Whole Genome Approaches to Complex Kidney Disease
February 11-12, 2012 Conference Videos

Sunday Plenary Discussion and Working Lunch
Reports From Cohorts, ELSI, and Nephrology Practice Groups

Video Transcript

1
00:00:00,000 --> 00:00:09,766
JEFFREY KOPP: I think we’ll begin our reports now. We have three sessions to hear from. I imagine the three presenting individuals or teams

2
00:00:09,766 --> 00:00:17,966
will want to spend something like five minutes, maybe, going over points and I guess I’ll let them decide if they want to have discussion point-

3
00:00:17,966 --> 00:00:28,432
by-point or wait until the end of their presentation and open it up. And so I hope those who couldn’t attend that particular session may have additional

4
00:00:28,433 --> 00:00:43,599
comments and maybe those who are in the session may want to elaborate. So, I think we’ll start with cohorts first: Martin and Linda.

5
00:00:43,600 --> 00:00:56,466
MARTIN POLLACK: So, we had the task of discussing cohorts, what we should be collecting in terms of samples, patients, information, and we

6
00:00:56,466 --> 00:01:13,932
solved all of these problems in 45 minutes. We talked about a bunch of different things in terms of collection size, in terms of what we should be

7
00:01:13,933 --> 00:01:22,666
collecting in terms of biomaterials, in terms of simple straightforward things like 24-hour urine collections. It was pointed out that this is

8
00:01:22,666 --> 00:01:35,866
important not just for the usual measurement of estimated GFR but there’s a lot of information regarding nutritional status, nutritional history of

9
00:01:35,866 --> 00:01:47,999
participants that can be obtained which might be useful for gene-environment kinds of studies. There was a fair amount of discussion about sort

10
00:01:48,000 --> 00:01:56,933
of technical difficulties that perhaps are somewhat inherent in kidney studies having to do with kidney tissue and correlating biological

11
00:01:56,933 --> 00:02:10,733
changes in kidney tissues and kidney cells with genetic data. It came up that NIDDK has a protocol on standardized urine collection for research

12
00:02:10,733 --> 00:02:23,066
purposes which is available on the Web. There was some discussion about measuring kidney function, eGFR, urine albumin-to-creatine ratios,

13
00:02:23,066 --> 00:02:32,599
and perhaps other biomarkers that might represent certain underlying phenotypes that are hard to assess better, the longitudinal phenotypes

14
00:02:32,600 --> 00:02:47,000
that normally take a long time to define such as GFR trajectory. There was some discussion about proteomics: whether and how to include

15
00:02:47,000 --> 00:02:51,400
urine proteomics and metabolomics in large-scale studies.

16
00:02:51,400 --> 00:03:05,300
LINDA KAO: And linking existing studies to other databases for long-term follow up and other complication outcomes. And then in terms of more

17
00:03:05,300 --> 00:03:16,566
novel phenotypes, there was discussion about non-invasive imaging. We heard from Wendy Hoy about assessing glomerular number; that there is

18
00:03:16,566 --> 00:03:27,299
great heterogeneity in that. And so, again I think there needs to be more exploration and more uniform data collection; I think that it’s one

19
00:03:27,300 --> 00:03:35,766
common topic that was discussed in great deal about uniform data collection. Even if people weren’t in large consortium studies, even if

20
00:03:35,766 --> 00:03:48,499
people were just doing small studies, it would be good for everybody to use similar protocols so that results can be more comparable.

21
00:03:48,500 --> 00:03:58,833
MARTIN POLLACK: Also there was some discussion of tissue resources for follow up on genetic studies, GWAS-based studies, and

22
00:03:58,833 --> 00:04:12,633
discussion of the NDRI resource for kidney tissue and the quality of RNA and the problems with that, the GTEX resource for tissue from cadavers

23
00:04:12,633 --> 00:04:22,566
and associated genotyping. Information was discussed but it was mentioned by some that there are quality issues in terms of the tissue of

24
00:04:22,566 --> 00:04:43,966
kidney, the issue of kidney tissue, the problems inherent in studying kidneys and getting kidney tissue came up. We didn’t solve that problem. The

25
00:04:43,966 --> 00:04:57,632
Neptune resource was mentioned as an ongoing prospective collection of well-phenotyped tissue for a variety of possible ancillary studies was

26
00:04:57,633 --> 00:05:12,166
mentioned, and questions were raised about the availability of archived samples for the general community, what’s out there, and can information

27
00:05:12,166 --> 00:05:23,699
about the various databanks that are available to the community be collated somehow for use in the community? We also talked about issues

28
00:05:23,700 --> 00:05:36,366
having to do with diversity in collecting information from other ethnic groups. Using genetics, it’s in some ways easier to classify

29
00:05:36,366 --> 00:05:47,032
people than self-identification in terms of at least geographic history, but the problem is getting participants of diverse ethnicities to participate in

30
00:05:47,033 --> 00:05:55,599
our studies as being the biggest challenge in terms of ethnic diversity.

