrence of a disease.

Are therapies based on pure probabilities even reimbursable?
This is not an easy question. Initial case law suggests that such interventions can be part of the GKV's catalog of services if one considers very rare exceptions such as the treatment of BRCA1 and BRCA2 mutations. The so-called Angelina Jolie effect can therefore be understood as a gateway to reimburse a medical intervention based on a probabilistic prognosis.

Probability in law?
It is clear: If you want to approach this topic, it will always be a development with many intermediate steps.

Starting with what?
First of all, risk appetite will need to be discussed. This alone will have many scientists and professional societies talking hotly. Which is no wonder, because the question of what level of probability of an outbreak of a disease, which will occur in an unknown number of years, should or even must be intervened is certainly not a trivial one. All the more so if, due to a probability, one would have to start a therapy - perhaps even lifelong - which can also have the potential for harm. The noble principle of “primum nihil nocere”, which comes from the Hippocratic tradition, will be put to the test.

If there is a therapeutic answer.
Of course. Without that you don't even need to start. But if a pharmaceutical manufacturer were to actually find a biomarker and a suitable medication for a disease that has only a one percent probability of occurring and whose occurrence can be prevented by the medication, it would be absolutely insane from an economic point of view to implement such an innovation Powerful
to be included in the GKV's health insurance catalogue.

What would the case look like with 30 %, with 60 or even 90 %?
So far there are no social answers to this. That's why we need to enter into a professional and social discussion at this point. But please do this on an evident basis and also keep the whole thing as open-ended as possible.
What is the core question?
It is as follows: At what probability does a disease risk become manifest and so clinically relevant that it needs to be treated and therefore regularly included in the catalog of services?

If the question is answered negatively...
... the model is dead. But then it's not a shame. Or maybe it is, because such an approach won't get a second chance any time soon. That's why you should be very careful not to burn such visions because you want to enter the market too early and/or with too little evidence.

However, if the question were answered positively...
...society will probably not be able to avoid recognizing disease interception as the prevention of disease before it breaks out through the combined use of diagnostics and therapeutic response. This is required by Section 1 of SGB V, in which health insurance has the task of “maintaining or restoring the health of the insured or improving their state of health”, but also because it is enshrined in the Basic Law for the Federal Republic of Germany that “ Everyone has the right to life and physical integrity.

Nevertheless, the self-administration will certainly have a hard time dealing with such a meta-system innovation if it one day becomes reality.
True. The fundamental reason for this is that there is no clear legal responsibility within SGB V for such a comprehensive system innovation.

What about “companion diagnostics” that have already been reimbursed?
These are diagnostic-drug dualities that only come into play when the disease in question has already broken out or the first diagnosable symptoms have appeared. However, the second part of the pairing, the drug, will always have to be subject to the early benefit assessment according to Section 35, in which an additional benefit is compared to the respective intended purpose.
moderate comparator therapy must be proven in RCTs. It is not enough that different biomarker reactions can be associated with disease probabilities, because it must be demonstrated how the reduction or, in the best case, zeroing of the disease probability caused by medication affects hard and, above all, patient-relevant endpoints.

Because the biomarker is a mere surrogate parameter...
...with which the G-BA, especially the IQWiG, had problems with its validity and still has problems with some of them today.

How can you get out of this dilemma if you want to?
First of all, you have to want it. If you want it and the basic ethical questions raised at the beginning have been answered, you will find ways and means. However, today's problems with the benefit assessment and the reimbursement mechanism suggest that the legislature will probably be required to change the current system if politicians even recognize that it is necessary to act and not react.

Isn't that exactly what the legislature is doing when it comes to the law for more safety in the supply of medicines with Article 12 No. 2
§ 35 should application-related data also be taken into account for orphan drugs?
That may be true and certainly indicates a broader acceptance of register data. However, at the moment the legal situation is such that without clearly proven additional benefits that have been granted by the G-BA, an adequate reimbursement amount can only be achieved in exceptional cases. In the case of orphans, which the amendment to the law is primarily aimed at, a starting price is accepted that enables the manufacturer to remain on the market, but this can change if, through additional studies that the G-BA can impose in the future, long-term benefits are foreseeable - or not. Such studies will probably also be used for disease interception approaches, which will only be able to demonstrate their effectiveness over very long observation periods.

A thought experiment: Let's assume that Disease Interception is introduced into the healthcare system at some point and reimbursed.
This is truly a great vision because you no longer have to go to the doctor. Then diseases will no longer break out, the quality of life will be maintained and secondary diseases and the associated costs can be avoided.

What would be expected from a health economics perspective?
The narrow approach of the early benefit assessment, which only looks at a one-year period and ignores all collateral damage and costs in other social areas, such as work, care and pensions, will be completely overwhelmed. Here you also have to create a lifelong time horizon in order to be able to estimate the consequences of an illness that may not break out for ten or fifteen years, or not at all. Conversely, this means that this question cannot usually be answered using RCT study data because this long time horizon is absolutely unusual in clinical studies and will certainly be difficult to finance. However, an adequate health economic benefit assessment needs even more: namely an evident bridge from the validated surrogate to the end point - which has so far been missing - as well as health economic modeling in which the life time gained must be considered in correlation with the quality of life from the beginning of the intervention to the end of life. But this is not entirely trivial, because such an intervention, should it really be successful, will avoid secondary illnesses and associated costs on the positive outcome side, but on the “negative” side it will result in a longer life, the chance to develop new - Age-related illnesses will also increase.

Is health economics capable of modeling such complexities?
There will be no way around expanding the health economics toolbox that has so far been established in the statutory health insurance system in Germany. This is a good reason to think about whether health economic modeling like QALY can really continue to exist in the shadows. But this is primarily a political decision, which could and perhaps should be driven by society. If you want to give a system innovation like Disease Interception a chance, society, politics and self-administration must be prepared to make a kind of bet on a better future - probably in a kind of pay-per-outcome system. This would mean that - should Disease Interception actually be available and able to be introduced onto the market one day - for moral and ethical reasons alone, one would not have to wait until long-term study data with the harshest possible endpoints are available, because the quality of life and quality of life that is possible by then Lifetime would be wasted forever.

Prof. Wasem, thank you very much for the interview. <<

The interview was conducted by MVF editor-in-chief Peter Stegmaier.

Citation note: