COMMENTARY
Dermatologic applications of direct-to-consumer genomic analysis Alexander L. Fogel, BS, Nason Azizi, MD, Jean Tang, MD, PhD, and Kavita Y. Sarin, MD, PhD Stanford, California Key words: behavior change; genetic testing; genome-wide association studies; genomics; health care utilization; personalized medicine; skin cancer.
enome-wide association studies have discovered thousands of single nucleotide polymorphisms associated with disease. In dermatology, over 200 single nucleotide polymorphisms are associated with important conditions, such as basal cell carcinoma, melanoma, atopic dermatitis, alopecia, and adverse drug reactions (Table I).1 These discoveries are spurring a burgeoning industry of direct-to-consumer (DTC) genomic analysis. At low cost and without a physician’s prescription, patients can order single nucleotide polymorphismebased sequencing of their genome and receive purported risk assessments for disease. Over 300,000 patients have already undergone this genotypic analysis, and more than one third of these patients share their results with physicians.2 Dermatologists should expect that increasing numbers of patients will present with questions on genomic results. In December 2013, the Food and Drug Administration made a pivotal decision to mandate that the largest DTC company, 23andMe, discontinue its health prediction services until it fully achieved medical device approval. Cited concerns included proof of accuracy and unreasonable risk of harm to patients, specifically regarding inclusion of highimpact disease and drug response predictions.3 This warning provides an opportunity for reflection and dialogue within the medical community, as the continuously decreasing cost of genome sequencing, improved understanding of genomic information, and emphasis on preventative medicine are likely to make genomic risk assessment increasingly common.3
G
Effective treatment of patients presenting with genomic information is thus essential, and it is crucial to understand the key components of DTC genomic analysis. Firstly, the diagnostic validity and clinical utility of genomic analysis is still uncertain.3 Not all disease risk associations identified in genome-wide association studies have been validated in clinical studies, risk predictions algorithms vary across DTC companies, and many variants that contribute to disease risk remain undiscovered. In addition, most known variants have small effect on overall risk. One exception is in pharmacogenomics, where a particular HLA allele, in the appropriate context, can significantly alter risk of a severe drug reaction (Table I).4 Pharmacogenetic testing of thiopurine methyltransferase is widely used to assess drug metabolism before starting azathioprine. Despite these issues, genomic analysis may offer some benefits to patient care. Currently, few predictive genetic tests are used in dermatology, although it is plausible that genomic information may stimulate patients to make disease risk-reducing lifestyle modifications. Research has shown that melanomarisk genetic testing can stimulate photoprotection and self-screening behaviors.5 In addition, studies suggest that patients who share their genomic results with physicians are more likely to be self-motivated, health-conscious individuals who are receptive to educational health experiences.6 Furthermore, genomic analysis does not appear to cause psychological distress in patients.2 Dermatologists may thus have a potentially critical role in educating patients who have undergone
From the Department of Dermatology, Stanford University School of Medicine. Funding sources: None. Conflicts of interest: None declared. Reprint requests: Kavita Y. Sarin, MD, PhD, Department of Dermatology, Stanford University School of Medicine, 450
Broadway St, Pavilion C, 2nd Floor, Redwood City, CA 94063. E-mail: [email protected]. J Am Acad Dermatol 2014;71:993-5. 0190-9622/$36.00 Ó 2014 by the American Academy of Dermatology, Inc. http://dx.doi.org/10.1016/j.jaad.2014.04.066
993
Disease