31
00:05:55,600 --> 00:06:02,466
LINDA KAO: And even though there are some studies that already exist, I think many of them could be underutilized. I know personally from

32
00:06:02,466 --> 00:06:14,532
me, I learned about a couple of other fairly large cohorts of African American populations that are out there which I didn’t know about. And then,

33
00:06:14,533 --> 00:06:26,333
even if people do study genetics in other non-European ethnic groups, there’s a great deal of difficulty in confirming their findings and getting

34
00:06:26,333 --> 00:06:33,933
that additional replication sample. MARTIN POLLACK: And then finally, Wendy Hoy

35
00:06:33,933 --> 00:06:38,733
talked about…what? LINDA KAO: I have no bullet points about that.

36
00:06:38,733 --> 00:06:48,666
MARTIN POLLACK: Yeah. She talked about some of the resources in Australia that she’s been developing and interestingly, at least I thought it

37
00:06:48,666 --> 00:06:56,932
was very interesting, that it seemed that the genetic risk, at least for kidney disease in Aborigines in Australia seemed to be different

38
00:06:56,933 --> 00:07:07,399
from the genetic risk factors that have been observed in other ethnically-defined populations, suggesting there’s a lot more to learn from

39
00:07:07,400 --> 00:07:22,000
studies of kidney disease and its relationship to ethnicity.

40
00:07:22,000 --> 00:07:31,000
JEFFREY KOPP: You’re not off the hook quite yet. So, one thought occurred to me, and I was in the session but didn’t have the chance to talk about it,

41
00:07:31,000 --> 00:07:37,533
would be environmental factors to collect. So things like smoking, maybe non-steroidals if that’s important, illicit drug use, maybe toxins, heavy

42
00:07:37,533 --> 00:07:46,799
metals, maybe other environmental toxins. This gets into the idea that I know you guys touched at least briefly idea of proteomics and metabolomics

43
00:07:46,800 --> 00:07:57,066
and knowing how to store samples for these various assays and having the resources to bank them in plasma and serum and urine is not trifling,

44
00:07:57,066 --> 00:08:02,266
but something to think about. LINDA KAO: Yeah, you might have been out of

45
00:08:02,266 --> 00:08:12,766
the room. The Genome Institute had started a project called PhenX—P-H-E-N-X—and the goal there was to provide standardized protocols for

46
00:08:12,766 --> 00:08:24,232
the collection of environmental data in the context of gene-by-environment studies. So there were, I think, 20 different domains and some of the data

47
00:08:24,233 --> 00:08:36,299
are questionnaire-based data, others…I know I participated in a diabetes group. There we provided common protocols for collection of blood

48
00:08:36,300 --> 00:08:48,833
specimen, which A1C assay do you use? So, that might be a good place to start. It’s sort of the most…most of the factors considered in that

49
00:08:48,833 --> 00:09:00,633
program are sort of the simplest, maybe not the most high-tech factors that most studies can accomplish, but that’s a great point, I think.

50
00:09:00,633 --> 00:09:06,099
JEFFREY KOPP: Thank you. MALE: Thank you. A very nice discussion about

51
00:09:06,100 --> 00:09:12,700
what went on. Do you think it would be worthwhile for the survivors—everybody who’s really staying for the most interesting part of the

52
00:09:12,700 --> 00:09:23,833
conference—to talk about the longitudinal samples that have data that are important for kidney disease, like Framingham, CRIC, MESA,

53
00:09:23,833 --> 00:09:36,433
ARIC? It would be very useful to talk about them so people knew exactly where one could go to. I may have missed some in my little microphone

54
00:09:36,433 --> 00:09:39,099
delineation. MARTIN POLLACK: I don’t know if anyone

55
00:09:39,100 --> 00:09:52,233
wanted to mention particular studies, but I agree that longitudinal follow-up, what happens to the kidney over time is really the phenotype we care

56
00:09:52,233 --> 00:10:03,066
about most and, you know, it’s difficult. I mean, it’s not just important for genetic studies but I think it’s important for all sorts of clinical studies. It’s

57
00:10:03,066 --> 00:10:13,199
important in terms of development of therapeutics and in terms of relationships and studies with the pharmaceutical industry because what we really

58
00:10:13,200 --> 00:10:23,500
care about is not if a GFR is a little bit low, but if the GFR is low and getting worse over time and we want to be able to assess longitudinal

59
00:10:23,500 --> 00:10:35,766
outcomes without having…it would be nice to have a single point in time surrogate for what happens over 10 years so that we can really

60
00:10:35,766 --> 00:10:40,499
study the endpoints that we care most about in a more compact time frame.

61
00:10:40,500 --> 00:10:47,700
MALE: One of the nice things are because a product of this conference is the breakout group recommendations get put on our website and get

62
00:10:47,700 --> 00:10:56,900
widely distributed, hopefully, is to encourage studies that are supported by other funding agencies to get some simple, cheap

63
00:10:56,900 --> 00:11:04,366
measurements that would be very useful for the public health regarding kidney disease, such as the urine protein-to-creatinine ratio, the urine

64
00:11:04,366 --> 00:11:14,999
albumin-to-creatinine ratio, creatinine, and we should encourage that from our brother and sister institutions and other investigators.

