1
00:00:00,330 --> 00:00:07,350
Welcome to the Oxford University Psychology Podcast series. My name is Daniel Morgan and I've got Professor Paul Harrison with me today.

2
00:00:07,350 --> 00:00:15,870
Good morning, Paul. Thank you very much for joining me. You have a very interesting research field into translational neuroscience where

3
00:00:15,870 --> 00:00:22,060
you have quite a broad range of research areas where you way from the lab work or

4
00:00:22,060 --> 00:00:26,640
work through to potential clinical applications and will perhaps would be interesting

5
00:00:26,640 --> 00:00:32,770
to begin by you just giving a broad overview of your of your research area.

6
00:00:32,770 --> 00:00:39,060
Well. So my research area originally was was, in fact, neuropathology.

7
00:00:39,060 --> 00:00:43,500
So having done my research degree and outside in the early part of Alzheimer's disease,

8
00:00:43,500 --> 00:00:49,770
I then wanted to apply the same kind of neuropathological techniques to schizophrenia, which is a notoriously difficult area.

9
00:00:49,770 --> 00:00:54,930
And we did a number of studies over 10 or 15 year period, which I think told us some things,

10
00:00:54,930 --> 00:01:02,690
but increasingly became apparent to me that the answer to really unravelling the pathophysiology of disease or schizophrenia didn't lie entirely,

11
00:01:02,690 --> 00:01:06,510
and perhaps even largely in studying the brains of people who'd had the illness.

12
00:01:06,510 --> 00:01:12,150
And so as a result of that, over the past 10 years or so, the research has broadened to include, as you suggested,

13
00:01:12,150 --> 00:01:23,850
a range of other approaches and techniques which inevitably taken away from questions about the biological correlates of our psychiatric syndromes,

14
00:01:23,850 --> 00:01:30,360
will get us more into understanding that that that treatments and their clinical implications.

15
00:01:30,360 --> 00:01:35,670
So that's really where the translation side is. And so, in a sense, when we see a clinical picture,

16
00:01:35,670 --> 00:01:48,840
we know that there's a whole of several layers of sort of understanding that feeds into that behaviour from neurotransmitters to psychological

17
00:01:48,840 --> 00:01:56,520
frameworks to understand some behaviours that have been specifically looking at and some of the genetic components of of that behaviour.

18
00:01:56,520 --> 00:02:02,220
So could you tell us a bit about that? Yes, that's right. So perhaps that's been the main shift in focus.

19
00:02:02,220 --> 00:02:06,450
And so because originally we did in sense case control studies where you take a series

20
00:02:06,450 --> 00:02:10,590
of brains from patients who are analysis necessarily the brains of people without.

21
00:02:10,590 --> 00:02:17,370
And you're looking for things that the occurrence of the diagnostic differences by the problems of those kind of studies, really,

22
00:02:17,370 --> 00:02:21,750
whether you do them in post-mortem brains or you do them in established living patients with analysis,

23
00:02:21,750 --> 00:02:25,620
it's very difficult to separate causes from consequences of illness.

24
00:02:25,620 --> 00:02:30,050
And the same in many ways applies to environmental studies of the risk factors for psychiatric illness.

25
00:02:30,050 --> 00:02:34,980
One of attractions in the genetics is not only that we know the major psychiatric

26
00:02:34,980 --> 00:02:39,000
standards have a substantial and often underestimated genetic component,

27
00:02:39,000 --> 00:02:43,920
but if we find genetic discoveries, essentially they can't be confounded by anything.

28
00:02:43,920 --> 00:02:52,410
You may still be very puzzled as to what they mean. But at least if you find a robust genetic association with a botanical syndrome,

29
00:02:52,410 --> 00:02:56,280
you can stop for matters into peckerwoods, which you can hang on to our studies.

30
00:02:56,280 --> 00:03:03,360
So that's really why we moved to understand the genetic basis of mechanisms like schizophrenia and bipolar disorder.

31
00:03:03,360 --> 00:03:12,750
OK, thank you. So what do we know about these sort of very broad, perhaps umbrella diagnoses of bipolar and schizophrenia?