Alopecia areata
1
Keloid1
Male-pattern baldness1
Psoriasis1
Melanoma1
Allopurinol-induced SCAR4y
RAF
P value
HLA-DQA2 ULBP3,ULBP6 CTLA4, ICOS IL2RA DUSP10-QRSL1P2 C3orf72-PRR23A NEDD4 EDA2R PAX1, FOXA2 HDAC4 HDAC9 TARDBP SPPL2C, MAPT-AS1 AUTS2 SETBP1 HLA-C TRAF3IP2 TRAF3IP2 IFIH1 TYK2 IL12B LCE3D IL23A IL23R MC1R TYR CASP8 ATM MX2 CDKN2A, MTAP HLA B*5801 Cw*0302 HLA B*5701
rs9275572-G rs9479482-A rs1024161-A rs3118470-G rs873549-C rs1511412-A rs8032158-C rs2497938-T rs6047844-T rs9287638-A rs2073963-G rs12565727-A rs12373124-T rs6945541-C rs10502861-C rs10484554-A rs240993-A rs458017-G rs17716942-A rs12720356-A rs3213094-A rs4112788-T/C rs2066808-A rs11209026-G rs258322-A rs1393350-A rs13016963-A rs1801516-A rs45430-G rs7023329-A HLA B*5801 Cw*0302 HLA B*5701
0.59 0.57 0.4 0.3 0.28 0.08 0.36 0.85 0.46 0.562 0.53 0.789 0.438 0.539 0.775 NR 0.25 0.06 0.86 0.9 NR NR NR NR 0.11 0.28 0.37 0.87 0.61 0.51 0.02-0.08 NR NR
1 3 10-35 4 3 10-19 4 3 10-13 2 3 10-12 6 3 10-23 2 3 10-13 6 3 10-13 2 3 10-91 2 3 10-39 1 3 10-12 1 3 10-12 9 3 10-11 5 3 10-10 2 3 10-9 3 3 10-9 4 3 10-214 5 3 10-20 2 3 10-16 1 3 10-13 4 3 10-11 5 3 10-11 3 3 10-10 2 3 10-7 7 3 10-7 3 3 10-27 2 3 10-13 9 3 10-10 3 3 10-9 3 3 10-9 7 3 10-9 8.1 3 10-18 2.5 3 10-13 \.0001
HLA B*1502
HLA B*1502
NR
1.4 3 10-21
OR [95% CI]
2.21 1.65 1.44 1.41 1.77 1.87 1.51 2.2 1.6 1.31 1.29 1.33 1.33 1.27 1.28 4.66 1.25 1.37 1.29 1.4 1.39 1.29 1.49 1.49 1.7 1.3 1.14 1.19 1.14 1.2 393.5 62.3 117
[1.98-2.47] [1.48-1.83] [1.30-1.59] [1.27-1.56] [1.58-1.99] [1.58-2.21] [1.35-1.69] [2.04-2.37] [1.49-1.72] [1.21-1.41] [1.20-1.38] [1.22-1.45] [1.21-1.45] [1.18-1.38] [1.18-1.39] [4.23-5.13] [1.16-1.34] [1.22-1.54] [1.17-1.43] [1.23-1.61] [1.26-1.53] [1.19-1.40] [1.28-1.73] [1.27-1.74] [1.54-1.87] [1.21-1.39] [1.09-1.19] [1.12-1.27] [1.09-1.18] [1.14-1.28] [NR] [NR] [29-481]
895 [50-15,869]
Population
European
Japanese
European
European
European
Han Chinese Korean, Thai European, North American, Australian (ethnically mixed) Han Chinese, Southeast Asian
CI, Confidence interval; NR, not reported; OR, odds ratio; RAF, relative allele frequency in controls; SCAR, severe cutaneous adverse reactions; SJS, Stevens-Johnson syndrome; SNP, single nucleotide polymorphism; TEN, toxic epidermal necrolysis. y Pharmacogenomic analysis involves sequenced based typing and is not included with full accuracy in direct-to-consumer genomic analysis reports.
NOVEMBER 2014
Carbamazepine-induced hypersensitivity reaction (including TEN/SJS)4y
SNP
J AM ACAD DERMATOL
Abacavir hypersensitivity syndrome4y
Gene
994 Fogel et al
Table I. Dermatologic diseases and representative associated single nucleotide polymorphisms
J AM ACAD DERMATOL VOLUME 71, NUMBER 5
genomic analysis. Although traditional disease-risk management is still advised, patients may be receptive to education on beneficial lifestyle changes, and should exert additional effort in this capacity. DTC genomic analysis is likely to improve with time, and it may someday become an important component of personalized health care. It is recommended that dermatologists stay abreast of this budding field. REFERENCES 1. Welter D, MacArthur J, Morales J, Burdett T, Hall P, Junkins H, et al. The NHGRI GWAS Catalog, a curated resource of SNP-trait associations. Nucleic Acids Res 2014;42:D1001-6.
Fogel et al 995
2. Bloss CS, Wineinger NE, Darst BF, Schork NJ, Topol EJ. Impact of direct-to-consumer genomic testing at long term follow-up. J Med Genet 2013;50:393-400. 3. Annas GJ, Elias S. 23andMe and the FDA. N Engl J Med 2014; 370:985-8. 4. Kaniwa N, Saito Y. Pharmacogenomics of severe cutaneous adverse reactions and drug-induced liver injury. J Hum Genet 2013;58:317-26. 5. Aspinwall LG, Leaf SL, Kohlmann W, Dola ER, Leachman SA. Patterns of photoprotection following CDKN2A/p16 genetic test reporting and counseling. J Am Acad Dermatol 2009;60: 745-57. 6. Darst BF, Madlensky L, Schork NJ, Topol EJ, Bloss CS. Characteristics of genomic test consumers who spontaneously share results with their health care provider. Health Commun 2014;29:105-8.
