1
00:00:00,750 --> 00:00:03,750
So can you just start by saying your name and what your position is here?

2
00:00:04,290 --> 00:00:11,490
My name is Martin Landry. I'm Professor of medicine and epidemiology in the Nuffield Department of Population Health, University of Oxford.

3
00:00:11,670 --> 00:00:18,360
Great. And without giving me your entire life history, can you just give me a quick summary of how you got to where you are now?

4
00:00:19,230 --> 00:00:23,790
Yes. I grew up not far from here. I grew up 20 miles west of here in a village called Brampton.

5
00:00:23,820 --> 00:00:29,910
My father was the village GP. My mother was a part time anaesthetist, so medicine was all around me.

6
00:00:31,530 --> 00:00:38,440
I went to school, finish it up into school again, just down the road, and then went to medical school.

7
00:00:38,460 --> 00:00:46,320
I knew I wanted to be a doctor from the age of about ten or 11 and then from medical school getting ready to go to medical.

8
00:00:46,350 --> 00:00:51,430
I went to medical school in University of Birmingham, right South 30th Reunion tomorrow.

9
00:00:52,800 --> 00:00:58,950
A dean of medicine at Birmingham left sufficient money for every time the there

10
00:00:58,950 --> 00:01:04,230
to be a 30th anniversary every year my parents went to their 30th 30 years ago.

11
00:01:04,290 --> 00:01:10,050
Now my turn this time. So I recommend that to whatever Dean of Medicine wants to do with their legacies this time around.

12
00:01:10,440 --> 00:01:15,760
It's a very nice tradition. So I went to Birmingham University to study medicine.

13
00:01:16,350 --> 00:01:23,160
Wasn't entirely sure what I wanted to do, but increasingly got interested in in-hospital medicine and increasingly in prescribing.

14
00:01:23,340 --> 00:01:26,840
How do you know which drugs to prescribe? When to prescribe, how much to prescribe?

15
00:01:26,850 --> 00:01:31,500
How do you know it works? How do you know about what the side effects are, what to look out for, and so on?

16
00:01:32,100 --> 00:01:35,550
And from that you quickly get into clinical trials because that's how you find out.

17
00:01:36,330 --> 00:01:41,250
Can you just give me an idea of what the what the state of prescribing was at that time?

18
00:01:43,310 --> 00:01:48,440
Well, you know, we did it perfectly. No, say the prescribing in the time.

19
00:01:48,450 --> 00:01:55,020
I mean, I had a brilliant clinical pharmacology lecturer in Birmingham called Professor Martin Kendall,

20
00:01:55,710 --> 00:02:02,820
whose approach to teaching was it was ahead of its time.

21
00:02:02,820 --> 00:02:06,930
It was very much, he imagined, a real life situation,

22
00:02:07,500 --> 00:02:16,469
a ladies sitting in front of you in your surgery saying that she's got chest pain and you're going to prescribe well, what are you going to get?

23
00:02:16,470 --> 00:02:22,830
And what? He'd ask the question, what are the ten things you're going to tell the lady about this prescription for Duchenne?

24
00:02:23,490 --> 00:02:28,319
And he used to get used to these roadshows. You get so gitanas a treatment for angina.

25
00:02:28,320 --> 00:02:30,120
It's a spray that you put under your tongue.

26
00:02:31,380 --> 00:02:35,910
I can give you the ten things it can cause headache, but it can deal with the angina and so on and so forth.

27
00:02:36,450 --> 00:02:39,090
But the way he did it was he would have these roadshows.

28
00:02:39,510 --> 00:02:47,460
We never went on the road, but he put ten students each week up on the stage and asked them a series of these sorts of questions,

29
00:02:48,030 --> 00:02:52,590
and those ten students knew who they were going to be in advance. They sort of knew what the questions were going to be in advance.

30
00:02:52,980 --> 00:02:59,460
But it very much put that sort of combination of you had to understand how drugs worked and the sort of science of it.

31
00:03:00,180 --> 00:03:08,880
You also had to understand how medicine works, and you were also under some element of pressure because they were with the big professor and,

32
00:03:08,940 --> 00:03:15,690
you know, 200 medical students in front of you. It was a very, very good training for thinking about the decision.

33
00:03:15,690 --> 00:03:19,950
That is a fundamental decision of medicine. Do I prescribe a treatment or don't I?

34
00:03:20,370 --> 00:03:25,440
And if I do prescribe a treatment, what treatment I want to prescribe, how do I prescribe it, etc.?

35
00:03:26,670 --> 00:03:32,310
So that's sort of that. That was really what got me interested into clinical pharmacology.

36
00:03:32,940 --> 00:03:37,560
And after a few years of junior doctor jobs, I went back to Birmingham.

37
00:03:38,190 --> 00:03:42,780
In fact, at his lecture as his clinical lecturer, he took an enormous risk on me.

38
00:03:42,780 --> 00:03:48,120
I didn't have a PhD, I'd only just passed my membership of all college positions.

39
00:03:48,120 --> 00:03:54,930
So the necessary exam to be a registrar. I had no publications, no papers, no no grants, no nothing.

40
00:03:56,910 --> 00:04:04,229
Now, he was either had incredible foresight or took a huge gamble or was just desperate.

41
00:04:04,230 --> 00:04:13,170
But whatever way it was he he took me on as his as his as registrar, as clinical lecturer, as his PhD student.

42
00:04:13,170 --> 00:04:16,350
And over the subsequent four and a half years or so,

43
00:04:17,580 --> 00:04:23,730
I was able to get my specialist training so I could become a consultant in clinical pharmacology in general medicine.

44
00:04:23,970 --> 00:04:25,890
I was able to get a Ph.D. from Birmingham,

45
00:04:26,790 --> 00:04:35,399
and I was also able to do the sort of higher education teaching qualifications that I came as part of as well.

46
00:04:35,400 --> 00:04:44,400
So I wanted to do those things in five years whilst living still in West Oxfordshire with a with by then two children under the age of two.

47
00:04:46,020 --> 00:04:49,410
Well I think it probably fair to say my wife had two children under the age of two.

48
00:04:51,060 --> 00:04:53,940
So I'm incredibly grateful to her for her support all the way through.

49
00:04:54,600 --> 00:04:58,649
But that really got me into into as I say, into the further into clinical pharmacology.

50
00:04:58,650 --> 00:05:03,640
But I got to a. Well, I thought it's all about these big clinical trials.

51
00:05:03,800 --> 00:05:10,390
Where are those big clinical trials being done? They're doing done in Oxford by Rory Collinson and Richard Richard Peto.

52
00:05:10,390 --> 00:05:20,380
Really? And some point during that, so ten years between going to medical school and finishing up Ph.D. and and and so on.

53
00:05:21,370 --> 00:05:25,870
I've increasingly started reading the results of the big clinical trials as they were coming out.

54
00:05:25,900 --> 00:05:37,719
There was one from the late 1980s called the Aces to Trial, named after the paper, which was well known as groundbreaking, is really large,

55
00:05:37,720 --> 00:05:43,690
really simple, very little paper work, and got incredibly impressive results and completely changed the face of cardiology.

56
00:05:43,690 --> 00:05:46,750
How we treat acute heart attack within weeks.

57
00:05:47,200 --> 00:05:50,230
So what? What? Just. That's just a little bit.

58
00:05:52,120 --> 00:05:56,260
How was heart attack treated previously? What was the rationale?

59
00:05:56,440 --> 00:05:58,810
What were the drugs being tested and what's the. Sure.

60
00:05:58,840 --> 00:06:10,510
So so prior to that, if you go back to the 1960s, well before my time that whilst my father was a GP,

61
00:06:11,950 --> 00:06:18,310
patients with a heart attack which were given bedrest at home come the early 1970s or so.

62
00:06:18,550 --> 00:06:24,370
Then coronary care units started to be getting established to bring these patients into hospital.

63
00:06:24,710 --> 00:06:29,560
When you brought them into hospital, you didn't do much more than give them bedrest.

64
00:06:30,160 --> 00:06:32,470
People were often in for ten days or a fortnight,

65
00:06:34,060 --> 00:06:41,020
and there wasn't much in the way of drugs that you could give people by the time it got to the mid 1980s.

66
00:06:41,260 --> 00:06:44,560
We hadn't advanced a whole load from there really.

67
00:06:44,920 --> 00:06:56,440
But this trial said if we thin the blood with aspirin, maybe knows aspirin, but it has an action against the platelets, stops and clumping.

68
00:06:56,770 --> 00:07:00,010
And if we were to dissolve the clot, which is the cause of the heart attack,

69
00:07:00,020 --> 00:07:05,200
a clot forming in the in the coronary arteries so that blood doesn't get to a particular bit of heart muscle.

70
00:07:05,620 --> 00:07:11,920
We could dissolve that clot. Maybe that would actually improve your chances of survival from the acute heart attack.

71
00:07:12,430 --> 00:07:17,620
Now, there were a lot of small trials that had been done perhaps up to that point,

72
00:07:17,620 --> 00:07:21,279
which if you added them all together, did a meta analysis, would say, actually,

73
00:07:21,280 --> 00:07:24,490
both of those treatments look like they're good treatments,

74
00:07:25,180 --> 00:07:31,630
but a collection of small trials hadn't been enough to actually convince doctors that this was the right thing to do.

75
00:07:32,410 --> 00:07:34,940
Not entirely surprised by that, because these are drugs.

76
00:07:35,140 --> 00:07:42,820
The idea of giving to somebody a drug that dissolves blood clots together with a drug that prevents blood clots forming.

77
00:07:43,390 --> 00:07:48,510
That's that sort of shouts. There's going to be bleeding and sure enough, there is bleeding.

78
00:07:48,520 --> 00:07:53,620
So when you see as a pay, as a doctor, one patient and you give them those treatments,

79
00:07:54,010 --> 00:07:58,540
you either see nothing happen because they get better and they and they survive and they go home from hospital,

80
00:07:58,690 --> 00:08:04,190
in which case that sort of just what you expecting to happen. So nothing is a success in that sense.

81
00:08:04,750 --> 00:08:11,110
Or they have these they can have a massive catastrophic bleed and require blood transfusion or even die from bleeding.

82
00:08:11,470 --> 00:08:13,090
So as a doctor, seeing one patient,

83
00:08:13,090 --> 00:08:19,810
you don't you can't tell the benefits because all that the benefit is is just the person continues to recover and survive.

84
00:08:20,080 --> 00:08:22,570
But you do see the harms in the harms are quite nasty.

85
00:08:22,900 --> 00:08:29,500
So it's not a surprise that doctors weren't to my mind that doctors weren't entirely comfortable with giving these treatments.

86
00:08:30,220 --> 00:08:34,330
Well, Richard, Peter and Roy Collins did was say, well, okay, this looked promising.

87
00:08:34,330 --> 00:08:44,049
Let's do a really big trial of I think was 18,000 people or so and randomise them toss a coin between getting to coin age or placebo and also

88
00:08:44,050 --> 00:08:49,630
between getting aspirin or placebo and do it so that some people get both and some people get neither and some people get one or the other.

89
00:08:49,990 --> 00:08:54,970
And when they did that, they found that the streptococci still in the blood.

90
00:08:55,270 --> 00:09:00,399
And so dissolving the clot reduced the risk of heart attack and the aspirin sending the blood reduced the risk of heart attack.

91
00:09:00,400 --> 00:09:07,180
And if you get in both together, they both reduce the risk of heart attack and you almost halve the risk of dying following an acute heart attack.

92
00:09:07,420 --> 00:09:12,520
And that was so clear that then practice changed within the following six months.

93
00:09:12,520 --> 00:09:18,820
And if you look at sales of the drugs or prescribing or whatever of our doctor surveys from about that time,

94
00:09:19,360 --> 00:09:27,370
this one study changed from huge uncertainty and a promising treatment that nobody was using into essentially it was definitive,

95
00:09:27,700 --> 00:09:32,590
almost to the point that you were considered a not very good doctor if you weren't following that evidence.

96
00:09:33,010 --> 00:09:35,350
So the key thing about that study was its size.

97
00:09:35,770 --> 00:09:43,090
It was size, it was randomised this coin toss and it was really simple to do on the front page of the protocol.

98
00:09:43,390 --> 00:09:51,969
It said by far the greatest contributor to success of this trial will be how easy it is for

99
00:09:51,970 --> 00:09:58,210
front line busy doctors and nurses to actually enter this their patients into this study,

100
00:09:58,450 --> 00:10:05,330
hence the workload. Has to be as absolutely minimal the extra workload.

101
00:10:05,720 --> 00:10:11,120
So it has to be practical because I mean, I was around those times perhaps a couple of years after.

102
00:10:11,120 --> 00:10:17,090
I mean, I was seeing exactly those patients. We were not getting much sleep, we didn't have much time.

103
00:10:19,010 --> 00:10:24,680
And the idea of doing these extra, extra things on top, it was fine if it was, you know,

104
00:10:24,710 --> 00:10:31,040
one page of questions, but if it had been a book or whatever, it would have just never, never got done.

105
00:10:31,430 --> 00:10:36,139
However, how much one was paid or shouted about it or whatever else.

106
00:10:36,140 --> 00:10:39,380
So the simplicity was also a really key driving point.