65
00:11:15,000 --> 00:11:22,133
MALE: You mentioned in the first part of your presentation 24-hour urine collections and I’m just wondering if you guys discussed best practices

66
00:11:22,133 --> 00:11:30,933
for that. I’ve had a couple of cord studies and the thing that I hear back most from my coordinators what limits participation and returning for future

67
00:11:34,333 --> 00:11:38,599
visits is a lot of participants do not like doing 24-hour urine collections. You can do almost anything to them, but if you ask them to do a

68
00:11:38,600 --> 00:11:46,333
24-hour urine collection then it’s suddenly…I mean, that really crosses a line. So, I’m just wondering if people have suggestions on how

69
00:11:46,333 --> 00:11:50,399
to… LINDA KAO: We heard about the Netherlands.

70
00:11:50,400 --> 00:12:01,400
GERJAN NAVIS: Actually, I’m the one who came up with this suggestion. I’m very well-aware of the skepticism about 24-hour urine. If you are part

71
00:12:01,400 --> 00:12:09,966
of a, let’s say, clinical trial or let’s say a large-scale screening, yes, it’s difficult and cumbersome. But if you have renal patients that

72
00:12:09,966 --> 00:12:18,566
come back to your outpatient clinic again and again and again and you tell them that this is an investment in the future because it gives you

73
00:12:18,566 --> 00:12:27,766
non-biased information on the lifestyle that’s impossible to get otherwise because food labeling is very poor, and when you give them

74
00:12:27,766 --> 00:12:38,766
feedback so that it helps them to comply with your dietary restrictions, in our hands it works. So, you have to integrate it in your clinical

75
00:12:38,766 --> 00:12:47,099
practice and you use it for feedback to your patients to show them that it helps improve their lifestyle. So then it’s a tool for let’s say supporting

76
00:12:47,100 --> 00:13:01,100
something that’s very difficult, changing your lifestyle. So, the 24-hour urine is your friend. So, that’s what I tell my patients and then it works.

77
00:13:01,100 --> 00:13:07,833
ROBERT KLETA: What we have found very useful is if they get a small fee, actually, if they complete a 24-hour urine collection to significantly

78
00:13:07,833 --> 00:13:10,966
increase the compliance. MARTIN POLLACK: It occurs to me it might be

79
00:13:10,966 --> 00:13:22,899
even useful to develop sort of a Wiki for best practices for various phenotyping. One issue that occurred to me in the course of a study I’m

80
00:13:22,900 --> 00:13:33,700
involved with in terms of kidney stone-related phenotypes is time of 24-hour collection within season of the year and its potential effect on

81
00:13:33,700 --> 00:13:43,833
sunlight, length of day, and vitamin D metabolism; something that hadn’t occurred to me at all until deep into the study. So if there was a mechanism

82
00:13:43,833 --> 00:13:54,166
for ongoing discussion and refinement of best practices that was some kind of, you know, Wiki-like web-based forum, it might be useful to the

83
00:13:54,166 --> 00:13:58,366
community. MALE: I just wanted to follow up on what Paul

84
00:13:58,366 --> 00:14:07,166
said about longitudinal. I think all those studies are great to think about but we also need to think about using the new resources with electronic

85
00:14:07,166 --> 00:14:15,599
records and being able to look across systems now, you know, hospital-based systems to follow patients. I mean, if we can re-identify samples,

86
00:14:15,600 --> 00:14:24,666
those are rich resources for longitudinal follow-up, both retrospectively and potentially prospectively that I think we’re just beginning to

87
00:14:24,666 --> 00:14:34,899
understand. I really think that’s where the future’s going because it’s so expensive to do longitudinal studies in the way we’re used to it, but now

88
00:14:34,900 --> 00:14:39,666
hospital systems are collecting this information and, granted, there’s problems with people talking to each other but even in Cleveland where all of

89
00:14:39,666 --> 00:14:47,066
the hospitals hate each other, I’m able to look in the Cleveland Clinic and they can look back for clinical reasons now, and we can look into Kaiser

90
00:14:47,066 --> 00:14:55,866
and back, and we can get 70% of the region’s information. So, that’s a possibility to get that information you want, too.

91
00:14:55,866 --> 00:15:07,766
MARTIN POLLACK: I don’t think we’re going to be able to have one method of collecting all this data on all cohorts. I mean, I’m a big fan that individual

92
00:15:07,766 --> 00:15:16,966
investigators and groups should design the studies that they want to study and I think that’s been extremely fruitful in genetic studies of the

93
00:15:16,966 --> 00:15:30,266
kidney to date, but to me, one of most important things would be to unify what the various tests and assays mean, even whether labs at two

94
00:15:30,266 --> 00:15:40,966
different institutions mean the same thing by a parathyroid hormone level or a creatinine level is not even that clear. So to me, that’s unifying

95
00:15:40,966 --> 00:15:59,366
descriptors of phenotypes is among the biggest issues that would help.

