WEBVTT

1
00:00:00.600 --> 00:00:08.199
<v Speaker 1>All of us in this room are involved with clinical trials, writing, delivering,

2
00:00:09.039 --> 00:00:14.800
<v Speaker 1>participating in clinical trials, and we know that often in

3
00:00:14.839 --> 00:00:19.920
<v Speaker 1>those trials we need to make changes. So today I'm

4
00:00:19.960 --> 00:00:23.879
<v Speaker 1>going to talk to you about to amend or not

5
00:00:24.079 --> 00:00:32.200
<v Speaker 1>to amend. That truly is the question. So what's the

6
00:00:32.280 --> 00:00:37.520
<v Speaker 1>problem right in our world? We need to be flexible,

7
00:00:37.560 --> 00:00:40.799
<v Speaker 1>We need to be innovative. We need to be able

8
00:00:40.920 --> 00:00:46.439
<v Speaker 1>to move and change our studies as the information comes

9
00:00:46.439 --> 00:00:51.039
<v Speaker 1>in that we require it. But what we've seen is

10
00:00:51.840 --> 00:00:56.960
<v Speaker 1>that certainly within ROSH, we traditionally have had a high

11
00:00:57.039 --> 00:01:02.960
<v Speaker 1>level of protocol amendments, even compare to our peers. Now,

12
00:01:03.000 --> 00:01:05.519
<v Speaker 1>some of my colleagues would say, well, because we're on

13
00:01:05.599 --> 00:01:12.079
<v Speaker 1>the forefront of innovation, a science demands that we are flexible.

14
00:01:12.680 --> 00:01:18.920
<v Speaker 1>And that's true, But is there a case for some

15
00:01:19.079 --> 00:01:25.280
<v Speaker 1>protocol amendments that are not needed, They don't need that process.

16
00:01:26.879 --> 00:01:29.879
<v Speaker 1>Can we group some of those protocol amendments so you

17
00:01:30.040 --> 00:01:36.359
<v Speaker 1>only do it once? Can we learn as we're designing

18
00:01:36.400 --> 00:01:39.159
<v Speaker 1>studies from the protocol amendments we've done in the past,

19
00:01:40.200 --> 00:01:45.359
<v Speaker 1>so you don't make that mistake again. So the first

20
00:01:45.359 --> 00:01:49.239
<v Speaker 1>thing we did was we started to think about what's

21
00:01:49.280 --> 00:01:54.159
<v Speaker 1>the problem why should we change? When we looked at

22
00:01:54.200 --> 00:01:58.159
<v Speaker 1>actually what is the cost and the burden of doing

23
00:01:58.239 --> 00:02:05.120
<v Speaker 1>a protocol amendment, we really found some surprising data. Yes,

24
00:02:05.200 --> 00:02:07.920
<v Speaker 1>there's the direct costs. You may need to pay fees

25
00:02:08.560 --> 00:02:12.879
<v Speaker 1>to submit a protocol amendment. You know that you need

26
00:02:12.919 --> 00:02:18.719
<v Speaker 1>to put it through some approval process, and that takes time.

27
00:02:20.919 --> 00:02:25.120
<v Speaker 1>But when we added up the amount of resource it

28
00:02:25.159 --> 00:02:29.719
<v Speaker 1>took for the people in house, but importantly the burden

29
00:02:29.800 --> 00:02:33.000
<v Speaker 1>it pays it placed on the sites that we're trying

30
00:02:33.000 --> 00:02:39.520
<v Speaker 1>to implement our protocols, we found that the amount of

31
00:02:39.759 --> 00:02:46.400
<v Speaker 1>both direct and indirect costs through the time, delays and

32
00:02:46.439 --> 00:02:50.439
<v Speaker 1>the resources we used, we're in the region of up

33
00:02:50.439 --> 00:03:00.680
<v Speaker 1>to half a million dollars per amendment. We showed that.

34
00:03:00.719 --> 00:03:02.879
<v Speaker 1>One of the first things we did in our change

35
00:03:02.919 --> 00:03:08.319
<v Speaker 1>management process was we just formed a simple infographic that

36
00:03:08.479 --> 00:03:13.479
<v Speaker 1>explained how many protocol amendments we were doing per study,

37
00:03:14.240 --> 00:03:18.840
<v Speaker 1>what the cost of those amendments were, and how possibly

38
00:03:18.919 --> 00:03:24.960
<v Speaker 1>we could redirect that cost, that time, and that effort

39
00:03:25.719 --> 00:03:31.520
<v Speaker 1>into things that were going to add more value to patients.

