Proteomics – the study of the proteome – is widely used within drug discovery. A portmanteau of ‘protein genome’, the proteome is the entire set of proteins expressed by a cell or organism. By using techniques like mass spectrometry, researchers can determine which proteins are implicated in which diseases, and use these as targets or biomarkers for new drugs.
Up till now, though, the technology has been underutilised in other areas – not least clinical trials. In March, outgoing FDA commissioner Scott Gottlieb stated that modernising clinical trials was a priority for the FDA, remarking that there is “a continued reluctance to adopt innovative approaches among sponsors and clinical research organisations”.
He cited proteomics as one such approach, pointing out that in an age of precision medicine, ‘enrichment’ strategies of this kind are important for guiding patient selection.
Let’s say you are trialling a new biologic, which is targeted towards patients with a certain disease and biomarker. If your study cohort includes all patients with that disease, you might have only a few responders. But if you’ve limited the study population to those with the biomarker, you won’t need so many patients, and the trial will be cheaper and more efficient. Proteomics is one way of finding out who has the biomarker.
What the study involved
Dr Axel Ducret, a scientist at Roche, recently made a bold step forward in this area. In April, he attended the MSCAL 2019 Conference (Mass Spectrometry Applications to the Clinical Lab), where he presented the results of a recent study conducted by Roche and proteomics specialists Biognosys and Caprion. The study explored how proteomics might be used in clinical trials, and concluded that the technology was ready to be applied more broadly.
“We found out that proteomics studies could be done within the time frame of a clinical trial,” says Ducret. “And it’s the first sign we have that there shouldn’t be any additional difficulty in monitoring the protein content of cells, rather than just the mRNA or genetic information.”
The study consisted of the analysis of 30 purchased colorectal cancer FFPE tissue samples, representative of trial samples. The method development, analytical validation and sample measurement were all performed as rigorously as if this were a real clinical trial.
In the discovery proteomics arm of the study, researchers used a label-free quantification mass spectrometric strategy developed by Biognosys. By comparing the signals generated against large libraries of reference data, they could generate a unique ‘fingerprint’ for each sample.
“This was an attempt to monitor the basal status of those tumours on a small population of people,” says Ducret. “We were able to follow the abundance of almost 9,000 proteins, which is at a comparable dynamic range to what we would obtain using RNAseq methods.”
As he explains, what you’d typically do with discovery proteomics data is to compare a basal level of protein expression with a disturbed level (i.e. after a drug has been administered). This enables you to stratify a tissue according to tumour grade, or determine whether a certain pattern of proteins might be predictive for drug response.
“You try to understand the definitive aspects of a cell with regard to a certain change of proteins,” he says. “To do this, you need to characterise many thousands of relevant proteins and the ability to quantify the change in levels. Over the last five years, we’ve been able to do this in greater depth and the quantification methods have become more powerful.”
Here, proteins enabling differentiation of tumour grades were identified with high statistical power, suggesting the technique is suitable for analysing subtle changes in protein expression.
The study also included a targeted proteomics arm, conducted with Caprion, which used a more traditional mass spectrometry approach to quantify 12 pre-selected biomarkers. Caprion developed the assays within six months (considerably faster than the norm) and with high accuracy. For 10 of the 12 biomarkers, the measurements were reported at a degree of precision comparable to a conventional ELISA assay.
“The flexibility to be able to develop and validate custom protein assays in such a short time is invaluable,” says Ducret.
A pivotal moment
So why haven’t proteomics been used in clinical trials up till now? Ducret thinks it comes down to two factors – the lack of technical knowhow and the lack of companies offering this kind of service.
“First you need to have the proper technology capability to do so – so the ability to collect tissues and analyse them reliably have really been the game changers,” he explains. “Typically in a clinical trial you get just a few micrometres of tissue that are put on the slide, so you have to have the sensitivity to analyse those samples. The other aspect is if you take materials from a clinical trial, you have to do that under GCP compliance, and of course there aren’t many companies that can do that.”
As he explains, this study was meant to demonstrate that it can be done in this context – and this is what makes it so new and exciting. Biognosys and Caprion, the companies that worked on the study with Roche, believe that discovery proteomics could soon become a more integrated part of clinical trials.
“We work with many of the most innovative pharma and biotech companies, who use our discovery proteomics services to advance their early-stage research programs. We increasingly see an interest in using our technology further into the clinical trial phase,” said Dr Claudia Escher, COO of Biognosys.
The future for proteomics in trials
Ducret adds that, while it’s undoubtedly useful to monitor the effects of drugs on proteins, it remains to be seen whether this is always the best and most relevant approach.
“One of the things we’re going to look at in the future is when is one ‘omics’ technology more relevant than the others?” he says. “If you have funding for only one ‘omics’ technology, which one is going to give you the answers that you need? I think in the future, the application in clinical trials will come when you have an exploratory arm to your approach, because proteomics gives you additional information on what’s going on.”
The next steps for his own research will be to apply the technique during an actual clinical trial.
“We have demonstrated this is feasible and doable at a reasonable price point, so now we want to find out what is the added value of having this kind of approach in clinical trials,” he says. “We might find that in some cases it’s very useful and in other cases it might not be.”
While his research is certainly novel, he points out that he is far from an ‘eccentric pioneer’ in the field. In fact, mass spectrometry is already being used in clinical trials, and his work builds on a large body of existing research.
“It’s been very slow progress, but I think in ten years we will be surprised that it took so long to use proteomics in clinical trials,” he says. “For me, it’s been a natural step to use this kind of technology in the clinic. It won’t replace all the assays that are used at the moment, but it will certainly provide additional information that we have been missing so far.”