Download Reporting Incidental Findings in Genomic Scale Clinical Sequencing

The Journal of Molecular Diagnostics, Vol. 17, No. 2, March 2015
jmd.amjpathol.org
SPECIAL ARTICLE
Reporting Incidental Findings in Genomic Scale
Clinical SequencingdA Clinical Laboratory Perspective
A Report of the Association for Molecular Pathology
Madhuri Hegde,*yz Sherri Bale,*x Pinar Bayrak-Toydemir,*{k Jane Gibson,*,** Linda Jo Bone Jeng,*yy Loren Joseph,*zz Jordan Laser,*xx
Ira M. Lubin,*{{ Christine E. Miller,*k Lainie F. Ross,*kk*** Paul G. Rothberg,*yyy Alice K. Tanner,*yz Patrik Vitazka,*x and Rong Mao*{k
From the Incidental Findings Working Group of the Association for Molecular Pathology (AMP) Clinical Practice Committee and the Whole Genome Analysis
Working Group,* Bethesda, Maryland; the Department of Human Genetics,y Emory University School of Medicine, Atlanta, Georgia; the Emory Genetics
Laboratory,z Emory University, Decatur, Georgia; GeneDx,x Gaithersburg, Maryland; the Department of Pathology,{ University of Utah School of Medicine,
Salt Lake City, Utah; the Department of Molecular Genetics,k ARUP Laboratories, Salt Lake City, Utah; the Department of Clinical Sciences,** University of
Central Florida College of Medicine, Orlando, Florida; the Division of Endocrinology, Diabetes and Nutrition, Department of Medicine, the Department of
Pathology, and the Division of Human Genetics, Department of Pediatrics,yy University of Maryland School of Medicine, Baltimore, Maryland; the Departments
of Pathologyzz and Pediatrics,kk University of Chicago, Chicago, Illinois; the Division of Cytogenetics and Molecular Pathology,xx North Shore Long Island
Jewish Health System, New Hyde Park, New York; the Division of Laboratory Programs, Standards, and Services,{{ Centers for Disease Control and Prevention,
Atlanta, Georgia; the MacLean Center for Clinical Medical Ethics,*** University of Chicago, Chicago, Illinois; and the Department of Pathology and
Laboratory Medicine,yyy University of Rochester School of Medicine and Dentistry, Rochester, New York
Accepted for publication
October 21, 2014.
Address correspondence to
Madhuri Hegde, Ph.D.,
Department of Human Genetics,
615 Michael St., Whitehead
Biomedical Research Bldg,
Emory University School of
Medicine, Atlanta, GA 30022.
E-mail: [email protected].
Advances in sequencing technologies have facilitated concurrent testing for many disorders, and the results
generated may provide information about a patient’s health that is unrelated to the clinical indication,
commonly referred to as incidental findings. This is a paradigm shift from traditional genetic testing in which
testing and reporting are tailored to a patient’s specific clinical condition. Clinical laboratories and physicians
are wrestling with this increased complexity in genomic testing and reporting of the incidental findings to
patients. An enormous amount of discussion has taken place since the release of a set of recommendations
from the American College of Medical Genetics and Genomics. This discussion has largely focused on the
content of the incidental findings, but the laboratory perspective and patient autonomy have been overlooked.
This report by the Association of Molecular Pathology workgroup discusses the pros and cons of nextgeneration sequencing technology, potential benefits, and harms for reporting of incidental findings,
including the effect on both the laboratory and the patient, and compares those with other areas of medicine.
The importance of genetic counseling to preserve patient autonomy is also reviewed. The discussion and
recommendations presented by the workgroup underline the need for continued research and discussion
among all stakeholders to improve our understanding of the effect of different policies on patients, providers,
and laboratories. (J Mol Diagn 2015, 17: 107e117; http://dx.doi.org/10.1016/j.jmoldx.2014.10.004)
The Incidental Findings Working Group is a joint working group of the
AMP Clinical Practice Committee and the Whole Genome Analysis
Working Group. The 2013 Clinical Practice Committee consisted of:
Matthew Bankowski, Milena Cankovic, Jennifer Dunlap, Larissa Furtado,
Jane Gibson, Jerald Gong, Thomas Huard, Linda Jeng, Loren Joseph (Chair
2013 to 2014), Marilyn Li, Paul Rothberg, M. Fernanda Sabato, and Patrik
Vitazka. The 2013 Whole Genome Analysis Working Group consisted of:
Nazneen Aziz, Pinar Bayrak-Toydemir, Daniel Farkas, Jane Gibson
(Chair), Bert Gold, Andrea Ferreira-Gonzalez, Timothy Greiner, Loren
Joseph, Roger Klein, Debra Leonard, Marilyn Li, Ira Lubin, Elaine Lyon,
Copyright ª 2015 American Society for Investigative Pathology
and the Association for Molecular Pathology.
Published by Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jmoldx.2014.10.004
Karen Mann, Rong Mao, Victoria Pratt, Iris Schrijver, and Patrik Vitazka.
Disclosures: M.H. is on the Scientific Advisory Board for Ingenuity/
Qiagen and Oxford Gene Technologies; S.B. is on the Scientific Advisory
Board for Ingenuity/Qiagen and Raindance Technologies; Standard of
practice is not defined by this article, and there may be alternatives. See
Disclaimer for further details.
The findings and conclusions in this report are those of the author(s)
and do not necessarily represent the views of the Centers for Disease
Control and Prevention/the Agency for Toxic Substances and Disease
Registry.
Hegde et al
The introduction of new genomic technologies in clinical
medicine,1 particularly next-generation sequencing (NGS) of
entire exomes and genomes, generates an abundance of data
only some of which is relevant to the clinical question that
prompted the investigation.2 Analysis of evidence that is
based on targeted gene panels can also result in the generation of data not relevant to the clinical question posed
because some genes are associated with multiple medical
conditions.3e5 These additional findings present difficulties to
the laboratory, most importantly deciding which information
should be reported. A number of factors go into this decision,
including patient preferences usually communicated through
the consent process and documented on the consent form, the
clinical relevance of findings, the potential for false-positive
or -negative results, and the application of professional
practice standards. In addition, the laboratory must grapple
with the issues of either confirming potentially important
incidental findings or reporting results with a lower quality
threshold than is usually required.3 Finally, if the laboratory
does have a policy of reporting potentially important incidental findings, a predictable unintended consequence could
be that the ordering provider misinterprets a report with no
pathogenic variants to mean meaning that there are actually
no pathogenic variants in a specific gene, when, in fact, the
gene sequence and its disease-causing mutation spectrum
may not have been fully covered during sequencing; therefore, pathogenic variants could have been missed.6 This
document was prepared by a Joint Working Group of the
Association for Molecular Pathology Clinical Practice Committee and Whole Genome Analysis Working Group to
discuss these issues from a laboratory perspective and to
develop recommendations for laboratories considering policies for reporting incidental findings.
By way of definition, the term incidental finding is used in
this report to include all findings that are clearly or expected
to be pathogenic but do not address the clinical question that
motivated testing. This is the same intent as was used in the
recently published policy statement from the American
College of Medical Genetics and Genomics (ACMG).7 The
ACMG Working Group on the reporting of incidental
findings from clinical exome and genome sequencing recommended that incidental findings for 56 genes that are
clearly pathogenic and clinically actionable be reported by
the laboratory. For example, identifying patients with an
inactivating mutation in the APC gene that causes familial
adenomatous polyposis permits increased surveillance and
early detection of cancerdwhen it can be best treated. The
ACMG document urged that this policy apply regardless of
the age of the patient, except for fetal samples, and also
included any germline testing done as part of cancer-related
genomic scale sequencing. The ACMG document is a significant milestone in the development of clinical genomic
sequencing services because it clearly defines the issues and
challenges and makes a major contribution to setting policies in this area. An ACMG workgroup recommendation
specifically excludes giving the patient the opportunity to
108
refuse or opt-out of receiving a report that details pathogenic
mutations identified in these 56 genes, catalyzing much
discussion in the field.3,8e13 The goals of the ACMG report,
the subsequent discussions, and this document are to inform
the development of policies which maximize the benefits
and minimize the harms that can result as a consequence of
the reporting of incidental findings to inform clinical and
personal health care decisions.
Here, we discuss several issues in the reporting of incidental findings in clinical genomic sequencing from the
laboratory perspective. In particular, we explore the potential
benefits and harms that could result from different decisions
for reporting incidental findings, including the effect on both
the laboratory and the patient. We consider the relation
between the reporting of incidental findings in genomics and
other areas of medicine. We also address how to include
patient decision making in the testing process to preserve
respect for patient autonomy. We make suggestions about
how to shape the pretest counseling/informed consent process and the laboratory report to accurately convey what was
and was not found. Finally, this document emphasizes the
need for continued research and discussion among all
stakeholders to improve our understanding of the effect of
different policies on patients, providers, and laboratories.
From Sanger Sequencing to NGS
Sanger sequencing is used routinely by clinical laboratories
for diagnosis of heritable conditions and to detect somatic
mutations in cancer. Although Sanger sequencing is
considered the gold standard, its relatively low throughput
and high cost have been major obstacles to its use in
sequencing large regions of the human genome in the clinical
setting. Thus, NGS, also called massively parallel
sequencing, has become an attractive alternative for applications that require assessment of multiple genes, exomes, or
genomes in situations in which the clinical presentation is
ambiguous and/or previous diagnostic testing did not lead to
identification of the underlying molecular basis for the
patient’s clinical presentation.1,14e17 In general, genomic
DNA used for library preparation for NGS can be either used
with (panels and exomes) or without (genomes) enrichment.
Each of the target enrichment approaches has advantages and
drawbacks. For example, all enrichment procedures can
suffer from analytical limitations, including but not limited to
the sequencing of pseudogenes, variation of amplification
efficiency across numerous targets, GC-rich area biases, and
allelic drop out. For the sequencing of a gene panel, lowcoverage or low-quality segments can be resolved by analysis with alternative methods including Sanger sequencing,
either preemptively or after such low-coverage regions are
found during review of sequencing data. Resolving all lowcoverage segments for whole-exome sequencing (WES) is,
however, highly impractical.18 WES is considered to be a
screening tool because not all genes are covered equally well
jmd.amjpathol.org
-
The Journal of Molecular Diagnostics
Reporting Incidental Findings in Genomics
and some genes or exons are not covered at all (eg, SMN1 gene
and triplet repeat disorders). WES has less than optimal
analytical sensitivity but offers relatively high clinical sensitivity. Although whole genome sequencing (WGS) has gained
popularity in the research area, because of its high associated
costs for sequencing and data storage, it is often performed at a
significantly lower coverage than the WES, reducing its
technical sensitivity. Another advantage of WES is the relative
ease of interpretation of variants because WES targets only the
exome (ie, approximately 92% of all known coding exons and
exon-intron boundaries, approximately 180,000 exons of
approximately 20,500 genes), which represents approximately
1% to 2% of the total human genome (or approximately 30 Mb
of sequence). Therefore, WES represents a compromise
between the number of genes that are assessed (all known
genes), albeit at imperfect coverage, and a limited number of
well-covered genes, compared with more analytically sensitive/specific panels and WGS.
exceptions apply. Much more difficult to interpret are
missense variants, especially those which are not located in
evolutionarily well-conserved regions, not found in certain
functional motifs or domains, or not known to be in residues
that are subject to posttranslational modification. In addition,
variants for which functional predictions with the use of
programs such as Sorting Intolerant from Tolerant and
Polymorphism Phenotyping, favor indeterminate or benign
variants that result in a conservative change, or variants
outside of the canonical splice site for which splicing predictors are mixed are also difficult to interpret. Erroneous
entries in public databases that are not curated or clinically
validated, insufficient evidence because of poorly designed
scientific studies (insufficient functional evidence or underpowered segregation analyses), isolated reports of association
of a particular gene with a phenotype or presence of a
phenotype that has not been previously associated with the
gene, and lack of clinical indications/notes can also hinder
interpretation of identified variants.20
Interpretation of Genomic Variants
Genome-wide sequencing approaches usually yield tens of
thousands variants when the sequenced data are aligned to the
reference genome. Not all variants are pertinent to the patient’s
clinical presentation and, in fact, usually only one (for dominant disorders) or two variants (for recessive disorders) cause
disease.19 It is impossible to evaluate every single variant
manually; therefore, bioinformatics analysis must be used to
sort (filter) variants to narrow the number of relevant candidates. In the filtering process, the variants are assessed with
several considerations, including population frequency of the
variant, presence of the variant in clinical or sequencing databases such as Human Gene Mutation Database, ClinVar,
Online Mendelian Inheritance in Man, National Heart, Lung,
and Blood Institute Grand Opportunity Exome Sequencing
Project, and Locus-Specific Databases; likely mode of inheritance; phenotypic features and genes associated with these
features; list of genes likely to explain the clinical presentation;
predicted functional severity of the sequence change; variant
type such as synonymous, missense, nonsense, frameshift, inframe; and pathogenic versus benign variants (ie, variants in
internal databases that violate the Hardy Weinberg Equilibrium, variants present in unaffected homozygotes); tissue
expression of a gene with a variant; relevance to the clinical
phenotype of pathways in which the product of the variant
gene participates, and so forth. Filtering processes can have
inherent flaws and may contribute to false-negative results.