32
00:03:12,750 --> 00:03:17,260
What do we know about the genetic contributions of that? And just broadly speaking?

33
00:03:17,260 --> 00:03:22,920
Well, the glass around half full and half empty. The simple answer is we still don't understand it at all.

34
00:03:22,920 --> 00:03:28,110
On the other hand, it is clear the research the last 10 years is Regenesis particularly to produce,

35
00:03:28,110 --> 00:03:30,660
firstly, that there is a substantial genetic component to,

36
00:03:30,660 --> 00:03:38,520
depending how you measure it, between about two thirds and three quarters of whatever is causing these illnesses is in our genes.

37
00:03:38,520 --> 00:03:45,840
We also know that about two thirds of the genes that contribute to schizophrenia also contribute to bipolar disorder and vice versa.

38
00:03:45,840 --> 00:03:50,940
And there's also an overlap genetically with autism, with learning disability and with major depression.

39
00:03:50,940 --> 00:03:56,580
So certainly a clinical sense of these drugs that really caused nature to join this.

40
00:03:56,580 --> 00:04:00,630
Many cases that are a mixture of two things is only one pure case of any diagnosis.

41
00:04:00,630 --> 00:04:08,940
That's certainly what the genetics is looking like in terms of what is the nature of the genetic basis of these various syndromes.

42
00:04:08,940 --> 00:04:11,730
Again, it's not one gene. There probably isn't even 100 genes.

43
00:04:11,730 --> 00:04:20,190
There are probably hundreds, if not thousands of genes contributing individuals small amounts to the genetic risk of these disorders.

44
00:04:20,190 --> 00:04:25,650
There might be a few rare genes, which then, if they happen to be present, are rare genetic abnormalities, which,

45
00:04:25,650 --> 00:04:34,380
if they happen to be presented to a patient, may make that person's illness slightly more attributable to a small number of penetration.

46
00:04:34,380 --> 00:04:45,030
So there's many genes that have perhaps a small impact on on on the risk of developing these these conditions.

47
00:04:45,030 --> 00:04:52,110
So, I mean, that makes your job very complex, doesn't it, because you're looking at so many different genes and all of if you look at any particular

48
00:04:52,110 --> 00:04:58,050
one gene that it might lead to somebody having a condition or not having condition.

49
00:04:58,050 --> 00:05:03,980
Are you looking at how they say. They say with one another or you're looking at them individually.

50
00:05:03,980 --> 00:05:07,730
So you're right, you've you've you've touched upon what is really on the key issues in matter

51
00:05:07,730 --> 00:05:12,710
of just how do you prioritise the many genetic discoveries and implications,

52
00:05:12,710 --> 00:05:20,810
because it's now quite quick to discover or to survey the genome with the Human Genome Project, the techniques that go with that.

53
00:05:20,810 --> 00:05:27,020
So it's very easy to identify hundreds or thousands of genes which are individually, statistically doing something.

54
00:05:27,020 --> 00:05:33,650
But then following those up and understanding the biological implications is very difficult and does need a lot of prioritisation.

55
00:05:33,650 --> 00:05:39,410
And the questions of what kind of grounds do you prioritise and what kind of models do you have to have for school,

56
00:05:39,410 --> 00:05:45,320
the genetic architecture of these conditions to help you make rational decisions about what to study.

57
00:05:45,320 --> 00:05:50,720
Now, some researchers are going for these rare, very rare, but if present,

58
00:05:50,720 --> 00:05:56,750
very penetrant genetic factors on the grounds that that gives you attraction to biology.

59
00:05:56,750 --> 00:06:02,720
Other people say from an epidemiological perspective that they're not important enough overall to invest it.

60
00:06:02,720 --> 00:06:07,070
We need to go for these common genes. And even that effects are very small.

61
00:06:07,070 --> 00:06:11,870
That's ultimately where the real breakthroughs need to be made. You can then say,

62
00:06:11,870 --> 00:06:19,430
do I take an individual gene or do I hypothesise that these genes interact with each other and you go for some kind of higher level or sometimes

63
00:06:19,430 --> 00:06:28,670
good points of convergence of genes and you go for what you think of the biochemical pathways that be downstream of a lot of these genes.

