Plasma Proteomics May Predict Disease Manifestations in ALS

WEBVTT
1
00:00:05.730 --> 00:00:08.029
My name is Michael Banata. I'm a neurologist at the
2
00:00:08.089 --> 00:00:11.970
University of Miami. I serve as the executive director of
3
00:00:12.050 --> 00:00:16.190
our ALS center and also run our neuromuscular program more broadly.
4
00:00:17.170 --> 00:00:21.489
I've been working in the field for almost 25 years,
5
00:00:21.570 --> 00:00:25.699
I think, and increasingly spend my time focused on ALS
6
00:00:25.730 --> 00:00:27.660
research and trying to develop better therapies.
7
00:00:29.589 --> 00:00:33.490
What was the key question behind this study and why
8
00:00:33.770 --> 00:00:34.950
is it so important?
9
00:00:36.350 --> 00:00:39.780
We're really trying to understand why it is that we've
10
00:00:39.820 --> 00:00:43.399
had so little success in developing therapies that really work
11
00:00:43.479 --> 00:00:47.780
for patients with ALS. And one of our key concerns
12
00:00:47.909 --> 00:00:52.630
is that experimental therapeutics are being brought to bear too
13
00:00:52.689 --> 00:00:56.119
late in the course of disease. And so this led
14
00:00:56.219 --> 00:00:59.560
us to ask the question, when does the biology of
15
00:00:59.640 --> 00:01:04.500
disease begin? And can we predict when symptoms are likely
16
00:01:04.599 --> 00:01:08.750
to arise? And if we can do those things, might
17
00:01:08.829 --> 00:01:14.269
we think about intervening therapeutically much earlier in the course
18
00:01:14.310 --> 00:01:16.849
of disease? And so in order to be able to
19
00:01:17.250 --> 00:01:21.000
answer that question, that series of questions, we had to
20
00:01:21.060 --> 00:01:25.519
study people at elevated risk for ALS. And so we
21
00:01:25.680 --> 00:01:28.739
set upon the idea of studying people who carry a
22
00:01:28.799 --> 00:01:32.359
genetic risk that puts them at a high lifetime risk
23
00:01:32.400 --> 00:01:37.219
of developing ALS and to study them longitudinally before symptoms
24
00:01:37.299 --> 00:01:41.280
ever develop. And then once they've developed symptoms, we can
25
00:01:41.319 --> 00:01:43.799
go back and look at the data that we've collected
26
00:01:44.400 --> 00:01:48.040
in this case, biofluids and proteomic data that we've generated
27
00:01:48.120 --> 00:01:52.000
from those biofluids, and ask if there was something measurable
28
00:01:52.060 --> 00:01:55.099
that could have told us that disease was coming.
29
00:01:55.939 --> 00:01:59.870
What did you learn about the biology of ALS before
30
00:01:59.989 --> 00:02:01.129
symptoms appear?
31
00:02:02.049 --> 00:02:04.650
So one of the things we learned from the PREFAL study,
32
00:02:04.689 --> 00:02:07.310
which is the source from which these data come, we
33
00:02:07.349 --> 00:02:10.610
learned about 10 years ago that a protein called neurofilament light,
34
00:02:10.689 --> 00:02:13.830
or NFL, goes up in the blood a year or
35
00:02:13.889 --> 00:02:17.710
two before people develop symptoms of disease. What we learned
36
00:02:17.810 --> 00:02:21.569
here is that it's not just neurofilament light. It's the
37
00:02:21.629 --> 00:02:25.870
concentration of almost 100 proteins that are measurable and that
38
00:02:25.889 --> 00:02:30.189
go up before symptoms of disease. And importantly, some of
39
00:02:30.250 --> 00:02:35.409
those proteins, their concentration is changing 8, 9, 10 years
40
00:02:35.590 --> 00:02:39.590
before symptoms emerge. So we have a much better handle
41
00:02:39.689 --> 00:02:44.169
on sort of what the architecture of the proteome is
42
00:02:44.270 --> 00:02:47.629
and what it looks like in blood a decade before
43
00:02:47.669 --> 00:02:49.349
people develop symptoms of ALS.
44
00:02:50.669 --> 00:02:55.960
What makes these proteomic findings promising as biomarkers?
45
00:02:57.240 --> 00:02:59.800
It's a good question. It really depends on what your
46
00:02:59.840 --> 00:03:03.139
understanding of a biomarker is. So let me drill into
47
00:03:03.180 --> 00:03:06.370
that a little bit. A biomarker for me is something
48
00:03:06.389 --> 00:03:09.669
that we can measure and quantify that tells us about
49
00:03:09.909 --> 00:03:15.250
biology or the risk of developing a certain outcome. And
50
00:03:15.289 --> 00:03:18.550
the FDA underscores, as does the NIH, that we have
51
00:03:18.569 --> 00:03:21.889
to think about biomarkers that are fit for purpose. And
52
00:03:21.949 --> 00:03:25.550
so in the context of studying people at genetic risk
53
00:03:25.610 --> 00:03:29.310
for ALS who've not yet developed symptoms of ALS, we're
54
00:03:29.330 --> 00:03:33.969
really talking about what are called susceptibility or risk biomarkers.