This is especially true for dominant missense variants that have
not been described in the literature, have been inherited from
either parent (as opposed to occurring de novo in the proband),
or have incomplete penetrance and/or variable expressivity.20
Another concern is the ability to interpret identified variants, especially missense and silent variants. Frameshift,
nonsense, and canonical splice site variants, and deletion
and duplication of several bases in regions within functional
domains are considered likely pathogenic, although some
The Journal of Molecular Diagnostics
-
jmd.amjpathol.org
Laboratory Perspective
Evidence-Based Targeted Gene Panels/WES/WGS
Since the implementation of NGS in clinical laboratories, the
focus has been on developing large (>50) gene panels.6,14,21
More recently the trend has shifted toward using WES and
WGS. Although it is assumed that variants identified on gene
panels are likely relevant to the patient’s phenotype, a finding
not related to the primary clinical indication may still be
identified. Such incidental findings are a certainty in WES and
WGS. The chance of identifying incidental findings increases
when the genes being tested increase in number and decrease in
specificity with respect to the disease. For example, a highly
focused gene panel test, consisting of only well-vetted genes
for a narrow and specific phenotype (eg, five genes associated
with Cornelia de Lange Syndrome), is unlikely to reveal information not directly related to the patient’s phenotype.
However, when a large panel designed to evaluate patients who
have a broad, or ill-defined phenotype (eg, ataxia and spasticity) is used, the chance of identifying an incidental finding is
increased. For example, in the process of testing all genes
known to cause ataxia, one might identify a heterozygous
variant in the ATM gene, which confers a significantly
increased risk of developing breast cancer. Both WES and
WGS provide at least some information on 90% to 95% of all
known genes, even though the purpose of doing the test is to
identify the underlying cause of a more-or-less specific
phenotype in a patient. Because all of the data must be filtered
in a variety of ways to get to the molecular basis of the disease,
many variants in many genes must be examined to obtain a
definitive answer. It is very likely that heterozygosity (carrier
status) for autosomal recessive or X-linked disorders will be
uncovered, and there is also a chance of identifying genes that
convey risk of developing a late-onset disorder, including
disorders with devastating neurologic effects, or cancer.
109
Hegde et al
Although WGS provides more data than WES, the risk of
identifying incidental findings is not expected to be much
higher because most of the additional data present in a WGS
test is not currently interpretable with respect to human disease.
From the laboratory’s perspective, the reporting of a specific
set of secondary findings (as recommended by the ACMG) has
several implications from patient consent, laboratory analysis/
workflow, and cost standpoints. It is generally understood that
obtaining consent for testing is the responsibility of the
ordering health care provider and not of the laboratory. If the
laboratory has decided to provide an opt-out for patients, it will
be necessary to modify the analysis to reveal or mask information on the specific genes such as those recommended for
reporting by ACMG. As recommendations change, laboratories must implement workflow changes continuously. The
review and interpretation of variants in the 56 recommended
ACMG genes requires significant effort from the bioinformaticists/analysts, genetic counselors, and laboratory directors as they work to determine which variants are reportable
and which do not meet criteria for reporting. In addition, each
reported incidental finding should be confirmed by Sanger
sequencing, incurring further expense and increasing turnaround times. The recommendation to report incidental findings in all persons who undergo WES is an additional burden
on laboratories that perform full WES on parents and/or siblings only in the process of attempting to identify the proband’s
disease-causing variant(s). To perform additional analyses on
those persons (and interpret and confirm the findings) involves
more effort and expense, and puts those laboratories at a
disadvantage compared with laboratories that only perform
targeted analysis of family member samples for segregation
analysis of presumptive disease variants in the proband.
Current limitations of NGS include major challenges in
interrogating medically significant genes with highsequence homology. These include genes with complex
sequence contexts such as pseudogenes, genetic rearrangements, or a high GC content. Exclusion from analysis
is a possibility but clearly not optimal for genes of high
medical relevance, especially those for which ACMG recommends returning results regardless of the patient’s clinical phenotype/indication for testing. Out of the 56 genes on
the ACMG recommended list, at least eight have one or
more pseudogenes associated with them (MYH7, MYLK,
PKP2, PMS2, PTEN, SCN5A, SDHC, and SDHD). These
regions should be sequenced with great attention, using
more stringent criteria, to ensure reporting of variants only
from the active copy of the gene and may require additional
confirmatory testing.
The Patient Perspective
Decision Making and Insurance Considerations
Studies by Tabor et al22,23 on the informed consent process in
research WGS found that, although all family members
involved wanted to receive results related to the condition for
110
which they were being tested, they differed on their desire to
receive incidental findings, even among members of the same
family. Several participants also felt that it was important to be
able to change their preferences between the time of consent
and the time of receiving results. A study of institutional review board perspectives on the return of genomic research
results noted respondents were firm in their position that the
research participants’ right to know/not know information
about themselves be respected.24,25 The National Heart, Lung,
and Blood Institute issued updated guidelines in 2010 on
reporting genetic research results to study participants.26 They
recommend that genetic results should be offered to study
participants if the results meet certain criteria, including during the informed consent process or subsequently, the study
participant has opted to receive his or her individual genetic
results. An overriding regard for the right of the participant/
patient in deciding what results to receive is evident in these
studies. Although these studies were conducted with persons
undergoing WGS on a research basis, it is reasonable to expect
similar results from clinical WES/WGS participants.
