0:10
Right.
0:11
So thank you for that introduction and hello everyone.
0:14
Let's begin.
0:15
So why are we here?
0:19
Biomarkers, right?
0:20
It's not a philosophical question.
0:22
And as we've seen over the last couple of days, the search for biomarkers remains the Holy Grail in a number of disease areas.
0:30
And in fact, the future of personalised health interventions rests squarely on the ability to discover and validate new biomarkers.
0:38
Now, as we have seen, biomarkers themselves can be nearly anything that distinguishes 1 phenotype from another, one individual from another.
0:47
So biomarkers can be based on a diagnostic test, for example, glucose or cholesterol measurements, a physical characteristic like body mass index, genetics, SNPs or any other distinguishing characteristic.
1:00
Age, diet.
1:02
Now unfortunately, general screening methods have been extremely challenging and there's been some challenges including technological challenges, mathematical challenges and the limited availability of proper sample types and numbers.
1:19
Now one recent and ever evolving technology which has the potential to overcome a number of these issues is metabolomics.
1:28
So my name is Alessandro Busetti.
1:30
I'm one of the senior field application scientists for Metabolon and the title of my presentation today is Biomarker Discovery and Validation using Global Metabolomics.
1:41
And in today's presentation, I will briefly introduce what is metabolomics.
1:46
I will then go into a bit more detail regarding our platforms and key deliverables before jumping into two interesting case studies highlighting the value of global and targeted metabolomics when it comes to the identification of biomarkers of response and disease progression.
2:03
And hopefully by the time I'm done, you will have a firm understanding of why this technology can be a game changer for your research and you will be running to our booth to plan your metabolomics studies.
2:16
So what is metabolomics?
2:18
Well, as we all know, metabolism is the conversion of food, so our fuel to the energy needed to run cellular processes.
2:26
And these enzyme catalysed reactions allow organisms to grow and reproduce and maintain their structures and most importantly to respond to their environments.
2:36
Now the intermediary and chemical endpoints of these enzyme catalysed reactions are called metabolites.
2:42
And you see here on the slide for examples.
2:45
I'm sure you're all familiar with glucose, a sugar, phenylalanine and amino acid, cholesterol, a lipid, and oleic acid, a fatty acid.
2:54
Now the metabolome is the global collection of all low molecular weight metabolites that are produced by cells during metabolism and Metabolomics is the scientific study of chemical processes involving metabolites.
3:10
So the small molecule substrates, intermediates and products of cell metabolism.
3:15
Now specifically, metabolomics is the systematic study of the unique chemical fingerprints that specific cellular processes live leave behind, so the study of their small molecule metabolite profiles.
3:33
Now importantly, the risks for getting most diseases actually arise from your metabolism, your environment and your lifestyle.
3:40
So, metabolomics provides a snapshot of the entire Physiology of a biological system, and it reflects biochemical activity within cells.
3:50
Thereby, it actually represents the system's functional status during disease and treatment.
3:57
Now notably, metabolites derive not only from the biological activity of the cells, reflecting the genomic variation of the individuals through proteins and enzymatic activities, but also from the environment, including the effects of drugs, diet, lifestyle, and not the least the effects of the microbiome on the metabolome.
4:18
And in fact, metabolomics provides the closest data set to the phenotype.
4:23
It comes downstream of the central dogma of biology.
4:26
So you have the genome, what could be transcript, the transcriptome, what is being expressed, the proteome, what is being translated.
4:34
But again, what do you know about those proteins?
4:36
Are they active, non active?
4:39
How accurately have they been annotated?
4:41
And metabolomics has the potential of integrating all these other omics and many other sources of data, for example, clinical measurements.
4:50
So metabolomics really describes the ultimate molecular phenotype and as such, it is one of the most effective tools to discover and validate biomarkers.
5:03
But what is it that we do?
5:04
And most importantly, how will this be a game changer for your research?
5:10
Well, at Metabolon, over the last 23 years, we have developed three main solutions for metabolomics.
5:17
Our global discovery panel, which is based on untargeted metabolomics analysis.
5:23
It is by far the most powerful approach for biomarker discovery that we can offer.
5:29
We can profile with a high degree of chemical accuracy up to 5400 metabolites, allowing us to cover all pathways and super pathways in mammalian systems, but more broadly in all biological systems.
5:43
So bacteria, fungi, plants, insects, fish, reptiles.
5:48
It is a comparative approach.