96
00:16:09,600 --> 00:16:25,166
SARA HULL: So, I was in the breakout group that focused in on the ethical, legal, and social implications and the optimal way to conduct

97
00:16:25,166 --> 00:16:33,599
whole genome research. It was a very small—we’ll call intimate---group and we joked that was because our speakers were just so

98
00:16:33,600 --> 00:16:41,733
tremendously clear, that there was nothing left to say. We actually solved all of our problems before we even got into the room, and of course

99
00:16:41,733 --> 00:16:52,699
that’s never the case when it comes to ethics. So, each of the speakers that we heard from yesterday and today in this theme presented

100
00:16:52,700 --> 00:17:05,033
what I think are a very nice list of points to consider and action items to keep in mind for researchers. In our session we delved a little

101
00:17:05,033 --> 00:17:14,499
more deeply into the idea of these action items and points and sort of generate specific researcher obligations that carry certain

102
00:17:14,500 --> 00:17:26,000
challenges and burdens and opportunities, and so we tried to explore some of those in a little more detail. I threw this together and hopefully it

103
00:17:26,000 --> 00:17:37,933
will make some coherent sense. Around the issue of data sharing, which we heard about in detail yesterday but was really a cross-cutting theme

104
00:17:37,933 --> 00:17:48,199
that came up again and again, of course there’s recognition of the need for collaboration and sharing and this starts with the physical act of

105
00:17:48,200 --> 00:17:58,933
depositing data in repositories like dbGaP and we actually talked through some of the requirements and nuances associated with doing so. There

106
00:17:58,933 --> 00:18:06,899
was some discussion yesterday and again in the session about special population considerations, so developing a plan that’s acceptable for

107
00:18:06,900 --> 00:18:15,600
different communities. It’s going to look very different what the requirements under a Syrian regime are going to be as compared to a

108
00:18:15,600 --> 00:18:29,100
historically black college where a cohort’s being generated or recruited from, and so there’s those considerations to pay attention to. We also tuned

109
00:18:29,100 --> 00:18:39,633
in to the fact that consent and the adequacy of consent for doing data deposition is a big part of this and one of the themes that hadn’t come out

110
00:18:39,633 --> 00:18:49,799
explicitly during the presentations but that we paid attention to is the concept of re-consent and for existing cohorts who have already given their

111
00:18:49,800 --> 00:19:00,833
consent to participate in some kind of research, whether it’s research related to kidney disease or genetic research in some form of a description of

112
00:19:00,833 --> 00:19:09,633
genetic. Is that going to be sufficient to allow the sharing of whole exome and whole genome data and how do we make decisions about when re-

113
00:19:09,633 --> 00:19:19,933
consent is going to be required and also engaging with IRBs on these questions, too, and their transition? Investigators can actually play a role in

114
00:19:19,933 --> 00:19:32,433
helping to educate IRBs about those realities as much as the other way around. And then an opportunity, I think, to consider—for the research

115
00:19:32,433 --> 00:19:42,566
community to consider—is really tuning-in to and paying attention to evolving policies. We heard that very clear policies exist in the context of

116
00:19:42,566 --> 00:19:53,566
GWAS SNP data and this is a moving target and evolving issue for broader genomic data sharing types, at least within the NIH and NIH-funded

117
00:19:53,566 --> 00:20:00,332
research, what the mandates will go along with that. There’s going to be Federal Register notices, there’s going to be an opportunity for public

118
00:20:00,333 --> 00:20:09,266
discourse, and it would be extremely useful for the research community to share its experiences and perspectives on some of the success

119
00:20:09,266 --> 00:20:18,366
stories, as well as challenges, around GWAS and some of the emerging—and I mean “emerge” not as the consortium, but in the original sense of

120
00:20:18,366 --> 00:20:30,266
the word—some of the experience around sharing exomic data as is happening in at least some institutes and with some cohorts. Okay, so

121
00:20:30,266 --> 00:20:39,799
management of results. There’s this also emerging sense that there may be some obligation to really pay attention to and figure out

122
00:20:39,800 --> 00:20:49,833
what to do with results that are inevitably going to come up that aren’t necessarily directly related to research questions around kidney disease that

123
00:20:49,833 --> 00:20:58,499
investigators might be asking but are definitely related to the technology being used. We delved into a little bit about the sort of personal decision

124
00:20:58,500 --> 00:21:06,766
that any given investigator is going to have to make about whether to do this as it comes up if they happen to see if or whether there’s a more

125
00:21:06,766 --> 00:21:17,266
active obligation to engage in the kind of interrogation that Ben Solomon and others have talked about. I thought a really nice point was

126
00:21:17,266 --> 00:21:29,066
made about the need for additional support and resources and we talked about all of the entities that are trying to develop lists of actionable