There are a number of reasons that persons may be hesitant to
receive incidental finding results. First, the expressivity of genetic variants known to be highly penetrant in high-risk communities is unknown for low-risk communities. In this vein, the
US Preventive Services Task Force recommended against
BRCA gene testing in low-risk populations (Risk Assessment,
Genetic Counseling, and Genetic Testing for BRCA-Related
Cancer in Women, Topic Page. US Preventive Services Task
Force; http://www.uspreventiveservicestaskforce.org/Page/
Topic/recommendation-summary/brca-related-cancer-riskassessment-genetic-counseling-and-genetic-testing, last
accessed February 9, 2015).
Patients may not want information that may cause more
distress than clinical benefit.8 Second, it is possible that
incidental findings revealed by WES/WGS testing may prevent a person from obtaining life insurance, long-term care
insurance, and/or disability insurance. Although the Genetic
Information Nondiscrimination Act protects against health
insurance and employment discrimination, it does not extend
to these other areas of insurance. Patients should be informed
of this possibility before testing. Patients should be given the
opportunity to carefully consider the risks associated with
possible incidental findings versus the possible risks of insurance discrimination and then decide for themselves
whether they want to receive the incidental findings. Without
an opt-out option, patients only have the choice of undergoing
WES/WGS, which could potentially diagnose their condition
and at the same time leave them unable to obtain life/
disability/long-term care insurance, or declining the opportunity for diagnosis to prevent insurance discrimination.
Public Databasesefrom AMP Letter to Presidential
Commission
In 2013, AMP submitted public comments to the Presidential Commission for the Study of Bioethical Issues on
jmd.amjpathol.org
-
The Journal of Molecular Diagnostics
Reporting Incidental Findings in Genomics
the topic of genetic and genomic testing. In that letter, the
creation and maintenance of public databases for the
curation of genomic variants was addressed. To reduce the
number of variants of uncertain significance in both primary and incidental findings, information must be gathered
on the possible clinical consequences of individual variants. As most genetic conditions are rare, information from
any single health care provider’s patients will likely be of
limited value. International databases of de-identified data
will provide much more valuable information that can be
used to interpret variants. Resources must be dedicated to
the creation, development, and maintenance of centralized,
curated databases of genomic variants, to which all laboratories should be strongly encouraged to submit
de-identified data, including individual variants identified,
relevant associated clinical information, and acquired
information about potential pathogenicity to publicly
available databases such as ClinVar database from the
National Center for Biotechnology Information. Data must
be submitted with appropriate safeguards to protect patient
privacy. Resources should be provided to allow for
continual curation and updating of the information contained within the database to ensure accurate and timely
classification of variants.
Comparison with Incidental Findings in Other
Areas of Medicine
Reporting of incidental findings is not a new problem in
medicine. Incidental findings in radiology are common. A
typical example would be a scan ordered to assess the kidneys, which necessarily includes the adrenal glands (present
at the top of the kidneys).27 If a lesion is present in the
adrenal gland, the radiologist cannot avoid seeing it even if
the examination was undertaken only because of suspected
kidney disease. Most such incidentalomas prove benign but
can trigger expensive anxiety-inducing work-ups. Similarly, in the field of genetics incidental findings are not
uncommon. For example, chromosome analysis of leukemia might find an unexpected complement of sex chromosomes, or evaluation of the segregation of a mutation in a
family may reveal misattributed paternity. A feature common to all these examples is that the clinician, although not
specifically looking for the incidental finding, cannot avoid
the observation.
The ACMG report, supported by some commenters, argues that incidental findings from WES/WGS are similar to
incidental findings in other areas of medicine and must be
reported, as in common practice.7,13 However, other commenters, as well as this document,28 point out that additional
findings, beyond the genes analyzed to answer the clinical
question that prompted testing, are not evident without
significant extra effort directed toward that end.8,10 The use
of the term incidental finding for both types of results is
unfortunate, because the two types of findings are
The Journal of Molecular Diagnostics
-
jmd.amjpathol.org
fundamentally different, one being unavoidable and the
other requiring additional analysis and interpretation.
Genetic Counseling and Consenting Perspective
The consenting process for WES and WGS is critically
important and should be performed by a person highly
trained in genetics such as a geneticist or genetic counselor.
Because WES or WGS is much more complex than consenting to single-gene analysis for a Mendelian condition,
patients and their families will need to be educated about the
possible types of results that could be returned. The medical
professional needs to be familiar with the testing protocol and
consent form used by the laboratory, because this will affect
the information the professional provides when obtaining
consent from the patient/family.29 Medical professionals who
order WES/WGS should also know that laboratory policies
for analysis and reporting of incidental findings may change
as they carefully consider advances in technology and the
effect of implementing the ACMG recommended panel as
well as the recommendations of other groups.
Each laboratory should clearly indicate on their consent
form if it routinely performs analysis for incidental findings
such as the ACMG recommended additional 56 genes or
whether it explores an even further expanded list of incidental
findings unrelated to phenotype (such as pediatric-onset
conditions, adult-onset conditions, carrier status for recessive conditions, or pharmacogenetic markers). Laboratories
should also be as specific as possible in listing the genes to be
analyzed and in defining the types of incidental findings they
routinely report, especially if they deviate from the ACMG
recommended list, because this will aid the genetics professional in obtaining the most accurate informed consent
possible. If they exclude certain types of conditions from
being reported as incidental findings (ie, adult-onset neurodegenerative conditions with no cure) that should be specified in the consent. The consent should clearly delineate
whether the patient or other family members have the option
of choosing not to receive incidental results. If several family
members are undergoing exome sequencing to help aid in the
variant analysis and diagnosis of the proband, each of these
persons should give consent on a separate written document
so that their independent desire to opt-in or opt-out of testing
for incidental findings can be respected. They should also be
informed about how they are to receive incidental findings if
they are not the proband, because the results from the incidental findings in other family members should remain
confidential. In addition, some laboratories that typically
provide incidental findings may feel strongly about not
violating the principle of a minor’s autonomy by not
reporting incidental findings for adult-onset conditions for a
patient who is younger than 18 years. They may decide not to
test for adult-onset conditions in minors. In a joint policy
statement and technical report, the ACMG and the American
Academy of Pediatrics continue to support deferring the
testing of late-onset conditions until adulthood.30
111
Hegde et al
Table 1
Education Material in Genomics
Target
audience
Ordering
practitioners
Concept to address
Materials/resources
testing laboratory
may provide
Clinical utility
Scientific references
Limitations of the technology
Practice guidelines
Scientific references
Likelihood of incidental findings Scientific references
Potential effect of incidental
findings on the patient
Example of potential
scenarios
Practitioner/patient discussion
NA
Genetic counseling partnership
Recommended
source for materials
Comments
General: manufacturer/
References should meet
professional
appropriate levels of evidencesocieties; testbased medicine
specific: laboratory*
Overview of potential for falsely
positive/negative results.