107
00:10:39,680 --> 00:10:45,770
So that that chart, that so much trial came out was a medical student, there was a series of other studies.

108
00:10:45,770 --> 00:10:49,760
It was somewhat like it bit different over the over the subsequent ten years.

109
00:10:50,210 --> 00:10:53,570
But when I saw that paper, everybody was saying, well, look at the results.

110
00:10:53,570 --> 00:10:56,570
And yeah, the results are really impressive and have a big effect.

111
00:10:56,570 --> 00:11:01,820
And, you know, everybody look the results over and over and over again. And I've been tested on them more times than you could imagine.

112
00:11:03,350 --> 00:11:08,260
But I looked at it and I thought, Isn't it interesting that I looked at it?

113
00:11:08,330 --> 00:11:12,830
Yeah. They also said something about the John Carter of the John Ratcliffe Hospital, Oxford,

114
00:11:13,550 --> 00:11:20,780
and I thought there's some there's two or three people sitting in a portacabin outside the john in the car park of the John Ratcliffe Hospital,

115
00:11:21,200 --> 00:11:24,770
who one day came to work and they said, you know, what's needed is a study this big.

116
00:11:25,820 --> 00:11:29,420
Okay. And number two, we're going to do it. And number three, they then went on to do it.

117
00:11:29,420 --> 00:11:35,330
And I was wrong about the portacabin. And it turns out that they were in a disused X-ray room and in the John Ratcliffe Hospital,

118
00:11:35,540 --> 00:11:38,839
which I think they weren't they'd commandeered and weren't supposed to be in.

119
00:11:38,840 --> 00:11:42,130
But the rest of the story is pretty much how I imagined it.

120
00:11:42,140 --> 00:11:48,650
Pretty much true. So it was the what I suddenly got interested in was how does that actually happen?

121
00:11:48,650 --> 00:11:51,800
It's not yet. The result is really impressive and the result matters.

122
00:11:53,510 --> 00:11:56,360
But I was interested to go behind sort of behind the scenes.

123
00:11:56,360 --> 00:12:04,460
It's sort of I guess there are people who see a wonderful West End performance, which is incredible, but they go behind the scenes and they think,

124
00:12:04,820 --> 00:12:14,090
I wonder how the, you know, the puppeteer actually works or, you know, what did the stage designer have in his mind when he put that together?

125
00:12:14,090 --> 00:12:17,270
How did the producer think I could assemble all these people and pull this off?

126
00:12:17,570 --> 00:12:22,340
And it was I guess it was that sort of that sort of intrigue which got me into it.

127
00:12:22,430 --> 00:12:26,239
So you were interested in the boxes and the barcodes and. Yeah, yeah, yeah, yeah.

128
00:12:26,240 --> 00:12:32,209
How did you you know, you had exactly the boxes, the barcodes, the, you know, don't ask more questions.

129
00:12:32,210 --> 00:12:40,580
You knew then you have to and so on. So when I came to this, we're now through to the late, very late 1990s.

130
00:12:41,450 --> 00:12:45,139
I was finishing off my by this time my clinical lectureship in Birmingham.

131
00:12:45,140 --> 00:12:53,660
I was finishing off my PhD in Birmingham and whatever, partly because my PhD was on why people with kidney disease get heart disease.

132
00:12:53,660 --> 00:13:01,730
And I've been speaking to Colin Agents who is here in Oxford about that because it was an area he had an interest in.

133
00:13:01,970 --> 00:13:11,209
So I sort of knew him a bit, but it became clear that the right thing to do if I wanted to pursue my interest in these really big clinical trials,

134
00:13:11,210 --> 00:13:15,980
was go and work for the people who pretty much invented them.

135
00:13:15,980 --> 00:13:23,120
I don't like me for saying that, but pretty much invented these sort of large, simple, so-called simple randomised trials.

136
00:13:23,960 --> 00:13:31,520
People used to call them large and simple. I think the aim is that the simplicity is what is how it feels on the ground.

137
00:13:31,760 --> 00:13:36,020
The complexities all dealt with so that the people on the ground don't have to worry about it.

138
00:13:36,980 --> 00:13:41,150
And that's what one really learns. I mean, again, it's like a, you know, a great theatre performance.

139
00:13:41,540 --> 00:13:49,580
There's huge energy and activity going on before it happens, during it afterwards and whatever behind the scenes.

140
00:13:49,820 --> 00:13:56,180
But as the audience, all you see is, is, is, is, is the the simplicity, the beauty, if you like.

141
00:13:58,230 --> 00:14:04,560
So was a job advertised all that you fancy?

142
00:14:04,590 --> 00:14:11,549
Can we do it? Yeah. So the job was advertised. Actually, I'm just gonna take the microphone off a moment,

143
00:14:11,550 --> 00:14:15,870
and I'm going to read you a piece of paper just because I need to get it from the other side.

144
00:14:16,140 --> 00:14:20,590
Okay. Yeah.

145
00:14:20,890 --> 00:14:27,520
So I've got a piece of paper in my, in my, in front of me now dated the 11th of November 1999,

146
00:14:27,520 --> 00:14:33,130
from calling the agents here in Oxford to probably Collins and Richard Peter really Dr. Martin Landry.

147
00:14:34,900 --> 00:14:41,860
And it goes sort of starts with an introduction and about what my background was and so on.

148
00:14:41,920 --> 00:14:45,459
They said, I think it would be useful to get him down to meet us.

149
00:14:45,460 --> 00:14:53,440
When you were both around, could I go ahead and organise a meeting in December and underneath it is is in Richard Peters.

150
00:14:53,440 --> 00:15:00,310
Individual handwriting is okay as long as it doesn't get taken as a definite a job offer are paid.

151
00:15:02,680 --> 00:15:08,180
So I showed that I took that out and showed that to Richard. Not no.

152
00:15:08,650 --> 00:15:14,139
Not long after I got my knighthood last year said you might not remember this yet.

153
00:15:14,140 --> 00:15:23,740
So then there was a job advertised for people, for someone to come and play a sort of junior clinical role on one of the big trials,

154
00:15:24,090 --> 00:15:27,220
which Rory had been planning on cholesterol lowering the heart protection study.

155
00:15:28,130 --> 00:15:33,700
So I joined for the last year of that they actually the memory I think they

156
00:15:33,700 --> 00:15:39,579
interviewed four people they they pointed to one was Professor Louise Bowman,

157
00:15:39,580 --> 00:15:42,790
who is also here and and then me.

158
00:15:42,790 --> 00:15:50,200
And so we started within a month of each other. And the reason I have worked very closely for the last 22 years or whatever it is,

159
00:15:52,420 --> 00:15:57,010
it was a very, very good time to join the unit's clinical trial service unit.

160
00:15:58,180 --> 00:16:05,950
The reason for that is that the heart protection study was studying whether simvastatin a statin would not only

161
00:16:05,950 --> 00:16:13,719
lower cholesterol but would reduce the risk of cardiovascular disease in a whole range of different sorts of people,

162
00:16:13,720 --> 00:16:17,980
people who had or hadn't had a heart attack and people who had or hadn't had a stroke and so on.

163
00:16:18,490 --> 00:16:24,459
And in particular, perhaps one of the most important questions was, was this a treatment with lowering cholesterol,

164
00:16:24,460 --> 00:16:28,330
something you only did to people with high cholesterol or something you could also do with

165
00:16:28,330 --> 00:16:32,590
people who got medium cholesterol or what we thought of at the time as low cholesterol.

166
00:16:33,370 --> 00:16:38,530
And when the results came out, it turned out that didn't really matter which sort of people you took.

167
00:16:38,950 --> 00:16:47,859
Lowering their cholesterol was a good thing to do. This was this study was in its final year or so for like ten or 15 months.

168
00:16:47,860 --> 00:16:57,580
When Louise and I joined, they had a there's a process in clinical trials called clinical events adjudication.

169
00:16:57,760 --> 00:17:04,030
This is basically a form gets filled in saying by the research to the to the participants.

170
00:17:04,030 --> 00:17:09,009
So very nice to see you. Can you tell me, have you ever had a heart attack since we last saw you?

171
00:17:09,010 --> 00:17:11,200
A stroke since we last show you and so on and so forth.

172
00:17:11,740 --> 00:17:18,370
And then on the basis that question, if you take a yes box, then somebody said, well, that's interesting, but that's not exactly definitive.

173
00:17:18,700 --> 00:17:25,359
So we now need to go and get all the hospital notes from whichever hospital this person is in, photocopy them, anonymize them and all that.

174
00:17:25,360 --> 00:17:29,259
Send them into the coordinating centres to sue.

175
00:17:29,260 --> 00:17:36,970
And somebody needs to go through those and by standard criteria say this meets our definition or the trial's definition of a heart attack and so on.

176
00:17:37,810 --> 00:17:42,850
Now by this stage for memory, they had something like 18,000 of these reports.

177
00:17:43,830 --> 00:17:49,510
From memory, they hadn't done many more than 1003 thousand of the people in the trial had died.

178
00:17:49,510 --> 00:17:55,390
So we had death certificates and all the details on that, too. I didn't see any of those and they all need to be done in duplicates.

179
00:17:55,810 --> 00:17:59,410
So Louise and I were essentially appointed to clear that backlog,

180
00:17:59,530 --> 00:18:11,260
and we spent months and many, many hours going through each of those each of those reports.

181
00:18:12,100 --> 00:18:17,679
But what it meant was we really saw the sort of the progression of a trial.

182
00:18:17,680 --> 00:18:22,450
We saw some of that back in back stage activity, but we saw this trial come to fruition.

183
00:18:22,720 --> 00:18:31,090
We then saw in November, November 2001, when the results were announced, the extraordinary impact they had.

184
00:18:32,710 --> 00:18:37,450
And again, people put their trust in me and then us.

185
00:18:37,840 --> 00:18:41,829
So because the results were being announced at the American Heart Association's big annual

186
00:18:41,830 --> 00:18:48,139
meeting in front of tens of thousands of cardiologists in in Los Angeles to see the data.

187
00:18:48,140 --> 00:18:54,970
And we had Internet, but we certainly didn't have any sort of team Zoom or any other form of sort of remote broadcasting and so on,

188
00:18:56,470 --> 00:19:00,910
Rory and Jane Armitage and others, we're going to present the results in this meeting.

189
00:19:02,050 --> 00:19:08,260
And they were they're going to cover the US media, which included the US science and and pharma media and so on,

190
00:19:08,260 --> 00:19:17,230
which is really important to get the message out. Louise and I went for the sort of the weekend of preparations for this.

191
00:19:19,090 --> 00:19:24,270
Then we were told we had to come back to the UK to present them to the media.

192
00:19:24,280 --> 00:19:31,120
So the UK, the leader leading the stories for the interviews with Fergus Walsh and all those.

193
00:19:32,470 --> 00:19:38,560
Suddenly it was it was me and Louise up in front of the cameras announcing these incredible results.

194
00:19:38,950 --> 00:19:44,110
To which we had contributed one year's hard work, but one year's work out of this entire story.

195
00:19:44,680 --> 00:19:51,070
But it really struck me that for, you know, this has been Rory's big project at the time.

196
00:19:51,400 --> 00:19:54,560
And for him to then say, look, actually, you will take will,

197
00:19:54,580 --> 00:20:02,409
will will that those two relatively junior characters actually from the media and from the story and everything else in the

198
00:20:02,410 --> 00:20:08,620
UK because we need to do the US and you couldn't in those days be in both places at once is very different in that today.

199
00:20:09,940 --> 00:20:14,079
That was again a huge opportunity and then people wanted to hear about themselves.

200
00:20:14,080 --> 00:20:17,950
All the academic meetings, all the professional meetings, scientific meetings all around the world,

201
00:20:17,950 --> 00:20:25,690
all sorts of societies of atherosclerosis and cardiovascular disease and whatever, some of which are well known, some of which are less well known.

202
00:20:26,230 --> 00:20:28,690
There were dozens and dozens of these meetings.

203
00:20:28,690 --> 00:20:36,459
And again, Rory and Jane would go and do the really important ones, quite rightly, that, you know, as an example,

204
00:20:36,460 --> 00:20:47,830
I was sent off to Ecuador for two to meetings Ecuador, Quito and Guayaquil, then Caracas, then Bogota and then Puerto Rico.

205
00:20:49,870 --> 00:20:53,710
We did Japan, I mean, but again, it was this trust that was put in us,

206
00:20:54,250 --> 00:21:04,120
but also the experience of how do you present what was quite complex data and distil it down into a message that the clinical audience could really,

207
00:21:04,300 --> 00:21:10,450
really grasp, because the aim of what we do isn't to get a publication.

208
00:21:10,450 --> 00:21:18,850
The publication is a stepping stone along the way is not just to get a regulatory approval that's a stepping stone along the way.

209
00:21:19,000 --> 00:21:23,050
So the FDA approves the drug or whatever or nice approved to pay for it or something.

210
00:21:23,470 --> 00:21:30,680
You've got to get it all the way through the clinical guidelines and actual frontline doctors being prepared to prescribe it and yeah,

211
00:21:30,700 --> 00:21:36,410
patients being prepared to take it. So you've actually got to think about all of those steps you have.