5:49
So Group One versus group 2, healthy versus diseased versus treatment or time point 1 versus time point 2 versus time point 3 if it's a longitude and old study.
6:01
And we can work with virtually every sample type and biological matrix, from the most common systemic matrices like blood, serum, plasma, saliva, urea, to more complex systems like 3D models like organoids, but also biopsies.
6:21
And we've worked with very exotic matrices as well, including dinosaur bones and tartar from ancient Egyptian mummies.
6:29
So once we run our analysis, we generate an incredible amount of data.
6:36
And this is not a joke, but we don't leave you alone.
6:39
So we always provide you with a report addressing the biological questions of the study.
6:44
So not just what went up and what went down, but what does this mean in terms of the biology of the study?
6:51
Now the second deliverable that we can offer is our targeted lipidomics analysis.
6:57
And as I'm sure you all know, lipids are not just energy storage molecules, they have a wide variety of bio activities.
7:04
They can be signalling molecules, they can be immune modulators.
7:07
They play an important roles in conditions like neurodegeneration, cancer, liver conditions.
7:15
And so we can, with our complex lipid panel, we can provide accurate quantification of over 1000 lipid species across 14 different classes.
7:25
And finally, the third deliverable is our targeted metabolomics analysis.
7:30
And as the term suggests, it is an analysis on a much smaller panel of biomarkers.
7:38
It's typically done after you've done untargeted metabolomics.
7:41
So you do the untargeted metabolomics to identify your biomarkers and then you use the targeted approach to validate them.
7:49
It is fully quantitative and you can choose from either are off the shelf panels which have been put together based on biomarkers from literature or based on the results of the untargeted metabolomics that we've run for you.
8:05
We can actually put together a personalised panel of metabolites to look at and what you see on the right here are some of the tools our clients have access to through their My Metabolon portal where the data is stored.
8:19
So apart from the raw data and the reports, there are tools for pathway visualisation, pathway enrichment, generation of heat maps, PCA plots, volcano plots and this is just to name a few.
8:31
So basically, you can keep on mining your data forever.
8:36
And this is a quick overview of our modus operandi, and it all begins with the study design.
8:43
So you would be sitting down with someone like me, we would be discussing what are the biological questions that you're keen to answer with your study?
8:52
What are the sample types which are available?
8:54
Are they the right ones to address these biological questions?
8:58
What are the sample groups that we should put together to address these biological questions?
9:03
What is the study power?
9:05
How many biological replicates do we need of this of these samples in order to ensure that the changes we see actually reach statistical significance?
9:14
And obviously then statistical analysis, what are the comparisons which are most appropriate for the study?
9:21
Now, once we're happy with the study design, you run your experiment, you collect the samples and you send them over to us to Raleigh, NC in the United States.
9:33
And we receive them.
9:34
We extract them.
9:35
Typically it's a methanol extraction which crashes out all the macromolecules like proteins.
9:42
And then we take these extracts and we load them onto our platforms.
9:46
Now let me be clear, we're very much aware to that to resolve the biological complexity of biological samples, a single method is not enough.
9:55
So every single sample we received is run with multiple methods.
10:00
So we have methods for compounds that ionise only in positive mode or negative mode on our mass specs.
10:06
We have chromatographic methods for compounds who are very polar, so highly polar or apolar compounds or compounds that require Helix separation.
10:18
And once they have been run, we generate this analysis generates an incredible amount of ion features and we go through three quality control steps.
10:27
So the first one is done by an AI software that we have developed in house, which is able to identify the peaks based on the compounds, the physical standards in our library based on their mass, their retention time on their columns, as well as their fragmentation patterns.
10:44
So I want to stress this, we provide Tier 1 identification.
10:48
There's no uncertainty in what we provide.
10:51
The second step of QC is done by a team of 20 curators.
10:55
And then a third step of QC is done by the study directors themselves, who are actually the people who have the institutional knowledge to then translate the metabolomics results into the biological insights that we give back to you.
11:11
Now, metabolomics is really applied from basic research with the discovery of molecular pathways leading to disease through to translational and Preclinical Research with biomarkers helping with the understanding of the mechanism of action in PKPD studies and those selection.
11:28
But ultimately, biomarkers are also crucial during clinical trial phases in which biomarkers can help with patient stratification, patient selection, trial quality, so identifying outliers or addressing compliance issues and site issues, as well as dose selection, assessment of safety and efficacy and studying phenotypic exploratory endpoints.