Specific genetic aberrations
not identified (ie, large indels)
Types: carrier status, inheritance
pattern; particular attention
should be paid to effect on
patient and other family
members
Medical effect (management);
psychological effect (anxiety/
relief); financial effect (life,
disability insurance, etc)
Provide adequate time and
information for an informed
decision
Expertise available to the
provider and/or patient before
and after testing
Find a genetic counselor
through the National
Society of Genetic
Counselors (http://www.
nsgc.org, last accessed
December 1, 2014)
Federal laws that govern genetic References to the Genetic Federal protections to patients
about the use of genetic
testing
Information
information for employment
Nondiscrimination Act
and health insurance
and patient protection
and Affordable Care Act
Local and state laws that govern
Specific local requirements may
genetic testing
affect consent forms,
disclosure information, etc
Uphold patient autonomy
Explanation for the
Patient empowerment for final
necessity of excellent
decision making
communication skills
Same as ordering practitioners
See above
Referring
More detailed summary of
Scientific references/
laboratorians/ Limitations of the technology
limitations (ie, exon drop out)
laboratory-specific
pathologists
and how the laboratory handles/
limitations (method
reports such limitations
based)
Patients
Uphold patient autonomy
Educational packet; online
materials for patients
General consenting process to
Educational packet for
include an explanation of:
patients
General description of the test
Purpose of the test
Offer pretest genetic counseling
Results may indicate a
predisposition to a disease,
further testing may be pursued
General: manufacturer/
professional
societies; testspecific: laboratory
Laboratory
Professional societies
Professional societies
NA
Professional societies
Professional societies/
laboratory
Professional societies
Laboratory
Professional Societies
Professional societies
local/state
requirements:
laboratory
Laboratory
Referring physician
Referring physician
Physician
(table continues)
112
jmd.amjpathol.org
-
The Journal of Molecular Diagnostics
Reporting Incidental Findings in Genomics
Table 1
Target
audience
(continued )
Materials/resources
testing laboratory
may provide
Concept to address
Incidental genomic findings,
general discussion
Variety of options for reporting
incidental findings
Comments
Recommended
source for materials
Laboratory
Patient has the right to choose Laboratory
what information should be in
the report (ie, targeted only vs
targeted and incidental etc)
Laboratory
Level of certainty the findings
predict a disease state
Names/categories of persons and
organizations to whom the
results may be disclosed
Identification of nonparental
Particularly when parentrelationships
proband triads are tested
Support services before and after Referrals to appropriate
Genetic counselors, medical
testing
professionals, as needed
geneticists, medical
specialists, therapists
Laboratory
Laboratory
Laboratory
*Laboratory provides the source of information to the referring physician who educates patient.
NA, not applicable.
Patients/families should be counseled on the numerous
limitations of exome sequencing. First, in the context of the
primary indication for the test, if a causative gene variant(s) is
not identified for a disorder in question, a genetic etiology has
not been excluded. They should understand that there are
many genes not amenable to variant detection by using NGS
technology. Next, should incidental findings be sought, not
all genes or gene regions will be analyzable by NGS, so a
negative result does not rule out a variant in the list of ACMG
genes or any other expanded list of incidental findings. For
example, the ACMG list includes the Lynch syndrome gene,
PMS2.31,32 Since there are up to 16 pseudogenes for some
exons in the PMS2 gene with a high level of extreme
sequence similarity, it may not be possible to determine
whether variants identified in the PMS2 gene are actually in
the PMS2 gene or a pseudogene. Because of all these problematic genes, the recommendation from this workgroup is
that pathogenic incidental findings should be confirmed by
Sanger sequencing before reporting the result. In addition the
common causative mutation in other ACMG genes may be a
large deletion (eg, SMN1, which causes spinal muscular atrophy); large deletions are not amenable to detection by either
NGS or Sanger sequencing and require other methodologies
for detection.
Laboratories should clearly describe on the consent
document the type of variants reported for the primary
indication, for example, pathogenic and likely pathogenic
variants, variants of unknown significance identified in
genes known to be associated with disease, and variants in
genes of unknown function. They should describe how this
differs from the types of variants reported for incidental
findings, which should be restricted to prior reported known
pathogenic variants and expected pathogenic variants that
are of the type expected to be causative for the disorder. The
The Journal of Molecular Diagnostics
-
jmd.amjpathol.org
consent should clearly state the psychological risks and risks
of insurance discrimination that could be the sequelae of
reporting incidental findings. This is important because of
the real difficulty persons who are carriers of cancer predisposition gene variants may face in getting life or
disability insurance.
Oversight and Regulation
Laboratories must comply with federal and state regulatory
requirements. At the federal level, the Clinical Laboratory
Improvement Amendments (CLIA) regulations provide minimal standards for laboratory practice.33 The Centers for Medicare and Medicaid Services provide oversight and enforcement
of the CLIA regulations. States such as New York and Washington have developed their own regulations comparable with
federal oversight, and the Centers for Medicare and Medicaid
Services has deemed these comparable with the federal
requirements which allow these states to use their own
oversight mechanisms (http://www.wadsworth.org/labcert/
clep/clep.html and http://www.hhs.gov/regulations, last
accessed December 2, 2014). Some states impose additional
requirements on laboratories beyond what is described in
the CLIA regulations (http://www.healthit.gov/sites/default/
files/290-05-0015-state-law-access-report-1.pdf, last accessed
December 2, 2014). For example, several states have regulations that address which person(s) is authorized to receive reports of clinical laboratory test results. A recently announced
policy from the US Health and Human Services indicates that
patients should be able to receive results directly from the
testing laboratory. The US Food and Drug Administration
primarily regulates manufacturers of tests that are sold to
clinical laboratories for clinical testing (http://www.fda.
113
Hegde et al
Table 2 Recommendations of the AMP Work Group on Incidental Findings
1. Test advantages and limitations should be made available to the physician and clearly stated on the laboratory reports.