55
00:03:34.520 --> 00:03:36.860
And what that means is a biomarker that we can
56
00:03:36.939 --> 00:03:40.590
measure that tells us something about the future risk of
57
00:03:40.659 --> 00:03:44.129
developing disease. And what we were able to show in
58
00:03:44.189 --> 00:03:48.710
this study, in both the discovery data set, but also
59
00:03:48.770 --> 00:03:52.210
in the replication cohort, and we can talk about that
60
00:03:52.250 --> 00:03:56.889
if we have time, it's an imperfect replication cohort. But
61
00:03:56.909 --> 00:03:58.889
what we were able to show is that the set
62
00:03:58.930 --> 00:04:02.780
of proteins that we can measure do indeed predict outcomes
63
00:04:03.180 --> 00:04:07.520
the likelihood or not of developing disease over the relatively
64
00:04:07.620 --> 00:04:12.360
short term. And so this combination of proteins then really
65
00:04:12.400 --> 00:04:16.939
does help to serve that purpose of identifying short-term risk
66
00:04:17.100 --> 00:04:20.079
of developing ALS. And I should just say, as I
67
00:04:20.180 --> 00:04:23.939
say that, that I definitely still see this at the
68
00:04:23.980 --> 00:04:28.189
discovery stage, even though we have a replication cohort. There's
69
00:04:28.209 --> 00:04:30.050
still quite a bit of work that needs to be
70
00:04:30.089 --> 00:04:34.110
done to validate this finding and get this ready for
71
00:04:34.269 --> 00:04:35.990
clinical use or for prime time.
72
00:04:37.129 --> 00:04:42.199
How could this research change how clinicians identify or monitor
73
00:04:42.259 --> 00:04:44.600
patients at risk for ALS?
74
00:04:45.740 --> 00:04:48.300
So I don't think it's yet ready for clinicians to
75
00:04:48.519 --> 00:04:50.939
use in the clinic. I think this is still a
76
00:04:51.000 --> 00:04:55.740
research tool. So to illustrate that, When we first discovered
77
00:04:55.769 --> 00:04:59.459
about 10 years ago that blood concentrations of neurofilament go
78
00:04:59.540 --> 00:05:03.540
up a few years before people develop symptoms, we went
79
00:05:03.620 --> 00:05:06.620
on to work with an industry partner in designing a
80
00:05:06.680 --> 00:05:10.660
prevention trial where we would monitor levels of neurofilament. And
81
00:05:10.680 --> 00:05:13.480
when they go up, people would be randomized to an
82
00:05:13.540 --> 00:05:18.720
experimental drug or placebo to see if we can prevent disease. Now,
83
00:05:18.779 --> 00:05:22.680
what we know now is that neurofilament alone is insufficient
84
00:05:22.839 --> 00:05:26.879
for many, if not most, of the genetic forms of ALS.
85
00:05:27.360 --> 00:05:29.939
And so what we envision this as a research tool
86
00:05:30.000 --> 00:05:32.750
is that we might be monitoring a panel of proteins
87
00:05:33.230 --> 00:05:36.589
and seeing how they change and using that potentially to
88
00:05:36.649 --> 00:05:39.649
enroll people in a clinical trial. But we are not
89
00:05:39.810 --> 00:05:42.029
yet at the point where this is ready for use
90
00:05:42.170 --> 00:05:46.620
by clinicians in the clinical care of people at risk
91
00:05:46.639 --> 00:05:47.160
for disease.
92
00:05:48.759 --> 00:05:53.069
What are the next steps towards translating these findings into
93
00:05:53.189 --> 00:05:54.310
clinical practice?
94
00:05:55.829 --> 00:05:58.230
It's a great question. I think there are several next steps.
95
00:05:58.750 --> 00:06:00.910
The one is we want to look not just in blood,
96
00:06:00.970 --> 00:06:04.329
but also in spinal fluid. The spinal fluid is closer
97
00:06:04.360 --> 00:06:06.579
to the brain and spinal cord, and there might be
98
00:06:06.620 --> 00:06:11.170
additional signals in spinal fluid that we can measure. We
99
00:06:11.230 --> 00:06:14.449
also want to bring to bear other proteomic techniques. We
100
00:06:14.470 --> 00:06:16.730
haven't touched on this yet, but in this paper, we
101
00:06:16.850 --> 00:06:20.389
used something called Olink, but there are other proteomic approaches
102
00:06:20.430 --> 00:06:22.689
that could be brought to bear. And we want to
103
00:06:22.769 --> 00:06:26.639
explore those techniques to see what are the similarities, what
104
00:06:26.660 --> 00:06:29.259
are the differences that would help give us confidence in
105
00:06:29.319 --> 00:06:32.300
some of the targets. And then once we really have
106
00:06:32.339 --> 00:06:35.420
that short list of proteins that we think are most informative,
107
00:06:35.899 --> 00:06:38.040
There's some work that needs to be done to develop
108
00:06:38.060 --> 00:06:42.560
perhaps more traditional immunoassays that are really robust and can
109
00:06:42.620 --> 00:06:48.649
be used in central laboratories or clinical laboratories that ultimately
110
00:06:48.750 --> 00:06:52.230
could get this small set of proteins out into much
111
00:06:52.310 --> 00:06:52.910
broader use.