212
00:21:36,650 --> 00:21:41,650
If the aim of doing a trial that says cholesterol lowering works regardless of your cholesterol,

213
00:21:41,650 --> 00:21:47,650
if you've had a heart attack, is when you've got the answer. The answer is to make sure that that turns into practice.

214
00:21:47,980 --> 00:21:58,660
So it's a huge, you know, learning experience at that time, you know, and I'm grateful to the opportunity that we were given.

215
00:21:59,530 --> 00:22:02,710
So that's taken us. Let's take this to 2000.

216
00:22:03,010 --> 00:22:11,770
So and then from then on, we're into the sorts of trials that gradually I was running and I was designing with Colin,

217
00:22:11,800 --> 00:22:18,490
agent, with Louise, with Jane Armitage, with Richard Haynes and so on over the last 20 years.

218
00:22:18,490 --> 00:22:22,500
Just give me a couple examples of things that we we don't have time to go into.

219
00:22:23,270 --> 00:22:24,969
So, so I'll give you example.

220
00:22:24,970 --> 00:22:36,070
So in I did my PhD in VI patients with kidney disease I get heart disease call invasion was also interested in that area understanding.

221
00:22:36,280 --> 00:22:40,899
Yes. So observing it is one thing doing something about it is different that needed a trial.

222
00:22:40,900 --> 00:22:45,910
So we did a trial of lowering cholesterol in 9009 and a half thousand people with kidney disease.

223
00:22:46,630 --> 00:22:50,020
It was the biggest trial in kidney disease ever done at the time,

224
00:22:50,020 --> 00:22:55,420
probably since is at least three times bigger than the next trial that was even running at the time.

225
00:22:55,750 --> 00:23:02,229
It was the first time that this unit had done a trial, which was both international and with long term,

226
00:23:02,230 --> 00:23:05,620
with sort of chronic daily daily tablets and therapy.

227
00:23:06,880 --> 00:23:12,640
And Colin and I designed that trial from the ground up, including thinking about actually, how do you make it work?

228
00:23:12,910 --> 00:23:19,030
How do you make sure that the procedures are being done in a clinic in in the southern states are the same

229
00:23:19,030 --> 00:23:23,920
as the features that are being done in Malaysia or New Zealand and how did you get the data and so on.

230
00:23:24,100 --> 00:23:28,089
And a lot of that was around how do you use technology.

231
00:23:28,090 --> 00:23:33,549
So web, this was modern web where you could actually enter stuff and you can order stuff.

232
00:23:33,550 --> 00:23:37,720
You know, the sort of things that we use all the time now was only just being born at that point.

233
00:23:37,900 --> 00:23:43,180
Could you use that to do the trials? So that was a lot of what that was.

234
00:23:43,660 --> 00:23:47,530
Another one was a trial of a drug called niacin.

235
00:23:47,530 --> 00:23:48,970
It's a vitamin B vitamin.

236
00:23:49,300 --> 00:24:00,040
It's on the side of your cornflake packet in small doses, in big doses, know probably 100 times bigger also in gram doses than it lowers cholesterol.

237
00:24:00,970 --> 00:24:06,010
And for 50 years, it's been considered to be a treatment for lowering cholesterol to prevent heart disease.

238
00:24:06,010 --> 00:24:09,540
So we did a trial and we found that one.

239
00:24:09,550 --> 00:24:15,580
It didn't lower the risk of heart disease. And two, it caused people to go into hospital more often with bleeding and infection.

240
00:24:15,700 --> 00:24:20,300
Nobody had known that in 50 years. And now. It doesn't get used for that reason.

241
00:24:22,190 --> 00:24:26,599
So yeah, those are two examples of these big cardiovascular trials.

242
00:24:26,600 --> 00:24:32,240
And even now we're doing some on, you know, once or twice yearly injections to lower cholesterol,

243
00:24:32,600 --> 00:24:37,340
which it would be a much more convenient way for many people, either in addition or instead of a statin.

244
00:24:38,930 --> 00:24:45,140
And just now, we've we're announcing results about a new treatment for people with kidney disease,

245
00:24:45,440 --> 00:24:52,459
which lowers their risk of heart of heart attack and particularly slows their decline in kidney function.

246
00:24:52,460 --> 00:24:57,860
You know, the big thing if you've got kidney disease is you really want to delay or avoid dialysis.

247
00:24:58,580 --> 00:25:00,930
And this treatment appears to certainly delay that.

248
00:25:01,540 --> 00:25:08,060
So there's been a series, a series of these trials over the last 20 years that have been sort of the backbone of my clinical trials work.

249
00:25:08,870 --> 00:25:13,790
So clinical trials have have got bigger and bigger and bigger, but they've also got more bureaucratic.

250
00:25:13,790 --> 00:25:17,719
And I understand you've been quite critical of some of the hurdles that you're

251
00:25:17,720 --> 00:25:22,970
forced to leap over by government regulations in different parts of the world.

252
00:25:23,030 --> 00:25:31,370
Yes. So I've just given you examples of the work and just think of them as examples of what can happen.

253
00:25:31,460 --> 00:25:37,610
Those trials, each of them is roughly ten fold cheaper and as I've indicated, often bigger,

254
00:25:37,610 --> 00:25:44,860
not always and just as reliable, if not more than how industry would do them, but ten fold cheap enough drugs.

255
00:25:45,020 --> 00:25:52,460
If the trial is ten fold cheaper, you can do more trials, you can include a greater range of patients and so on and so forth.

256
00:25:52,640 --> 00:26:00,020
You can actually bring more interesting drugs through to to that stage, because if a trial costs you $1,000,000,000,

257
00:26:00,770 --> 00:26:05,090
you an awful lot of interesting drugs get killed off even before they ever hit a patient.

258
00:26:06,560 --> 00:26:14,030
So the if one looks at it in in context that ices to trial it might be for much of one of the

259
00:26:14,030 --> 00:26:19,260
students being double Richard and Rory was essentially two pieces of paper one at the beginning,

260
00:26:19,280 --> 00:26:24,260
one at the end. And that was essentially all they were all there was to that trial.

261
00:26:25,760 --> 00:26:34,580
If you look at many trials now, there are thousands of dozens of pages of case report form,

262
00:26:35,030 --> 00:26:43,250
and there are hundreds of standard operating procedures and documentation and whatever else that goes with it,

263
00:26:44,270 --> 00:26:50,360
largely adding little value, but really being driven by the rules and regulations.

264
00:26:50,360 --> 00:27:00,770
So in the late 1980s and through to about mid 1990s, a set of rules were developed called good clinical practice.

265
00:27:02,000 --> 00:27:05,330
The old joke goes that they're not good, they're not clinical and they're not practical.

266
00:27:05,780 --> 00:27:11,540
Unfortunately, the old joke joke is still current, so nothing much has changed in those 25 years,

267
00:27:12,440 --> 00:27:17,360
but they take a very rigid view of how the world works.

268
00:27:17,780 --> 00:27:25,750
So they were written in the days when a sort of carbon copy in a fax machine with state of the art technology.

269
00:27:25,760 --> 00:27:28,820
So you've got a top copy in a bottom copy and they should really match.

270
00:27:29,240 --> 00:27:34,100
And then you send the bottom copy in the post or your fax it or something to

271
00:27:34,100 --> 00:27:37,969
somebody else who then types it in the machine because they typed it into machines.

272
00:27:37,970 --> 00:27:41,180
Somebody else has to type it into the machine to check that you've got the two answers.

273
00:27:42,470 --> 00:27:49,700
And because you're not sure that the top coffee matches the bottom copy, you send monitors out to site to check that those two things really do match.

274
00:27:50,210 --> 00:27:52,400
And because you're not sure that the monitors have done a good job,

275
00:27:52,400 --> 00:27:55,730
you send more monitors to check that the monitors have done the monitoring correctly.

276
00:27:56,690 --> 00:27:59,240
And so you layer and lower and lower this stuff on,

277
00:27:59,750 --> 00:28:04,159
which makes this story is the difference between what you're saying now and what the ICE trial did.

278
00:28:04,160 --> 00:28:10,340
Because the drugs that we were testing in the ICC trial, they were testing in the ICC trial, it were already approved.

279
00:28:10,790 --> 00:28:15,169
No, no. They say this is not about I mean, of course,

280
00:28:15,170 --> 00:28:20,990
there are differences in different types of trials and chemotherapy treatment is a completely different beast to,

281
00:28:21,440 --> 00:28:26,300
you know, studying antioxidant vitamins to see whether they prevent heart attack.

282
00:28:26,480 --> 00:28:34,490
They don't. We done that one. So you will of course, you have to tailor to the clinical issues and what you know about the drug and so on.

283
00:28:35,240 --> 00:28:42,920
But so much of the process is just process and paperwork without actually really focusing on what matters.

284
00:28:43,910 --> 00:28:51,080
If you've got two randomised groups, we did a bit about how trials work, which would be obvious to some and less obvious to others.

285
00:28:51,440 --> 00:28:57,410
But essentially you take a group, a large enough group of patients and for each patient you toss a coin,

286
00:28:57,860 --> 00:29:01,610
you get the drug tails, you get something else. Let's call it placebo for now.

287
00:29:03,080 --> 00:29:08,750
And then you know that come the end of the study, whatever happens to those people,

288
00:29:09,020 --> 00:29:13,760
the differences between those who came down hard and those who came down tails, there were two things that could have happened.

289
00:29:14,120 --> 00:29:18,450
One is the drug cause. That difference reduced the risk of health.

290
00:29:18,510 --> 00:29:22,590
To tackle whatever it might be or increase the risk of going into hospital with bleeding.

291
00:29:22,890 --> 00:29:29,880
The second is chance. Well, we have statistical tests to tell us whether something is likely to be purely due to the play of chance.

292
00:29:30,090 --> 00:29:34,830
So we can sort of deal with that issue. And that issue becomes a lot easier to deal with the bigger your study.

293
00:29:35,780 --> 00:29:42,089
You know, if you've got very small numbers, you get imbalances purely by the play, by the play of chance.

294
00:29:42,090 --> 00:29:45,330
But if you got very big numbers, then actually that all washes out.

295
00:29:45,870 --> 00:29:53,600
But it also tells you that how accurate does any one data point to need to be in that context?

296
00:29:53,610 --> 00:29:59,070
So if I take 10,000 people, toss a coin started versus placebo to see what happens to them.

297
00:29:59,940 --> 00:30:07,770
If I decide if I sort of mis code or just miss the fact that one person got a heart attack in one of the arms versus the other,

298
00:30:08,130 --> 00:30:12,030
there's probably 400 heart attacks in one arm and 300 in the other arm.

299
00:30:12,330 --> 00:30:18,030
Missing one heart attack is just noise in the system. Now, I've just I've literally just come from cardiology this morning,

300
00:30:18,450 --> 00:30:23,520
missing one heart attack this morning, face to face with one patient is absolutely critical.

301
00:30:24,180 --> 00:30:28,110
So that's a different situation. But that's not what we're doing.

302
00:30:28,110 --> 00:30:30,810
We're not trying to diagnose heart attack when we do these trials.

303
00:30:31,050 --> 00:30:38,040
We're trying to assess whether the strategy of giving a statin versus not cypher zebo actually makes a difference.

304
00:30:38,280 --> 00:30:44,940
So all that's that you don't need ultimate precision in order to get an incredibly precise answer.

305
00:30:46,590 --> 00:30:56,940
So the rules are written in ways which don't don't recognise and completely misunderstand that they want every data point to be absolutely precise.

306
00:30:57,600 --> 00:31:05,549
Now, if you said to me you can have perfectly precise data individuals, every data point is correct and it costs you no more.

307
00:31:05,550 --> 00:31:12,060
In terms of anything than slightly imprecise data, I'd say, well, yeah, precise data would be better,

308
00:31:12,570 --> 00:31:19,350
but if you if I have to now trade that off for a now, I get a smaller study and I get a more select group of people.

309
00:31:19,350 --> 00:31:24,030
I only get men and women and I don't get people from other ethnic groups and so on and so forth.

310
00:31:24,480 --> 00:31:27,180
I've suddenly lost a huge amount of information.

311
00:31:27,720 --> 00:31:33,120
Or if I only get a shorter study because getting that position while someone's in hospital might be feasible,

312
00:31:33,120 --> 00:31:36,749
but for five years afterwards might be infeasible. Awful.

313
00:31:36,750 --> 00:31:41,790
If you think of some treatments now you're thinking about what happens to these people in 20 years time.

314
00:31:42,480 --> 00:31:46,500
It's impossible to still keep getting that level of precision all the way through.

315
00:31:47,250 --> 00:31:50,909
But it's not only impossible, it's unnecessary. And that's the thing.

316
00:31:50,910 --> 00:31:58,319
It's not a question of reducing the quality or reducing the standard is a question of actually understanding what you're trying to do.

317
00:31:58,320 --> 00:32:06,870
And you're trying to work out whether there's a separation between two groups of people who one lot who by chance were giving the start in one lot,

318
00:32:06,870 --> 00:32:08,430
who by chance were given placebo.

319
00:32:08,730 --> 00:32:18,510
And whether that has made it made a difference and it's not really it doesn't really matter whether it's 399 versus 501 or it's 400 versus 500.

320
00:32:19,050 --> 00:32:21,630
Makes no difference to the to the physician.