2. Laboratories should develop a consent form, which clearly explains the test advantages and limitations, and the categories of variants
reported (eg, diagnostic, carrier, pharmacogenetic markers).
3. Choice of opt-in or opt-out be given to patients and their families (in case of minors) in the precounseling session.
4. Laboratory clearly states which genes (ACMG-defined list, laboratory-defined list, or all genes) will be analyzed in incidental findings with
the disclaimer that not all regions of all genes are always covered.
5. Only known pathogenic or likely pathogenic variants should be reported.
6. Reported pathogenic variants should be confirmed with an alternate technology. The method used for confirmation should be stated under
the methodology section in the reports.
7. Laboratories should actively submit the list of pathogenic variants identified in their defined list of genes for which they choose to report
incidental findings to public databases such as ClinVar, Human Gene Mutation Database, and Leiden Open Variation Database.
ACMG, American College of Medical Genetics and Genomics; AMP, Association for Molecular Pathology.
gov/MedicalDevices/ProductsandMedicalProcedures/InVitro
Diagnostics/default.htm, last accessed December 2, 2014). To
date, clinical sequencing tests are developed in the laboratory
and do not come under the purview of the Food and Drug
Administration. In 2013, the Food and Drug Administration
cleared four NGS products, two cystic fibrosis tests and a
platform in combination with a universal reagent kit. To date,
the regulatory agencies described here have not addressed the
issue of reporting incidental findings. Finally, the College of
American Pathologists accredits medical laboratories in a
voluntary system by using inspections according to checklists
that are laboratory specific and regularly updated (http://www.
cap.org/apps/cap.portal?_nfpbZtrue&_pageLabelZaccred
itation, last accessed December 2, 2014). The Molecular
Pathology Checklist (www.cap.org/molecularpathology
checklist), updated July 31, 2012, states that laboratories
should have a “policy regarding reporting of clinically significant genetic findings unrelated to the clinical purpose for
testing” but does not specify what that policy should be. Thus,
the laboratories are responsible for developing their own
individual policies.
CLIA requires that clinical laboratories establish the performance specifications of tests offered. These include accuracy, precision, analytic sensitivity and specificity, reference
range, reportable range, and other relevant characteristics. This
requirement applies equally to all reported findings of a clinical test with no distinction between test results that address the
reason the test was ordered or are incidental. However, this
standard does not require that all test results meet identical
criteria, just that the performance specifications and limitations
of the test are documented. In practical terms, if incidental
findings are reported, the methods used must be validated with
accepted protocols that meet regulatory standards and, if
applicable, accreditation standards. It is clearly not permitted
to report incidental findings if the laboratory has not established that the performance specifications of this aspect of
testing meets these standards.
At the present time, regulatory and accreditation
requirements do not distinguish between primary and incidental findings. Therefore, assay development for both should
be comparable, and the reporting of an incidental finding
should be as reliable as the reporting of a primary test result.
For example, the validation protocols should be comparable.
In addition, the direct costs of validating, identifying, interpreting, and reporting incidental findings may also be significant, and it is not clear to what extent payers will reimburse for
aspects of testing that are not explicitly ordered.
114
jmd.amjpathol.org
The Need for Education in Genomics
As genomic medicine increases in clinical utility and enters
the everyday vernacular of the medical community, there is a
tremendous need to ensure that the appropriate information
and education reach the correct audiences. The education
needs to be tailored to all of the stakeholder groups influenced by genomics. Stakeholders will likely include multidisciplinary care teams who can appropriately manage
patients in an era of precision medicine. These groups include
i) ordering practitioners (physicians, physician assistants,
nurses, genetic counselors), ii) referring laboratorians and/or
pathologists (eg, molecular laboratory geneticists certified by
the American Board of Medical Genetics and Genomics), iii)
patients (and family members, when applicable), iv) policy
makers, and v) payers.
In each of the groups listed in the previous paragraph,
there clearly will be a wide range of educational levels,
socioeconomic backgrounds, and abilities to access supportive services. Thus, a one-size-fits-all approach to education is wholly insufficient to deal with the particular
questions and concerns of each stakeholder population.
The importance of the education process in genomic
testing and the possibility of incidental findings cannot be
overstated. All stakeholders should be comfortable with the
intent, likelihood, and potential effect of incidental findings
before ordering any genomic evaluation.
With the large amount of educational materials that should
be provided to the variety of stakeholders, it is important to
address potential sources for these materials (Table 1). Some
materials, such as those directly related to a particular test,
should be provided by the performing laboratory. For
laboratory-developed procedures, there may not be any other
resource accurately describing the test characteristics. Testspecific educational information should be a component of
-
The Journal of Molecular Diagnostics
Reporting Incidental Findings in Genomics
the informed consent process; in fact, some states (ie, New
York) have civil rights laws that mandate this information be
provided at the time of consent. For more general information,
such as the limitations of a particular method, the molecular
diagnostic community could look to the manufacturers and/or
professional societies for the development of educational
materials. Clearly, a robust education program should be
developed in conjunction with professional societies.
Workgroup Conclusion on Reporting of
Incidental Findings
The detection of genetic variants that may be of clinical
significance but are unrelated to the indication for testing
may result from NGS analysis of gene panels, whole
exomes, and genomes. A great deal of controversy exists
about whether these variants should be reported to the
ordering physicians and the patients.
The ACMG Working Group on Incidental Findings initially
recommended that known pathogenic and predicted to be
pathogenic mutations in a set of 56 genes be reported without
regard to the age of the patient or the initial reason for testing,
except for fetal samples. After extensive feedback, ACMG
issued a clarification, stating that patients should be given an
opportunity to opt-out of receiving incidental findings.
This workgroup is of the opinion that each laboratory that
provides NGS services must develop a policy for analysis and
reporting of known and predicted pathogenic variants from the
list of genes recommended by ACMG and other known and
predicted pathogenic variants (Table 2). The laboratories’
policy should be based on its ability to provide this information with sufficient accuracy. We also support the practice that
persons undergoing WES or WGS should have the opportunity to opt-in or opt-out of receiving a report that documents
variants that are not relevant to the initial reason for testing.