127
00:21:29,066 --> 00:21:37,032
variants and the fact that it would be burdensome for any one investigator to have to come up with this list independently, based on the particular

128
00:21:37,033 --> 00:21:47,433
population and disease and research questions that they’re focused on and that a more central effort to do this would be really valuable and help

129
00:21:47,433 --> 00:21:59,333
relieve some of the burden on investigators, and that the ability to both identify and to be able to disclose these results require both funding and

130
00:21:59,333 --> 00:22:06,933
additional infrastructure, which are institutional considerations and funding agency considerations. But then there was sort of a

131
00:22:06,933 --> 00:22:17,733
cautionary—I wouldn’t call it push-back—but a balancing to drive home the point that lists might be helpful but they’re not sufficient; that having a

132
00:22:17,733 --> 00:22:25,699
list of actionable variants is a little bit of an oversimplification. It’s not just a checklist that you can go back to any given person and say “if you

133
00:22:25,700 --> 00:22:34,166
have this, this, and this.” The “this” is much deeper. It involves making decisions about which variants within a region and what it means based

134
00:22:34,166 --> 00:22:43,166
on family history; what it means in terms of the personal, the family history as well as the individual’s medical history; and how they’ll

135
00:22:43,166 --> 00:22:51,932
receive that information, and preserving the opportunity for choices when that’s appropriate. One additional issue that was flagged is coming

136
00:22:51,933 --> 00:23:01,833
up with ways to share. So when there are these lists and investigators do make choices about what is actionable and make a case-by-case

137
00:23:01,833 --> 00:23:10,233
decision about what to return or what not to return, communicating this back to the broader community would be another area for sharing

138
00:23:10,233 --> 00:23:20,333
and for more openness, and right now the way the human…HGM…the medical…the mutation database—I can’t remember the acronym—that

139
00:23:20,333 --> 00:23:28,866
conveys the current literature on the status of various mutations, to be able to annotate that database in some way and let people know what

140
00:23:28,866 --> 00:23:39,432
parts of the literature have been on a case-by-case basis sort of overruled in individual clinical decision-making. That conversation is something

141
00:23:39,433 --> 00:23:50,533
that has to occur in a more robust way and presents another obligation but also an opportunity for investigators. Third was this area

142
00:23:50,533 --> 00:24:02,699
of paying attention to the role of race and ethnicity and engaging with communities. So, we talked this morning about an obligation is for

143
00:24:02,700 --> 00:24:10,866
investigators who recognize the need for diversity and the collection of data in the recruitment of participants, and there’s sort of

144
00:24:10,866 --> 00:24:20,032
this theme hanging in the air—the point that was made yesterday—that we have exome chips that are only suitable for use in European populations

145
00:24:20,033 --> 00:24:28,199
just by virtue of how the data were curated and pulled together. And so to help us move beyond that kind of technology, I think there’s a

146
00:24:28,200 --> 00:24:38,233
consensus that we have to pay attention to this, but how exactly we do this and what this means in terms of figuring out how to engage with

147
00:24:38,233 --> 00:24:51,399
communities was something we talked about in our discussion—in our breakout session—and really the need to look to others, partnering with

148
00:24:51,400 --> 00:24:59,233
people who have been successful at doing this is a step that requires investigators to kind of step out of maybe the view of the world and the

149
00:24:59,233 --> 00:25:07,333
particular slice of any given population that they’re used to dealing with, and this might even involve stepping outside of the world of genomics

150
00:25:07,333 --> 00:25:20,599
and looking at other kinds of research and other ways that we engage with diverse communities. But this is a huge area that I think some more

151
00:25:20,600 --> 00:25:32,900
robust conversation and attention to will benefit. And then finally there was this theme of informed consent and we talked a little bit about the idea of

152
00:25:39,466 --> 00:25:43,099
moving beyond the, “What do I do? Just tell me what to do and I’ll do it. Give me the consent form template and I’ll cut-and-paste it, “ to the need for

153
00:25:43,100 --> 00:25:51,300
broader conversations, and I keep on saying this idea of broadening the conversation, but here with respect to informed consent we identified a

154
00:25:51,300 --> 00:26:01,533
few key stakeholders. Communication with subjects: when you have a potential subject sitting before you thinking about the decision

155
00:26:01,533 --> 00:26:09,766
about whether or not to enroll in research and making all of the issues that Julie Sapp talked about transparent, that’s one challenge,

156
00:26:09,766 --> 00:26:19,099
remembering that some of these participants are also viewing the world from a consumer perspective. Some of them are coming to the

157
00:26:19,100 --> 00:26:25,933
table with a great command of information that’s available on the internet, they might have even through 23andMe and have some of their

158
00:26:25,933 --> 00:26:32,799
sequence data in-hand when they show up and how that changes the conversation. Conversation with communities: define both

159
00:26:32,800 --> 00:26:42,533
those diverse populations as well as advocacy groups and thinking about new ways to engage with populations and to help bring everybody on

