Name: Jean-Charles Sanchez
Position: Head of the Biomedical Proteomics Group at the Faculty of Medicine, Geneva University
Background: PhD, Buckingham University; founder, Swiss Proteomics Society and European Proteomics Association
At the Human Proteome Organization's 11th annual meeting this week in Boston, the publisher Elsevier announced the launch of a new journal, Translational Proteomics.
Focused on connecting basic proteomics research to its ultimate clinical applications, the journal aims to provide a space for publications detailing proteomics experiments from early discovery through to validation and the bedside.
The journal's editor-in-chief is Jean-Charles Sanchez, head of the Biomedical Proteomics Group at the Faculty of Medicine at Geneva University and a founder of both the Swiss Proteomics Society and the European Proteomics Association.
ProteoMonitor sat down with Sanchez this week at HUPO to discuss the new journal, what niche it hopes to fill, and the general status of proteomics' long march to the clinic.
Below is an edited version of the interview.
What niche does Translational Proteomics aim to fill? Why does the field need a journal specifically for this area?
There are all the other journals – [for instance], Proteomics and Journal of Proteomics Research, which are more or less in the same niche. Then you have Molecular & Cellular Proteomics, which is more cellular-based, biology-based proteomics with a higher impact. And then you have journals like Proteomics Clinical Applications and Clinical Proteomics. But my feeling has been in the last year that there was missing something that can be a channel for broader studies that really combine the basic science and the discovery part of a study with large validation in cohorts of patients – and not just in 10 patients. So really [the journal] is to bridge the basic sciences to clinical research … to publish multidisciplinary papers and approaches to translate really the discovery to the patient.
That has been my main focus for more than 20 years. For example, we have a paper coming out on a marker for stroke with a very precise utility that we did in a large cohort of patients. This comes from discovery by proteomics.
So the discovery was [published] in one paper, and now the validation is in another paper, because if you send it to MCP they will tell you, "Very good, but it is too clinically orientated, so you should go to Clinical Chemistry," for example. And if you go to Clinical Chemistry, but you also include the discovery [portion], they will say, "This is good, but you should cut it to just the stats and clinical parts." So my view is that these two papers should be just one, showing the translation from the first step of discovery to the use at the bedside with the patients.
There has been a lot of frustration over the years on the part of both proteomics researchers and clinicians in terms of moving proteomic discoveries into the clinic. With the launch of this journal, are you saying you think the field is ready to overcome these challenges?
The whole field wasn't there [previously]. I think the people who were doing proteomics were not really aware of the limitations of the studies, [and the difficulty] of being able to translate them further to the patients. And the same on the clinical side – they were really looking at it as a technology rather than something that can really bring added value to the patient.
But I think things are changing. Everyone always says, "Can you tell me a biomarker found by proteomics that is now used in patients?" But in some ways that is a bad question because you can do that with any field. Just look at the [US Food & Drug Administration] approval each year of markers – there is one each year. So is that really [a] worthwhile [question]? Is there really something statistical to say if there is an impact or not an impact [from proteomics]?
I think the question is different. People are always looking at biomarkers as a diagnostic marker. [But for] many diseases you don't need a pure diagnostic marker. If somebody has a stroke, and they are in the hospital, you don't need a diagnostic marker to say if it is a stroke. The physician knows that in 99 percent of the cases. What [they] need, for example, is [to know] when was the stroke. One hour ago? Or six hours ago? Because that has a very big impact on treatment. These are the sorts of niche questions where the biomarkers are starting to bring something. All these very specific niche questions … the basic scientists doing the discovery should really be aware of, and that is why you really need multidisciplinary teams to be able to translate something.
So do you envision most of the publications in the journal being from large, multi-group studies that bring together basic researchers, clinicians, etc.?
You have no choice. You need to have the physicians in for sure. So it's not just one team, it's several teams. You need statisticians, biostatisticians, mass spectrometrists, the clinicians with the right question. It's not just one group in proteomics that will go all the way through the process.
And there is also the commercial aspect. You cannot have a marker that goes to the patient if it is not commercialized. So it has to follow all the QC and acceptance [procedures] and so on. Industry is part of [the term 'translational']. They are part of the pipeline, part of translating a marker or a protein to the patient. Because Abbott, Roche, and many other small diagnostic companies are the ones that will decide if it can go to the patient, and for different reasons.
[For instance], is the market big enough to develop? Because the cost to develop a marker is huge in comparison to the payoff for the companies. Looking at some of the well-known markers that are used – troponin, for instance – if you look at the window from discovery to when they are approved by FDA or sold by the company to patients, the [patent] window is 20 years, and until the marker finds its real niche and real intended utility, it takes a lot of time. It's much more complicated than we thought in the beginning — there are so many things happening in humans and very few that are really specific to one organ or disease. For example, [prostate-specific antigen] — now it would never be commercialized, probably.
You say you want to publish studies as one paper that right now would likely be broken up into two or three different papers. How do you think researchers will take to this idea, given that it will potentially lower their number of publications?
Perhaps when you are a young scientist you need publications and you want to speed up [that process] but I think we are all learning that the number of papers is not the issue; that is not the goal. So I would prefer to have really full and complete studies. I don't want just potential markers, [so] that at the end of the paper we have just another potential marker of this kind of cancer. No. I want that at the end a commercial company will look at it and say, "Oh, gosh, we have to test it," because [the researchers] did enough studies. They looked at, say, 400 patients. I'm not asking for studies with 1,500 multi-cohort patients, because that would be impossible, but to have one or two cohorts with large populations that people can look at and say that is potentially very interesting. This is not just a potential marker, it is a marker.
In speaking to scientists at commercial clinical labs like Quest or Laboratory Corporation of America, it seems many times they are frustrated with what they see as the low quality of protein biomarker studies coming out of academia. Do you think this journal will help to remedy that?
The example that I know well, because I am working in it, is stroke. And, sure, at the beginning when we were going to see some of the diagnostic companies, we were very naive. I think we have had a time window to learn in the last ten years. And I think a number of people have become mature [in terms of understanding diagnostic companies' needs and expectations]. For example, in stroke, throughout the whole process we have companies involved, and they are giving feedback on what they think, and then we are trying also on our side to modulate their view. Because many of them are not completely clear on what will be the intended use. So that is also up to us and the specialists in the disease and clinicians in that disease to try to move a little bit the view and [establish] what is the real utility for the physician and the patient at the end of the day. So this type of work should be fully iterative among the different actors and the patient associations as well.
You do [a biomarker discovery study] in proteomics and you have a nice verification, but then people don't follow the next step of the story [because] it comes out two years later [in a different journal] and people forget about it. So that's why I think we should move a little bit [toward having it all in one journal], and I think things are moving a little bit [in that direction]. I think the momentum is right for that.