321
00:32:22,440 --> 00:32:29,580
I said earlier on that the reason I when we first came here, we were employed essentially to clear that backlog of clinical trial adjudication.

322
00:32:29,820 --> 00:32:33,570
These piles and piles of documents that had to be coded as to, yes,

323
00:32:33,570 --> 00:32:38,129
they definitely had a heart attack or didn't accord its own criteria after we'd been through

324
00:32:38,130 --> 00:32:41,670
that and after we presented the results and it had a big impact all around the world and so on,

325
00:32:41,670 --> 00:32:45,930
and prescribing patterns change and everybody now knew what a statin was, which they didn't at the time.

326
00:32:46,920 --> 00:32:54,900
That was a common issue. After we'd done all that, we went back and we looked and said, What if Louise and Martin hadn't done all that work?

327
00:32:55,290 --> 00:33:00,840
What if we just taken the patients report when the when the nurse said, Tell me, have you had a heart attack since I last saw you?

328
00:33:01,080 --> 00:33:04,400
They said, Yes, we believed it was yes. And if you said no, we believed it.

329
00:33:04,440 --> 00:33:08,400
No, we got almost precisely the same answer.

330
00:33:09,540 --> 00:33:19,650
We were out by the second or third decimal point on the risk ratio, the confidence interval, the p value, whatever.

331
00:33:19,890 --> 00:33:26,730
So statistically, we got essentially the same answer regulatory wise in terms of would get someone to give it a label,

332
00:33:27,090 --> 00:33:31,049
give it a, an indication we allowed it to be prescribed and got the same answer clinically.

333
00:33:31,050 --> 00:33:40,530
We got the same answer if we'd just taken the patient's words which are not precise, it's not a precise diagnosis,

334
00:33:40,530 --> 00:33:47,790
we'd have got the same answer we'd have known that started work for substantially less cost much quicker.

335
00:33:48,300 --> 00:33:56,370
And because of that, we could have done the study even bigger or even more diverse, or even longer or even better in a variety of ways.

336
00:33:56,670 --> 00:34:04,380
Because now scale becomes becomes possible when you say, well, a big study is just like a tiny study, how do you do it?

337
00:34:04,920 --> 00:34:12,120
Exactly the same things many, many more times over that becomes infeasible and therefore you actually misinformation.

338
00:34:12,780 --> 00:34:18,450
So if you think of it from a regulator's point of view, I'm not a regulator. I work with lots then their job really.

339
00:34:18,510 --> 00:34:20,420
It is not to regulate trials.

340
00:34:20,430 --> 00:34:26,820
They need to do some of that because there are the trials need to do a job, but their job really is to improve the health of the nation.

341
00:34:27,870 --> 00:34:34,110
So they need better evidence to allow them to make better decisions about whether to licence a drug or not.

342
00:34:35,490 --> 00:34:39,510
So actually what they want is trials that give them better evidence,

343
00:34:39,780 --> 00:34:48,960
not trials that fill some sort of slightly obsessive version of what the truth might be.

344
00:34:49,650 --> 00:34:54,510
If I win that, all that story forwards and say, you look at look at today.

345
00:34:55,470 --> 00:35:05,100
Well, there's health care information all around us. The NHS Digital will collect information on every hospital admission, every diagnosis,

346
00:35:05,100 --> 00:35:10,409
every operation and all that for business planning reasons and reimbursement reasons and all those sorts of things.

347
00:35:10,410 --> 00:35:13,830
And with the right consents and tactical operations and all those other things in place,

348
00:35:15,930 --> 00:35:19,440
which is a non-trivial undertaking with those things in place,

349
00:35:19,440 --> 00:35:24,870
you can use you can do trials where you randomise patients to a treatment versus not,

350
00:35:25,440 --> 00:35:30,780
and you can link through to those data and you can use all those routine data that are already out there.

351
00:35:30,780 --> 00:35:33,060
They're being collected for free. Where do they come from?

352
00:35:33,240 --> 00:35:39,840
Well, they came from somebody typing in the answers from the clinical notes which were written when each patient was in hospital.

353
00:35:40,380 --> 00:35:41,340
And you can use that.

354
00:35:41,340 --> 00:35:48,240
And again, you'll find that there's a little bit of mismatch between that and whichever page of the the hospital notes you look at.

355
00:35:48,780 --> 00:35:52,379
Believe me, different pages will have different different pieces of information on them,

356
00:35:52,380 --> 00:35:57,270
some of them contradictory, but they won't be exactly the data.

357
00:35:57,270 --> 00:36:01,710
And in then it's just digital and the paper records will be exactly aligned.

358
00:36:02,190 --> 00:36:08,820
But if you do, if you link to the NHS digital data, you will get exactly the same answers as if you looked at the routine data.

359
00:36:09,150 --> 00:36:13,500
But again, you can do it bigger. You can do it when patients move around the country and they move away from

360
00:36:13,500 --> 00:36:17,040
my hospital into somewhere else where they're never seen in hospital again.

361
00:36:17,460 --> 00:36:22,860
You know, the huge difference between somebody, nothing has happened to somebody.

362
00:36:23,310 --> 00:36:26,910
And we have no idea whether something has happened to anybody because both of those

363
00:36:26,910 --> 00:36:30,920
look like sort of missing pieces of information and we can do it for much longer.

364
00:36:30,930 --> 00:36:42,959
So again, that sort of is bringing us forward to we can use we don't have to be absolutely precise in order to get a very precise answer and actually

365
00:36:42,960 --> 00:36:50,550
a better answer and a more useful answer in more contexts and in more ways than one would've done if I'd only done it mechanically.

366
00:36:50,940 --> 00:36:56,340
And yeah, we have done trial, we have done studies where we've said, What happens if you just do it the old fashioned way?

367
00:36:56,550 --> 00:37:00,300
You know, have you had a heart attack since I last saw you? Or we go to NHS Digital and say,

368
00:37:00,540 --> 00:37:05,909
Tell me whether any of these people have had a heart attack with vaccinations and they give you exactly the same answer.

369
00:37:05,910 --> 00:37:10,200
So the rules are not written in that way.

370
00:37:10,200 --> 00:37:14,700
They don't emphasise the principles, they emphasise the operational details.

371
00:37:14,700 --> 00:37:20,579
And the way the world works today is so different from 1995, how we talk to each other,

372
00:37:20,580 --> 00:37:26,970
how we communicate, you know, the the smartphones and and the like,

373
00:37:27,240 --> 00:37:35,820
how data is collected and manage the relationships between patients in their and their and and the medical profession is wildly different.

374
00:37:36,360 --> 00:37:42,750
And if you think about where most patients spend most of their time, it's thankfully nowhere near any doctor.

375
00:37:43,260 --> 00:37:51,089
Even long term diseases like diabetes, most people with diabetes spend most of their time, if you like, out in the world,

376
00:37:51,090 --> 00:37:57,510
living their everyday life, doing whatever they do without having to go into hospitals, without going to the GP or to diabetes.

377
00:37:57,750 --> 00:38:04,770
Of course they come to those met periodically. So these there are new opportunities to actually bring trials to the patient rather than making

378
00:38:04,770 --> 00:38:10,500
the you're trying to insist that the patient somehow comes to this magical clinical trial site,

379
00:38:10,920 --> 00:38:15,810
but they're only possible if the rules are based on principles and the principles are so easy.

380
00:38:16,020 --> 00:38:19,560
Number one, get a reliable result, actually.

381
00:38:20,010 --> 00:38:23,729
Recap. Number one, ask a question that you care about the answer.

382
00:38:23,730 --> 00:38:31,290
Number two, answer it. In the meantime, you need to make sure you look after the rights and wellbeing of the patients who are in the trial.

383
00:38:32,190 --> 00:38:38,909
So in a sense, think about it as you go to look after the people who in the trial and you can look after the results because they will,

384
00:38:38,910 --> 00:38:42,840
one way or another influence many, many more people who are not in the trial.

385
00:38:44,310 --> 00:38:49,290
I would add to those rules and say we need actually doing things efficiently is

386
00:38:49,290 --> 00:38:54,490
actually we should be driving to avoid wastage because we can't afford wastage.

387
00:38:54,550 --> 00:38:58,860
Costs go up, we do less trials, we get less evidence, we do worse medicine,

388
00:38:59,760 --> 00:39:03,489
doing things in a way that does involve the communities that people come from,

389
00:39:03,490 --> 00:39:09,360
that both the clinical communities that people come from and the the communities that patients come from.

390
00:39:09,360 --> 00:39:13,349
And of course, many people don't think of themselves as patients most of the time. Think about vaccines.

391
00:39:13,350 --> 00:39:17,910
Most people are vaccinated. There's sort of nothing wrong with them. They have vaccine.

392
00:39:18,140 --> 00:39:22,190
To stop something that might happen either in the short term or in the long term.

393
00:39:23,360 --> 00:39:31,129
So I think there are many ways in which we could have much better rules which actually explain what the principles are,

394
00:39:31,130 --> 00:39:34,160
actually what we're trying to achieve. What does good look like?

395
00:39:34,550 --> 00:39:41,420
But then allow people to, if you like, innovate and find solutions that are fit for the for the particular context,

396
00:39:41,600 --> 00:39:46,190
which is what you needed to do, which is very difficult when you are doing it.

397
00:39:46,190 --> 00:39:49,470
So we really need to get on to it, I think, find out how we're doing. Right.

398
00:39:49,500 --> 00:40:00,020
Okay. So can you remember? I know you can't because I've heard the story before, but let's tell it again how you first heard about that.

399
00:40:00,290 --> 00:40:02,839
There was something going on in China that looked as if it might get serious.

400
00:40:02,840 --> 00:40:07,760
And and how long did it take you before you realised this was something actually you were going to have to get involved with?

401
00:40:08,660 --> 00:40:12,049
But I do remember seeing some very early, I guess,

402
00:40:12,050 --> 00:40:22,730
tweets or whatever from Jeremy Fowler and a few others in early January of that year along the lines of, Yeah, this doesn't look good.

403
00:40:24,490 --> 00:40:35,930
I am the cardiovascular doctor. I don't do infectious disease, I don't do overseas, you know, emerging outbreaks, you know,

404
00:40:35,930 --> 00:40:40,159
sales and those and all those things are there are other people who do those and do them very well.

405
00:40:40,160 --> 00:40:41,120
So it wasn't my area.

406
00:40:41,600 --> 00:40:49,520
And so it was sort of, if you like, mentally a problem for somebody else to sort out that wasn't do didn't have a direct impact on me.

407
00:40:49,940 --> 00:40:55,580
Of course, as we all knew, ties in the same position probably as most other people in the UK or elsewhere.

408
00:40:56,490 --> 00:41:00,350
Of course we saw them move to Iran and then to northern Italy and by the time

409
00:41:00,350 --> 00:41:03,650
we got to northern Italy it was beginning to feel a little bit close to home.

410
00:41:04,190 --> 00:41:07,819
Yeah, not least because those two countries are not obviously connected.

411
00:41:07,820 --> 00:41:12,590
So you're suddenly seeing this thing. This is this is entirely amateur for my perspective.

412
00:41:12,920 --> 00:41:17,600
You're studying this thing pop up, clearly spread in a number of different places.

413
00:41:17,600 --> 00:41:22,100
So this was this isn't just like it sort of, you know, gradually migrating down the corridor.

414
00:41:22,100 --> 00:41:25,400
It's actually it's clearly getting on jet planes.

415
00:41:27,380 --> 00:41:34,730
It was it was as things in February started to sort of really become more obvious again in the north of Italy.

416
00:41:35,330 --> 00:41:38,690
And late February 28th of February,

417
00:41:38,690 --> 00:41:44,420
I was on a train on the way back from somewhere in the north of England where I

418
00:41:45,020 --> 00:41:48,350
had a meeting with NHS Digital about clinical trials and data analyst stuff,

419
00:41:49,940 --> 00:42:00,559
and I was imagining Jeremy Pharaoh I was doing work with him at the time about or for him about clinical trials and the guidelines I emailed out.

420
00:42:00,560 --> 00:42:09,950
Jemmy Fallon said at the bottom of the piece, at some point, is anybody is anybody thinking about randomisation?

421
00:42:09,950 --> 00:42:12,049
I said, at some point people are going to start wanting to.

422
00:42:12,050 --> 00:42:19,550
So treatments and vaccines and whatever it people and we really need to know if they if they work or they don't and if we don't randomise,

423
00:42:19,790 --> 00:42:23,000
we'll never find out. And that would be that that could be a disaster.

424
00:42:23,660 --> 00:42:29,299
And I'd also say, look, it's exactly at the time when there are no treatments,

425
00:42:29,300 --> 00:42:36,410
there is a widespread disease with a bad outcome is exactly the time when randomisation is the right way forward.

426
00:42:38,690 --> 00:42:44,389
And also it's a time when regulators and others are prepared to do some innovation.

427
00:42:44,390 --> 00:42:52,910
Because I have to I wasn't sure that I ever get a response, but I got one about 5 minutes later saying, speak to Peter Horby and Richard Peter.