112
00:06:54.649 --> 00:06:58.370
And what are next steps as far as the study
113
00:06:58.410 --> 00:07:01.209
goes and where do you take it from here?
114
00:07:02.300 --> 00:07:04.540
Well, the study is ongoing. One of the questions we
115
00:07:04.600 --> 00:07:08.120
could not answer in this initial look at the data is,
116
00:07:08.540 --> 00:07:11.779
are there different panels of proteins for people who carry
117
00:07:11.899 --> 00:07:15.149
different genetic risk? So if you carry gene A or
118
00:07:15.209 --> 00:07:19.089
gene B or gene C, is the biology different? And
119
00:07:19.129 --> 00:07:21.810
does the panel of proteins that we need to explore
120
00:07:21.889 --> 00:07:25.740
and monitor differ? And that's really a function. We were
121
00:07:25.860 --> 00:07:29.240
unable to address that question because we don't have enough
122
00:07:29.300 --> 00:07:33.079
people with those different genetic forms who've developed disease that
123
00:07:33.120 --> 00:07:36.029
we can answer those questions. So I think that's one
124
00:07:36.050 --> 00:07:39.069
of the important next steps, as are the other steps
125
00:07:39.110 --> 00:07:42.009
that I described, looking at spinal fluid and using some
126
00:07:42.029 --> 00:07:45.430
of these other proteomic approaches to really make sure that
127
00:07:46.189 --> 00:07:49.149
we are seeing the most informative panel of proteins.
128
00:07:50.629 --> 00:07:55.100
And for the ALS community, What hope can they take
129
00:07:55.259 --> 00:07:56.300
from this study?
130
00:07:57.379 --> 00:07:59.759
So I think there are many reasons for hope here.
131
00:07:59.779 --> 00:08:03.920
So the first is that this work represents almost 20
132
00:08:03.920 --> 00:08:06.930
years of data, but I think it's transformative in our
133
00:08:06.990 --> 00:08:11.670
understanding of disease that we can increasingly think about disease
134
00:08:11.730 --> 00:08:16.449
not based on symptoms, but based on biology. And ultimately,
135
00:08:16.509 --> 00:08:19.250
as we're developing therapeutics, we want to be able to
136
00:08:19.329 --> 00:08:23.480
intervene upon that biology. And so the hope really is
137
00:08:23.620 --> 00:08:28.670
for early intervention that's biologically informed. The analogy I'd give
138
00:08:28.689 --> 00:08:32.549
your listeners and readers is thinking about cancer. We don't
139
00:08:32.629 --> 00:08:36.169
like to treat metastatic cancer. We like to screen for cancer.
140
00:08:36.210 --> 00:08:39.509
We like to detect it early. We like to intervene
141
00:08:39.750 --> 00:08:44.519
early with therapeutics that are informed by that understanding of biology.
142
00:08:45.539 --> 00:08:48.279
In the field of ALS, we're not yet where the
143
00:08:48.299 --> 00:08:51.629
world of oncology or cancer is yet. But I see
144
00:08:51.669 --> 00:08:54.629
this as a really important step in moving us towards
145
00:08:54.669 --> 00:08:58.820
that goal of being able to think about early intervention
146
00:08:58.940 --> 00:09:00.519
and even disease prevention.
147
00:09:01.940 --> 00:09:04.139
Is there anything you would like to add? Is there
148
00:09:04.159 --> 00:09:06.519
a question I'm not asking or a topic that you
149
00:09:06.559 --> 00:09:09.759
would like people to know about the research?
150
00:09:11.379 --> 00:09:14.299
I think you've hit on the important things. You can
151
00:09:14.360 --> 00:09:17.809
probably tell from this, I'm cautious by nature and I
152
00:09:17.850 --> 00:09:21.919
don't like to overplay results. So I think The data
153
00:09:21.980 --> 00:09:24.840
are important. They tell us a lot. But what I
154
00:09:24.940 --> 00:09:27.399
also don't want is to people be going out and
155
00:09:27.440 --> 00:09:31.100
trying to get their neurologist or primary care doctor to
156
00:09:31.740 --> 00:09:34.860
order these tests in some shape or form. They are
157
00:09:34.919 --> 00:09:38.779
not yet ready for clinical use. But we are hard
158
00:09:38.820 --> 00:09:42.750
at work trying to move this forward in the same
159
00:09:42.799 --> 00:09:45.549
way that we did with neurofilament a decade ago to
160
00:09:46.049 --> 00:09:48.250
get this ready for use in the way that we're
161
00:09:48.330 --> 00:09:52.429
currently using neurofilament. in a prevention trial. So I think
162
00:09:52.769 --> 00:09:55.740
we're moving forward. There's reason for hope, but I also
163
00:09:55.779 --> 00:09:57.350
don't want to hype results.