11:53
And this is just to name a few.
11:57
So finally, let's jump into two quick but juicy case studies highlighting the value of metabolomics when it comes to biomarker discovery and validation.
12:08
Now, the first study I wanted to talk to you about today is a study in the field of immune oncology and more specifically on the discovery of biomarkers of response to immunotherapy.
12:20
Now, as you all know, immuno oncology is different from regular cancer therapy.
12:25
It's not just about the tumour, it's about the whole system.
12:28
And most of all, it's about unleashing a system response.
12:32
Now given the complexity of the immune system and the seemingly binary response to immune checkpoint blockade, there is a fierce motivation to identify biomarkers that can predict response.
12:45
Now, one proposed solution is based on the idea that therapeutic response to immune checkpoint blockade operates like a critical state transition of a complex system.
12:55
So these complex systems are highly sensitive to the initial state and while notoriously difficult to predict in advance, can be detected closer to the tipping point of response.
13:07
Hence, profiling dynamic markers at a systems level has the potential to identify response hubs, and these hubs can double up as biomarkers for response and drug targets.
13:18
So the first study I will be presenting today is a study by Arthur Frankel et al, and it was published in the journal Neoplasia in 2017.
13:29
One of the reasons I chose this study is because it was the first prospective pilot study aimed at the discovery of novel biomarkers of response to ICT as well as looking at the effects of the human gut microbiome and metabolites on immune checkpoint inhibitor therapy response in metastatic Melanoma patients.
13:50
What is the background to this study?
13:51
Well, very simply immune checkpoint therapy is able to achieve durable remission in up to half of the patients with metastatic Melanoma, but the prognosis for ICT non responders remains very poor.
14:06
So as I said, the main goals of the study were to discover novel biomarkers of response to ICT therapy as well as to understand the role of the gut microbiome in ICT response.
14:18
Now what you see in this slide is the study design as well as the summary of the clinical characteristic of Melanoma patients who underwent checkpoint inhibitor therapy and also an excerpt of a table summarising individual clinical characteristics of these Melanoma patients.
14:35
So briefly, there were 39 adult Melanoma patients and therapy consisted of one of four regiments.
14:43
The first one was out and they're all based on humanised monoclonal antibodies.
14:49
So the first one was outpatient ipilimumab every three weeks for four doses.
14:55
The second one is nivolumab with ipilimumab every three weeks for four doses, followed by nivolumab every two weeks.
15:05
The third one is nivolumab every two weeks and the fourth one is pembrolizumab.
15:10
So pembrolizumab or Keytruda every three weeks.
15:15
So the first thing that the researchers did was to perform metagenomic shotgun sequencing on patient faecal samples prior to ICT.
15:25
And the results of metagenomic shotgun sequencing were used by the researchers to identify specific bacterial species that are enriched in the gut microbiomes of Melanoma patients who are responding to ICT therapy.
15:41
And what we will see now are some of the results of the metagenomic shotgun sequencing.
15:45
So in fact, from a taxonomic point of view, amongst all treated ICT patients, responder microbiomes were found to be significantly enriched with Bacteroides coccus, Streptococcus parasanguinis compared to those with progression.
16:01
Amongst those patients treated with only ipilimumab and nivolumab, responder microbiomes were enriched with the Firmicute phylum members Faecalibacterium prausnitzii and Holdemania filiformis and the Bacteroidetes phylum member Bacteroides thetaiotaomicron.
16:18
Whereas amongst the patients treated with pembrolizumab alone, the responder microbiomes were enriched with Dorea formicigenerans.
16:27
Now, interestingly, despite distinct gut microbiota signatures, the overall microbiome diversity was not significantly different between responders and those with progressive disease.
16:43
Importantly, metagenomic shotgun sequencing also revealed differences in microbiome gene content.
16:50
So by identifying the presence or the absence and calculating the abundance of microbial functional pathways between responders and those with progressive microbiomes.
17:02
Now amongst all ICT recipients, responder microbiomes were found to be significantly enriched with bacterial enzymes involved in fatty acid synthesis.
17:12
Whereas amongst ipilimumab and nivolumab recipients, responder microbiomes were enriched with bacterial enzymes involved in inositol phosphate metabolism.
17:24
So overall, using metagenomic shotgun sequencing, researchers were able to identify a number of taxa that were significantly enriched in either responders or non responders in all ICT patients treated with a combination of ipilimumab and nivolumab or pembrolizumab alone.