Each laboratory must also determine what information it is
willing to provide to parents or other third parties, and whether
these third parties will have the opportunity to opt-in or optout. This decision should be part of the patient education
and informed consent process. Most patients will likely opt-in,
given the opportunity to receive additional information that
could be used to improve their health. However, health care
professionals must be sensitive to the possibility that patients
dealing with a difficult medical issue that led to the need for
WES or WGS may not wish to have an additional burden of
concerns placed on them about future health issues. In addition, the needs of specific cultural and ethnic groups must be
accommodated. Every laboratory that performs NGS testing
for clinical applications should be aware of the ethical and
patient management issues associated with incidental findings.
Laboratories should have policies in place that clearly define
their approach to reporting the ACMG gene list and other
incidental findings.
NGS technology is evolving rapidly, and the rate of
confirmation of single nucleotide variants is very high, but
The Journal of Molecular Diagnostics
-
jmd.amjpathol.org
the technology is not accurate for detecting copy number
variants. It is likely that once a pathogenic variant is
reported, additional family members will obtain targeted
testing for themselves or prenatal diagnosis. Targeted
testing will likely be performed with alternate method such
as Sanger sequencing. For these reasons, this workgroup
recommends that reported variants identified by NGS be
confirmed with an alternate method. As the NGS technology evolves, it is likely that these recommendations
will change.
Only the known pathogenic and expected pathogenic
variants should be included in the incidental finding report if
the laboratory chooses to report these. The interpretation of
variant pathogenicity should follow the ACMG Recommendation for Interpretation of Sequence Variants.19 The
determination of pathogenicity for known variants is based
on scientific literature, variant databases, and clinically
validated/curated locus-specific databases, which support
the assertion that the variants in question are causative of
disease or not. For variant(s) that have not been reported
previously, the known mechanism of disease and the
ACMG guidelines should be followed. All of the known
pathogenic and expected pathogenic variant(s) should be
reported in accordance with the nomenclature recommended
by the Human Genome Variation Society (http://www.hgvs.
org/mutnomen, last accessed December 2, 2014). The report
should include relevant references to the literature and references to variant databases to support the classification of
the pathogenicity of the variant.
Incidental findings should be clearly labeled as such and
included in the full report sent to the ordering physician.
One ongoing consideration is that often the ordering
physician is not experienced in interpretation of reports
that involve gene variants and their implications for disease. It will be very useful to include information in the
report about the variant’s association with disease and to
provide recommendations for clinical follow-up. The
report should be written so that it can be understood
without the need for a high level of genetic knowledge and
should avoid the use of scientific jargon. The report should
clearly outline the limitations of the analysis. For example,
some genes, or parts of genes, may not be adequately
sequenced (some region of the genes could have low or
absent sequencing coverage), and some variants that are
discovered may not be clearly interpretable for risk of
clinical disease. This is important because the ordering
physician and the patient might believe that the lack of
reporting of an incidental finding equates with a lack of
pathogenic variants.
This workgroup appreciates that the field of genomic/
precision medicine continues to evolve rapidly and the
clinical, laboratory, and ethical issues surrounding reporting
of incidental findings needs continued evaluation from the
multidisciplinary clinical community. It is likely that recommendations from professional societies will evolve as
NGS technology matures, genomic education improves, and
115
Hegde et al
regulatory authorities and other groups refine the standards
for laboratory medicine with respect to genomic analysis.
Disclaimer
The AMP Clinical Practice Guidelines and Reports are
developed to be of assistance to laboratory and other health
care professionals by providing guidance and recommendations for particular areas of practice. The Guidelines or
Report should not be considered inclusive of all proper
approaches or methods, or exclusive of others. The Guidelines or Report cannot guarantee any specific outcome, nor
do they establish a standard of care. The Guidelines or
Report are not intended to dictate the treatment of a
particular patient. Treatment decisions must be made based
on the independent judgment of health care providers and
each patient’s individual circumstances.
AMP makes no warranty, expressed or implied, regarding
the Guidelines or Report and specifically excludes any
warranties of merchantability and fitness for a particular use
or purpose. AMP shall not be liable for direct, indirect,
special, incidental, or consequential damages related to the
use of the information contained herein.
References
1. Grody WW, Thompson BH, Hudgins L: Whole-exome/genome
sequencing and genomics. Pediatrics 2013, 132:S211eS215
2. Kohane IS, Masys DR, Altman RB: The incidentalome: a threat to
genomic medicine. JAMA 2006, 296:212e215
3. Park JY, Fortina P, Kricka LJ: Genomic test validation for incidental
findings. Clin Chem 2014, 60:292e293
4. Schrijver I, Aziz N, Farkas DH, Furtado M, Gonzalez AF, Greiner TC,
Grody WW, Hambuch T, Kalman L, Kant JA, Klein RD, Leonard DG,
Lubin IM, Mao R, Nagan N, Pratt VM, Sobel ME, Voelkerding KV,
Gibson JS: Opportunities and challenges associated with clinical
diagnostic genome sequencing: a report of the Association for Molecular Pathology. J Mol Diagn 2012, 14:525e540
