>> “The view is on the one hand very individual, on the other hand systematic and systemic,” said DKG General Secretary Dr. Johannes Bruns, together with the DGHO and the network against colon cancer, patron of the specialist symposium, which was, as always, well attended, should be considered. Even long after the introduction of the AMNOG, the topic of endpoints is a discussion of different perspectives, although it is still important to bring the different worlds together. However, since this would not be possible within the next five years, he sybellineically predicted that a tenth event would probably be necessary.
The effects of different perspectives - on the one hand the individual, on the other the systemic - can be seen in many places. On the one hand, there is evidence-based medicine itself, where from the perspective of the clinical researcher or the researching clinician - according to Prof. Dr. Michael Hallek, Chairman of the Board, German Society for Hematology and Medical Oncology eV and Director of Clinic I for Internal Medicine at the University Hospital in Cologne - care and research will merge more and more. In order to visualize this thesis, he showed two pictures at the beginning of his lecture: a large truck weighing station, next to it a small, fine gold scale: an analogy that - as he later stated - PD Dr. Stefan Lange, deputy director of the IQWiG institute, would not have fully understood.
It's actually quite simple: What Hallek wanted to express with this is the difference between what can be measured with large RCTs with high exclusion criteria and what cannot be evaluated with this instrument. This is primarily what he calls “mechanism-based evidence”. Hallek: “If you have a clear mechanism in front of you that you understand, you may have to make do with gold scales.”
He makes this clear using the example of lung cancer, which is becoming increasingly genetically fragmented. Even though he certainly doesn't want to deny that randomized controlled trials (if they are done correctly, see page 20) are still important and remain the gold standard of evidence, RCTs are simply not possible for certain mutations with clear mechanisms. This is, among other things, for ethical reasons alone: For example, with the so-called ROS1 mutation, there is an impressive response to the active ingredient critozinib (“Xalkori”) within a few weeks. Where the remaining lifespan would normally be nine months, perhaps a year, lung cancer patients treated with this active ingredient lived up to eight years after the initial diagnosis, in some cases completely without any symptoms.
But now, in its assessment in June 2016, the G-BA found an indication of considerable additional benefit in the treatment of advanced, non-small cell lung cancer in the main reasons for the decision on “Xalkori” – which was brought onto the market by Pfizer in January 2016 awarded, but not in the subsequent assessment in March 2017 for the extension of the indication to ROS1-positive lung cancer: Here, an additional benefit compared to the appropriate comparator therapy was considered “not proven”.
The problem is for the G-BA, which in turn is based on an identical assessment by IQWiG: “There are no randomized controlled studies with patients with ROS1-positive advanced NSCLC in a direct comparison of crizotinib versus the appropriate comparator therapy.”
The fact that this is not possible for ethical reasons because, according to Hallek, the comparative therapy would be chemotherapy is negated. Hallek: “You can never measure such a clear situation with the technology of randomized trials, and yet people live longer just because they get the drug.” But since all assessments in the AMNOG process require RCT, “nothing would other than trying to put the gold dust on the truck scales and then being surprised that you can't measure anything."
However, this approach has serious implications for care, explained Professor Dr. med. Michael Untch, chief physician at the Clinic for Gynecology and Obstetrics and head of the interdisciplinary breast center at the HELIOS Klinikum Berlin-Buch, in the subsequent panel discussion. It was only in September of this year that he admitted a patient who had already received every chemotherapy option on the market. After a test, a ROS1 mutation was detected, whereupon a corresponding treatment application was submitted to the health insurance company, but the response from the MDK took four weeks: “Critozimib” was approved for the treatment of lung cancer, but not for that of the ROS1 mutation. Be it so
Untch’s words – because of the “verdict of the G-BA”
not possible to give the medication.
Untsch's accusation: "The young woman, a mother of four children, has almost been sentenced to death, although she would have the opportunity to live longer if she could take this medication for three or six months on a trial basis." Of course that is Evidence from RCT “the big scale”. But Untch advocates that “we also need the small scales,” because – with a view to the IQWiG represented by Lange – “they will not be able to incorporate this gold dust, this rarity, into a large randomized study.”
This is exactly where another world comes together: that of individual and system considerations. "We can't talk about individual cases, that's not possible at the system level," said Lange, but added: "That sounds technocratic and harsh, and I'm terribly sorry." But that's the applicable principle, which IQWiG must adhere to; and that is also well founded. Lange: “We have to prepare our knowledge in a structured and transparent manner and draw insights from it.”
This not only applies to medical evidence, but also to other parameters, such as PRO1. To do this, Dr. Ethan Basch from the University of North Carolina2 published in “JAMA” in 2017 that where PROs have been raised – and where appropriate has been responded to – patients live longer. According to Lange, one could even say: “PRO save lives”; which is remarkable because this would almost be a breakthrough designation for the FDA due to the comparatively low costs.
However, things are different when it comes to progression-free survival (PFS3). For Lange, the actual core question behind PFS is: “Does the delay in the disease process lead to relief for the patient?” And further: Does an effect on the surrogate PFS say anything about the relief for the patient? One can try to substantiate these questions with data and must do so.
The problem: A current meta-analysis
With almost 14,000 patients from 38 RCTs that investigated the connection between the effect on PFS and the effect on quality of life, Lange would have come to the “catastrophic conclusion” that there was no effect at all. “What is often rumored, that a treatment effect on the surrogate PFS has something to do with influencing the quality of life, is simply not true,” concludes the deputy head of IQWiG from this meta-analysis.
Eva Schumacher-Wulf, editor-in-chief of the breast cancer magazine “Mamma Mia!”, published by Springer Medicine, doesn’t want to leave that alone in the panel discussion. She speaks to Lange directly and says: “Just as you have been working at IQWiG for 14 years, I have been seeing patients for 14 years.” During this time, she has never met a patient who said, “Oh, that's not the case “It’s a shame if the tumor continues to grow.” As a rule, a change in therapy is then necessary, almost always accompanied by a loss of quality of life (QoL). For them, the denial that the progression of the disease has an impact on the quality of life is completely incomprehensible; You should also first take a close look at what is asked about in the QOL surveys. Schumacher-Wulf: “It doesn’t take much science to imagine that a progressive cancer patient is one step closer to death.” She “wouldn’t know where to imagine that this wouldn’t have an influence on QoL.” “I want to get this fantasy.”
However, the authors of the article4, which only appeared in “JAMA” in August of this year, see it completely differently: “We failed to find a significant association between (effect on) PFS and (effect on) HRQoL5 in cancer clinical trials. These findings raise questions regarding the assumption that interventions prolonging PFS also improve HRQoL in patients with cancer.”
According to Lange, this insight can no longer be simply dismissed. He simply cannot change the result of this analysis – there are probably no others. Lange: “The data on PFS simply doesn’t go in the direction that you indicated.” He doesn’t want to downplay personal experiences, but “these are not the basis on which we work.” And, he should add, they also have to work in accordance with the current law and the current procedural rules based on it. <<
by: MVF editor-in-chief Peter Stegmaier