17:42
But let's get finally to the biomarkers identified using our global untargeted metabolomics platform.
17:49
And let's take a quick step back.
17:50
It's well known that gut microbiota derived products such as short chain fatty acids, inositol phosphates, secondary bile acids, they can induce very strong physiological changes, the attenuation of colitis, or they can have anti tumour effects, for example through the modulation of epigenetic mechanisms.
18:09
And this has been observed also in the absence of bacteria, so for example, in vitro models or in the autobiotic animal models.
18:17
Hence, the researchers performed unbiased untargeted metabolomic profiling on the same patient faecal samples used for metagenomic shotgun sequencing to determine if we could detect significant differences in gut metabolites between responders and those with progressive disease.
18:36
So what did we find?
18:38
Well, of the 1901 identified compounds so known and named biochemicals amongst all ICT recipients, 83 metabolites were significantly different when comparing the responder group to the progressive group and 49 of these metabolites increased and 34 decreased.
18:59
And I'll go into a bit more detail about these metabolites in a second.
19:03
Amongst those patients treated only with ipilimumab, so the anti CTLA 4 and nivolumab, so the anti PD 1, the responder metabolomes were significantly enriched in 45 and depleted in 22 metabolites.
19:18
And finally, amongst those patients treated with pembro alone, responder metabolites were significantly enriched in nine and depleted in five metabolites.
19:28
And this is just an overlap of the different treatments.
19:32
Now interestingly as you can see and it's pretty small, I apologise for that, significantly enriched and depleted metabolites in responder metabolomes.
19:42
We're involved in a number of metabolic pathways.
19:45
So amino acid metabolism, lipid metabolism, peptide metabolism, diazoglycerols and triazoglycerols and xenobiotics but strikingly 15, 2 anacardic acid levels were markedly increased in all ICT responders, so 62 fold.
20:05
And in patients receiving ipilimumab and nivolumab that was a 94 fold increase in the responders.
20:12
So anacardic acid is an alkyl derivative of salicylic acid and it is produced in the nutshell of cashews and also in mangoes.
20:22
And interestingly, anacardic acid stimulate phagocytes and kind of mint bactericidal activity.
20:28
Since anacardic acid is considered a xenobiotic and not known to be of bacterial origin, we expanded patient histories to include queries about their diet and specifically the consumption of plant related products with high levels of anacardic acid.
20:47
And of note, 5 of 6 patients with the highest (15, 2) anacardic acid levels reported consuming cashews at least weekly.
20:56
So as I said at the start, this was the first detailed report of human gut microbiome metagenomic and metabolomic profiling in Melanoma patients treated with a combination of anti PD-1 and anti CTLA 4 immunotherapy as well as anti PD-1 therapy alone.
21:13
Using a combination of metagenomic shotgun sequencing and unbiased metabolomic profiling, the researchers were able to identify specific gut microbial species and numerous gut metabolites that were associated with response to ICT therapy in both all patients as well as subsets of patients treated with ipilimumab and nivolumab and pembro alone.
21:36
Now we performed unbiased gut metabolomic profiling in an attempt to gain greater functional insight.
21:42
And surprisingly, amongst the 1901 valuable metabolites, the most dramatic correlation was seen with a plant xenobiotic, so 15, 2 anacardic acid.
21:54
And as noted previously, anacardic acids stimulate neutrophils and macrophages and similar to the effects of a particular of specific bacterial species, the activation of macrophages and dendritic cells may enhance T cell recruitment to tumour metastases and consequently augment the response and enhance the response to ICT.
22:19
In fact, anacardic acid has been shown to have anti-tumour effects in several preclinical models and further preclinical and clinical studies are currently on the way on this very potential simple therapeutic intervention.
22:35
Now while these preliminary observation do not establish a causal connection between gut microbiota and gut metabolite and ICT therapy, this contributed in spurring the initiative to pursue larger follow up studies and more detailed laboratory investigation.
22:56
So these studies laid the foundation to optimise the host response to ICT therapy.
23:03
And just to conclude, 1 anecdote, I went out last night, I wanted to buy mangoes and cashews as props for this talk, and I went to three supermarkets.
23:15
I didn't find any.
23:16
So I am pretty sure that's because everyone here in Manchester is familiar with this research.
23:22
OK, that must be the reason.
23:23
So I won't go into the second case study because I think there's a time issue, a time constraint.
23:29
But please come and visit our booth afterwards and we'll be happy to, well engage with talks with you.