428
00:42:53,600 --> 00:43:02,360
So the following week, so something like the 4th of March or so, Peter and I met with Richard, Peter,

429
00:43:02,360 --> 00:43:11,179
in which his office said have some we'd met Peter and I'd had I I'd had a about in total about 20 minutes of discussion split

430
00:43:11,180 --> 00:43:16,700
over about three conversations on the telephone and had met him once up to that point over the last couple of couple of years.

431
00:43:17,330 --> 00:43:23,180
And that was that was that was the limit of of our of our knowledge of each other.

432
00:43:25,520 --> 00:43:32,870
So we met in which his office and started talking about what a trial should look like over the next four or five days.

433
00:43:32,870 --> 00:43:36,890
So that took us over the weekend. It wasn't clear W.H.O. were trying to do so.

434
00:43:37,010 --> 00:43:43,400
I mean, they obviously did in the end, but they were trying to do something. Should the UK be in with that and so on and so forth.

435
00:43:43,640 --> 00:43:50,600
And then it got and then it got to and I asked you, could we provide an IT system for the trial because you've got some data collection.

436
00:43:51,590 --> 00:43:56,360
So I did a little bit of discussion talking around various people in the Parliament about how we might go about that.

437
00:43:58,070 --> 00:44:07,639
Come the Monday I was on the famous number 18 bus with Jeremy heading between Marylebone and and the Wellcome Trust,

438
00:44:07,640 --> 00:44:15,400
a ten minute journey and the stories of the horrors that were happening in northern Italy were all over the serious newspapers.

439
00:44:15,420 --> 00:44:19,790
The stories of what the tabloids considered. B the horrors of running out of lavatory.

440
00:44:19,790 --> 00:44:25,070
Well will over the front of the red tops seriously that that that those were the new stories of the day.

441
00:44:26,760 --> 00:44:35,629
And and I had also had heard from colleagues in northern Italy and so on just how horrendous it was,

442
00:44:35,630 --> 00:44:38,240
people being taken off ventilators, not put on ventilators,

443
00:44:38,810 --> 00:44:43,370
having three refrigerated lorries in the hospital carparks because they didn't have the space in the mortuary,

444
00:44:43,370 --> 00:44:47,930
all this sort of thing, I mean, really sounded to me was horrendous.

445
00:44:48,320 --> 00:44:52,280
And Jeremy said, look, I think London's going to be like this in two weeks. It's coming here.

446
00:44:52,280 --> 00:44:59,989
And this we are on a crowded bus. And he asked me about, you know, what the discussion for the trial.

447
00:44:59,990 --> 00:45:06,740
He said, look, we've got to get this guy up and running within two weeks. We agreed that in the 10 minutes bus ride and I got two welcome.

448
00:45:09,320 --> 00:45:13,729
And the first call I made actually was to two or three Collins as head of department.

449
00:45:13,730 --> 00:45:18,650
I said, look, we've, you know, we talked about the it I spoke to Richard and whatever else,

450
00:45:19,100 --> 00:45:26,720
but I really think we've actually got to go completely in this post I'm back know there's no point being half and half

451
00:45:27,350 --> 00:45:35,450
and I said my judgement is we should just go for this because I don't think anybody else will be able to deliver.

452
00:45:37,430 --> 00:45:41,150
And I said, oh and by the way there wasn't any money.

453
00:45:44,690 --> 00:45:54,769
So at that point I said, Look, I think having spoke to Jeremy, given the situation I think we should go for this is his department,

454
00:45:54,770 --> 00:45:59,750
not mine, but I think we should sort of essentially underwrite this.

455
00:46:01,280 --> 00:46:06,830
I said, I'm pretty confident that if we if we get going, the money will follow at some point.

456
00:46:07,160 --> 00:46:10,399
I just can't tell you today I've got the grant and I've got the money.

457
00:46:10,400 --> 00:46:13,490
And, you know, we to have to do this in a different order. We haven't got time.

458
00:46:13,820 --> 00:46:17,960
And he agreed. Now, I turned out on Monday to answer your question on money,

459
00:46:18,890 --> 00:46:26,120
the Peter who put in for grants to study treatments for COVID in China way back in the February.

460
00:46:26,120 --> 00:46:32,149
But grant to Nigel is it taken in job for six weeks whatever it was to turn this through

461
00:46:32,150 --> 00:46:36,120
by which time there wasn't any cobra in China because of the lockdown had worked.

462
00:46:36,390 --> 00:46:42,860
That first lockdown worked very well. And so they said, yes, you can have the money, but you must study to do it in the UK.

463
00:46:43,340 --> 00:46:47,060
And so actually the grant did come through for a slightly different route.

464
00:46:48,320 --> 00:46:53,930
But yeah, we were prepared to do this, you know, at risk.

465
00:46:53,930 --> 00:47:03,139
We had to move at risk and that was what was obvious to me. And so yeah, ninth and 10th of, of that march we broke the protocol.

466
00:47:03,140 --> 00:47:06,980
I had a copy of the study protocol, this large, simple trial on my desk.

467
00:47:07,580 --> 00:47:16,130
I had a copy of a paper Richard and Rory and Selim Yusuf had written from 1984, which says Why we need some large, simple trials.

468
00:47:16,570 --> 00:47:22,130
It was all about heart disease. But if you substituted the word heart disease for COVID, the signal slotted into place.

469
00:47:23,390 --> 00:47:31,910
And I looked at one of my usual protocols for one of my usual cardiovascular five year trials, and I thought, that's not the way to do it.

470
00:47:31,910 --> 00:47:39,080
We've got to do it simply because this is this is going to be really difficult at the coalface.

471
00:47:39,080 --> 00:47:43,850
And the trick is absolutely going to be can we minimise the additional effort for the frontline staff?

472
00:47:43,850 --> 00:47:50,150
Otherwise we'll never happen. And number two, the second trick is we've got to get it in now before the NHS works out.

473
00:47:51,170 --> 00:47:58,129
It's sort of standard protocol to whatever who gets assessed in A&E, how do they get trials, what treatments do they get, all that sort of thing.

474
00:47:58,130 --> 00:48:02,600
We had to get it in so doing. The trial was part of just what they did in in this context.

475
00:48:03,640 --> 00:48:08,690
So the big question is how did you know what drugs to try?

476
00:48:08,690 --> 00:48:12,110
So this was, yes, a trial of treatments is not to do with prevention.

477
00:48:12,110 --> 00:48:17,239
This is to do this is to treat. So my answer one question is, why did we choose this particular context?

478
00:48:17,240 --> 00:48:23,510
I mean, we could have done a trial of prevention. We could have done a trial of an in doctors and nurses who are looking after patients with COVID.

479
00:48:23,510 --> 00:48:30,500
They're at high risk of getting us another version of prevention. We could have done a trial in the community to prevent people getting into hospital,

480
00:48:32,480 --> 00:48:39,590
but it seemed to me that the number one problem we were facing and Italy was facing was there were a lot of people dying.

481
00:48:39,740 --> 00:48:44,390
The hospitals were overwhelmed and there were a lot of people dying and a lot of people needing ventilators.

482
00:48:45,110 --> 00:48:51,500
And so, you know, when the house is on fire, the first thing you try to do is put out the house, put out the fire.

483
00:48:51,740 --> 00:48:55,920
And, you know, of course you want to do prevention and all those other things they did.

484
00:48:56,000 --> 00:49:01,940
Yeah. And that's where in particular the vaccines come in. But we never knew whether we're ever going to get a vaccine, let alone when.

485
00:49:03,410 --> 00:49:10,790
So it seemed to me that the major challenge was we've got patients who are declaring themselves sick because they're sick enough to get into hospital.

486
00:49:11,120 --> 00:49:16,550
So diagnosis and the problem and we know what the outcome is, one in three, one in four were dying.

487
00:49:16,970 --> 00:49:23,220
You've got to do some. About that before we do anything else. And that's that's why we focussed on the treatment side.

488
00:49:23,910 --> 00:49:27,360
We had some help. W.H.O. had done some shortlisting of drugs they come up with.

489
00:49:27,540 --> 00:49:32,010
I say short if they prioritise but they the list was 100 drugs long or something.

490
00:49:35,070 --> 00:49:43,139
Nervtag, which Peter was on, had also done some prioritisation and with particularly with regards dexamethasone,

491
00:49:43,140 --> 00:49:51,360
there had been a a trial sort of protocol in case of pandemic flu, which has been a trial or was going to be a trial of steroids.

492
00:49:51,370 --> 00:49:54,179
So what's the argument for using steroids in an infectious. Well,

493
00:49:54,180 --> 00:50:00,690
the argument for use for using them is that the inflammatory response to the virus is

494
00:50:00,690 --> 00:50:06,479
so it becomes so great that the inflammatory response itself starts to cause damage.

495
00:50:06,480 --> 00:50:08,370
So you start to get secretions and so on,

496
00:50:08,370 --> 00:50:13,050
and then you get a gas exchange in the lungs and then you become hypoxic and then you need a ventilator and so on.

497
00:50:13,710 --> 00:50:18,750
The argument against using them is that steroids suppress the immune system at a time

498
00:50:18,750 --> 00:50:21,450
when you're trying to fight an infection for which there were no other treatments.

499
00:50:23,640 --> 00:50:28,140
It seemed to me that if you don't know what you're doing, you're better off randomising or finding out.

500
00:50:28,140 --> 00:50:32,700
Finding out there are others.

501
00:50:32,730 --> 00:50:37,139
It has to be said, who had very strong views in in one direction or another.

502
00:50:37,140 --> 00:50:39,300
And in fact, once we got the dexamethasone up and running,

503
00:50:39,810 --> 00:50:47,910
there were a group of eminent professors in London who wrote to the MHRA, the chief medical officer,

504
00:50:47,910 --> 00:50:51,180
US and various others saying You shouldn't even be studying steroids,

505
00:50:51,180 --> 00:50:56,610
you should be studying dexamethasone because it is dangerous to suppress the immune system in people fighting infection.

506
00:50:57,090 --> 00:51:02,639
We had to by that time argue very strongly with the MHRA that actually nobody knew.

507
00:51:02,640 --> 00:51:07,200
We needed to find out that this was the quickest way of finding out gravity.

508
00:51:08,520 --> 00:51:13,200
So yes, we started with hydroxychloroquine. That was an easy choice and it was everywhere.

509
00:51:13,200 --> 00:51:16,349
Everybody was saying wonderful, but that's a malaria drug.

510
00:51:16,350 --> 00:51:20,309
So why would anybody think that? Where is it a malaria drug? Into a aetiology drug.

511
00:51:20,310 --> 00:51:32,680
But in the laboratory. If you if you if you put hydroxychloroquine on on infected cell cells with almost any virus, then they look unhappy.

512
00:51:32,700 --> 00:51:39,960
So in the laboratory for a whole range of viruses, hydroxychloroquine has been touted as an antiviral.

513
00:51:41,130 --> 00:51:44,370
No one's ever yet found a viral disease for which is useful.

514
00:51:44,370 --> 00:51:49,620
To the best of my knowledge, Peter's more the expert on on the choice of drugs than me.

515
00:51:50,760 --> 00:51:56,399
But yeah, there were. And then there was a paper that would come out in The Lancet,

516
00:51:56,400 --> 00:52:02,760
a series of about 15 or 18 people who'd been given hydroxychloroquine as if the most known antibiotic.

517
00:52:04,710 --> 00:52:09,210
And this was proclaimed as wonderful treatment. Now, actually, in that paper, half the data was missing.

518
00:52:09,420 --> 00:52:14,370
It was too small and it wasn't roundabout, and it wasn't it was a a really poor paper.

519
00:52:16,800 --> 00:52:21,960
And actually the world was dominated by some pretty poor papers in those early days, probably ever since.

520
00:52:23,610 --> 00:52:29,880
So, you know, and Trump was claiming that this is a miracle cure.

521
00:52:30,930 --> 00:52:39,030
The French were very minded to use it. Bolsonaro in Brazil was saying this is a must for treatment.

522
00:52:39,060 --> 00:52:43,530
It happened in India. All over the world. People were using it. The UK had a stockpile of it.

523
00:52:45,540 --> 00:52:47,520
And the question was, should they use it?

524
00:52:47,570 --> 00:52:54,420
We we said to them, no, the thing to do is randomise this randomised and find out quickly whether it's the right thing to do or not.

525
00:52:54,990 --> 00:52:59,639
Otherwise, if you use it, you'll run out at some point and you never know whether you need more or actually

526
00:52:59,640 --> 00:53:03,920
you didn't use any at all that it happened with a drug called Tocilizumab in Italy.

527
00:53:04,380 --> 00:53:10,040
They do so much of it that they had run out so that they couldn't even use it for, Oh, we're running low.

528
00:53:10,060 --> 00:53:13,470
They're treating people with arthritis, which is where it's normally used.

529
00:53:14,070 --> 00:53:20,940
And at the end of having burnt through quite a lot of money, but also this huge stockpile of tocilizumab,

530
00:53:21,360 --> 00:53:26,400
they had no idea whether they need to buy more or whether they just wasted a whole load of drug.

531
00:53:26,730 --> 00:53:32,580
We said, look, randomised, randomised, randomised all the way to make randomisation part of your public health strategy,

532
00:53:33,720 --> 00:53:40,110
you do R&D it seems like on the fly while you while whilst you're fighting the pandemic,

533
00:53:40,110 --> 00:53:43,860
you won't regret it actually learning which ones work and which ones don't.