160
00:26:42,533 --> 00:26:49,899
board with what’s going on in the research arena. Conversations with the public in a broader way as well, and this dovetailed, really, with the

161
00:26:49,900 --> 00:26:58,833
idea of conversations with the media, helping the public develop better media literacy, and to interpret any given finding that makes variancy

162
00:26:58,833 --> 00:27:05,933
much more important than maybe they are in reality, and helping scientists figure out how to better engage with the media so that the media

163
00:27:05,933 --> 00:27:12,933
represent our findings with some more accuracy. This is all part of a bigger picture that will help with the informed consent process and general

164
00:27:12,933 --> 00:27:24,166
understanding and transparency and trust in the research enterprise. I hope that fairly represents and organizes the themes that we covered in that

165
00:27:24,166 --> 00:27:36,166
session. Do we want to open it up for people to follow up on these points? Add to things that I missed? Oh, I’m sorry. One more point. Jeffrey

166
00:27:36,166 --> 00:27:42,432
made the very important point that a lot of the value of this conference is the conversations that occur in the hallway and one important

167
00:27:42,433 --> 00:27:49,666
conversation I had in the hallway raised a point that didn’t come out explicitly anywhere but I just took the liberty to stick up—it’s really a cross-

168
00:27:49,666 --> 00:27:58,299
cutting theme—that some of the discussion about data sharing across geographic and political boundaries runs up against an issue where

169
00:27:58,300 --> 00:28:05,500
certain communities might be hesitant to share their own resources with the broader community, at least without a conversation about what’s in it

170
00:28:05,500 --> 00:28:15,333
for them and how to protect their own interests in this information and people have started to talk about the idea of sovereignty and national

171
00:28:15,333 --> 00:28:28,666
concerns, but that’s another theme to be tuned-in to and to pay attention to as we talk about the goals of data sharing.

172
00:28:28,666 --> 00:28:36,832
JEFFREY KOPP: Thank you for that nice summary. So we’ve heard about various resources. Papers have been mentioned and so

173
00:28:36,833 --> 00:28:43,766
forth. There are local resources at each of our institutions, but if somebody wants to continue this conversation, can I suggest that you or

174
00:28:43,766 --> 00:28:51,332
others might be willing to have phone conversations or point to people who would be willing to do that?

175
00:28:51,333 --> 00:28:55,366
SARA HULL: Absolutely. JEFFREY KOPP: Because you have, particularly

176
00:28:55,366 --> 00:29:01,799
at the NIH Clinical Center, I think there’s a depth of expertise in coping with, for example, whole exome informed consent issues. Some

177
00:29:01,800 --> 00:29:09,400
institutions are already there and others may be just starting, so I think that would be a wonderful resource to have available.

178
00:29:09,400 --> 00:29:18,466
SARA HULL: I think all of our contact information is in the packets that you received when you checked in, but absolutely, and I will more likely

179
00:29:18,466 --> 00:29:26,332
than not be able to help refer you to the expert who knows. But we do have a good amount of experience; we’ve been looking at the whole

180
00:29:26,333 --> 00:29:38,533
exome and genome studies. The first hint that this was coming for our IRB at NHGRI was in 2006, so the IRB has some experience and the benefit

181
00:29:38,533 --> 00:29:48,299
of people like Julie Sapp who have been—I’m pointing to where she was sitting before, she’s not here—who had been teaching us about the

182
00:29:48,300 --> 00:29:57,866
best ways to do this, so I think we do have the advantage of some early experience. But I think, as was pointed out, we’re not necessarily the

183
00:29:57,866 --> 00:30:06,799
most generalizable of places where research is occurring, and so I think I appreciate all the comments that I’ve heard here about the specific

184
00:30:06,800 --> 00:30:15,600
challenges, funding and otherwise, at each individual site and it’s context-specific. So, I just throw out that as a limitation to my knowledge

185
00:30:15,600 --> 00:30:22,066
base, but absolutely please follow up. We might have one more comment.

186
00:30:22,066 --> 00:30:35,166
FEMALE: In addition to standardized protocol of how to collect phenotype, is it possible for us to have some consent form that’s being evaluated

187
00:30:35,166 --> 00:30:45,499
by experts in bioethics and genetics, test writing one or two studies already so we know what language to use when we, too, want to work or

188
00:30:45,500 --> 00:30:49,666
when we address the issue of incidental findings and re-contacting.