Dermatologic applications of direct-to-consumer genomic analysis Alexander L. Fogel, BS, Nason Azizi, MD, Jean Tang, MD, PhD, and Kavita Y. Sarin, MD, PhD Stanford, California Key words: behavior change; genetic testing; genome-wide association studies; genomics; health care utilization; personalized medicine; skin cancer.
enome-wide association studies have discovered thousands of single nucleotide polymorphisms associated with disease. In dermatology, over 200 single nucleotide polymorphisms are associated with important conditions, such as basal cell carcinoma, melanoma, atopic dermatitis, alopecia, and adverse drug reactions (Table I).1 These discoveries are spurring a burgeoning industry of direct-to-consumer (DTC) genomic analysis. At low cost and without a physician’s prescription, patients can order single nucleotide polymorphismebased sequencing of their genome and receive purported risk assessments for disease. Over 300,000 patients have already undergone this genotypic analysis, and more than one third of these patients share their results with physicians.2 Dermatologists should expect that increasing numbers of patients will present with questions on genomic results. In December 2013, the Food and Drug Administration made a pivotal decision to mandate that the largest DTC company, 23andMe, discontinue its health prediction services until it fully achieved medical device approval. Cited concerns included proof of accuracy and unreasonable risk of harm to patients, specifically regarding inclusion of highimpact disease and drug response predictions.3 This warning provides an opportunity for reflection and dialogue within the medical community, as the continuously decreasing cost of genome sequencing, improved understanding of genomic information, and emphasis on preventative medicine are likely to make genomic risk assessment increasingly common.3
G
Effective treatment of patients presenting with genomic information is thus essential, and it is crucial to understand the key components of DTC genomic analysis. Firstly, the diagnostic validity and clinical utility of genomic analysis is still uncertain.3 Not all disease risk associations identified in genome-wide association studies have been validated in clinical studies, risk predictions algorithms vary across DTC companies, and many variants that contribute to disease risk remain undiscovered. In addition, most known variants have small effect on overall risk. One exception is in pharmacogenomics, where a particular HLA allele, in the appropriate context, can significantly alter risk of a severe drug reaction (Table I).4 Pharmacogenetic testing of thiopurine methyltransferase is widely used to assess drug metabolism before starting azathioprine. Despite these issues, genomic analysis may offer some benefits to patient care. Currently, few predictive genetic tests are used in dermatology, although it is plausible that genomic information may stimulate patients to make disease risk-reducing lifestyle modifications. Research has shown that melanomarisk genetic testing can stimulate photoprotection and self-screening behaviors.5 In addition, studies suggest that patients who share their genomic results with physicians are more likely to be self-motivated, health-conscious individuals who are receptive to educational health experiences.6 Furthermore, genomic analysis does not appear to cause psychological distress in patients.2 Dermatologists may thus have a potentially critical role in educating patients who have undergone
From the Department of Dermatology, Stanford University School of Medicine. Funding sources: None. Conflicts of interest: None declared. Reprint requests: Kavita Y. Sarin, MD, PhD, Department of Dermatology, Stanford University School of Medicine, 450
Broadway St, Pavilion C, 2nd Floor, Redwood City, CA 94063. E-mail: [email protected]. J Am Acad Dermatol 2014;71:993-5. 0190-9622/$36.00 Ó 2014 by the American Academy of Dermatology, Inc. http://dx.doi.org/10.1016/j.jaad.2014.04.066
993
Disease
Alopecia areata