40
00:03:33.319 --> 00:03:36.680
<v Speaker 1>So then the question became, so what is driving these

41
00:03:36.719 --> 00:03:42.879
<v Speaker 1>protocol amendments? And I would often hear the conversation. Well,

42
00:03:42.879 --> 00:03:46.039
<v Speaker 1>I would never do a protocol amendment that wasn't absolutely needed.

43
00:03:47.599 --> 00:03:50.159
<v Speaker 1>My regulatory colleague might make me do it because this

44
00:03:50.280 --> 00:03:56.879
<v Speaker 1>is a very regimented industry. Well, I would never do

45
00:03:57.000 --> 00:03:59.719
<v Speaker 1>a protocol amendment that wasn't needed. It's a lot of

46
00:03:59.759 --> 00:04:03.680
<v Speaker 1>works and a lot of process. But my safety colleague

47
00:04:04.520 --> 00:04:09.439
<v Speaker 1>might think that it's necessary. I would never do a

48
00:04:09.439 --> 00:04:13.120
<v Speaker 1>protocol amendment that's not needed. It's only for the safety

49
00:04:13.120 --> 00:04:16.279
<v Speaker 1>of the patient. But for some of these smaller issues,

50
00:04:17.439 --> 00:04:19.600
<v Speaker 1>the quality department will say we need to do a

51
00:04:19.600 --> 00:04:24.879
<v Speaker 1>protocol a member. So it really was the time to

52
00:04:25.000 --> 00:04:29.079
<v Speaker 1>address this as an issue in how we work as

53
00:04:29.120 --> 00:04:36.480
<v Speaker 1>an organization. So what did we do. The first thing

54
00:04:36.560 --> 00:04:40.639
<v Speaker 1>we did was we got together a team of real

55
00:04:40.759 --> 00:04:46.399
<v Speaker 1>experts cross functionally so that we could figure out what

56
00:04:46.639 --> 00:04:52.160
<v Speaker 1>the key questions you need to ask of the change

57
00:04:52.240 --> 00:04:55.839
<v Speaker 1>that you want to make in your protocol that will

58
00:04:55.879 --> 00:04:59.680
<v Speaker 1>help you decide whether an amendment is needed or not.

59
00:05:01.920 --> 00:05:08.040
<v Speaker 1>It comes out into a pretty simple decision framework. We

60
00:05:08.120 --> 00:05:11.560
<v Speaker 1>could implement that in a system that is able to

61
00:05:11.639 --> 00:05:15.920
<v Speaker 1>pick up the metadata about that study easily, so the

62
00:05:15.959 --> 00:05:18.439
<v Speaker 1>teams don't have a lot of burden in the process

63
00:05:18.480 --> 00:05:24.959
<v Speaker 1>that they're following, it guides them through these questions. If

64
00:05:25.000 --> 00:05:30.560
<v Speaker 1>they've got a particularly complex case, then it directs them

65
00:05:30.680 --> 00:05:35.000
<v Speaker 1>to this team of experts who can all get round

66
00:05:35.000 --> 00:05:38.920
<v Speaker 1>a table. So you've got your regulatory input, your safety input,

67
00:05:39.279 --> 00:05:45.519
<v Speaker 1>your quality input, legal if necessary, all round the table

68
00:05:45.600 --> 00:05:51.839
<v Speaker 1>and can give that team advice In the moment. The

69
00:05:52.160 --> 00:05:54.920
<v Speaker 1>decision whether they make a protocol amendment or not is

70
00:05:54.920 --> 00:05:58.839
<v Speaker 1>still down to that study team, and so they can

71
00:05:58.879 --> 00:06:06.879
<v Speaker 1>pull this through. By standardizing this amendment approach, it helped

72
00:06:06.959 --> 00:06:12.360
<v Speaker 1>us to reduce unnecessary amendments and collect key data and

73
00:06:12.480 --> 00:06:16.199
<v Speaker 1>insights so that we can learn from that in the future.