189
00:30:49,666 --> 00:30:56,599
SARA HULL: Yeah, so I think that’s a really interesting idea, the idea of developing some kind of resource for sharing consent forms and I

190
00:30:56,600 --> 00:31:05,666
found myself very curious to compare. We heard about, I think, an Australian cohort study and maybe one from the Netherlands where there

191
00:31:05,666 --> 00:31:16,299
were differences in language and the experience and success with respect to getting consent. I often think that when people say “no” to a

192
00:31:16,300 --> 00:31:23,966
consent form, I still consider that a success story because maybe the consent form really laid out some of the nuances and let people make a good

193
00:31:23,966 --> 00:31:34,099
decision, but maybe it’s concerning as well if we’re scaring them. But I think sharing consent forms is a great idea and there is a couple

194
00:31:34,100 --> 00:31:42,533
of…maybe I’ll make sure to communicate these resources that were there. There is a website that NHGRI has developed that has some model

195
00:31:42,533 --> 00:31:49,866
consent form language around genomics and research for certain kinds of studies and that’s evolving, but maybe this group wants to make a

196
00:31:49,866 --> 00:31:59,566
decision, too, to share its own consent forms that they’ve been using and develop a sense of best practices. I would endorse that as long as people

197
00:31:59,566 --> 00:32:08,099
promise to resist the temptation simply to cut-and-paste and to actually think about how it pertains to their specific cohorts and how it plugs into the

198
00:32:08,100 --> 00:32:11,333
broader conversation. What? MALE: Maybe a PDF.

199
00:32:11,333 --> 00:32:20,033
SARA HULL: Right, we’ll share our PDFs. MALE: [inaudible comment]

200
00:32:20,033 --> 00:32:47,466
SARA HULL: Scanned PDFs you can’t cut-and-paste; it’s only the ones that you generate on the computer. Okay. That sounds like a great idea.

201
00:32:47,466 --> 00:33:04,366
JOHN SEDOR: Our group was charged to talk about where we are in the kidney community and incorporating genetic data into practice. We really

202
00:33:04,366 --> 00:33:14,499
focused on: is there anything clinically actionable now rather than what could potentially happen down the road? So, I think there was a

203
00:33:14,500 --> 00:33:22,900
consensus that we’re certainly not ready for community-based, I should say, screening yet, prediction or pharmacogenomics. There’s been a

204
00:33:22,900 --> 00:33:33,666
lot of nice studies across a number of diseases, but recently and including and applying the CDKGen data sets, it suggests that right now

205
00:33:33,666 --> 00:33:41,066
genetic information that we have isn’t going to be useful for those sorts of things and we really focused on the possibility of using genetic tests

206
00:33:41,066 --> 00:33:49,866
as a biomarker or when we brought up the risk stratification, almost similar to using the LDL and we focused on really probably the only potentially

207
00:33:49,866 --> 00:34:02,566
clinically actionable finding we have is the c22 APOL1/MYH9 variants, and wanted to make the point that studies of the clinical utility of these

208
00:34:02,566 --> 00:34:11,599
findings should be proceeding in parallel with people interested in the underlying biology of the findings and what it means in terms of disease

209
00:34:11,600 --> 00:34:23,366
mechanism. We really tried to focus on potential studies that the community could be thinking about now and we came up with three different

210
00:34:23,366 --> 00:34:34,166
areas that related to the APOL1 finding. One would be HIVAN. Would this would potentially be something we could use to initiate antiretroviral

211
00:34:34,166 --> 00:34:45,599
therapy earlier than would otherwise be dictated by CD4 or viral load findings? That may go out the window if the HIV clinical community decides that

212
00:34:45,600 --> 00:34:53,333
everybody’s going to need to be treated early now with the findings about preventing progression, but it’s still a possibility. We

213
00:34:53,333 --> 00:35:05,966
discussed transplantation. Barry had to leave, but of course his provocative paper that genotyping the donors had a predicted outcome in the

214
00:35:05,966 --> 00:35:14,199
recipients and necessarily needs to be confirmed and I think this is potentially a major issue in the African American donor community. If they’re

215
00:35:14,200 --> 00:35:23,300
really at increased risk, would it be worthwhile knowing their genotype? It’s certainly something that would amenable to a good study. Then we

216
00:35:23,300 --> 00:35:31,600
discussed in some detail potentially the use…and I should put a non-diabetic chronic kidney disease in the African American community. Can you use

217
00:35:31,600 --> 00:35:43,933
APOL1 allele information? Erwin proposed an interesting idea to feed back to clinicians through electronic records to see how that would affect

218
00:35:43,933 --> 00:35:54,499
implementation of standard goals that we have in terms of both process. Are you measuring albumins, urinary albumins? Are you measuring

219
00:35:54,500 --> 00:36:06,133
blood pressures, as well as looking at outcomes? Obviously, all these things have a lot of ELSI issues. I thought that, without having to state the

220
00:36:06,133 --> 00:36:15,599
obvious, I think there would be a lot of interest and people interested in those issues that would be important and this clearly is a gene that’s

221
00:36:15,600 --> 00:36:27,633
linked to the African American community and there’s all kinds of issues in terms of racism and the way the public can interpret these findings in

222
00:36:27,633 --> 00:36:36,999
negative ways, and I think those are ripe for study also. Jeffrey brought up the issue, as we were moving on to a different talk, that is

223
00:36:37,000 --> 00:36:44,833
potentially developing a curated database of kidney disease genetic information: where you can get the test done, what does it mean, is any

224
00:36:44,833 --> 00:36:56,133
of that clinically actionable? The potissin mutation came up as potentially, should everybody be genotyped now before you initiate therapy and