64
00:06:28,670 --> 00:06:30,530
The metaphor that sometimes uses you know,

65
00:06:30,530 --> 00:06:36,260
you try and go after every trick or treat when you go for the big river or all the tributaries have joined up to have a bigger effect.

66
00:06:36,260 --> 00:06:39,590
And again, at the moment, I think there's different people applying different strategies.

67
00:06:39,590 --> 00:06:45,680
And it remains to be seen which if mean, is the more fruitful and comes to fruition.

68
00:06:45,680 --> 00:06:50,030
So a lot of exploration needs to be done about these many different genes and

69
00:06:50,030 --> 00:06:57,710
their contribute contribution towards what we recognise as psychiatric diagnoses.

70
00:06:57,710 --> 00:07:03,260
But I guess the question comes as to the nature nurture question,

71
00:07:03,260 --> 00:07:11,150
and are you doing any any sort of investigations on on how particular house Constellation's

72
00:07:11,150 --> 00:07:20,660
or groups of genes might interact with with adverse circumstances in people's lives?

73
00:07:20,660 --> 00:07:24,920
For sure. As you said, it's another level of complexity because the evidence is that many of these

74
00:07:24,920 --> 00:07:28,580
genes are probably not having a direct and immediate effect on disease risk.

75
00:07:28,580 --> 00:07:30,350
They may well be affecting individuals.

76
00:07:30,350 --> 00:07:38,390
Responses to environmental events and environmental events could be anything from our infection to smoking cannabis to having a head injury.

77
00:07:38,390 --> 00:07:44,480
And, of course, experimentally gets even more difficult when you try and combine genetic factors with environmental factors, as you can imagine,

78
00:07:44,480 --> 00:07:50,810
because of things with combinations of things rapidly multiply beyond any meaningful control,

79
00:07:50,810 --> 00:07:57,680
we as a generally have stayed away from the environmental side of things just to try and retain a degree of focus.

80
00:07:57,680 --> 00:08:02,470
But with these temperatures are all CitiPower here we have an interesting interaction

81
00:08:02,470 --> 00:08:06,020
and one of the particular genes with with tetrahydrocannabinol the active ingredient of

82
00:08:06,020 --> 00:08:11,540
cannabis and exploring some of the mechanisms that may underlie the apparent association

83
00:08:11,540 --> 00:08:16,520
between your genetic makeup and the effect of smoking cannabis earlier in your life,

84
00:08:16,520 --> 00:08:20,180
has your risk of developing schizophrenia later. So that's an example.

85
00:08:20,180 --> 00:08:32,240
Some of got a gene environment interaction. So the nature nurture question remains sort of a lofty, lofty ideal and research at the moment.

86
00:08:32,240 --> 00:08:33,080
It certainly does.

87
00:08:33,080 --> 00:08:39,860
The problem, as I say, is that it's now possible to measure the genome very quickly in very large numbers of people to measure the environment,

88
00:08:39,860 --> 00:08:44,030
the people's environment, to the necessary level of detail.

89
00:08:44,030 --> 00:08:52,160
And longitudinally, not just cross-section, of course, is a far more expensive and long term and frankly, unfeasible operation.

90
00:08:52,160 --> 00:08:55,990
So I think the genes, the environment, bit of risk for psychiatric illnesses will go on,

91
00:08:55,990 --> 00:09:02,570
be more mysterious in some ways than the genetics will be just for that practical reason.

92
00:09:02,570 --> 00:09:12,620
So moving on to thinking about practical implications of your your research on genetics and how that might apply to the clinical room,

93
00:09:12,620 --> 00:09:22,790
you've written a bit about how what you've understood about the genetic contributions to mental illness might impact on on Ansonia medications.

94
00:09:22,790 --> 00:09:25,130
Could you tell us a bit about that? Yes.

95
00:09:25,130 --> 00:09:32,720
So I think potentially genetics can help treatment, but not necessarily exclusively drug treatments in several ways.

96
00:09:32,720 --> 00:09:38,570
One is it may turn out there are genetic predictors of response to the drugs we have available at the moment.