5. Shahmirzadi L, Chao EC, Palmaer E, Parra MC, Tang S,
Gonzalez KD: Patient decisions for disclosure of secondary findings
among the first 200 individuals undergoing clinical diagnostic exome
sequencing. Genet Med 2014, 16:395e399
6. Rehm HL: Disease-targeted sequencing: a cornerstone in the clinic.
Nat Rev Genet 2013, 14:295e300
7. Green RC, Berg JS, Grody WW, Kalia SS, Korf BR, Martin CL,
McGuire AL, Nussbaum RL, O’Daniel JM, Ormond KE, Rehm HL,
Watson MS, Williams MS, Biesecker LG; American College of
Medical Genetics and Genomics: ACMG recommendations for
reporting of incidental findings in clinical exome and genome
sequencing. Genet Med 2013, 15:565e574
8. Burke W, Matheny Antommaria AH, Bennett R, Botkin J,
Clayton EW, Henderson GE, Holm IA, Jarvik GP, Khoury MJ,
Knoppers BM, Press NA, Ross LF, Rothstein MA, Saal H,
Uhlmann WR, Wilfond B, Wolf SM, Zimmern R: Recommendations
for returning genomic incidental findings? We need to talk! Genet Med
2013, 15:854e859
9. McGuire AL, Joffe S, Koenig BA, Biesecker BB, McCullough LB,
Blumenthal-Barby JS, Caulfield T, Terry SF, Green RC: Pointcounterpoint. Ethics and genomic incidental findings. Science 2013,
340:1047e1048
116
10. Wolf SM, Annas GJ, Elias S: Point-counterpoint. Patient autonomy
and incidental findings in clinical genomics. Science 2013, 340:
1049e1050
11. Ross LF, Rothstein MA, Clayton EW: Mandatory extended searches in
all genome sequencing: “incidental findings,” patient autonomy, and
shared decision making. JAMA 2013, 310:367e368
12. Vayena E, Tasioulas J: Genetic incidental findings: autonomy
regained? Genet Med 2013, 15:868e870
13. Green RC, Lupski JR, Biesecker LG: Reporting genomic sequencing
results to ordering clinicians: incidental, but not exceptional. JAMA
2013, 310:365e366
14. Hegde MR: Marching towards personalized genomic medicine. J
Pediatr 2013, 162:10e11
15. Gargis AS, Kalman L, Berry MW, Bick DP, Dimmock DP,
Hambuch T, et al: Assuring the quality of next-generation sequencing
in clinical laboratory practice. Nat Biotechnol 2012, 30:1033e1036
16. Park JY, Kricka LJ, Fortina P: Next-generation sequencing in the
clinic. Nat Biotechnol 2013, 31:990e992
17. Korf BR, Rehm HL: New approaches to molecular diagnosis. JAMA
2013, 309:1511e1521
18. Rehm HL, Bale SJ, Bayrak-Toydemir P, Berg JS, Brown KK,
Deignan JL, Friez MJ, Funke BH, Hegde MR, Lyon E; Working
Group of the American College of Medical Genetics and Genomics
Laboratory Quality Assurance Committee: ACMG clinical laboratory
standards for next-generation sequencing. Genet Med 2013, 15:
733e747
19. Richards CS, Bale S, Bellissimo DB, Das S, Grody WW, Hegde MR,
Lyon E, Ward BE; Molecular Subcommittee of the ACMG Laboratory
Quality Assurance Committee: ACMG recommendations for standards
for interpretation and reporting of sequence variations: revisions 2007.
Genet Med 2008, 10:294e300
20. Bale S, Devisscher M, Van Criekinge W, Rehm HL, Decouttere F,
Nussbaum R, Dunnen JT, Willems P: MutaDATABASE: a centralized and
standardized DNA variation database. Nat Biotechnol 2011, 29:117e118
21. Ankala A, Hegde M: Genomic technologies and the new era of
genomic medicine. J Mol Diagn 2014, 16:7e10
22. Tabor HK, Berkman BE, Hull SC, Bamshad MJ: Genomics really gets
personal: how exome and whole genome sequencing challenge the
ethical framework of human genetics research. Am J Med Genet A
2011, 155A:2916e2924
23. Tabor HK, Stock J, Brazg T, McMillin MJ, Dent KM, Yu JH,
Shendure J, Bamshad MJ: Informed consent for whole genome
sequencing: a qualitative analysis of participant expectations and perceptions of risks, benefits, and harms. Am J Med Genet A 2012, 158A:
1310e1319
24. Clayton EW, Smith M, Fullerton SM, Burke W, McCarty CA,
Koenig BA, McGuire AL, Beskow LM, Dressler L, Lemke AA,
Ramos EM, Rodriguez LL; Consent and Community Consultation
Working Group of the eMERGE Consortium: Confronting real time
ethical, legal, and social issues in the Electronic Medical Records and
Genomics (eMERGE) Consortium. Genet Med 2010, 12:616e620
25. Dressler LG, Smolek S, Ponsaran R, Markey JM, Starks H, Gerson N,
Lewis S, Press N, Juengst E, Wiesner GL; GRRIP Consortium: IRB
perspectives on the return of individual results from genomic research.
Genet Med 2012, 14:215e222
26. National Heart, Lung, and Blood Institute working group, Fabsitz RR,
McGuire A, Sharp RR, Puggal M, Beskow LM, Biesecker LG,
Bookman E, Burke W, Burchard EG, Church G, Clayton EW,
Eckfeldt JH, Fernandez CV, Fisher R, Fullerton SM, Gabriel S,
Gachupin F, James C, Jarvik GP, Kittles R, Leib JR, O’Donnell C,
O’Rourke PP, Rodriguez LL, Schully SD, Shuldiner AR, Sze RK,
Thakuria JV, Wolf SM, Burke GL: Ethical and practical guidelines for
reporting genetic research results to study participants: updated
guidelines from a National Heart, Lung, and Blood Institute working
group. Circ Cardiovasc Genet 2010, 3:574e580
27. Berland LL, Silverman SG, Gore RM, Mayo-Smith WW,
Megibow AJ, Yee J, Brink JA, Baker ME, Federle MP, Foley WD,
jmd.amjpathol.org
-
The Journal of Molecular Diagnostics
Reporting Incidental Findings in Genomics
Francis IR, Herts BR, Israel GM, Krinsky G, Platt JF, Shuman WP,
Taylor AJ: Managing incidental findings on abdominal CT: white
paper of the ACR incidental findings committee. J Am Coll Radiol
2010, 7:754e773
28. Townsend A, Adam S, Birch PH, Friedman JM: Paternalism and the
ACMG recommendations on genomic incidental findings: patients
seen but not heard. Genet Med 2013, 15:751e752
29. Richardson HS, Cho MK: Secondary researchers’ duties to return
incidental findings and individual research results: a partialentrustment account. Genet Med 2012, 14:467e472
30. Ross LF, Saal HM, David KL, Anderson RR; American Academy of
Pediatrics; American College of Medical Genetics and Genomics:
The Journal of Molecular Diagnostics
-
jmd.amjpathol.org
Technical report: ethical and policy issues in genetic testing and
screening of children. Genet Med 2013, 15:234e245
31. Chadwick RB, Meek JE, Prior TW, Peltomaki P, de La Chapelle A:
Polymorphisms in a pseudogene highly homologous to PMS2. Hum
Mutat 2000, 16:530
32. Niessen RC, Kleibeuker JH, Jager PO, Sijmons RH, Hofstra RM:
Getting rid of the PMS2 pseudogenes: mission impossible? Hum Mutat
2007, 28:414. author reply 415
33. Centers for Medicare and Medicaid Services. US Department of Health
and Human Services. Part 493 Laboratory Requirements: Clinical
Laboratory Improvement Amendments of 1988. 42 C.F.R. Section
493: 1443e1495
117