14 December 2012. Johan Luthman of Merck & Co., Inc., West Point, Pennsylvania, heads the company’s BACE inhibitor development program, which began enrolling last month for an 18-month Phase 2/3 trial of its lead compound (MK-8931) in mild to moderate Alzheimer’s disease (see ARF related news story; EPOCH).
Q&A With Johan Luthman. Questions by Esther Landhuis.
Q: ClinicalTrials.gov says the company will test the compound in 1,960 mild to moderate AD patients. Is that correct?
A: That is what we’re aiming for. We are starting out with 200 subjects in the initial Phase 2 component. Once we have a three-month safety readout in this subgroup, we will slide into Phase 3 with about 1,700-1,800 subjects. How long the Phase 2 takes will depend on how fast we recruit those 200 subjects. We expect to have a safety readout by mid-2013.
Q: How long do you expect enrollment to take?
A: We generally do not talk about enrollment time—more about when we hope to finish a trial, in this case, late 2016, beginning 2017. Enrollment can be remarkably hard. Getting to and from the trial site is a big obstacle, as is the number of procedures for screening. You can tire patients out with all the different measures they need before going into the trial. But you need as many readouts as you can. It’s a tricky balance.
Q: How much will the EPOCH trial cost?
A: We do not disclose the costs for our trials; however, trials in Alzheimer’s disease are very costly.
Q: MK-8931 is a disease-modifying compound. The field has moved toward earlier-stage trials, partly out of recognition that amyloid reduction at the dementia stage is too little too late. Why are you doing this trial in mild to moderate AD, after all the recent trial failures in that patient group?
A: There are a couple of factors. One, BACE inhibitors have not yet been tested in mild to moderate Alzheimer’s patients, the population with the greatest medical need. There is discussion about “a point of no return.” But that is a theoretical discussion right now. We should leave no patient behind. We think this group of patients deserves to be helped, and we have a responsibility to study the drug in this group.
As we learn from the 200-patient safety cohort, we will look at the earlier stages. Besides the ongoing Phase 3 trial of Roche’s humanized Aβ antibody gantenerumab, no one has really executed a big trial in earlier-stage patients. The field is just now developing the technology and tools for use in big trials. It’s pioneering work.
Q: Are there plans in the EPOCH trial for separate analyses of mild versus moderate AD?
A: One is always interested in different subgroups such as mild versus moderate AD, ApoE4 versus non-E4. I cannot comment on how the study has been powered for subgroup analyses.
Q: Are plans in the works for a trial in prodromal AD?
A: We are planning for a prodromal AD trial. We will look across the spectrum, but at this moment we are not sharing any details on our next steps in the development program of MK-8931.
Q: In the prodromal AD trial, do you consider using the new diagnostic criteria, which combine an episodic memory deficit with an abnormal biomarker result for a clinicobiological diagnosis of Alzheimer’s in the absence of dementia?
A: This is the kind of approach everyone is taking now. We are not much different.
Q: In the EPOCH trial, do you expect effects on ADAS-Cog with your inhibitor?
A: Certainly. That’s a primary readout in the trial. But we don’t expect to see effects early. That is why we have an 18-month trial. We will have interim analyses—for example, the safety readout at three months—but the main efficacy readout is the complete analysis at the end of the trial.
Q: Will the EPOCH trial enrich for mild AD, for example, by requiring a high MMSE range for inclusion?
A: We are not biasing our enrollment in the trial. This is a mild to moderate AD trial. Of course one could have some filters, for example, more mild or more moderate, but this is not our intention right now.
Q: Why not? Finding an effect only in mild AD is widely seen as having saved the solanezumab Phase 3 program.
A: I agree. We all believe earlier may be better. However, that doesn’t exclude the possibility that a BACE inhibitor would work in more advanced cases, like moderate AD.
Also, I don’t think you can draw many conclusions from a trial of an antibody with an unclear mechanism of action. We are talking about a very different mechanism of action. We have a clear readout for our BACE inhibitor’s mechanism of action. In Phase 1, it lowered CSF Aβ more than 90 percent.
Q: Will the EPOCH trial confirm the clinical diagnosis by way of a lumbar puncture or an amyloid scan at inclusion?