97
00:09:38,570 --> 00:09:45,540
Let's call pharmaco genomics. And again, there are some data, but nothing which at the moment means that, as you know,

98
00:09:45,540 --> 00:09:52,220
in clinical practise, we routinely would measure something genetic to help us decide which drug to use.

99
00:09:52,220 --> 00:10:00,060
But that's more interesting. You had given that we know current drugs have many, many limitations, is whether the disease could help us identify new.

100
00:10:00,060 --> 00:10:03,600
Or new drug targets, and I think, again,

101
00:10:03,600 --> 00:10:10,950
it's too early to say if you look at a few of the many genes that we have to think schizophrenia as an example,

102
00:10:10,950 --> 00:10:15,730
although there are many genes, they're not independent of each other in terms of what we think those genes are doing.

103
00:10:15,730 --> 00:10:20,190
So it was a clear convergence of many of the genes on certain core pathways,

104
00:10:20,190 --> 00:10:25,140
some of which were already potential drug targets for other reasons and others which are novel.

105
00:10:25,140 --> 00:10:33,570
So, for example, treatment and energy receptor signalling and synaptic plasticity seem to be common effects of a number of the risk genes

106
00:10:33,570 --> 00:10:41,470
and strategies to try and normalise or improve glutamate transmission had already been under development in the field.

107
00:10:41,470 --> 00:10:47,430
So one would imagine, I would hope that the genetic discoveries may both increase the focus on that and

108
00:10:47,430 --> 00:10:52,050
sharpen the focus to decide exactly where within what is an exciting pathway.

109
00:10:52,050 --> 00:10:56,940
The drugs of the next generation can more usefully targeted civilians and have a

110
00:10:56,940 --> 00:11:00,360
focus on some of the genes across schizophrenia and bipolar disorder and calcium.

111
00:11:00,360 --> 00:11:05,460
Certainly so calcium channel antagonists, for example, which we already have in medicine,

112
00:11:05,460 --> 00:11:10,740
cardiology, for example, if they have any as yet untapped potential, is novel drug treatments.

113
00:11:10,740 --> 00:11:13,320
But whether, you know, whether that's going to prove to be the case,

114
00:11:13,320 --> 00:11:19,950
whether those drugs will turn out to be effective and safe and meet all the usual problems of developing new psychiatric drugs remains to be seen.

115
00:11:19,950 --> 00:11:26,160
And I think the genetics, at the very least, is giving the field of psychiatric drug discovery a well needed boost.

116
00:11:26,160 --> 00:11:32,730
And it's helping, I think, reverse the previous decline of interest by the pharmaceutical industry in our field.

117
00:11:32,730 --> 00:11:43,410
We work in psychiatry using a descriptive diagnostic classification system where we have these well, perhaps some some might argue,

118
00:11:43,410 --> 00:11:56,310
quite old fashioned diagnoses of schizophrenia and bipolar and perhaps depression and anxiety in your in your work of understanding

119
00:11:56,310 --> 00:12:07,800
genetics and how this sort of quite a common thread of genetic contributions and sort of mental disorder across the spectrum.

120
00:12:07,800 --> 00:12:13,650
Have you had any thoughts about how or any thoughts about our current diagnostic system?

121
00:12:13,650 --> 00:12:21,360
Do you think it needs to be shaped in light of your understanding of genetics?

122
00:12:21,360 --> 00:12:26,700
Well, I think most people I'm quite sure our current system is wrong, but I'm not quite sure what I would do to make it any better.

123
00:12:26,700 --> 00:12:31,350
We can make it different from what any better because the genetics is not deterministic.

124
00:12:31,350 --> 00:12:38,640
It's not like a single gene disorder where you can start to redefine the syndrome by the genetic mutation, for example.

125
00:12:38,640 --> 00:12:48,030
It's naive to think we're ever, ever going to get to that sort of level of redescribed or syndromes on some biological basis.

126
00:12:48,030 --> 00:12:55,470
However, I think we can wield a lot better than the current time descriptive form of classification.

127
00:12:55,470 --> 00:13:01,680
So, for example, if we I mean, let's say we compare psychosis as a broad category with the book category,

128
00:13:01,680 --> 00:13:07,860
we still use the term dementia as a descriptive syndrome to describe see the constellation of symptoms.