74
00:06:18.480 --> 00:06:21.439
<v Speaker 1>So you may ask, yeah, that's all very nice and

75
00:06:21.920 --> 00:06:26.920
<v Speaker 1>pretty logical, but how do you change the mindset of

76
00:06:27.040 --> 00:06:33.759
<v Speaker 1>people to actually make a difference. Well, it's important that

77
00:06:33.879 --> 00:06:40.839
<v Speaker 1>this works for them. The features of this tool is

78
00:06:40.879 --> 00:06:47.079
<v Speaker 1>that is integrated into their daily work pattern. They would

79
00:06:47.360 --> 00:06:50.600
<v Speaker 1>normally have written up a brief of what their amendment

80
00:06:50.600 --> 00:06:54.160
<v Speaker 1>would be and ask for some medical writing support and

81
00:06:54.720 --> 00:07:00.680
<v Speaker 1>other support to put the work through the system. This

82
00:07:00.759 --> 00:07:05.800
<v Speaker 1>feeds into that and the automation enables the process to

83
00:07:05.839 --> 00:07:11.240
<v Speaker 1>follow three freely. It creates a dashboard so that we

84
00:07:11.360 --> 00:07:15.199
<v Speaker 1>can see the overview of all of the submissions the

85
00:07:15.360 --> 00:07:21.000
<v Speaker 1>status and we can get insights from them, and we

86
00:07:21.079 --> 00:07:24.800
<v Speaker 1>have this deddicated support and continuous improvement through our in

87
00:07:24.839 --> 00:07:34.399
<v Speaker 1>house team. So did it work. We've established a huge

88
00:07:34.439 --> 00:07:40.279
<v Speaker 1>amount of uptake across the organization, in particularly in our

89
00:07:40.360 --> 00:07:44.839
<v Speaker 1>late stage trials where this is focused. And if you

90
00:07:44.879 --> 00:07:47.279
<v Speaker 1>look at the graphic in the middle, when we started

91
00:07:47.319 --> 00:07:51.480
<v Speaker 1>this process in twenty twenty three, the blue blocks show

92
00:07:51.519 --> 00:07:54.040
<v Speaker 1>you that there was a really high proportion of our

93
00:07:54.079 --> 00:07:59.319
<v Speaker 1>studies would submit an amendment in any one year and

94
00:07:59.720 --> 00:08:05.439
<v Speaker 1>all of them were processed. When we started this initiative

95
00:08:05.480 --> 00:08:10.399
<v Speaker 1>in twenty twenty four, we still saw a high proportion

96
00:08:10.519 --> 00:08:15.399
<v Speaker 1>of our studies submitting an amendment, but we were able

97
00:08:15.439 --> 00:08:19.360
<v Speaker 1>to say more than twenty five percent of them were

98
00:08:19.399 --> 00:08:25.279
<v Speaker 1>not necessary and we avoided the teams going through that process.

99
00:08:26.680 --> 00:08:31.000
<v Speaker 1>As we've got into twenty twenty five, we've actually seen

100
00:08:31.199 --> 00:08:36.240
<v Speaker 1>a mindset shift. The teams are not coming to us

101
00:08:36.879 --> 00:08:40.440
<v Speaker 1>requesting to do a protocol amendment in the first place.

102
00:08:41.399 --> 00:08:46.759
<v Speaker 1>They've learned from this behavior, they've learned what the guidance is,

103
00:08:48.120 --> 00:08:53.159
<v Speaker 1>and we're not seen as many come through, and even

104
00:08:53.360 --> 00:08:55.799
<v Speaker 1>of those that do come through, we're now seeing that

105
00:08:55.879 --> 00:09:02.679
<v Speaker 1>we're avoiding about a third of those protocol amendments. So

106
00:09:02.720 --> 00:09:06.480
<v Speaker 1>what does that mean, Well, if you want to put

107
00:09:06.519 --> 00:09:12.960
<v Speaker 1>it in the sort of crudest terms, I suppose in

108
00:09:13.080 --> 00:09:17.919
<v Speaker 1>terms of money, in terms of the avoided direct costs,

109
00:09:18.960 --> 00:09:23.480
<v Speaker 1>indirect costs, and the time we've saved in getting our

110
00:09:23.639 --> 00:09:28.879
<v Speaker 1>trials to patients more quickly. We have saved money in

111
00:09:28.919 --> 00:09:32.799
<v Speaker 1>the region of tens of millions of dollars that can

112
00:09:32.840 --> 00:09:37.639
<v Speaker 1>get reinvested into places where it will really make a difference.