A: We are not aiming to have strong biomarker bias at inclusion. In some recent trials, there was some discussion about a third of the participants not, in fact, being AD patients, based on amyloid PET imaging. [Editor’s note: See ARF CTAD story.] PET amyloid is a biomarker. It is an indication of how your brain looks, but it is not established as the “standard of truth” measure, which is still postmortem pathology. Data are being generated with PET ligands and looking postmortem. But those are small datasets in late-stage patients. We don’t fully understand what “PET amyloid-negative” really means. Are they 100 percent not Alzheimer’s? Or does it depend in part on the sensitivity of the biomarker? We have to be careful saying that people who are PET amyloid-negative are not AD patients. They may have weaker forms of amyloid pathology that are not easily detected.
I am on the ADNI Private Partners Scientific Board (PPSB). We plan to dedicate one session of the ADNI PPSB March meeting at AAN in San Diego to diffuse amyloid imaging.
There is big discussion about whether one should use biomarkers in a mild to moderate AD trial. I think it’s too early to say.
Another approach is to power your study to make up for the patients who may be misdiagnosed.
Q: Did you decide not to test the inhibitor in a prodromal population now because it was so hard to screen for such patients in Bristol-Myers Squibb’s avagacestat and Roche’s gantenerumab trials?
A: Just because there is a strong rationale to look at earlier stages, we should not forget mild to moderate AD patients. It is relevant to test a BACE inhibitor in this population. We really don’t know how the compound will work in these patients.
Regarding a prodromal AD trial, the field is just now building the infrastructure to run these trials. If you use biomarkers for enrichment, they have to go through clinical qualification, in some cases technical development as well, before we are comfortable enough to use them in a trial.
CSF Aβ and tau biomarkers are starting to be developed to a level where they can be used in Phase 3. Then, it is a matter of standardization across different assays and multiple sites. For PET amyloid imaging, there is one ligand approved, and others in development. They have not yet been used extensively for enrichment in trials. As far as volumetric MRI, we are at the cusp of getting those techniques ready to be used. It’s going to be very exciting in the next two to three years when all these tools can be used in larger efforts.
Q: If EPOCH misses its primary endpoints but generates signals on the secondary endpoints and biomarkers, will Merck continue to study this inhibitor?
A: It very much depends on what we see. The EPOCH trial does not have a strong emphasis on biomarkers. The main emphasis is the clinical readout.
It is also based on how much we know about the biomarker changes. Time will tell us more and more what those changes represent.
Q: In which direction would volumetric MRI have to change in order to be a positive sign? Many of our readers are unsure about that because in two different antibody programs, hippocampal volume seems to have decreased in response to treatment.
A: The progression of the disease is characterized by loss of volume, something we generally assume reflects neurodegeneration. The assumption is, therefore, to expect less reduction in brain volume if a mechanism would be neuroprotective. On the other hand, specific mechanisms such as the antibody-mediated plaque clearance approach could have a different effect; if you deplete the amount of amyloid plaque, there may be a further loss of volume. However, the relationship between loss of plaque and potential counteracting effect by stopping the brain damage is not known. With a BACE inhibitor, you would not necessarily expect much of an effect on established plaques and, thus, one would assume a reduction in the progressive brain volume loss, if the drug slowed the ongoing neurodegeneration.
Q: MK-8931 Phase 1 has shown strong CSF Aβ lowering in healthy volunteers. Is it an aim of the EPOCH trial to find out which direction CSF Aβ will go in people with brain amyloid, whose CSF Aβ is already low?
A: We believe we have sufficient information from the intended population (see Phase 1 AD trial). We will communicate this at the AD/PD meeting in March.
Q: Then, in essence, this trial could validate the direction of CSF Aβ as a surrogate outcome marker.
A: There really are no data yet showing that lowering Aβ peptides, even as much as we have shown, will have an effect on the disease itself. This is what we are going to test.
Q: We have received many questions on this point. If MK-8931 strongly changes CSF Aβ but misses the primary outcome in this trial, would that result not leave open the possibility that the same Aβ change would be associated with a clinical benefit at earlier stages, when the disease is thought by some to be more strongly driven by Aβ? It seems our readers wonder if Merck will answer this question even if this trial was to fall short of meeting primary endpoints.
A: As I mentioned before, we are intending to run a prodromal AD trial as well, pending the EPOCH safety cohort results. However, even at this earlier stage there is a lot of established brain pathology, including advanced amyloid pathology, in many subjects. I guess the question is if treating prodromal AD with an Aβ peptide-lowering therapy is even too late to affect the disease. The frank answer is, we do not know—we have to learn from the initiated trial and planned trials in prodromal AD. The effect in even earlier stages of the disease has to be studied as well to fully understand if there is a disease stage with a point of no return for this mechanism. The field is rapidly establishing the possibility to run larger trials in very early populations, so-called asymptomatic AD.
Q: Thank you for this interview.