1
Keloid1
Male-pattern baldness1
Psoriasis1
Melanoma1
Allopurinol-induced SCAR4y
RAF
P value
HLA-DQA2 ULBP3,ULBP6 CTLA4, ICOS IL2RA DUSP10-QRSL1P2 C3orf72-PRR23A NEDD4 EDA2R PAX1, FOXA2 HDAC4 HDAC9 TARDBP SPPL2C, MAPT-AS1 AUTS2 SETBP1 HLA-C TRAF3IP2 TRAF3IP2 IFIH1 TYK2 IL12B LCE3D IL23A IL23R MC1R TYR CASP8 ATM MX2 CDKN2A, MTAP HLA B*5801 Cw*0302 HLA B*5701
rs9275572-G rs9479482-A rs1024161-A rs3118470-G rs873549-C rs1511412-A rs8032158-C rs2497938-T rs6047844-T rs9287638-A rs2073963-G rs12565727-A rs12373124-T rs6945541-C rs10502861-C rs10484554-A rs240993-A rs458017-G rs17716942-A rs12720356-A rs3213094-A rs4112788-T/C rs2066808-A rs11209026-G rs258322-A rs1393350-A rs13016963-A rs1801516-A rs45430-G rs7023329-A HLA B*5801 Cw*0302 HLA B*5701
0.59 0.57 0.4 0.3 0.28 0.08 0.36 0.85 0.46 0.562 0.53 0.789 0.438 0.539 0.775 NR 0.25 0.06 0.86 0.9 NR NR NR NR 0.11 0.28 0.37 0.87 0.61 0.51 0.02-0.08 NR NR
1 3 10-35 4 3 10-19 4 3 10-13 2 3 10-12 6 3 10-23 2 3 10-13 6 3 10-13 2 3 10-91 2 3 10-39 1 3 10-12 1 3 10-12 9 3 10-11 5 3 10-10 2 3 10-9 3 3 10-9 4 3 10-214 5 3 10-20 2 3 10-16 1 3 10-13 4 3 10-11 5 3 10-11 3 3 10-10 2 3 10-7 7 3 10-7 3 3 10-27 2 3 10-13 9 3 10-10 3 3 10-9 3 3 10-9 7 3 10-9 8.1 3 10-18 2.5 3 10-13 \.0001
HLA B*1502
HLA B*1502
NR
1.4 3 10-21
OR [95% CI]
2.21 1.65 1.44 1.41 1.77 1.87 1.51 2.2 1.6 1.31 1.29 1.33 1.33 1.27 1.28 4.66 1.25 1.37 1.29 1.4 1.39 1.29 1.49 1.49 1.7 1.3 1.14 1.19 1.14 1.2 393.5 62.3 117
[1.98-2.47] [1.48-1.83] [1.30-1.59] [1.27-1.56] [1.58-1.99] [1.58-2.21] [1.35-1.69] [2.04-2.37] [1.49-1.72] [1.21-1.41] [1.20-1.38] [1.22-1.45] [1.21-1.45] [1.18-1.38] [1.18-1.39] [4.23-5.13] [1.16-1.34] [1.22-1.54] [1.17-1.43] [1.23-1.61] [1.26-1.53] [1.19-1.40] [1.28-1.73] [1.27-1.74] [1.54-1.87] [1.21-1.39] [1.09-1.19] [1.12-1.27] [1.09-1.18] [1.14-1.28] [NR] [NR] [29-481]
895 [50-15,869]
Population
European
Japanese
European
European
European
Han Chinese Korean, Thai European, North American, Australian (ethnically mixed) Han Chinese, Southeast Asian
CI, Confidence interval; NR, not reported; OR, odds ratio; RAF, relative allele frequency in controls; SCAR, severe cutaneous adverse reactions; SJS, Stevens-Johnson syndrome; SNP, single nucleotide polymorphism; TEN, toxic epidermal necrolysis. y Pharmacogenomic analysis involves sequenced based typing and is not included with full accuracy in direct-to-consumer genomic analysis reports.
NOVEMBER 2014
Carbamazepine-induced hypersensitivity reaction (including TEN/SJS)4y
SNP
J AM ACAD DERMATOL
Abacavir hypersensitivity syndrome4y
Gene
994 Fogel et al
Table I. Dermatologic diseases and representative associated single nucleotide polymorphisms
J AM ACAD DERMATOL VOLUME 71, NUMBER 5
genomic analysis. Although traditional disease-risk management is still advised, patients may be receptive to education on beneficial lifestyle changes, and should exert additional effort in this capacity. DTC genomic analysis is likely to improve with time, and it may someday become an important component of personalized health care. It is recommended that dermatologists stay abreast of this budding field. REFERENCES 1. Welter D, MacArthur J, Morales J, Burdett T, Hall P, Junkins H, et al. The NHGRI GWAS Catalog, a curated resource of SNP-trait associations. Nucleic Acids Res 2014;42:D1001-6.
Fogel et al 995
2. Bloss CS, Wineinger NE, Darst BF, Schork NJ, Topol EJ. Impact of direct-to-consumer genomic testing at long term follow-up. J Med Genet 2013;50:393-400. 3. Annas GJ, Elias S. 23andMe and the FDA. N Engl J Med 2014; 370:985-8. 4. Kaniwa N, Saito Y. Pharmacogenomics of severe cutaneous adverse reactions and drug-induced liver injury. J Hum Genet 2013;58:317-26. 5. Aspinwall LG, Leaf SL, Kohlmann W, Dola ER, Leachman SA. Patterns of photoprotection following CDKN2A/p16 genetic test reporting and counseling. J Am Acad Dermatol 2009;60: 745-57. 6. Darst BF, Madlensky L, Schork NJ, Topol EJ, Bloss CS. Characteristics of genomic test consumers who spontaneously share results with their health care provider. Health Commun 2014;29:105-8.