534
00:53:45,060 --> 00:53:52,590
And that was an argument we had to keep banging on about over and over again and still do as well as people who thought dexamethasone was dangerous.

535
00:53:52,800 --> 00:53:56,520
But there were probably some who thought, Well, if you think it works, we should just give it to everybody.

536
00:53:56,700 --> 00:54:01,109
There were some it was it was, it was less the lobby in favour of oh, just them, the drug,

537
00:54:01,110 --> 00:54:04,589
it can't do anything which was very strongly a sort of hydroxychloroquine thing.

538
00:54:04,590 --> 00:54:10,829
Right. Less strongly a dexamethasone thing I would say, although some I believe when we set up the trial we said, well,

539
00:54:10,830 --> 00:54:17,400
if you have a really big belief that this particular patient needs dexamethasone or must have dexamethasone, then.

540
00:54:17,890 --> 00:54:20,530
Don't randomise them into that bit of the trial randomised into the race.

541
00:54:20,530 --> 00:54:26,290
But then it wasn't because the job was not the right thing to do, but most people most of the time.

542
00:54:26,290 --> 00:54:32,410
So we don't actually know what we're doing. One of the wonderful things we face, that's a long phrase.

543
00:54:33,040 --> 00:54:42,009
One of the unusual things about COVID was it was perfectly legitimate for doctors to say, we don't know what we're doing.

544
00:54:42,010 --> 00:54:47,710
We don't know how to treat this. That's very unusual in my field of cardiology.

545
00:54:49,870 --> 00:54:53,499
Of all the international guidelines published by the American Heart Association,

546
00:54:53,500 --> 00:54:57,549
European Society of Cardiology, and so on, if you look at all those guidelines,

547
00:54:57,550 --> 00:55:00,460
Rockliff, who is now FDA commissioner a few years ago,

548
00:55:00,730 --> 00:55:05,950
did a review of all those guidelines and says which of these are based on good evidence from randomised trials?

549
00:55:06,790 --> 00:55:11,470
15% were based on good evidence from randomised trials, 85% were not.

550
00:55:11,500 --> 00:55:18,100
Now a few of those things that are not were things like is smoking, is smoking harmful for the heart?

551
00:55:18,280 --> 00:55:21,970
You don't do randomised trials of that, doesn't need a randomised trial and so on.

552
00:55:22,180 --> 00:55:28,749
So it's a bit extreme. But the point is that even in a evidence based speciality like cardiology,

553
00:55:28,750 --> 00:55:35,320
which is probably at the top of the tree along with some types of cancer, huge amounts of treatment decisions are based on.

554
00:55:35,560 --> 00:55:42,340
We sort of don't know. We don't know. We've got experimental evidence, theoretical evidence, some experience, whatever.

555
00:55:42,340 --> 00:55:46,690
But bottom line, we don't actually know. Think about the niacin example I gave earlier.

556
00:55:47,050 --> 00:55:51,640
Seems it lowers cholesterol. It's a vitamin. It's been around 50 years.

557
00:55:51,640 --> 00:55:56,800
Nobody's ever noticed it causing anything other than Flushing. Why not?

558
00:55:57,130 --> 00:56:03,370
Well, as it turns out, it's actually makes people go. Of course, if people are going to hospital with, in fact, nasty infection and bleeding.

559
00:56:03,370 --> 00:56:11,919
So this assumption that we know what we're doing is sort of touching but misplaced in some ways.

560
00:56:11,920 --> 00:56:16,239
I don't want to put patients off in COVID.

561
00:56:16,240 --> 00:56:20,200
It was perfectly legitimate to say we just don't know what the right treatments are.

562
00:56:20,290 --> 00:56:25,850
There are no trials. We don't know what we're doing. So how did you go about recruiting the patients for the trial?

563
00:56:25,870 --> 00:56:32,780
So, so, so, so just in brief, the early bit because it has been talked about in a number of before.

564
00:56:34,420 --> 00:56:36,610
Number one, we had to get the trial up and running as quickly as possible.

565
00:56:36,610 --> 00:56:41,530
That took us nine days, which as opposed to normally nine months, 18 months.

566
00:56:44,110 --> 00:56:50,600
My first worry was, how do we get started? My second my second worry was how do we get it into every hospital or as many hospitals we can?

567
00:56:50,620 --> 00:56:56,739
I thought initially if we could get into 60 hospital GP doing well, but when we set up trials here, it can take us,

568
00:56:56,740 --> 00:57:01,629
you know, we might do, I don't know what it is, five or ten hospitals a month if we're lucky or something.

569
00:57:01,630 --> 00:57:07,620
It takes a long time working around hospital, my hospital, getting the paperwork sorted out and all the rest and making sure you've sent

570
00:57:07,640 --> 00:57:11,200
those little barcodes and labels to the right address and all those other things.

571
00:57:11,950 --> 00:57:23,049
And I thought we had that long. And Lucy Fletcher, who's an experienced trauma as you're from here and and others in her team, did a wonderful job.

572
00:57:23,050 --> 00:57:28,150
They got every hospital in the country set up within about within about eight weeks.

573
00:57:28,150 --> 00:57:36,820
So about within within four weeks we had three course of those hospitals already set up, including many, many hospital.

574
00:57:37,930 --> 00:57:43,750
They all every hospital but many of this hospital had never done clinical trials, never done these sorts of clinical trials before.

575
00:57:45,910 --> 00:57:49,240
And, of course, a trial at those hospitals in diverse parts of the country.

576
00:57:50,090 --> 00:57:53,790
Did you have to go up to get authority to go down to the hospitals?

577
00:57:53,800 --> 00:57:59,590
Yeah. Well, you have to you have to get ethics approval and regulatory approval and so on and so forth.

578
00:58:00,810 --> 00:58:08,230
We did so. We did. We got a letter from the chief chief medical officer and the director of.

579
00:58:09,220 --> 00:58:19,930
So the Chief Medical Officer for each bit of the UK. So each nation plus the medical director of the NHS to write a letter which basically said that

580
00:58:19,930 --> 00:58:25,210
the trial is to be deemed as part of standard of care and not an optional extra or on the side.

581
00:58:25,600 --> 00:58:30,040
And we strongly encourage you to take part. And that letter was extraordinarily useful.

582
00:58:30,040 --> 00:58:40,149
So it is easy to get with that. But yes, I mean, I would say I would say yes, in in that circumstance.

583
00:58:40,150 --> 00:58:44,139
I mean, it takes some takes some effort, it takes some persuasion.

584
00:58:44,140 --> 00:58:47,830
You've got to put your case and whatever else. But yes,

585
00:58:48,310 --> 00:58:56,770
I think it was and that letter mattered a lot because suddenly this was now part of the day job at a time

586
00:58:56,770 --> 00:59:01,870
when hospitals were working out how the [INAUDIBLE] they were going to cope with an unprecedented crisis,

587
00:59:02,260 --> 00:59:06,879
how they were going to deal with staff numbers, how they're going to do with wards, how they get more ICU beds,

588
00:59:06,880 --> 00:59:12,190
or whether they're going to contribute to the Nightingale Ward, hospitals, all that stuff as well.

589
00:59:12,430 --> 00:59:17,290
That letter was very, very helpful. So that was a big part of it and.

590
00:59:17,430 --> 00:59:25,630
To train the doctors who are in these hospitals. So we did all the training online with short ten minute videos for different bits.

591
00:59:25,650 --> 00:59:29,460
If you want to take consent, you look at this page, if you want to fill in this form, you look at that bit and so on.

592
00:59:29,940 --> 00:59:32,429
We made the trial open to as many staff as we could.

593
00:59:32,430 --> 00:59:38,130
We realised that you couldn't just have one member of staff who was the only person in the hospital trained to do the trial.

594
00:59:38,430 --> 00:59:44,700
We had to make sure that you had the frontline doctors who are junior doctors, many of them, that they could also take part in the trial.

595
00:59:44,940 --> 00:59:57,690
We ended up with nigh on 10,000 doctors, nurses, pharmacists, our back office, R&D staff, whatever, all who have contributed in some way to the trial.

596
00:59:59,340 --> 01:00:03,450
They were all listed in the in the authorship list, in the appendix to all the papers.

597
01:00:04,710 --> 01:00:08,280
We had to make them silly. Question But did you need an NHS to do that?

598
01:00:08,280 --> 01:00:11,429
If we hadn't have the NHS, would it have been possible?

599
01:00:11,430 --> 01:00:19,320
It doesn't sound like a stupid question. It's not a question of they often ask, but often friends in the United States DNA, the NHS helps enormously.

600
01:00:21,300 --> 01:00:25,360
Having data helps enormously. But it would be possible.

601
01:00:25,440 --> 01:00:34,500
I think it would be possible to do the same mechanism if you didn't actually have a whole NHS and if you didn't actually have the data that we have.

602
01:00:36,150 --> 01:00:38,549
If you look at in the United States, for example,

603
01:00:38,550 --> 01:00:47,610
Kaiser Permanente is one of the big health insurance systems out there and I come with exactly how many patients they have,

604
01:00:48,360 --> 01:00:53,730
but I think it's probably more than the size of Scotland. You know, it's you know, it's a vast number.

605
01:00:54,960 --> 01:01:00,060
There's no reason why this sort of clinical trial couldn't have been done within one of those sort of health care systems.

606
01:01:00,690 --> 01:01:10,860
And in fact, actually, throughout the last few years, I've had many conversations with with leaders in the US about why they couldn't do it,

607
01:01:10,980 --> 01:01:14,309
why they didn't do it, why they couldn't do it, and how can they copy it now?

608
01:01:14,310 --> 01:01:21,600
In fact, even just this week, Rob, Caleb and two other senior leaders from FDA have written an opinion piece in

609
01:01:21,600 --> 01:01:28,440
JAMA exactly about the US needs to learn from recovery for clinical trials,

610
01:01:28,440 --> 01:01:33,120
what they call it at the point of care. In other words, at the front line too, I think it's the first time we mentioned recovery.

611
01:01:33,120 --> 01:01:39,810
So this trial was called recovery. Yes. Which stood for randomised evaluation of COVID therapies, I think.

612
01:01:42,270 --> 01:01:48,690
Yeah. Peter came up with the acronym. You need an interesting you have to have somebody who can think of the name and that's than me.

613
01:01:53,040 --> 01:02:04,050
So yeah, I think we've done consensus about testing existing drugs and we've done funding, we've done that and we've done that.

614
01:02:04,440 --> 01:02:11,759
So yes, so we you get the drugs from so for drugs like dexamethasone and hydroxychloroquine, the NHS has them anyway.

615
01:02:11,760 --> 01:02:15,239
So what we said was you might describe these anyway.

616
01:02:15,240 --> 01:02:21,389
So rather than prescribes and randomised and they would if you like, for regular NHS stock for some of the drugs.

617
01:02:21,390 --> 01:02:29,430
So Tocilizumab was donated by Roche. This was sort of six weeks or so into the study when the study was going well.

618
01:02:30,420 --> 01:02:37,049
They and I think that was a very good relationship and I think it's actually a sort of model that one would want to copy.

619
01:02:37,050 --> 01:02:42,870
It's not a question of, Oh, this is an industry trial or this is an academic trial, this is a genuine collaboration.

620
01:02:42,870 --> 01:02:48,090
We have with Oxford University is the sponsor of the trial. That's a sort of regulatory term responsible for the trial.

621
01:02:49,230 --> 01:02:56,120
Under all the rules, we're studying lots of drugs. But Roche have a drug that looked interesting.

622
01:02:56,120 --> 01:02:59,910
It was recommended, recommended as sort of prioritised, as worthy of study.

623
01:03:00,660 --> 01:03:04,110
They gave it to us. They didn't give us any money. That's fine.

624
01:03:04,120 --> 01:03:07,500
That we put it plugged it into the trial. We do the trial.

625
01:03:07,650 --> 01:03:11,219
We produce produced the data, we publish the data.

626
01:03:11,220 --> 01:03:15,330
We talk to the regulators and others and health policy people and so on.

627
01:03:15,330 --> 01:03:24,510
And whoever wants to know to explain the data, we also return it back to Roche and then Roche are able to use it for their own internal purposes,

628
01:03:24,540 --> 01:03:31,949
develop new treatments, whatever they want to do, but they're also able to use it to apply to regulators for a an extension to the drug licence.

629
01:03:31,950 --> 01:03:34,980
So it's not just for rheumatology, it's now for COVID as well.

630
01:03:35,580 --> 01:03:43,170
And that's fine. Actually, it's a very good, if you like, independent test of a particular treatment that happens to be come from industry.

631
01:03:43,170 --> 01:03:47,549
All treatments come from industry. Ultimately. The second one was the it was Regeneron.

632
01:03:47,550 --> 01:03:53,220
That was the second big example was Regeneron who donated antibodies.

633
01:03:53,220 --> 01:03:57,000
Monoclonal antibodies stick on the virus and nobody knew whether they would work.

634
01:03:57,330 --> 01:04:03,420
By the time you get to hospital is interesting the thoughts about what the disease had changed by this point.

635
01:04:03,420 --> 01:04:08,129
This is thought six months in or something because the dexamethasone or anybody said, oh, no, it's an inflammatory disease.