129
00:13:07,860 --> 00:13:14,490
But we have individual dementias, the group defining the 4G and increased according to their genetic origins.

130
00:13:14,490 --> 00:13:20,100
Now, I would imagine what will happen to psychosis is it would begin to be picked apart in that kind of way.

131
00:13:20,100 --> 00:13:27,420
The majority of cases will continue to be idiopathic in a causal sense for a while, and they will need to continue to be described as such.

132
00:13:27,420 --> 00:13:34,470
But there will be, I suspect, individual rare causes of psychosis which become defined by their aetiology.

133
00:13:34,470 --> 00:13:37,320
Let's take another very topical and controversial at the moment,

134
00:13:37,320 --> 00:13:45,810
the possibility of some cases of psychosis due to anti autoantibodies and essentially autoimmune disorder presenting as schizophrenia.

135
00:13:45,810 --> 00:13:49,920
Now, that raises fascinating scientific and most logical questions.

136
00:13:49,920 --> 00:13:54,870
But that's, I think, an example where it may be that a small portion of cases become viewed differently and,

137
00:13:54,870 --> 00:14:01,740
of course, treated differently if the science or the evidence that holds up is fascinating.

138
00:14:01,740 --> 00:14:09,810
Hearing you talk about potential futures of of understanding in psychiatric presentations,

139
00:14:09,810 --> 00:14:16,430
in clinical practise and how your work is informed that I am speaking to you,

140
00:14:16,430 --> 00:14:29,370
it is is clear to me that your your your research is is very informs your practise, even though it's very scientifically based.

141
00:14:29,370 --> 00:14:39,930
It's very almost a pure science. It still has an ability to sort of make you think in a fresh way about about what when you see patients.

142
00:14:39,930 --> 00:14:46,980
And it would be just really interesting to hear how how you feel about being an academic psychiatrist.

143
00:14:46,980 --> 00:14:50,520
What do you think it brings brings to your professional life?

144
00:14:50,520 --> 00:14:58,090
And what would you say to actually aspiring academic psychiatrists?

145
00:14:58,090 --> 00:15:06,640
Well, I love the combination, I think it's a real privilege to both be a doctor, a clinician and sort of have all the, I guess,

146
00:15:06,640 --> 00:15:15,700
the trappings that go with that whilst also having the academic freedom to pursue or to me are interesting and challenging intellectual problems.

147
00:15:15,700 --> 00:15:20,890
And of course, the problem is that in the end, something might make a difference to somebody if in the long term future.

148
00:15:20,890 --> 00:15:29,500
My my contribution has been minor. I think at least 50 battles have picked up in the morning and feel motivated to to go on working both tonight.

149
00:15:29,500 --> 00:15:35,050
So I would encourage anybody who's interested either in psychiatry or in my case, in the neuroscience.

150
00:15:35,050 --> 00:15:41,860
The combination really is very rewarding. And there are still, despite the difficulties in funding, both on the clinical and the research,

151
00:15:41,860 --> 00:15:45,940
there are still real opportunities for a few bright and often hard working enough.

152
00:15:45,940 --> 00:15:53,900
You know, I would I would certainly be dogged in your pursuit of that in terms of career advice.

153
00:15:53,900 --> 00:15:57,670
Again, I would encourage young people to to read around, to look around,

154
00:15:57,670 --> 00:16:03,010
to ask around, to meet people in my personal experiences that certainly work for me,

155
00:16:03,010 --> 00:16:06,550
finding a person or feel that inspires a particular thing,

156
00:16:06,550 --> 00:16:14,750
a person who really can get you to the right starting point for your career and then goes on inspiring you.

157
00:16:14,750 --> 00:16:21,940
I've had the privilege to work with a number of really inspiring people throughout my career, and I think that's my biggest advice being inspired.

158
00:16:21,940 --> 00:16:25,970
Professor Paul Arrison has been great speech here this morning. Thank you very much for your time.

159
00:16:25,970 --> 00:16:30,220
Pleasure. Thank you. And thank you for tuning in to the Oxford University Psychiatry podcast series.

160
00:16:30,220 --> 00:16:33,123
We hope you listen to some more. Thank you. Goodbye.