225
00:36:56,133 --> 00:37:07,933
decide if it was going to be appropriate or not in those patients? There’s obviously a lot of opportunity here for community engagement

226
00:37:07,933 --> 00:37:23,599
research, also. If we’re going to focus in on APOL1, it’s an ideal opportunity of getting feedback from the community about: are they

227
00:37:23,600 --> 00:37:32,133
interested in this information? How will it be used? What’s the best way for us to proceed? And then Rob, at the end, raised there’s a lot of

228
00:37:32,133 --> 00:37:39,499
other opportunities but we didn’t get into that. I know it would be interesting to know the genetic architecture of the number of other issues in

229
00:37:39,500 --> 00:37:47,800
kidney disease research but that was a little outside of what we were focusing on. For example, what are the genes that regulate rate of

230
00:37:47,800 --> 00:37:58,333
progression? Can you use genetic information to help people that are at high risk for stones? Another good one would be, AKI recovery or

231
00:37:58,333 --> 00:38:07,966
not? If we learn more about the genetic architecture, would there be a way to use that information in clinical practice? Hopefully, I

232
00:38:07,966 --> 00:38:15,999
summarized. It was a good discussion. I think a lot of people have been thinking interesting thoughts on this and I think it’s exciting that we

233
00:38:16,000 --> 00:38:24,933
have potentially a clinically actionable finding that we can start thinking of ways to test in research settings; how this might help us in our practice

234
00:38:24,933 --> 00:38:39,699
now and not be waiting for understanding the biology whenever that may come. So, I’ll stop there.

235
00:38:39,700 --> 00:38:49,633
MALE: Thank you, John. I came late to your discussion, I was in the cohorts discussion first and I just want to point out -that a very large,

236
00:38:49,633 --> 00:39:03,799
very simple clinical study that is being funded by several of the institutes at NIH—the SPRINT study, which is NHLBI, NIDDK, the Aging Institute and

237
00:39:03,800 --> 00:39:13,566
some other institute support as well—there is an ancillary study looking at genetics both in patients with chronic kidney disease and non-chronic

238
00:39:13,566 --> 00:39:26,199
kidney disease controls where the study is a study of various medications used with different blood pressure goals. It’s very clinical and it’ll be

239
00:39:26,200 --> 00:39:38,300
very interesting to see if any clinical utility comes out regarding the APOL1 variants and MYH9 variants.

240
00:39:38,300 --> 00:39:44,233
GERJAN NAVIS: I would like to come back to what you said about kidney transplants. I don’t think it’s that provocative. I think the kidney

241
00:39:44,233 --> 00:39:52,466
transplant setting is a unique setting, let’s say, to do genetics and actually we have some data—not in African Americans because we don’t have

242
00:39:52,466 --> 00:40:06,732
too much of them—but in our [---] large transplant cohort where we replicate findings from GWAS as risk locus for CKD for graft loss and we’ve

243
00:40:06,733 --> 00:40:13,999
also found differences in a new frequency between donors and recipients, so this is independent replication in a different setting

244
00:40:14,000 --> 00:40:25,233
within a single cohort. But of course, all the better journals ask for independent replication. So if we have people in the audience here with transplant

245
00:40:25,233 --> 00:40:36,766
cohorts who would be interested in, let’s say, to replicate our findings or to come up with findings of their own that we could replicate, I think it’s a

246
00:40:36,766 --> 00:40:44,632
very powerful tool. Also, it’s a dissection tool because if you’re not sure about the biology of risk allele and it follows the kidney, it’s an

247
00:40:44,633 --> 00:40:52,466
inter-renal effect. I realize that’s too simple, but… JOHN SEDOR: Yeah. Thank you. Interesting

248
00:40:52,466 --> 00:41:08,466
comments, I agree. Okay. JEFFREY KOPP: Okay. Thanks to all you hardy

249
00:41:08,466 --> 00:41:17,566
participants who stayed to the end. Again, I’d like to thank the committee that organized this, the speakers that have traveled quite a distance in

250
00:41:17,566 --> 00:41:27,399
some cases, not so far in other cases, and all of you. I think it’s been a very good meeting. It is being video-archived, as you know, and I think

251
00:41:27,400 --> 00:41:34,833
we’ll try to follow up on some of the suggestions to maybe take some of these workout group summaries and condense them and put them

252
00:41:34,833 --> 00:41:40,833
together in some kind of a format. So, thank you all and have a safe trip back home.

253
00:41:40,833 --> 00:41:45,999
FEMALE: [---] was the leadership of this conference and he [---].

254
00:41:46,000 --> 00:41:48,166
JEFFREY KOPP: Oh, thank you.===




Date Last Updated: 9/18/2012

General Inquiries may be addressed to:
Office of Communications and Public Liaison
NIDDK, NIH
Building 31, Rm 9A06
31 Center Drive, MSC 2560
Bethesda, MD 20892-2560
USA
Phone: 301.496.3583