636
01:04:08,130 --> 01:04:12,390
You don't need antiviral treatments, the virus isn't relevant, so it's completely switched their minds.

637
01:04:13,530 --> 01:04:16,950
So yeah. With these antibodies work in patients in.

638
01:04:17,030 --> 01:04:20,030
Hospitals. So they also gave us the drug.

639
01:04:21,110 --> 01:04:24,319
Why is it a genuine one versus whatever other ones were about?

640
01:04:24,320 --> 01:04:28,490
Well, it was the Vaccines Taskforce led by Kate Bingham,

641
01:04:29,000 --> 01:04:35,720
who actually was charged with doing the sort of prioritisation and and selection of which of the many antibodies were to be used,

642
01:04:36,920 --> 01:04:39,469
why they came on the vaccines versus something else, I have no idea.

643
01:04:39,470 --> 01:04:45,800
But they did a very good job at considering the different treatments that are available.

644
01:04:46,160 --> 01:04:50,600
So if you think just think about like it, does it actually stick to the version of viruses going around at the moment,

645
01:04:51,500 --> 01:04:57,020
but also how, how, how many people is it been used in so far and how much stock actually is there?

646
01:04:57,020 --> 01:05:00,980
Because, you know, if there's only enough for a few hundred people, well, that's no good to anybody.

647
01:05:03,020 --> 01:05:06,530
And yet more recently, other other companies have done similar.

648
01:05:06,530 --> 01:05:13,340
So GSK, for example, have given us SOTROVIMAB, which is another antibody which we're studying at the moment.

649
01:05:14,420 --> 01:05:18,260
So we we slightly skipped over the the results. Let's go back to that.

650
01:05:18,260 --> 01:05:24,050
So you've got the whole thing up and running really quickly. Oh, when did you first have something you could report?

651
01:05:24,320 --> 01:05:29,330
Well, the first results came out in in June of that year.

652
01:05:29,330 --> 01:05:38,360
So within so I said nine days to get the first patient in in less than 100 days, we had 12,000 patients.

653
01:05:40,520 --> 01:05:50,929
And the first two results, the first one to raise its head was hydroxychloroquine, which it was quite clear.

654
01:05:50,930 --> 01:05:54,200
The doctor told us we should look at the data.

655
01:05:54,350 --> 01:06:00,440
It was quite clear that this was a treatment that didn't have any benefit, any chance of benefits.

656
01:06:00,440 --> 01:06:04,040
It might even be harmful, but it certainly had no benefits.

657
01:06:06,170 --> 01:06:17,690
We announced that in I think it's something like the 6th of June that had the impact that I expected, number one, and I personally had been hopeful,

658
01:06:17,690 --> 01:06:24,559
which is that hydroxychloroquine largely stopped being used not only in the UK but around the world, not exclusively.

659
01:06:24,560 --> 01:06:31,970
Some people carried on. They would do but a number to the predictable backlash,

660
01:06:33,080 --> 01:06:39,380
which is there were a number of people who are or have been very, very strong hydroxychloroquine advocates.

661
01:06:41,990 --> 01:06:51,200
It is interesting that people hold on to almost beliefs, if you like, and some treatments just get a sort of a momentum behind them,

662
01:06:51,560 --> 01:06:56,540
as if by sheer willpower, a drug can suddenly can a virus and be good for patients.

663
01:06:57,350 --> 01:07:02,690
All the drugs we've studied, I've hope they work. I've had reason to think that they might work.

664
01:07:04,070 --> 01:07:09,380
But then you do the test, you do the trial to find out whether they actually work, how well they work, and in whom they work.

665
01:07:09,830 --> 01:07:14,750
And this one doesn't work. That's the result is a good job.

666
01:07:14,750 --> 01:07:21,620
We got a clear result. A promising drug is not is not it turns out not to be a useful drug.

667
01:07:21,860 --> 01:07:32,509
That's the and that's the end of that story. But it didn't stop some fairly vociferous comments and feedback from some quarters.

668
01:07:32,510 --> 01:07:40,220
As I say from yeah, for 90% of it it was a lot of people extremely grateful to actually know the answer to what was an important question.

669
01:07:40,850 --> 01:07:42,440
And then the second one was the dexamethasone,

670
01:07:42,440 --> 01:07:49,250
which we pitched on you pretty much at the same time within a few days or whatever, what the result was.

671
01:07:49,850 --> 01:07:55,190
But we then because of that result, we then had to spend a week really making sure that we got it right.

672
01:07:56,300 --> 01:07:57,560
When we first looked at the data,

673
01:07:57,560 --> 01:08:06,230
here was a result that had a very clear clinically and statistically significant reduction in mortality for for the sickest people,

674
01:08:06,230 --> 01:08:15,560
people on ventilators and people on oxygen. And at that stage, no treatments available, no vaccine in sight.

675
01:08:18,470 --> 01:08:24,290
We were still in the first lockdown here in the UK. Other parts of the world were really struggling.

676
01:08:24,290 --> 01:08:29,390
We didn't know quite how bad Africa was going to get and so on and so forth.

677
01:08:29,620 --> 01:08:36,979
And so we had this result for a drug that costs £5 even less in some parts of the world is on the essential medicines list,

678
01:08:36,980 --> 01:08:39,980
is in every pharmacy, in pretty much every hospital in the world.

679
01:08:41,330 --> 01:08:47,030
And we have this result that says it saves lives. And you know that the moment you open your mouth, the world changes.

680
01:08:47,240 --> 01:08:57,350
So we did have to spend a week really digging into the data and making absolutely certain that the that we understood the data,

681
01:08:57,350 --> 01:09:05,989
that we got the data right, we got the messages right. We knew what we had to make sure that we got it straight.

682
01:09:05,990 --> 01:09:12,020
And then on the 16th of June, we announced it at the lunchtime on a press conference through the Science Media Centre.

683
01:09:13,370 --> 01:09:16,700
We wanted to make sure that this got out as a science and health story.

684
01:09:16,840 --> 01:09:19,030
First, not a political story.

685
01:09:20,260 --> 01:09:30,190
We were asked if we would look we would say at first at ten Downing Street and we say, well, we're happy to come to ten Downing Street,

686
01:09:31,450 --> 01:09:37,600
but we're going to announce it to the science media first because we really want to make sure that we got it.

687
01:09:37,660 --> 01:09:41,649
Doctors need to know the site, the science story. They need to know the clinical story.

688
01:09:41,650 --> 01:09:51,160
They need to know the data that's much more important than the sort of the politics and whatever else.

689
01:09:51,640 --> 01:10:03,430
So, yeah, we announced it at lunchtime and by the time it was then we got another invite from number ten saying, Please come to the daily briefing.

690
01:10:04,540 --> 01:10:06,990
There's two of us. They said, Well, we can only take one.

691
01:10:07,000 --> 01:10:13,000
So I've been to number ten once before for something completely unrelated and low key picked out.

692
01:10:13,150 --> 01:10:22,690
So we decided we'd send Peter in my time. And so one of the artefacts that is going to go in is, is my tie, which was,

693
01:10:23,050 --> 01:10:27,700
which is also since appeared at Buckingham Palace three times when Peter went to get his knighthood.

694
01:10:27,700 --> 01:10:31,110
When I went to get my knighthood and when Richard Heinz went to get his MBA.

695
01:10:31,120 --> 01:10:35,760
So this tie, not not the one I'm wearing right now.

696
01:10:35,770 --> 01:10:39,100
No, it's too precious these days. It doesn't come out.

697
01:10:39,470 --> 01:10:42,630
And so, yeah.

698
01:10:42,640 --> 01:10:48,490
So Peter announced the results again with, with, uh, Boris Johnson.

699
01:10:49,590 --> 01:10:52,570
There's one of those 5:00 briefings. By the time he did so,

700
01:10:53,230 --> 01:11:04,360
the NHS had issued a statement to every hospital saying Dexamethasone is now to be considered standard of care in patients who are on oxygen,

701
01:11:04,370 --> 01:11:11,140
on ventilators. So nine days to get the study started. Less than 100 days to get the first results.

702
01:11:11,530 --> 01:11:18,090
3 hours to get it into policy. And that isn't just chance.

703
01:11:18,100 --> 01:11:22,570
That's part of the when you ask a good question, you will to answer it robustly,

704
01:11:22,750 --> 01:11:27,100
you want to answer it in a way that will change practice, that will change the world, if you like.

705
01:11:27,430 --> 01:11:30,040
It's a very it's a very unique set of circumstances.

706
01:11:31,240 --> 01:11:36,530
But that was always our objective was to get an answer that it was just clear cut what you should do.

707
01:11:36,550 --> 01:11:39,160
We did that with hydroxychloroquine and we did that with steroids.

708
01:11:40,090 --> 01:11:50,620
And over the subsequent few weeks, W.H.O., NIH, FDA, EMA, all these other people all accepted the results.

709
01:11:50,620 --> 01:11:56,860
And it was and it was adopted very rapidly. And is there an estimate of how many people's lives?

710
01:11:56,950 --> 01:12:02,559
Well, yeah. At the time the estimate came out, which was the following march, I wasn't terribly comfortable with it.

711
01:12:02,560 --> 01:12:07,150
But I think it must be true by now, which is that at least 2 million lives worldwide have been saved as a consequence.

712
01:12:09,160 --> 01:12:16,899
Impossible to know, because you can't count. You can't count how many people nothing happens to, if you like.

713
01:12:16,900 --> 01:12:21,730
They survived. But I think that's a that's a reasonable estimate.

714
01:12:22,420 --> 01:12:27,700
We've had three further treatments tocilizumab the genuine antibody in people who don't

715
01:12:27,760 --> 01:12:33,459
already have their own antibodies and baricitinib another arthritis drug since then.

716
01:12:33,460 --> 01:12:38,440
So we with over all we've studied included 48,000 people to date.

717
01:12:40,570 --> 01:12:49,090
We've got, I think it's ten or possibly 11 results by now for clear treatments that can be used in the sickest patients in particular circumstances.

718
01:12:49,510 --> 01:12:56,020
Six others that shouldn't be used. But that's good to know. And you have changed the way people have thought about trials.

719
01:12:56,020 --> 01:13:06,310
The feedback from frontline doctors, frontline patients has been remarkable in for the doctors.

720
01:13:06,640 --> 01:13:09,730
This was at last. Many of them had never done trials before.

721
01:13:10,030 --> 01:13:14,740
This is a chance to help generate an answer, find a solution to this crisis,

722
01:13:15,070 --> 01:13:19,420
rather than merely having to sort of suffering struggle their way through it.

723
01:13:20,680 --> 01:13:24,520
We had hospitals, as I say, who've never done trials before.

724
01:13:24,970 --> 01:13:29,500
Somewhere in South Tees, you know, nearly a thousand patients in one hospital,

725
01:13:30,010 --> 01:13:36,250
that one hospital put almost as many patients into the trial as entire trials in the United States.

726
01:13:37,750 --> 01:13:42,729
And one of the lessons has been in the United States is and this is a sort of analysis by Janet Woodcock,

727
01:13:42,730 --> 01:13:48,490
who was acting chief commissioner of the FDA at the time, is now chief deputy commissioner.

728
01:13:50,260 --> 01:13:56,319
She's analysis in all all the registered trials registries of ongoing clinical trials.

729
01:13:56,320 --> 01:13:59,980
It's not 2700 to 800 clinical trials of COVID.

730
01:14:00,910 --> 01:14:06,940
And she went through them, and 95% of them, in her view, never had a chance of answering the question.

731
01:14:06,940 --> 01:14:14,890
They were either not randomised or they were too small. And that I think is a lesson in that we don't need lots and lots more trials.

732
01:14:15,340 --> 01:14:22,040
We need more. The trials. And you have to ask yourself what the why is that come about?

733
01:14:23,390 --> 01:14:34,730
And in large part, that's because the academic credit for leading a trial that never produces an answer is substantially greater

734
01:14:34,730 --> 01:14:41,420
than the academic trial credit for contributing a very small part to a trial that changes the answer.

735
01:14:42,530 --> 01:14:49,249
Put it very crudely, if you think that recovery may have saved a million lives and I told you a little earlier that

736
01:14:49,250 --> 01:14:54,649
there might be 10,000 NHS staff of one sort or another who've contributed in one way or another,

737
01:14:54,650 --> 01:14:58,850
big or small or whatever. You divide one number by the other number.

738
01:14:59,390 --> 01:15:04,340
That's the contribution, the average contribution in terms of lives saved of each of those people.

739
01:15:04,880 --> 01:15:11,420
And you compare that with those 95% of 2700 trials on average is so well over 2000 trials

740
01:15:11,780 --> 01:15:18,049
where people have played a leading role in something that either has never got anywhere,

741
01:15:18,050 --> 01:15:24,170
would never had a hope of getting anywhere. And I think we have to think about in academia, in our reward structures.

742
01:15:24,170 --> 01:15:28,970
And so we have to think very, very differently. Peter and I have always been very clear.

743
01:15:29,840 --> 01:15:36,890
Peter and I have got lots of credit, lots of accolades and lots of attention, if you like.

744
01:15:37,040 --> 01:15:41,870
Not all of it good. But we have and we are sort of the frontmen for the recovery trial.