113
00:09:39.519 --> 00:09:45.639
<v Speaker 1>And for our teams, they've found this process easy to use, intuitive,

114
00:09:46.519 --> 00:09:51.480
<v Speaker 1>it goes with their normal workflow, and they can see

115
00:09:51.519 --> 00:09:59.720
<v Speaker 1>the impact that it's having directly. So what's next. We've

116
00:09:59.720 --> 00:10:03.919
<v Speaker 1>got lot of rich data. Now we are using AI

117
00:10:04.559 --> 00:10:09.440
<v Speaker 1>to take all of these amendments and categorize them into

118
00:10:09.480 --> 00:10:15.120
<v Speaker 1>a standard categorization framework. We can then take all of

119
00:10:15.159 --> 00:10:20.759
<v Speaker 1>those amendments from early and late stage development and learn

120
00:10:21.440 --> 00:10:26.279
<v Speaker 1>from these amendments. We intend to pull that right back

121
00:10:26.320 --> 00:10:33.399
<v Speaker 1>into the study design phase and so that really building

122
00:10:33.480 --> 00:10:40.519
<v Speaker 1>literally quality by design. We can avoid these amendments in

123
00:10:40.600 --> 00:10:48.159
<v Speaker 1>the first place by giving visualizations and data to the

124
00:10:48.240 --> 00:10:54.200
<v Speaker 1>teams as they put pen to paper. So I thank

125
00:10:54.240 --> 00:10:57.919
<v Speaker 1>you for your attention, and I don't know if there

126
00:10:57.919 --> 00:11:01.080
<v Speaker 1>are any questions. We've got it maybe a couple of minutes, Mary,

127
00:11:01.159 --> 00:11:04.200
<v Speaker 1>But other than that, thank you.

128
00:11:07.080 --> 00:11:11.600
<v Speaker 2>Take one question, how are you handling amendments? How are

129
00:11:11.600 --> 00:11:16.960
<v Speaker 2>you avoiding amendments under the EU CTR and the CTIS process.

130
00:11:18.080 --> 00:11:24.960
<v Speaker 1>So it's interesting some of the amendments that we do.

131
00:11:25.039 --> 00:11:27.480
<v Speaker 1>We know that you can't avoid all amendments. There are

132
00:11:27.480 --> 00:11:30.840
<v Speaker 1>some amendments that are required. What we're trying to do

133
00:11:30.960 --> 00:11:34.279
<v Speaker 1>is be more thoughtful about it, and is there a

134
00:11:34.320 --> 00:11:37.399
<v Speaker 1>way that we can group those the way that the

135
00:11:37.519 --> 00:11:41.080
<v Speaker 1>time procesing it's taking for those amendments to go through.

136
00:11:41.559 --> 00:11:43.960
<v Speaker 1>You don't want to be doing an amendment more than

137
00:11:44.240 --> 00:11:47.600
<v Speaker 1>once a year anyway. So what we've tried to do

138
00:11:47.799 --> 00:11:50.840
<v Speaker 1>is figure out which the ones that are absolutely required

139
00:11:51.919 --> 00:11:55.480
<v Speaker 1>versus those that can be handled in a different way,

140
00:11:56.120 --> 00:12:01.440
<v Speaker 1>for for instance, via clarification letter. But that wouldn't apply.

141
00:12:01.279 --> 00:12:06.399
<v Speaker 2>To that Based on this the last part, when you

142
00:12:06.519 --> 00:12:09.799
<v Speaker 2>were researching what kind of amendments were happening, do you

143
00:12:09.919 --> 00:12:13.679
<v Speaker 2>have any examples that you saw the specific areas that

144
00:12:14.240 --> 00:12:18.000
<v Speaker 2>came up a lot in your analysis.

145
00:12:20.200 --> 00:12:23.200
<v Speaker 1>Yeah, I think there's a couple of areas. One area

146
00:12:23.279 --> 00:12:29.960
<v Speaker 1>was inconsistencies across a protocol document. So where we have

147
00:12:30.559 --> 00:12:34.639
<v Speaker 1>the same topic discussed in various parts of the protocol document,

148
00:12:34.879 --> 00:12:37.840
<v Speaker 1>if you've got inconsistencies, then we were needing to correct them.

149
00:12:39.080 --> 00:12:41.639
<v Speaker 1>We've approached that in a couple of different ways. We've

150
00:12:42.080 --> 00:12:47.639
<v Speaker 1>simplified our protocol template using the transcelerate as a model,

151
00:12:48.840 --> 00:12:52.440
<v Speaker 1>and created a lot more standard content that we're autogenerating

152
00:12:53.000 --> 00:12:57.720
<v Speaker 1>so that we're removing those inconsistencies as we go. We're

153
00:12:57.759 --> 00:13:02.639
<v Speaker 1>also using an AI check to provide consistency checks across

154
00:13:02.639 --> 00:13:03.559
<v Speaker 1>the protocol as well.

155
00:13:05.279 --> 00:13:07.240
<v Speaker 2>M mm hmm.