745
01:15:42,590 --> 01:15:49,610
But the contribution of so many people has been is what is what made it possible.

746
01:15:49,610 --> 01:15:55,760
So 48,000 patients who've taken part in the trial in really difficult circumstances for them,

747
01:15:56,480 --> 01:16:03,350
10,000 doctors and nurses who were completely overwhelmed emotionally, time pressure, everything else.

748
01:16:04,100 --> 01:16:11,659
And then I don't know what it is, 50 people or something wobbly here in Oxford in the clinical trials unit over time,

749
01:16:11,660 --> 01:16:14,780
probably 25 at any one time who've contributed.

750
01:16:16,450 --> 01:16:24,049
It's a much bigger and broader teamwork than just who'll be in LANDRO again,

751
01:16:24,050 --> 01:16:28,130
if I think about it purely from sort of Oxford University or an academic perspective,

752
01:16:28,940 --> 01:16:32,870
I think that this would I know that this would have been impossible,

753
01:16:33,590 --> 01:16:43,460
impossible without the 20 years of experience of working out how to do clinical trials, number one, actually understanding the fundamental principles.

754
01:16:44,390 --> 01:16:52,550
There wasn't a protocol, a standard operating procedure or whatever that could prepare anybody for trying to do a trial in the middle of a pandemic.

755
01:16:53,120 --> 01:17:03,140
You had to understand the the basic principles of good design, of good ethics and so on and so forth.

756
01:17:03,440 --> 01:17:09,110
In order to work out how you designed a trial that would not only answer the question,

757
01:17:09,410 --> 01:17:16,010
but also that anybody could take part in seems true across all the sort of technical elements.

758
01:17:16,010 --> 01:17:22,210
So the sorts of people who built a computer system for the trial in nine days have been here for,

759
01:17:22,220 --> 01:17:26,600
I'm guessing, 20 years or so, long term sustainable funding.

760
01:17:26,960 --> 01:17:32,840
The people who were the trial managers, the people who were the data managers managing data,

761
01:17:32,840 --> 01:17:38,000
shuttling backwards and forwards between different computers here, or with NHS, digital or whatever else.

762
01:17:38,510 --> 01:17:41,810
All of these different types of people, the people who test the computers,

763
01:17:42,470 --> 01:17:50,360
those long term, that long term experience of technical staff is fundamental.

764
01:17:51,650 --> 01:17:54,830
I think that there's a there's something that we all have to reflect on,

765
01:17:54,830 --> 01:18:02,810
which is actually how do we ensure that these people are sustained, their careers are rewarding, that we can retain them?

766
01:18:03,350 --> 01:18:08,030
And, you know, a significant chunk of that is a threat in terms of salaries,

767
01:18:08,030 --> 01:18:14,120
in my view, academia, academic salaries are not the same as industry salaries.

768
01:18:15,170 --> 01:18:19,490
Those technical people are not on an academic pathway.

769
01:18:19,820 --> 01:18:26,240
If I interviewed on any one of those for a new job, I wouldn't be interested in how many people they published,

770
01:18:26,510 --> 01:18:29,900
how many students they supervised, or how much grants they got come in coming in.

771
01:18:30,800 --> 01:18:34,490
In fact, I wouldn't want them to do any of those things. Number one, they're probably not very good at them.

772
01:18:34,670 --> 01:18:38,480
Number two, I want them to do the things they are really good at, which is programming computers,

773
01:18:38,750 --> 01:18:42,620
managing data, sorting out ethics applications and the sorts of things.

774
01:18:42,620 --> 01:18:51,409
So I think one of the things that was was so true of recovery was in on the ninth and 10th of March, I had a series of phone calls.

775
01:18:51,410 --> 01:18:56,300
I'd ring people up and I say, Covid's coming, we've got this trial.

776
01:18:57,470 --> 01:19:02,870
I really think you could do with some help. Do you think you could do your bit?

777
01:19:03,570 --> 01:19:06,770
Oh, by the way, I need an answer by the end of this conversation.

778
01:19:07,070 --> 01:19:13,610
I don't mind whether it's yes or no. I really don't mind. But the time pressure was such that I need to know.

779
01:19:14,060 --> 01:19:20,940
And they all said yes, but it was big. Because those people were there and able to help, of course, other things.

780
01:19:21,570 --> 01:19:26,310
It wasn't a question of dropping other stuff. Other stuff was being dropped because nothing else much was going to be happening.

781
01:19:27,120 --> 01:19:34,500
So that's a slightly different scenario. But the reality is that those people here with their experience in there and and and their expertise.

782
01:19:34,920 --> 01:19:39,600
And so I think as we looking ahead, we have to think about as a university.

783
01:19:39,750 --> 01:19:43,140
It's not just Oxford University. It's true for across academia.

784
01:19:43,440 --> 01:19:53,370
How does one get that sustained long term investment in the skills of the technical people that allow the academic people to then flourish?

785
01:19:54,390 --> 01:19:59,670
Very good. That's usually where I end up. But I've just got a few things that if we can just skip through and we've only got 5 minutes left,

786
01:20:00,300 --> 01:20:04,500
which are really about the the impact of the pandemic on you personally.

787
01:20:06,780 --> 01:20:11,220
Well, the questions that I usually ask, I'll put them all at once, and then you can take them in whatever role do you like.

788
01:20:12,450 --> 01:20:16,410
How did it change the way you worked? I mean, presumably where you home based or did you come in?

789
01:20:17,970 --> 01:20:24,780
Did you feel personally threatened by the virus? And how did it how close did it come to you as an as an infection?

790
01:20:25,740 --> 01:20:31,440
And were you continuing to do clinical work in the hospital? And so were you having to go in and get ground up and do all that stuff?

791
01:20:31,740 --> 01:20:45,090
So I the first week or so of the planning, we had weekly we had daily meetings of the sort of growing recovery team.

792
01:20:46,050 --> 01:20:49,920
And by the time we got to the Thursday, I said, We can't do this anymore.

793
01:20:51,180 --> 01:20:55,590
We're about to be thrown out. We really can't be meeting like this.

794
01:20:56,020 --> 01:20:59,160
And from then on, we were everything was remote and we all went home.

795
01:21:01,080 --> 01:21:04,950
Some people didn't meet each other ever for over a year.

796
01:21:05,160 --> 01:21:09,560
So some people had never worked before before with each other before,

797
01:21:09,590 --> 01:21:14,820
never knew each other before and didn't actually meet each other physically for well over a year.

798
01:21:17,430 --> 01:21:25,770
So, yes, I did most of this from my desk at the bottom of the staircase in west Oxfordshire in the village I grew up in.

799
01:21:27,750 --> 01:21:35,130
In terms of my personal feelings, my clinical works, my clinical work,

800
01:21:36,240 --> 01:21:41,100
I was never at the front line, so I was never gowned up in all those things I did do.

801
01:21:42,210 --> 01:21:52,560
I dropped my clinics and got somebody else to cover them for about a month in the middle of this when it got when it was extreme but otherwise not.

802
01:21:52,570 --> 01:22:01,860
I didn't get I wasn't and that was I'm lucky and I recognise that in a sense it was as a sort of irf would say,

803
01:22:01,860 --> 01:22:10,980
you know, my wall was flying a desk and yes, I was taken out of that version of the front line.

804
01:22:11,700 --> 01:22:17,040
I go to a different version of the front line, which was the sort of ee political stress.

805
01:22:17,040 --> 01:22:21,930
I don't just mean I don't actually mean, you know, party politics or whatever.

806
01:22:22,440 --> 01:22:31,739
But there were big political issues all the way through policy issues and so on in health and in in finance across the whole piece.

807
01:22:31,740 --> 01:22:34,770
And so I did see a huge amount, amount of that.

808
01:22:34,770 --> 01:22:38,460
So I had my own version of stress. And how did you deal with that?

809
01:22:43,170 --> 01:22:50,400
Well, my my children are all adults, but they all came home and all three of them came home.

810
01:22:50,430 --> 01:22:55,400
My wife was at home. And I guess we sort of and of course, you can go anywhere.

811
01:22:55,410 --> 01:22:59,280
The dog got a lot of walking because that was the only thing you were allowed to do.

812
01:23:01,470 --> 01:23:06,330
But personally, I personally felt okay.

813
01:23:07,440 --> 01:23:13,770
I mean, I worried about the virus I had. I was admitted to intensive care unit with pneumonia 15 years ago or so.

814
01:23:13,980 --> 01:23:18,990
Completely different. Uh, but I did know what being a patient on it, you.

815
01:23:18,990 --> 01:23:21,930
You felt like I didn't get ventilated, but I did know what I felt like.

816
01:23:22,290 --> 01:23:28,650
And I had been a junior doctor doing 56 hour shifts back in Birmingham in the early 1990s.

817
01:23:29,190 --> 01:23:34,950
So I knew what it was like. Yeah. Being completely knackered at the front line for for the doctors.

818
01:23:36,960 --> 01:23:46,560
But it did seem to me that and the other thing that the thought kept going through my mind was that we learned about history going backwards,

819
01:23:46,560 --> 01:23:50,580
looking backwards. So we know how the story ended. Think about the Second World War.

820
01:23:52,200 --> 01:24:02,549
You know that, you know, in the middle of the blitz in 19 1940, 1941, you know how that story is going to end when you're actually there at the time.

821
01:24:02,550 --> 01:24:06,900
You don't know how it's going to end. You don't know when it's going to end and you don't know if you're going to be there to see it.

822
01:24:08,580 --> 01:24:15,660
Personally, I never was too worried about whether I would be there to see it, but I thought this year we just didn't know what the future looked like.

823
01:24:17,250 --> 01:24:24,629
And actually keeping that mind set was useful to actually think, you know, this is actually natural.

824
01:24:24,630 --> 01:24:30,150
This is understandable because I think many people were sort of thinking, well, you know, just give it another few weeks.

825
01:24:30,150 --> 01:24:36,690
And, you know, it was quite obvious to me that we were we were we were we were living this forward.

826
01:24:36,720 --> 01:24:40,170
And that's how history actually works to the people who live through history.

827
01:24:40,500 --> 01:24:46,650
If you live it, you live it forwards and you don't quite know where it's going to end and you don't quite know how it's going to end.

828
01:24:46,710 --> 01:24:53,670
You don't quite know whether you're going to be there. It's a it was a particular view or mindset that I had.

829
01:24:55,680 --> 01:25:07,260
So, I mean, the biggest stresses were were were issues around when were larger issues around things like hydroxychloroquine.

830
01:25:08,880 --> 01:25:12,660
We came under a lot of pressure to stop the trial early of issues like dexamethasone.

831
01:25:12,700 --> 01:25:18,210
We got a lot of pressure not to do the trial after we published results of hydroxychloroquine.

832
01:25:18,780 --> 01:25:30,480
You know, we got a lot of limited quarters, we got a lot of really quite adverse correspondence and so on,

833
01:25:31,080 --> 01:25:36,660
you know, complaints and, you know, nasty letters to high up people and all this sort of stuff.

834
01:25:37,410 --> 01:25:41,670
I was never thankfully actually directly threatened.

835
01:25:41,670 --> 01:25:46,050
I know colleagues who were pretty, pretty unpleasant.

836
01:25:47,370 --> 01:25:50,429
Uh, we studied drugs to find out what the right answer was.

837
01:25:50,430 --> 01:25:55,920
That was all I was interested in doing. The thing I was very pleased that we did, which we haven't mentioned,

838
01:25:56,250 --> 01:26:02,220
was that we took a view on day one that this had to be we had to be completely transparent and completely open.

839
01:26:02,760 --> 01:26:05,790
And so we put everything that we could on the public website.

840
01:26:05,790 --> 01:26:10,230
How many people have been recruited? All the protocols, all the ethics and letters, everything else, all on the on there.

841
01:26:10,860 --> 01:26:18,360
And also that we were very proactive and open on the sort of media front to, you know,

842
01:26:18,360 --> 01:26:25,649
taking all those interviews, to, you know, taking opportunities to explain why we needed a trial is hard.

843
01:26:25,650 --> 01:26:33,690
You know, help me, doctor. And the doctor says, well, when you toss a coin is a sort of odd sort of you know, it's an odd set up.

844
01:26:34,440 --> 01:26:36,569
But actually, it turns out that for some of those treatments,

845
01:26:36,570 --> 01:26:41,910
tossing a coin was an awful lot better than giving them hydroxychloroquine or convalescent plasma or a number of other things.

846
01:26:41,910 --> 01:26:47,219
You know, give it prescribing something is not is not necessarily doing somebody good.

847
01:26:47,220 --> 01:26:54,510
And yeah, some of these treatments are called considered to be prescribed, but on a compassionate use basis,

848
01:26:55,440 --> 01:27:01,800
and I've never quite understood that it's compassionate to give something. You got no idea whether it's any good or not, it.

849
01:27:02,280 --> 01:27:09,000
So being transparent, you're making sure we got this story out there was important.

850
01:27:09,480 --> 01:27:13,210
And as a result, I mean, people do know what a trial is now. Yeah.

851
01:27:14,940 --> 01:27:18,960
And I think we have opened up that area and opened up a lot of people minds.

852
01:27:21,530 --> 01:27:21,760
And.